text
stringlengths 87
880k
| pmid
stringlengths 1
8
| accession_id
stringlengths 9
10
| license
stringclasses 2
values | last_updated
stringlengths 19
19
| retracted
stringclasses 2
values | citation
stringlengths 22
94
| decoded_as
stringclasses 2
values | journal
stringlengths 3
48
| year
int32 1.95k
2.02k
| doi
stringlengths 3
61
| oa_subset
stringclasses 1
value |
|---|---|---|---|---|---|---|---|---|---|---|---|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences 10.1289/ehp.7845ehp0113-00136616203248ResearchEnvironmental MedicineChronic Beryllium Disease and Sensitization at a Beryllium Processing Facility Rosenman Kenneth 1Hertzberg Vicki 2Rice Carol 3Reilly Mary Jo 1Aronchick Judith 4Parker John E. 5Regovich Jackie 4Rossman Milton 41 Michigan State University, East Lansing, Michigan, USA2 Emory University, Atlanta, Georgia, USA3 University of Cincinnati, Cincinnati, Ohio, USA4 University of Pennsylvania, Philadelphia, Pennsylvania, USA5 West Virginia University, Morgantown, West Virginia, USAAddress correspondence to K. Rosenman, Michigan State University, 117 West Fee Hall, East Lansing, MI 48824-1316 USA. Telephone: (517) 353-1846. Fax: (517) 432-3606. E-mail:
[email protected] 2005 26 5 2005 113 10 1366 1372 13 12 2004 26 5 2005 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright. We conducted a medical screening for beryllium disease of 577 former workers from a beryllium processing facility. The screening included a medical and work history questionnaire, a chest radiograph, and blood lymphocyte proliferation testing for beryllium. A task exposure and a job exposure matrix were constructed to examine the association between exposure to beryllium and the development of beryllium disease. More than 90% of the cohort completed the questionnaire, and 74% completed the blood and radiograph component of the screening. Forty-four (7.6%) individuals had definite or probable chronic beryllium disease (CBD), and another 40 (7.0%) were sensitized to beryllium. The prevalence of CBD and sensitization in our cohort was greater than the prevalence reported in studies of other beryllium-exposed cohorts. Various exposure measures evaluated included duration; first decade worked; last decade worked; cumulative, mean, and highest job; and highest task exposure to beryllium (to both soluble and nonsoluble forms). Soluble cumulative and mean exposure levels were lower in individuals with CBD. Sensitized individuals had shorter duration of exposure, began work later, last worked longer ago, and had lower cumulative and peak exposures and lower nonsoluble cumulative and mean exposures. A possible explanation for the exposure–response findings of our study may be an interaction between genetic predisposition and a decreased permanence of soluble beryllium in the body. Both CBD and sensitization occurred in former workers whose mean daily working lifetime average exposures were lower than the current allowable Occupational Safety and Health Administration workplace air level of 2 μg/m3 and the Department of Energy guideline of 0.2 μg/m3.
berylliumchronic beryllium diseaseepidemiologyexposure–responselymphocyte proliferation testing
==== Body
Researchers early on recognized that chronic beryllium disease (CBD) occurred after both high and low levels of exposure and hypothesized that the disease was immunologically mediated (Sterner and Eisenbud 1951). Subsequent work has confirmed the importance of cellular immunity to beryllium in the pathogenesis of CBD (Rossman 2001). The factors that determine why some individuals develop cellular immunity to beryllium while others do not still need to be elucidated. Medical screenings of beryllium-exposed workers consistently demonstrate that a larger percentage of individuals will have a positive blood lymphocyte proliferation test to beryllium (become sensitized) than will be diagnosed with CBD (sensitization and granuloma in lung parenchyma) (Henneberger et al. 2001; Kelleher et al. 2001; Kreiss et al. 1993a; Stange et al. 2001). It is not known what proportion of individuals who are sensitized to beryllium will progress to develop CBD. Furthermore, there is a varied clinical presentation of patients with CBD and variability in its progression (Newman et al. 1996; Rossman et al. 1999).
The current occupational air standard for beryllium, first proposed in 1951, was based on the toxicity of other metals such as arsenic, lead, and mercury and modified to reflect beryllium’s lower atomic weight and concern about its greater toxicity (Eisenbud 1982). Epidemiologic health outcome and exposure studies were not used to develop the initial time-weighted average permissible exposure level of 2 μg/m3. Fifty years later, this remains the current air level that Occupational Safety and Health Administration (OSHA) enforces in the workplace.
Recent studies looking at beryllium disease and exposure have either used a surrogate of exposure (i.e., months of exposure, percent exposed to unfired beryllium oxide) or calculated exposure metrics and found increased disease with some parameters of increased exposure (Henneberger et al. 2001; Kelleher et al. 2001; Kreiss et al. 1993b, 1997; Viet et al. 2000). One study found an exposure–response relationship for sensitization with CBD but not for sensitization without CBD (Viet et al. 2000). Other work has addressed the possibility of particle size (McCawley et al. 2001), skin absorption (Tinkel et al. 2003), and/or genetic susceptibility (Saltini et al. 2001) as important factors that confound a straightforward exposure–response relationship.
We investigated possible exposure–response relationships separately for various measures of exposure, including mean, peak, and cumulative metrics and differing chemical and physical forms for the development of beryllium sensitization and for the development of CBD.
We have also assessed whether the current OSHA (2005) and Department of Energy (DOE 1999) permissible levels were protective against the development of CBD and sensitization.
Materials and Methods
The cohort was composed of workers from a beryllium production facility in eastern Pennsylvania, which operated from 1957 to 1978. The names of former workers with at least 2 days of work up to 31 December 1969 who had previously been identified from personnel records and matched with Social Security Administration Form 941 records by the National Institute for Occupational Safety and Health (NIOSH), as part of a seven-company mortality study, were obtained from NIOSH (Ward et al. 1992). The last owner of the facility provided the names of workers, social security numbers, demographic information, and the last known address of all individuals who began work from 1 January 1970 until the plant closed in 1978.
Because this study was a cooperative effort with NIOSH, addresses from the last income tax filing of members of the cohort were obtained by NIOSH from the Internal Revenue Service. NIOSH had previously ascertained the vital status of the cohort as of 31 December 1988 using the Social Security Administration, the Internal Revenue Service, post office cards mailed to the last known address, the Department of Veterans Affairs, the Health Care Finance Administration, and the National Death Index (Ward et al. 1992).
We mailed the initial invitation to participate in the medical screening program to the last known address of all members of the cohort not known to be deceased as of 31 December 1988. The mailing included a cover letter about the study, a fact sheet about beryllium, a one-page two-sided questionnaire, and a postage-paid envelope. The questionnaire requested demographic information and had questions about previous lung disease, smoking history, and work history at the beryllium facility.
We attempted to contact everyone who did not return the questionnaire. This included multiple phone contacts or actual visits to the person’s home if telephone contact was unsuccessful. Internet address searches using search engines such as Yahoo! and Netscape were performed to locate current mailing addresses of individuals with returned mailings. In addition, we used the Social Security Death Index (Ancestry.com 2005) to help determine vital status of individuals. Local staff in the two communities not only made visits to last known addresses but also asked the long-term workers to assist in identifying individuals who could not be located.
All individuals located, whether or not they participated in the medical screening or completed a questionnaire, received a subsequent mailing summarizing the results of the screening and notification of federal legislation passed in the fall of 2000 that provided compensation for workers with CBD and coverage for medical costs for follow-up of workers with beryllium sensitization from this facility.
All individuals located had the opportunity to have a blood lymphocyte proliferation test for beryllium (BeLPT), a posterior–anterior chest radiograph, and simple spirometry. Before the testing, we obtained consent to conduct testing from the individual. In addition, each participant completed a questionnaire on other work exposures that might contribute to respiratory deficiencies. This included other possible sources of beryllium exposure as well as exposure to asbestos, coal, and silica.
Medical testing was performed at two primary sites in the community in eastern Pennsylvania. For individuals who had moved to other parts of the country, medical testing was performed in a location convenient to the individual (i.e., personal physician, local medical centers, etc.).
Blood was collected Monday through Thursday and shipped for next morning delivery. All blood was processed the next day and analyzed. All BeLPT was performed at the University of Pennsylvania. Any individual with a positive BeLPT test was offered a repeat test. If an individual’s results were negative on the repeat test, then the individual was offered the opportunity to repeat the blood test 1 year later.
A panel of three “B” readers interpreted all chest radiographs. One B reader was a radiologist (J.A.), one a pulmonologist (J.E.P.), and one an internist and occupational medicine physician (K.R.). At least two B readers had to classify a radiograph with ≥1/0 profusion in order for a radiograph to be classified as positive for parenchymal disease.
Any individual who had two positive BeLPTs and/or a consensus chest radiograph reading of ≥1/0 for profusion was referred to the University of Pennsylvania for follow-up testing, which consisted of a posterior–anterior chest radiograph, a BeLPT, an electrocardiogram, a complete medical history including respiratory symptoms using a standardized collection instrument, and bronchoscopy with both bronchial biopsy and lymphocyte testing of lavage fluid for beryllium. All bronchoscopies were performed by a single pulmonologist (M.R.).
Whether or not an individual had CBD or beryllium sensitization was decided by consensus by the internist/occupational physician (K.R.) and pulmonologist (M.R.). Table 1 outlines the criteria used to categorize the medical testing results. However, only individuals who had bronchoscopy were used in the analysis describing the predictive power of radiographs or BeLPT.
All individuals received a letter with the results of their initial screening and, where applicable, a letter with the results of the follow-up testing. The Human Subject Review Boards of Emory University, Michigan State University, the University of Cincinnati, and the University of Pennsylvania approved this study.
Through discussions with long-term production and management employees, we identified major changes in the process and engineering/work practice controls. Trends in the exposures over time were evaluated in relation to dates of process changes and visually from plots of the data to identify other time points at which exposure measurements indicated a change in conditions.
Exposure had been monitored at the facility using a method that combined the concentration at each task performed by a worker, weighted by the duration in the shift of that task; the products of concentration and duration at all tasks performed as part of a job were summed and divided by the duration to the shift. This final value was called the daily weighted average (DWA) exposure. Data accumulated over the operating history of the plant were identified and computerized. Using this information, a task exposure matrix (TEM) and a job exposure matrix (JEM) were constructed (Chen 2001). Task-related exposure measurements were available for two time periods, 1957–1962 and 1971–1976. Because the data most closely followed a log-normal distribution, the geometric mean was calculated for each task-year combination. For years with no measurements, we estimated exposures by interpolating between the previous and subsequent values. For example, if measurements were available for 1957, 1958, and 1959 but not for 1960, the 1959 value was entered into the TEM. The plant history was used to develop a strategy for imputing values from 1963 to 1971. We used the mean of task estimates for 1962 and 1971 for the period 1963–1969; because of the engineering changes in 1970, the 1971 values were used for 1970. Estimates for 1976 were used for the remaining years of plant operation, based on employee interviews. For tasks never measured, the task in the same work area most similar to the unmeasured task was identified with the assistance of long-term employees; the exposure value for the measured task was entered into the TEM for the unmeasured task.
We completed the JEM by first calculating the geometric mean exposure for each year in which at least one DWA measurement was available. Exposure estimates for job–year combinations without measurements were estimated based on the plant history of engineering changes. In the absence of information showing production or control technology changes in years before or after measurement data, the measurements were assumed valid and extended to the empty cells in the JEM. Where increases or decreases in exposure were justified from the plant history, we used analysis of variance (ANOVA) to evaluate the significance of the change in exposure. Where statistically significant changes were identified, the new value was entered into the cell of the JEM.
For 39 of the 130 job titles, no measurements were available for the job in any year. For each of these jobs, we used information from the long-term workers to identify the job with tasks most similar to it with measurements. The time–activity pattern needed for the evaluation of exposure was developed and used to calculate a DWA estimate of exposure using data in the TEM.
The values were reviewed by a group of long-term employees who represented experience in all production areas of the facility, maintenance, and management. They were specifically asked to review the relative exposure values for production areas. For example, the exposure estimated for the fluoride furnace operator is slightly higher than the helper; this was confirmed to be correct because the helper stood away from the furnace and supplied materials to the perimeter only. The involvement of the group of long-term employees provided added confidence in our derived estimates.
The JEM and TEM were linked through the listing of the tasks in each job taken from the DWA calculation sheets. For jobs never sampled, the association was through the time–activity information developed with the help of long-term employees and, finally, put into DWA format.
For every job title in the JEM, the chemical and physical form of the exposure was listed. Chemical forms included beryl ore, beryllium metal, beryllium fluoride, beryllium hydroxide, and beryllium oxide; physical forms included dust, fume, or mixed (dust and fume). Individuals from the facility were assigned, based on jobs worked, the number of months exposed to three different chemical forms: nonsoluble beryllium compounds (beryllium metal and oxide), soluble beryllium compounds (beryllium fluoride and hydroxide), and mixed chemical forms. Individuals were similarly assigned to the number of months exposed to the three physical forms: dust (beryllium metal, hydroxide, or oxide), fume (beryllium fluoride), and mixed (mixed dust and fume). This allowed us to evaluate any differences in response due to very small particle size (fume) or larger particle size (dust or mixed).
We used chi-square tests to compare the groups (definite or probable disease vs. sensitized vs. no disease) with respect to discrete outcomes. ANOVA was used to compare the groups with respect to continuous outcomes (age, cumulative, mean, and peak exposure levels). For the three disease outcome group comparisons, a screening p-value was set at 0.25, below which the pairwise comparisons between groups (definite or probable disease vs. no disease, definite or probable disease vs. sensitized, sensitized vs. no disease) were further investigated. For the discrete outcomes, further chi-square tests were performed on the resulting 2 × k tables. For the continuous outcomes, the linear contrasts for these pairwise comparisons were examined in order to control for multiple comparisons. For ease of presentation, we also used two-sample t-tests to examine pairwise comparisons of the groups. These parametric tests were followed by the Wilcoxon rank-sum test, a nonparametric test used to ameliorate the effects of violations of the assumptions for the parametric tests (e.g., normal distribution).
We further explored exposure–response relationships with logistic regression analysis after adjustment for potential confounders (smoking, age, other beryllium exposure). In addition to an analysis where only cases with complete information were included, an analysis was carried out after multiple imputations (Rubin 1987) of cumulative and mean exposure values (missing on 65 of 574 individuals).
p-Values are presented as calculated. All analyses were performed using SAS statistical software (version 9.1; SAS Institute, Cary, NC). The results of the spirometry testing are not reported in this article.
Results
A total of 1,351 individuals were identified to have worked at this facility. A summary of the participation rate for this facility is shown in Table 2. Approximately one-fourth (24.4%) of the cohort died before the medical screening began, and another 10.8% could not be located. Among the 875 individuals located, 160 (11.8%) indicated either that they had worked for the company that owned the facility but at a different location, or that they had completed a job application and underwent a pre-employment physical for work at the facility but had either not been hired or had decided not to accept a job at that plant.
Of the remaining 715 former employees, the participation rate was 63.9% (457 of 715) for completion of all components of the medical screening and 91.3% (653 of 715) for completion of the questionnaire only. Five hundred twenty-eight individuals (73.8%) completed at least the blood and chest radiograph component. Reasons members of the cohort gave for not participating included that the individual a) had only worked for a short time; b) felt he or she was too old and that testing would not matter; c) did not have any health problems; d) did not want to jeopardize his or her current health insurance, especially with no compensation available (at the time the individual was contacted); and e) felt there was no effective treatment for beryllium disease.
Table 3 compares the demographics of medical screening participants with nonparticipants. Medical screening took place from the fall of 1996 through the summer of 2001. Participants were on average the same age as the nonparticipants, the same sex and race, last worked in a more recent year, and worked on the average 3.3 years longer. Participants were mainly male (91%), and almost all white. Seventy percent had ever smoked cigarettes.
Among the 577 individuals that were tested for beryllium, 110 were referred for follow-up testing at the Hospital of the University of Pennsylvania (Table 4). In addition to the 110 referred, 9 individuals from the facility had previously been diagnosed at the University of Pennsylvania with CBD. All 577 individuals, including the 9 previously diagnosed with CBD, were categorized per the criteria in Table 1. The results of this classification are shown in Table 5. Of the cohort, 7.6% (44) had probable or definite CBD, 2.1% (12) had possible CBD, 6.9% (40) were sensitized to beryllium, and 4.0% (23) were possibly sensitized.
Table 6 shows the predictive power of having unrecognized CBD documented by bronchoscopy based on the results of the screening tests performed. Having two positive BeLPTs and scarring on the chest radiograph, involving either all zones or the lower zones only, had the highest predictive value for the development of CBD (100%). In descending order for the other combination of tests, the predictive values for CBD were scarring on the radiograph in all zones with negative BeLPT (75%), positive BeLPT (48.3%), scarring in the upper zones with negative BeLPT (40%), and scarring on the chest radiograph just in the lower zones (7.7%).
There were 33 cases of definite/probable CBD among production workers, 5 among clerical/office workers, 3 in engineers, 1 in a supervisor/inspector, 1 in a laboratory worker, and 1 in an industrial hygiene technician. There were 27 cases of sensitization among production workers, 10 among clerical/office workers, 2 among engineers, and 1 in a nurse.
Table 7 shows the occurrence of definite and probable CBD and sensitization by first decade worked, last decade worked, and duration of years worked. The mean year of first exposure for definite/probable CBD was 1963, for sensitized cases it was 1965, and for the normal group it was 1964. Further, the mean year last exposed for definite/probable, sensitized, and normal individuals was 1973, 1968, and 1971, respectively. The mean duration of exposure for definite/probable, sensitized, and normal individuals was 9.4 years, 2.7 years, and 8.7 years, respectively.
Tables 8–12 show the occurrence of definite and probable CBD and sensitization by the peak, average, and cumulative exposure metric, by chemical and physical form of beryllium and the OSHA (2005) standard of 2 μg/m3 and the DOE (1999) standard of 0.2 μg/m3. Individuals who were sensitized had a lower total cumulative and peak exposure (Table 8), lower nonsoluble cumulative and average exposure (Table 11), and lower dust and mixed exposure (Table 10). Individuals with CBD had a lower soluble (Table 9) and fume exposure (Table 10). The mean beryllium exposure levels for the DWA categories in Table 11 were 0, 1.23, and 8.95 μg/m3, respectively and in Table 12 were 0.14, 1.19, and 4.76 μg/m3, respectively.
Discussion
The prevalence of CBD and sensitization to beryllium in former workers at this beryllium production facility in eastern Pennsylvania was high: 7.6% with CBD, 6.9% with sensitization, 2.1% with possible CBD, and 4.0% with possible sensitization. This facility operated from 1957 to 1978. Representative exposure estimates for tasks ranged from 0.9 to 84.0 μg/m3 in the 1960s, although most time-weighted averages were below the OSHA (2005) standard of 2 μg/m3, ranging from 1.1 to 2.5 μg/m3. Exposure estimates in the 1970s were lower, with representative tasks ranging from 0.5 to 16.7 μg/m3 and time-weighted averages ranging from 0.7 to 3.5 μg/m3.
The 14.5% prevalence of CBD and sensitization in the cohort we studied contrasts with overall prevalence reports of 3.3% among nuclear workers from Rocky Flats (Stange et al. 2001), 1.8–5.9% from beryllium ceramics manufacturing (Kreiss et al. 1993b, 1996), and 4.6% from a beryllium production facility (Kreiss et al. 1997). Our overall prevalence is similar to the prevalence reports for more highly exposed subgroups from these studies, such as machinists (Kreiss et al. 1996). Our higher overall prevalence rate reflects both the level and the widespread exposure to beryllium in the facility we studied, where 11 definite/ probable cases occurred among nonproduction workers such as clerical, supervisory, and engineering staff and 13 sensitization cases occurred in clerical/office personnel. Our mean and range of cumulative exposure, which was 199.25 μg-year/m3 (0.0–3970.61 μg-year/m3, are appreciably higher than estimates reported in other studies: 6.09 μg-year/m3 (0.15–10.64 μg-year/m3) (Kelleher et al. 2001), 1.35 μg-year/m3 (estimated range, 0–6.41 μg-year/m3) (Viet et al. 2000), and no mean provided (estimated range, 0.9–41.2 μg-year/m3) (Henneberger et al. 2001). An additional factor that probably contributes to the higher prevalence of CBD in our cohort is the long latency since last exposure, which would have allowed a higher proportion of individuals who were sensitized to progress on to CBD than in other cohorts that have been studied (Newman et al. 2005). Most previous prevalence studies of beryllium-exposed workers have been of current employees (Henneberger et al. 2001; Kelleher et al. 2001; Kreiss et al. 1996), or they have included former workers (Stange et al. 2001) but have not presented the results separately for current and former workers. One study similar to ours only had formerly exposed individuals (Kreiss et al. 1993b). This latter study, unlike ours, found no individuals with sensitizations alone without CBD. This would suggest that the higher prevalence of CBD in our study population was not solely related to the long latency since last exposure because we would have expected a lower rate of sensitization alone without CBD if increased prevalence of CBD was solely caused by the long latency.
Despite the fact that there is an overall increase of beryllium disease in working populations with higher exposure to beryllium, investigators have been unable to show a clear-cut exposure response between air concentrations of beryllium and CBD or sensitization (Henneberger et al. 2001; Kelleher et al. 2001; Viet et al. 2000). This has led researchers to examine the possible role of particulate size (Kelleher et al. 2001; McCawley et al. 2001) and skin exposure (Tinkle et al. 2003).
We found no difference in duration of exposure for individuals with CBD versus those who had no evidence of beryllium disease, but we did find that those who were sensitized had begun work later, last worked longer ago, and had a shorter duration of exposure than did those with CBD or those who tested normal (Table 7). This difference for individuals with sensitization was also true for cumulative and peak exposure (Table 8), cumulative mixed and cumulative and mean nonsoluble exposure (Table 9), cumulative and mean dust, and cumulative mixed exposure (Table 10). On the other hand, cumulative and mean soluble and cumulative and mean soluble fume exposures were lower for CBD (Table 10).
In sum, we either found no exposure response or the significant exposure responses we did find were in the opposite direction than expected, with individuals with CBD or sensitization having less estimated exposure than those with no beryllium disease. The risk of CBD compared with sensitization if a person’s mean exposure was below the current DOE (1999) permissible level of 0.2 μg/m3 was less than if their mean level was > 0.2 μg/m3 but below the current OSHA (2005) permissible exposure level of 2 μg/m3 (Table 11). However, only being exposed to beryllium less than either the DOE or the OSHA time-weighted average did not protect a worker from the development of CBD or sensitization. There were only two people in the cohort whose highest level of exposure was never above the 0.2 μg/m3 DOE standard. CBD and sensitization occurred even if the highest level of exposure was never greater than the 2 μg/m3 OSHA standard, and our data would suggest that peak exposure levels > 0.2 μg/m3 were as harmful as even higher peak exposure levels > 2 μg/m3 (Table 12).
A possible explanation for the failure to find an association between increased beryllium disease and sensitization and increased levels of exposure is that this analysis did not consider the role of genetic predisposition to both sensitization and disease. Because the genetic marker glu69 on HLA-DPB1 has been associated with 80–90% of cases of both CBD and sensitization, a better control group for this analysis would be HLA-DPB1 glu69–positive individuals who did not have CBD or sensitization. We have recently been funded to test our population for this marker and thus will eventually be able to determine the interaction of exposure and genetic predisposition.
The finding of higher working lifetime beryllium exposures in those with CBD compared with those who are just sensitized suggests that the body burden of beryllium might relate to the severity of disease in those with a genetic predisposition. Our finding that sensitized individuals compared with individuals with CBD had a higher exposure to beryllium in a soluble form and to fumes of beryllium supports this hypothesis (in this facility soluble beryllium and fume is practically equivalent, r = 0.94). Presumably the soluble forms of beryllium would be more likely to be mobilized and eliminated and result in a lower body burden of beryllium compared with a similar exposure to insoluble beryllium.
Individuals who have recently converted their PPD (purified protein derivative) skin test for tuberculosis to positive may, after treatment, revert to a negative PPD (Tager et al. 1985). Thus, with decreasing or elimination of the antigen, the cellular immune response (i.e., PPD reaction) may fade or be eliminated. Because the PPD reaction is similar to BeLPT, this suggests that a decreased immune response to beryllium may occur in individuals with a lower body burden of beryllium (i.e., antigen). Thus, a reduced immune response to beryllium may account for the association of beryllium sensitization with a lower body burden of insoluble beryllium or predominantly soluble beryllium exposure compared with individuals with CBD. An alternative explanation that less soluble beryllium exposure is confounded by elevated levels of other forms of beryllium is not supported by analyzing potential correlations between levels of exposures to the different forms of beryllium.
Other researchers have suggested the importance of skin exposures to the development of beryllium disease. We have no data to directly address whether skin exposure is of importance in the development of beryllium disease in this cohort. However, others have hypothesized that small particle size increases the likelihood of both inhalation and skin absorption and exposure (McCawley et al. 2001). Our data showed the opposite results with reduced levels of exposure to fume, which would be the smallest particle size form of exposure that occurred in this facility, and CBD (Table 10).
A limitation of our study is the uncertainty in the exposure estimates. The exposure metrics developed for study participants were based on relatively sparse data, with interpolation from measurement data for years when no data were available. Major gaps in the data were associated with the mid-1960s and from 1977 to 1981. Exposure estimates for the earlier time period were based on measurements in preceding or succeeding years; for the later time period, estimates for the mid-1970s were extended into the later years. These decisions were based on plant history and conversations with long-term workers. All interpolation was accomplished using preestablished rules and was independent of any knowledge of disease status. The use of professional judgment like this is often required in retrospective exposure assessment studies. Because the exposure estimates were created for jobs and tasks, without knowledge of a work history or disease status, it is likely that this mis-classification would be nondifferential, attenuating any ability to detect exposure–response relations (Checkoway et al. 1991; Copeland et al. 1977).
A further limitation relates to the effect of nonparticipants on study results. The overall participation rate was high and nonparticipants were generally similar to participants except their duration of exposure was less. However, it is possible that the 11% of the total cohort that did not participate had a lower rate of CBD because asymptomatic individuals might be less motivated to participate. On the other hand, the 24% of the cohort who were deceased at the initiation of our study and the 11% we could not locate might be expected to have a higher prevalence of disease.
A third limitation of our study is that we used a single laboratory for the blood lymphocyte testing for beryllium in a one-time screening. It has recently been reported that the use of a single laboratory results in false negative results of 20–30% (Stange et al. 2004). Because radiographs were part of our screening, we would expect the false negative rate for CBD to be lower than the potential false negative rate for sensitization. Because our cohort was no longer exposed to beryllium, it is less likely that repeat screening will identify additional cases of CBD or sensitization, as has been shown in currently exposed cohorts (Newman et al. 2001).
The participation level of individuals who warranted more extensive testing after the initial screening is another limitation of this study. Only 56 of the 110 (51%) individuals who screened positive by radiograph or BeLPT elected to have a bronchoscopy. Participation in more extensive testing was similar in those with positive radiographs (47%) and those with abnormal positive BeLPT only (57%). The lack of a biopsy and broncholavage in half of the individuals who were positive on the initial medical screening means we may have misclassified individuals into the definite/ probable CBD and sensitization groups. This misclassification would decrease the likelihood of finding an exposure–response or other relationship with CBD or sensitization. To minimize misclassification errors, we excluded cases classified as possible CBD or possible sensitization from both the disease and normal groups during analysis. However, we included four individuals classified as CBD because of a diagnosis at the University of Pennsylvania before our study, although these individuals never had evidence of sensitization in their bronchial lavage fluid or blood. We are not aware of any reason how misclassification could cause the inverse relationship between exposure and disease that we found.
A final limitation is that multiple comparisons were made in Tables 9 and 10. Adjustments for these multiple comparisons can be made by tripling the p-value reported, using the properties of the Bonferroni inequality. If this were done, a number of the associations would no longer be statistically significant in Tables 9 and 10. Given the consistent direction of the findings, our conclusions concerning soluble and nonsoluble forms of beryllium remain unchanged even if this adjustment were made.
In conclusion, this cohort is a high-risk group for the development of CBD and sensitization. The development of beryllium disease has continued to occur years after exposure has ceased. Former beryllium workers and their health care providers need to be aware of this ongoing risk. A combination of two positive BeLPTs and an abnormal chest radiograph on the initial medical screening was the best predictor of the presence of CBD. However, there were individuals who had CBD with an abnormal chest radiograph, involving all or just the upper lobes, and negative BeLPT (Table 6).
We were unable to show an exposure–response relationship. The inclusion of genetic data combined with exposure data may better define which individuals in this cohort are at a particularly high risk of development of CBD and/or sensitization and may account for the absence of the typical exposure–response seen with other environmental or occupational toxins. We are currently performing molecular typing of DRB1 and DPB1 alleles on individuals with CBD and sensitization and a sample of those who tested normal to investigate for a possible gene–exposure relationship.
This cohort is a high-risk group for the development of CBD and sensitization. The development of beryllium disease has continued to occur years after exposure has ceased. Former beryllium workers and their health care providers must be kept aware of this ongoing risk.
The results of this study show that current occupational health standards for beryllium do not provide adequate protection against the development of CBD or sensitization. Twenty-four percent of the workforce that was exposed to beryllium below the current OSHA (2005) allowable threshold limit value developed CBD or sensitization. Similar levels of adverse outcomes (21%) were seen in those exposed to beryllium below the time-weighted average DOE (1999) guideline of 0.2 μg/m3. Even the more protective time-weighted average of 0.02 μg/m3 proposed by the American Conference of Governmental and Industrial Hygienists (2005) did not eliminate adverse outcomes. The identification of cases of CBD and sensitization in this population at levels of cumulative exposure lower than current standards or guidelines underscores the need to more fully understand the determinants of exposure (e.g., peak, physical/chemical form) that may contribute to disease risk, so that these may be included in standard setting.
We acknowledge the dedicated, extensive work of T. Carey, A. Krizek, C. Vsetula, and C. Zamba in the identification, contacting, scheduling, and tracking of the medical screening participants; we thank R. Swank and B. Toth for their expert phlebotomy assistance in this project; and, most important, we thank the screening participants for their invaluable support.
This study was funded by grants from the National Institute for Occupational Safety and Health (U60/CCU512218) and the Department of Energy (DE-FG03-98EH98027) and was supported in part by U.S. Public Health Service research grant M01RR00040 from the National Institutes of Health.
M.R. reported competing financial interests: He has provided expert testimony for law firms representing companies and workers; he performs clinical evaluation of patients with chronic beryllium disease; and he is the director of the beryllium lymphocyte proliferation testing laboratory. The other authors declare they have no competing financial interests.
Table 1 Criteria for beryllium disease categories.
Disease category Bronchial lavage Biopsy granuloma Chest radiograph Blood LPT Spirometry
CBD + BAL LPT Positive
Not done Positive Two + LPTs
Not done Positive Upper lobe fibrosis
Probable CBD + BAL LPT Upper lobe fibrosis
Not done Upper lobe fibrosis Two + LPTs
Possible CBD Not done Upper lobe fibrosis – LPT
Not done Upper lobe fibrosis Single + LPT and no retest
Sensitization – BAL LPT Negative Normal Two + LPTs Normal
+ BAL LPT Negative Normal Normal
Not done Lower or midlobe fibrosis or normal Two + LPTs Normal
Possible sensitization Single + LPT and no retest or – LPT retests
Abbreviations: –, negative; +, positive; BAL, bronchoalveolar lavage; LPT, lymphocyte proliferation test.
Table 2 Summary of cohort participation.
No. (%)
Cohort 1,351
Deceased 330 (24.4)
Unable to locate 146 (10.8)
Contacted 875
Denied ever working at facility 160 (11.8)
Potential participants 715
Medical testing 653 (91.3)
Questionnaire only 79
Questionnaire and blood 22
Questionnaire and chest radiograph 6
Questionnaire, chest radiograph, and blood 71
Questionnaire, chest radiograph, and PFTs 12
Questionnaire, PFTs, and blood 6
All components (questionnaire, chest radiograph, blood, and PFTs) 457
Refusals 62 (8.7)
PFT, pulmonary function test.
Table 3 Demographics of medical screening participants versus those who completed questionnaire only.a
Characteristic Completed questionnaire onlyb Medical screening participantsc
Birth year (mean ± SE) 1936 ± 1.18 1935 ± 0.44
Male sex (%) 91.8 90.8
White race (%) 100.0 99.8
Duration (no.) of years worked (mean ± SE) 5.2 ± 0.85 8.5 ± 0.40*
Last year worked (mean ± SE) 1969 ± 0.89 1971 ± 0.32*
a Sixty-two individuals who refused to complete questionnaire and medical screening are not included.
b Includes 79 individuals who completed only the questionnaire.
c Includes 574 individuals who completed questionnaires and some part of medical screening.
*p < 0.05.
Table 4 Reason for referral for follow-up testing.
No. (%)
Blood
Two positive BeLPTs 53 (9.2)
Chest radiograph
Profusion > 1/0 per at least two B readers 50 (8.7)
Blood and chest radiographa 7 (1.2)
Total referred 110b (19.1)
a Met criteria for both blood and chest radiograph referral.
b Nine (1.6%) additional individuals in the cohort met the study’s criteria for CBD based on testing performed before the study’s medical screening.
Table 5 Disease categorization of medical test results (n = 577).
Disease category No. (%)
Definite CBD 32 (5.5)
Probable CBD 12 (2.1)
Possible CBD 12 (2.1)
Sensitized 40 (6.9)
Possibly sensitized 23 (4.7)
No CBD and/or sensitization 458 (79.4)
Table 6 Predictive power of having unrecognized CBD documented by bronchoscopy based on results of BeLPT and chest radiograph.
Total no. having bronchoscopy Confirmed CBD cases [No. (%)]
Blood
Two positive BeLPTs 29 14 (48.3)
Chest radiograph 22 6 (27.3)
Profusion > 1/0 at least two B readers
All zones 4 3 (75.0)
Upper zones only 5 2 (40.0)
Lower zones only 13 1 (7.7)
Blood and chest radiographa 5 5 (100.0)
All zones 3 3 (100.0)
Upper zones only 0 NA
Lower zones only 2 2 (100.0)
NA, not applicable.
a Met criteria for blood and chest radiograph referral.
Table 7 Development of definite/probable CBD and sensitization by decade of first and last exposure and duration of exposure.
Decade of first exposure [No. (%)]
Decade of last exposure [No. (%)]
Duration of exposure (years) [No. (%)]
Disease outcome 1950s 1960s 1970s 1950s 1960s 1970s 1980s < 1 1 to < 5 5 to < 15 ≥15
Definite/probable CBD 14 (34) 17 (41) 10 (24) 3 (8) 8 (21) 22 (56) 6 (15) 10 (24) 8 (20) 11 (27) 12 (29)
Sensitization 7 (18) 22 (56) 10 (25) 1 (3) 20 (57) 14 (40) 0 (—) 9 (23) 19 (49) 9 (23) 2 (5)
Normal 112 (27) 222 (54) 75 (18) 26 (7) 126 (33) 177 (46) 58 (15) 70 (17) 136 (33) 107 (26) 99 (24)
For the decade of first exposure, p = 0.03 for sensitization vs. normal. For the decade of last exposure, p = 0.03 for sensitization vs. definite/probable and p = 0.008 for sensitization vs. normal. For the duration of exposure, p = 0.008 for definite/probable vs. sensitization and p = 0.03 for sensitization vs. normal.
Table 8 Development of definite/probable CBD and sensitization by average cumulative, average mean, and peak exposure (± SE).
Disease outcome No. of individuals Mean cumulative exposure (μg-year/m3) Mean exposure (days) Mean average exposure (μg/m3) Mean peak exposure (μg/m3)
Definite/probable CBD 40 181 ± 29 3,483 ± 550 8.7 ± 0.8 81 ± 14
Sensitization 37 100a ± 23 1,934b ± 55 7.1 ± 0.9 53c ± 14
Normal 377 209 ± 16 3,359 ± 176 8.3 ± 0.3 87 ± 13
a p = 0. 03 for sensitization vs. definite/probable, and p = 0. 0003 for sensitization vs. normal.
b p = 0.047 for sensitization vs. definite/probable, and p = 0.02 for sensitization vs. normal.
c p = 0. 01 for sensitization vs. normal.
Table 9 Development of definite/probable CBD and sensitization by chemical form of beryllium, mixed, nonsoluble, and soluble: mean cumulative, mean average, and mean peak exposure levels.
Mixed
Nonsoluble
Soluble
Disease outcome No. Cumulative (μg-year/m3) Mean (μg/m3) Peak (μg/m3) Cumulative (μg-year/m3) Mean (μg/m3) Peak (μg/m3) Cumulative (μg-year/m3) Mean (μg/m3) Peak (μg/m3)
Definite/probable CBD 40 50 3.7 2.1 126 7.6 4.6 5.8a 0.8b 2.1
Sensitization 37 20c 2.3 4.4 61d 5.4e 2.8 19 2.3 4.4
Normal 377 49 3.4 3.5 128 7.4 4.5 26 1.6 3.6
a p < 0.0001 for definite/probable vs. normal.
b p = 0.02 for definite/probable vs. normal.
c p = 0.0005 for sensitization vs. normal.
d p = 0.04 for sensitization vs. definite/probable, and p = 0.003 for sensitization vs. normal.
e p = 0.02 for sensitization vs. normal.
Table 10 Development of definite/probable CBD and sensitization by physical form of beryllium, dust, fume, and mixed: mean cumulative, mean average, and mean peak exposure levels.
Dust
Fume
Mixed
Disease outcome No. Cumulative (μg-year/m3) Mean (μg/m3) Peak (μg/m3) Cumulative (μg-year/m3) Mean (μg/m3) Peak (μg/m3) Cumulative (μg-year/m3) Mean (μg/m3) Peak (μg/m3)
Definite/probable CBD 40 128 7.4 4.6 4a 0.7b 0.3 49 3.6 2.1
Sensitization 37 66c 5.1d 3.5 17 2.3 3.1 17e 2.4 4.4
Normal 377 138 7.1 5.4 20 1.4 1.3 46 3.3 3.5
a p = 0.0002 for definite/probable vs. normal.
b p = 0.03 for definite/probable vs. normal.
c p = 0.0021 for sensitization vs. normal.
d p = 0.009 for sensitization vs. normal.
e p = 0.0004 for sensitization vs. normal.
Table 11 Development of definite/probable CBD and sensitization by the American Conference of Governmental and Industrial Hygienists notice of intended change, current OSHA, and DOE DWA threshold levels.
Mean DWA exposure (μg/m3) [n (%)]
Disease outcome 0 to < 0.02 0.02 to < 0.2 0.2 to < 2 ≥2
Definite/probable CBD 1 (7) 0 (0) 4 (17) 35 (8)
Sensitization 2 (14) 0 (0) 2 (8) 33 (8)
Normal 11 (79) 0 (0) 18 (75) 348 (84)
Table 12 Development of definite/probable CBD and sensitization by highest exposure.
Highest exposure level (μg/m3) [n (%)]
Disease outcome 0 to < 0.2 0.2 to < 2 ≥2
Definite/probable CBD 0 (—) 18 (9.3) 22 (8.5)
Sensitization 0 (—) 19 (9.8) 18 (7.0)
Normal 2 (100) 157 (80.9) 218 (84.5)
==== Refs
References
American Conference of Governmental and Industrial Hygienists 2005. Annual Reports of the Committees on TLVs and BEIs for Year 2004. Cincinnati, OH:American Conference of Governmental and Industrial Hygienists.
Ancestry.com 2005. Social Security Death Index. Available: http://www.ancestry.com/search/db.aspx?dbid=3693 [accessed 24 August 2005].
Checkoway H Savitz D Heyer N 1991 Assessing the effects of nondifferential misclassification of exposures in occupational studies Appl Occup Environ Hyg 6 528 533
Chen MJ 2001. Development of Beryllium Exposure Metrics for Workers in a Former Beryllium Manufacturing Plant [PhD Thesis]. Cincinnati, OH:University of Cincinnati.
Copeland K Checkoway H McMichael A 1977 Bias due to mis-classification in estimation of relative risk Am J Epidemiol 105 488 495 871121
DOE (Department of Energy) 1999 Chronic beryllium disease prevention program: final rule Fed Reg 64 68853 68914
Eisenbud M 1982 Origins of the standard for control of beryllium (1947–1949) Environ Res 27 79 88 7040071
Henneberger PK Cumro D Deubner DD Kent MS McCawley M Kreiss K 2001 Beryllium sensitization and disease among long-term and short-term workers in a beryllium ceramics plant Int Arch Occup Environ Health 74 167 176 11355290
Kelleher PC Martyny JW Mroz MM Maier LA Ruttenber AJ Young DA 2001 Beryllium particulate exposure and disease relations in a beryllium machining plant J Occup Environ Med 43 238 249 11285872
Kreiss K Mroz MM Newman LS Martyny J Zhen B 1996 Machining risk of beryllium disease and sensitization with median exposures below 2 micrograms/m3 Am J Ind Med 30 16 25 8837677
Kreiss K Mroz MM Zhen B Martyny J Newman LS 1993a Epidemiology of beryllium sensitization and disease in nuclear workers Am Rev Respir Dis 148 985 991 8214955
Kreiss K Mroz MM Zhen B Wiedemann H Barna B 1997 Risks of beryllium disease related to work processes at a metal, alloy, and oxide production plant Occup Environ Med 54 605 612 9326165
Kreiss K Wasserman S Mroz MM Newman LS 1993b Beryllium disease screening in the ceramics industry J Occup Med 35 267 274 8455096
McCawley MA Kent MS Berakis MT 2001 Ultrafine number concentration as a possible metric for chronic beryllium disease risk Appl Occup Environ Hyg 16 631 638 11370941
Newman LS Lloyd J Daniloff 1996 The national history of beryllium sensitization and chronic beryllium disease Environ Health Perspect 104 937 943 8933038
Newman LS Mroz MM Balkissoon Mairer LA 2005 Beryllium sensitization progesses to chronic beryllium disease. A longitudinal study of disease risk Am J Respir Crit Care Med 171 54 60 15374840
Newman LS Mroz MM Maier LA Daniloff EM Bakissoon R 2001 Efficiency of serial medical surveillance for chronic beryllium disease in a beryllium machining plant J Occup Environ Med 43 231 237 11285871
OSHA 2005. Toxic and Hazardous Substances. 29CFR1910.1000, Table Z-1. Washington, DC:Occupational Safety and Health Administration.
Rossman MD 2001 Chronic beryllium disease: a hypersensitivity disorder Appl Occup Environ Hyg 16 615 618 11370939
Rossman MD Regovich J Atochina O Liang J Lee CW Lee J 1999 Progressive lung dysfunction in pre-clinical beryllium disease [Abstract] Am J Respir Crit Care Med 159 A621
Rubin DB 1987. Multiple Imputation for Nonresponse in Surveys. New York:J. Wiley & Sons.
Saltini C Richeldi L Losi M Amicosante M Voorter C VanderBerg-Loonen E 2001 Major histocompatibility locus genetic markers of beryllium sensitization and disease Eur Respir J 19 677 684 11716174
Stange AW Furman FJ Hilmas DE 2004 The beryllium lymphocyte proliferation test: relevant issues in beryllium health surveillance Am J Ind Med 46 453 462 15490468
Stange AW Hilmas DE Furman FJ Gatliffe TR 2001 Beryllium sensitization and chronic beryllium disease at a former nuclear weapons facility Appl Occup Environ Hyg 16 405 417 11297055
Sterner JH Eisenbud M 1951 Epidemiology of beryllium intoxication Arch Ind Hyg Occup Med 4 123 151
Tager IB Kalaidjian R Baldini L Rocklin RE 1985 Variability in the intradermal and in vitro lymphocyte responses to PPD patients receiving isoniazid chemoprophylaxis Am Rev Respir Dis 131 214 220 3882032
Tinkle SS Antonini JM Rich BA Roberts JR Salmen R DePree K 2003 Skin as a route of exposure and sensitiziation in chronic beryllium disease Environ Health Perspect 111 1202 1208 12842774
Viet SM Torma-Krajewski Rogers J 2000 Chronic beryllium disease and beryllium sensitization at Rocky Flats: a case-control study Am Ind Hyg Assoc J 61 244 254
Ward E Okun A Ruder A Fingerhut M Steenland K 1992 A mortality study of workers at seven beryllium processing plants Am J Ind Med 22 885 904 1463033
|
16203248
|
PMC1281281
|
CC0
|
2021-01-04 23:41:27
|
no
|
Environ Health Perspect. 2005 Oct 26; 113(10):1366-1372
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.7845
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences 10.1289/ehp.7766ehp0113-00137316203249ResearchEnvironmental MedicineCase Report: Occupationally Related Recurrent Varicella (Chickenpox) in a Hospital Nurse Ku Chih-Hung 12Liu Yu-Tien 3Christiani David C. 241 School of Public Health, National Defense Medical Center, National Defense University, Taipei, Taiwan2 Department of Environmental Health and Epidemiology, Harvard School of Public Health, Boston, Massachusetts, USA3 Department of Microbiology and Immunology, National Defense Medical Center, National Defense University, Taipei, Taiwan4 Department of Medicine, Harvard Medical School, Massachusetts General Hospital, Boston, Massachusetts, USAAddress correspondence to C-H. Ku, School of Public Health, National Defense Medical Center, National Defense University, Room 4328, No. 161, Section 6, Min-Chuan East Rd., Taipei City, Taiwan 114, Taiwan. Telephone/fax: 886-2-8792-9059. E-mail:
[email protected] authors declare they have no competing financial interests.
10 2005 15 6 2005 113 10 1373 1375 17 11 2004 14 6 2005 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright. Commonly accepted outcomes of varicella-zoster virus (VZV) infections include chickenpox (primary) and shingles (recurrence or latency), as well lifetime immunity against chickenpox. We report the case of a registered nurse who worked in a neurologic surgery ward in a general hospital in Taipei, Taiwan. While working there for approximately 1 year, she developed recurrent chickenpox after caring for a paraparesis patient, who had herpes zoster during hospitalization in August 2002. The varicella incubation period was 10 days, which matched the range (10–21 days). Recently negative specific serum IgM and positive specific serum IgG indicated a past VZV infection. The nurse did not get herpes zoster from the second episode of varicella on 9 August 2002 to 4 April 2005 and is now convalescing. We conclude that occupational VZV hazards exist in the health care environment and suggest testing for VZV antibody and a VZV vaccination program for susceptible health care workers.
chickenpoxindirect fluroscent antibodyoccupational exposurepolymerase chain reactionshinglesTaiwanvaricella-zoster virus
==== Body
Varicella (chickenpox), a common contagious disease of childhood, is caused by the varicella zoster virus (VZV) [Centers for Disease Control and Prevention (CDC) 2005]. VZV is characteristic of the alpha herpes viruses and establishes latency in the cells of the dorsal root ganglia after primary infection (Arvin 1996). The etiology of varicella and herpes zoster was first reported by von Bo′kay in 1888 from the observation that susceptible children often developed varicella after exposure to adults with herpes zoster (Arvin 1996; CDC 2005; von Bo′kay 1909). Varicella results from the primary VZV infection, whereas herpes zoster (shingles) is the result of reactivation (Arvin 1996; CDC 2005; Jumaan et al. 2002). Primary varicella infection usually results in lifetime immunity (CDC 2005), and second episodes of varicella are uncommon (CDC 2005; Gershon et al. 1984), but they may occur (CDC 2005). VZV disease history always indicates that varicella is the primary infection, and herpes zoster is a recurrence of the disease (Arvin 1996; CDC 1996, 1997
CDC 2005; Gershon et al. 1984; Jumaan et al. 2002), as well-documented second episodes of varicella are rare (Gershon et al. 1984). Here we report a case of apparent VZV reinfection with recurrent varicella infection in a nurse in a teaching general hospital in Taiwan.
Case Presentation
A 25-year-old nurse, who had childhood chickenpox, was diagnosed with varicella without mention of complication [International Classification of Diseases, Revision 9 (ICD-9) code 052.9; World Health Organization (WHO) 2001] by a dermatologic physician in a teaching general hospital after she cared for a 62-year-old male paraparesis patient who developed herpes zoster during hospitalization. She graduated from nursing school in July 2001, passed the licensing board, and then started to work in the neurologic surgery ward of a general teaching hospital in Taipei, Taiwan.
Toward the end of June, we conducted a study of occupational VZV hazards to health care workers in this hospital. The nurse was one of the volunteers who carried an air sampler for several hours in rotation with her colleagues on 9 July 2002. Saliva was collected simultaneously. Nested polymerase chain reaction (PCR) VZV DNA results were negative both in the personal air samples and in saliva.
On 13 July 2002, a 62-year-old man was sent to the emergency room due to paraparesis after he received Chinese traditional chiropractic treatment from a nonprofessional. He was diagnosed with spondylitis with a T8 compression fracture and T9 myelopathy, suspected tuberculosis (TB) of the spine, and paraparesis and was transferred to a neurologic (internal medicine) ward in the evening. The next day he underwent surgery for total laminectomy (from T8 to T9) and hook system (from T6 to T11), and was sent to the surgical intensive care unit (SICU) for 5 days. On 18 July 2002, the patient was transferred to the neurologic surgery ward where the nurse worked.
Because the patient was suspected of having TB of the spine, he underwent pleural and video-assisted thoracoscopic surgery (VATS) biopsy on 1 August. According to the nurse’s observation, the patient was worried, anxious, and under a great deal of stress regarding his health because his income was the primary financial resource of his family. On 3 August, multiple pruritic rash and vesicles were found over the patient’s abdomen and lower back flank area. Herpes zoster was confirmed by a dermatologic physician.
During the patient’s hospitalization in the neurologic surgery ward, the nurse attended the patient for several days (on 19–22 July, 29 July, 31 July–1 August, and 4–5 August 2002) before she resigned from the job on 12 August. Her regular nursing care tasks included measuring body temperature, blood pressure, and pulse; administering medicine; asking how the patient was feeling; and helping the patient to change his bed rest position.
At the age of 5 years, the nurse had been infected with chickenpox by her kindergarten-age sister, who herself was previously infected by her kindergarten classmates. At that time, the nurse, her sister, and her brother had multiple chickenpox vesicles on their faces simultaneously. Because of her previous chickenpox history, it was supposed that the nurse was immune to the VZV; while caring for the patient, she did not wear gloves, a mask, or an isolation gown before the herpes zoster was confirmed.
Because she had passed the admission examination for graduate study in a medical school, the nurse planned to resign her job to become a full-time student by 15 August. Before she resigned, she took a short vacation to her hometown in Ping-Tung County, in the south of Taiwan, on 6 August. However, on 8 August, she developed a high fever (39.5°C, 103.1°F), malaise, and headache. She went to a private clinic and was diagnosed with influenza by a physician in her hometown. The next day, in addition to the prodromal symptoms, pruritic rash started to appear on her face, neck, and trunk. She came back to the teaching general hospital in Taipei on 9 August and was diagnosed with varicella without mention of complication (ICD-9 code 052.9) by another dermatologic physician (not the herpes zoster patient’s dermatologic physician). The nurse received treatment with Allegra (fexofenadine, 60 mg twice per day; Hoechst, Kansas City, MO, USA), chlorpheniramine (4 mg three time per day; Mine Ta Chemistry Pharmacy Co., Ltd., Tai Chung, Taiwan), and Scanol (acetominophen, 500 mg three times per day; Scanpharm, Birkerød, Denmark) for 14 days. Because her clinical symptoms were very clear, no laboratory confirmation testing was performed at that time. However, to confirm this second varicella episode before reporting this case, we invited the nurse to take serologic tests for varicella antibody on 4 April 2005. The results indicated that VZV-specific IgM was negative [< 1:20; Merifluor VZV IgM indirect fluorescent antibody, (IFA); Meridian Bioscience Inc., Cincinnati, OH, USA], but the VZV IgG was positive (1:80 on the basis of > 1:10, Merifluor VZV IgG IFA), with high sensitivity (93.9% for IgM and 100% for IgG) and high specificity (100% for both antibodies) according to the manufacturer’s instructions.
Negative specific serum IgM and positive specific serum IgG indicated a past VZV infection. Because varicella and herpes zoster both resulted from the same antigen, the VZV IgG-positive reaction was excluded from the herpes zoster. It was also excluded because the nurse did not get herpes zoster from the second episode of varicella on 9 August 2002 to 4 April 2005. The nurse is now convalescing.
Discussion
The nurse was diagnosed with an acute chickenpox infection by a dermatologic physician at the teaching general hospital on the basis of her clinical symptoms and signs, including a high fever (39.5°C), malaise, and headache prior to the rash appearing on her face, neck, and trunk. The duration of the high fever and rash were 5 and 14 days, respectively. Fever and malaise may appear 1–2 days before rash onset in chickenpox primary infections in adults, whereas rash is usually the first sign of the disease in children (CDC 2005). According to the CDC case classification (Jumaan et al. 2002), a confirmed case is defined as one that is confirmed by laboratory testing or that meets the clinical case definition and is epidemiologically linked to a confirmed or a probable case (Jumaan et al. 2002). Recently, negative specific serum IgM and positive specific serum IgG indicated a past VZV infection. The nurse did not get herpes zoster from the second episode of varicella on 9 August 2002 to 4 April 2005. Our case matches both definitions because the vesicular illness preceded acute onset with diffuse maculopapulovesicular rash and without other apparent cause (Jumaan et al. 2002), and it was epidemiologically linked to the care of a confirmed herpes zoster patient.
The air sampling study was performed on 9 July 2002 in the neurologic surgery ward, and the paraparesis patient was admitted to the hospital on 13 July 2002 and transferred to the neurologic surgery ward on 18 July 2002. The patient’s rash was first reported on 3 August 2002. It is not surprising that the air sampling result was negative. In addition, it is unlikely that the nurse was infected from the air sampling equipment because the assembled cassettes and sampling pump with tubing were put in a laminar flow unit with ultraviolet light for 8 hr before sampling, and sampled cassettes were replaced with a new assembled cassette every day.
To investigate this uncommon case, we considered four parameters: the agent (VZV), infectious routes, the host (the nurse), and the hospital environment (hospital ward).
Agent.
Previous reports have mentioned the same etiology of varicella and herpes zoster (Arvin 1996; CDC 2005; von Bo′kay 1909). In 1925, children without varicella history developed varicella after being inoculated with fluid recovered from the herpes zoster lesions (infectious virus), which demonstrated the transmissibility of the agent (Arvin 1996; Kundratitz 1925). In the present case, the nurse developed a recurrence of varicella after nursing a herpes zoster patient, although she had been infected by her sister when she was 5 years of age.
Varicella has been reported as the disease most commonly confused with smallpox (Jumaan et al. 2002); however, smallpox was eradicated from Taiwan in 1955 (Center for Disease Control, Taiwan 2004a, 2004b), and the routine smallpox vaccination program in Taiwan ended in 1979 after the disease was eradicated worldwide following the WHO campaign (BiotechEast 2003; Center for Disease Control, Taiwan 2004a, 2004b; Lee 2003). Because the nurse was born in 1978, it is extremely unlikely that her rash at 5 years of age was smallpox.
New variants of VZV have been reported recently, including VZV-MSP, isolated in St. Paul–Minneapolis, MN (Santos et al. 1998, 2000), and VZV-BC, isolated in British Columbia (Tipples et al. 2002). We do not know whether there is a mutant strain of VZV in Taiwan; thus this is a good index case for further study. Another consideration is that the nurse’s immunity was insufficient against the second VZV attack, since she was infected initially over 20 years earlier by her sister. Unfortunately, it was not possible to check her VZV antibody titers before her recurrent varicella episode.
Infection routes and incubation period.
It has been documented that cases of herpes zoster are the infectious sources of chickenpox in susceptible persons (Arvin 1996; CDC 2005). VZV is transmissible by respiratory routes (Arvin 1996; CDC 2005). Possible infection routes while the nurse was caring for the patient included communicating with the patient and helping the patient change his bed rest position without employing protection such as a mask or gloves.
Previous PCR research showed that VZV is transmissible for 24–48 hr before rash onset, which is consistent with epidemiologic evidence (CDC 2005; Koropchak et al. 1991), or 4–5 days after a rash appears (CDC 2005).
The chickenpox incubation period is from 14 to 16 days after exposure (Arvin 1996; CDC 2005), with a range of 10–21 days (CDC 2005). In this case, the paraparesis patient had the rash on 3 August 2002. The nurse had cared for him on 31 July and on 1, 4, and 5 August. The most likely infection dates of this episode are 31 July and 1 August, since the rash appeared on the nurse’s face, neck, and trunk on 9 August, which is within the range of 10 days. In addition, recent VZV-IgG serologic tests indicated that the antibody titer is still high (1:80 on the basis of 1:10 of a positive criterion). This method was simple and highly sensitive and allowed for the rapid and reliable determination of immunity to VZV (Sauerbrei et al. 2004). Because varicella and herpes zoster resulted from the same antigen, VZV, the VZV IgG-positive reaction was excluded from herpes zoster. It was also excluded because the nurse did not get herpes zoster from the second episode of varicella on 9 August 2002 to 4 April 2005.
Host.
In addition to a lack of protection (gloves, mask) while caring for the patient, stress may be another important host factor contributing to recurrence. The nurse reported that before she resigned from the hospital, she felt stress both from the nursing practice and the logistics of her resignation, as well as anxiety about her impending study in graduate school. In fact, during our study of occupational VZV hazards to healthcare workers in early July 2002, the nurse reported performance anxiety about graduate school and stress from multiple clinical duties and logistic matters concerning her resignation, such as the fact that she was on duty on the date of graduate school registration, 5 August 2002. According to the nurse’s own report after the recurrence, we believe that stress played an important role in this recurrent varicella.
Environment.
Documented environmental VZV DNA included sampling from the air of the active patient’s room (Sawyer et al. 1994), or the active patient’s family, including air conditioner filter, table, television remote control, and door handle (Asano et al. 1999). In our study, 20.5% of the 44 air samples from different departments of the same hospital were VZV DNA positive. VZV is highly temperature sensitive (inactivated at 56–60°C) in the environment and is not infectious if the virion’s envelope is disrupted (Arvin 1996). The temperature in the hospital was controlled by the central operation department at about 25°C.
The nurse was a young and unmarried female, with only one year of work history. Her disease appears to be recurrent varicella, which is very uncommon, and it appears to be occupationally related. Documented VZV infection may cause significant morbidity or mortality, affecting the mother, the fetus, or the newborn (Arvin 1996; CDC 1996, 1997
CDC 2005; Harger et al. 2002; Jumaan et al. 2002), as well increasing the risk of premature delivery when infected during late pregnancy (Arvin 1996; Paryani and Arvin 1986). We report this case to emphasize that a varicella infection history may not be sufficient for determination of VZV immunity, contrary to the common belief that a reliable history of varicella is a valid measure of immunity because the rash is distinctive and subclinical cases rarely occur (CDC 1996, 2005; Jumaan et al. 2002). We think that VZV antibody testing may be necessary for health care workers, especially for new female workers.
Documented methods of VZV antibody detection include complement fixation, IFA, fluorescent antibody to membrane antigen (FAMA), neutralization, indirect hemagglutination, immune adherence hemagglutination, radioimmunoassay, latex agglutination, and enzyme-linked immunosorbent assay (CDC 1996). Although IFA, FAMA, neutrilization, and radioimmunoassay have been reported to be sensitive but time consuming (CDC 1996), a highly sensitive, specific, and rapid IFA test using VZV-infected A549 cells as antigen has been developed (Sauerbrei et al. 2004). The sensitivity and the specificity of this method are 100%, compared to the FAMA test, with the lowest limit of detection 50 mIU/mL versus 250 mIU/mL anti-VZV IgG for IFA and FAMA, respectively (Sauerbrei et al. 2004).
Conclusion
Occupational VZV hazards exist in the health care environment. The traditional concept of VZV lifetime immunity after the primary varicella infection may not be appropriate in a health care setting. We suggest checking serologic titers for VZV antibody, followed by a VZV vaccination for nonimmune health care workers.
==== Refs
References
Arvin A 1996 Varicella-zoster virus Clin Microbiol Rev 9 3 361 381 8809466
Asano Y Yoshikawa T Ihira M Furukawa H Suzuki K Suga S 1999 Spread of varicella-zoster virus DNA to family members and environments from siblings with varicella in a household Pediatrics 103 5 e61 Available: http://www.pediatrics.org/cgi/content/full/103/5/e61 [accessed 7 October 2004].10224205
BiotechEast 2003. Taiwan stocks up on smallpox vaccine. Available: http://www.biotecheast.com/modules.php?op=modload&name=News&file=article&sid=389 [accessed 7 October 2004].
CDC (Centers for Disease Control and Prevention) 1996 Prevention of varicella: recommendations of the Advisory Committee on Immunization Practices (ACIP) MMWR Recomm Rep 45 RR-11 1 36
CDC (Centers for Disease Control and Prevention) 1997 Immunization of health-care workers: recommendations of the Advisory Committee on Immunization Practices (ACIP) and the Hospital Infection Control Practices Advisory Committee (HICPAC) MMWR Recomm Rep 46 RR-18 1 42
CDC 2005. Varicella. In: Epidemiology and Prevention of Vaccine-Preventable Diseases (Atkinson W, ed). Atlanta, GA:Centers for Disease Control and Prevention, 159–175.
Center for Disease Control, Taiwan 2004a. Achievements. Taipei City, Taiwan:Center for Disease Control, Department of Health, Taiwan. Available: http://203.65.72.7/WebSite_En/AboutCDC/Achievements.htm [accessed 7 October 2004].
Center for Disease Control, Taiwan 2004b. History of Vaccination in Taiwan. Taipei, Taiwan:Center for Disease Control, Department of Health, Taiwan. Available: http://203.65.72.83/En/di/ShowPublication.ASP?RecNo=480 [accessed 7 October 2004].
Gershon AA Steinberg SP Gelb L 1984 Clinical reinfection with varicella-zoster virus J Infect Dis 149 2 137 142 6321605
Harger JH Ernest JM Thurnau GR Moawad A Momirova V Landon MB etal 2002 Risk factors and outcome of varicella-zoster virus pneumonia in pregnant women J Infect Dis 185 4 422 427 11865393
Jumaan A Hughes H Schmid S Galil K Plott K Zimmerman L 2002. Varicella. In: Manual for the Surveillance of Vaccine-Preventable Diseases (Wharton M, Hughes H, Reilly M, eds). Atlanta, GA:Centers for Disease Control and Prevention, 11–21.
Koropchak CM Graham G Palmer J Winsberg M Ting SF Wallace M 1991 Investigation of varicella-zoster virus infection by polymerase chain reaction in the immunocompetent host with acute varicella J Infect Dis 163 5 1016 1022 1850441
Kundratitz K 1925 Uëber die Aëtiologie des Zoster and über seine Beziehungen zu Varizellen [in German] Wien Klin Wochenschr 38 502 503
Lee M-L 2003. Public Health in Taiwan: Past, Present and Future. Available: http://www.gio.gov.tw/taiwan-website/4-oa/politics/hct20011205.htm [accessed 7 October 2004].
Paryani SG Arvin AM 1986 Intrauterine infection with varicella-zoster virus after maternal varicella N Engl J Med 314 24 1542 1546 3012334
Santos RA Hatfield CC Cole NL Padilla JA Moffat JF Arvin AM 2000 Varicella-zoster virus gE escape mutant VZV-MSP exhibits an accelerated cell-to-cell spread phenotype in both infected cell cultures and SCID-hu mice Virology 275 2 306 317 10998331
Santos RA Padilla JA Hatfield C Grose C 1998 Antigenic variation of varicella zoster virus Fc receptor gE: loss of a major B cell epitope in the ectodomain Virology 249 1 21 31 9740773
Sauerbrei A Farber I Brandstadt A Schacke M Wutzler P 2004 Immunofluorescence test for sensitive detection of varicella-zoster virus-specific IgG: an alternative to fluorescent antibody to membrane antigen test J Virol Methods 119 1 25 30 15109817
Sawyer MH Chamberlin CJ Wu YN Aintablian N Wallace MR 1994 Detection of varicella-zoster virus DNA in air samples from hospital rooms J Infect Dis 169 1 91 94 8277202
Tipples GA Stephens GM Sherlock C Bowler M Hoy B Cook D 2002 New variant of varicella-zoster virus Emerg Infect Dis 8 12 1504 1505 12498673
von Bo′kay J 1909 Uëber den aëtiologischen Zusammenhang der Varizellen mit gewissen Fällen von Herpes Zoster [in German] Wien Klin Wochenschr 22 1323 1326
WHO 2001. Manual of the International Statistical Classification of Diseases, Injuries, and Causes of Death. 9th Revision. Geneva:World Health Organization. Available: http://www.cdc.gov/nchs/about/major/dvs/icd9des.htm [accessed 26 August 2005].
|
16203249
|
PMC1281282
|
CC0
|
2021-01-04 23:41:28
|
no
|
Environ Health Perspect. 2005 Oct 15; 113(10):1373-1375
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.7766
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences 10.1289/ehp.8041ehp0113-00137616203250ResearchChildren's HealthMaternal Fish Consumption, Hair Mercury, and Infant Cognition in a U.S. Cohort Oken Emily 1Wright Robert O. 23Kleinman Ken P. 1Bellinger David 45Amarasiriwardena Chitra J. 3Hu Howard 35Rich-Edwards Janet W. 16Gillman Matthew W. 171 Department of Ambulatory Care and Prevention, Harvard Medical School and Harvard Pilgrim Health Care, Boston, Massachusetts, USA2 Department of Pediatrics, Boston Children’s Hospital and Harvard Medical School, Boston, Massachusetts, USA3 Channing Laboratory, Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts, USA4 Department of Neurology, Boston Children’s Hospital and Harvard Medical School, Boston, Massachusetts, USA5 Department of Environmental Health,6 Department of Epidemiology, and7 Department of Nutrition, Harvard School of Public Health, Boston, Massachusetts, USAAddress correspondence to E. Oken, Department of Ambulatory Care and Prevention, Harvard Medical School and Harvard Pilgrim, 133 Brookline Ave., Boston, MA 02215 USA. Telephone: (617) 509-9835. Fax: (617) 859-8112. E-mail:
[email protected] authors declare they have no competing financial interests.
10 2005 26 5 2005 113 10 1376 1380 24 2 2005 26 5 2005 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright. Fish and other seafood may contain organic mercury but also beneficial nutrients such as n-3 polyunsaturated fatty acids. We endeavored to study whether maternal fish consumption during pregnancy harms or benefits fetal brain development. We examined associations of maternal fish intake during pregnancy and maternal hair mercury at delivery with infant cognition among 135 mother–infant pairs in Project Viva, a prospective U.S. pregnancy and child cohort study. We assessed infant cognition by the percent novelty preference on visual recognition memory (VRM) testing at 6 months of age. Mothers consumed an average of 1.2 fish servings per week during the second trimester. Mean maternal hair mercury was 0.55 ppm, with 10% of samples > 1.2 ppm. Mean VRM score was 59.8 (range, 10.9–92.5). After adjusting for participant characteristics using linear regression, higher fish intake was associated with higher infant cognition. This association strengthened after adjustment for hair mercury level: For each additional weekly fish serving, offspring VRM score was 4.0 points higher [95% confidence interval (CI), 1.3 to 6.7]. However, an increase of 1 ppm in mercury was associated with a decrement in VRM score of 7.5 (95% CI, –13.7 to –1.2) points. VRM scores were highest among infants of women who consumed > 2 weekly fish servings but had mercury levels ≤1.2 ppm. Higher fish consumption in pregnancy was associated with better infant cognition, but higher mercury levels were associated with lower cognition. Women should continue to eat fish during pregnancy but choose varieties with lower mercury contamination.
child developmentcognitionenvironmental healthfishmercuryn-3 fatty acidsneurotoxinsseafood
==== Body
The developing brain is uniquely vulnerable to environmental insult (Rice and Barone 2000). During the last century, devastating community-level exposures to mercury-contaminated grain in Iraq and seafood in Japan established the sensitivity of the fetus to mercury’s toxic effects (Myers and Davidson 2000). More recently, two well-designed longitudinal studies have studied whether chronic low-level exposure to organic mercury from maternal seafood consumption during pregnancy is associated with adverse effects on fetal brain development, with conflicting results. Among 900 children in the Faroe Islands followed to 7 years of age, higher umbilical cord blood methylmercury was associated with lower scores on several developmental and cognitive tests (Grandjean et al. 1997). In contrast, in the Seychelles Child Development Study, no association of total maternal hair mercury with neurodevelopmental test performance was demonstrated in follow-up of > 700 children to 9 years of age (Davidson et al. 1998; Myers et al. 2003), despite the fact that mercury levels were similar to those in the Faroe Islands.
However, for women making dietary choices, the relevant question is whether eating fish and other seafood during pregnancy will impair their children’s development. Fish contain nutrients, including n-3 polyunsaturated acids, that may benefit mother and fetus (Lauritzen et al. 2001; Neuringer et al. 1994; Olsen et al. 1993). The net effect of the beneficial nutrients and harmful contaminants contained within fish has not been well studied and remains unclear (Chapman and Chan 2000). One recent study of 7,421 British children reported higher developmental scores at 15 months of age among offspring of women who consumed more fish during pregnancy, but found no association of umbilical cord tissue mercury levels with development among a subset of these participants (Daniels et al. 2004). In the present study, we examined associations of maternal fish and seafood intake and maternal hair mercury at delivery with 6-month infant cognition in a U.S. cohort.
Materials and Methods
Subjects and setting.
Study subjects were participants in Project Viva, a prospective cohort study of gestational diet and other exposures, pregnancy outcomes, and offspring health in eastern Massachusetts. Institutional review boards of participating institutions approved the study. All procedures were in accordance with the ethical standards for human experimentation established by the Declaration of Helsinki (World Medical Association 1997). Recruitment and retention procedures have been described previously (Gillman et al. 2004; Oken et al. 2004).
Briefly, we recruited women at their initial clinical obstetric appointment. Eligible participants were women who presented for their initial clinical visit at < 22 weeks of gestation, had a singleton pregnancy, were able to complete study forms in English, and did not plan to move out of the study area before delivery. We collected information about demographics, health history, and health habits by interview and self-administered questionnaire and obtained infant birth weight and gestation length from medical records (Oken et al. 2004). We calculated birth weight for gestational age using a U.S. national reference (Oken et al. 2003b).
At the second study visit, performed at 26–28 weeks of gestation, participants completed a semiquantitative food frequency questionnaire, which has been previously calibrated against blood levels of long-chain marine n-3 fatty acids (Fawzi et al. 2004). The questionnaire quantified average frequency of consumption of > 140 specified foods and beverages, including alcohol, during the preceding 3 months. The four questions regarding fish queried intake of “canned tuna fish (3–4 oz.)”; “shrimp, lobster, scallops, clams (1 serving)”; “dark meat fish, e.g., mackerel, salmon, sardines, bluefish, swordfish (3–5 oz.)”; and “other fish, e.g., cod, haddock, halibut (3–5 oz.).” Six response options ranged from “never/less than 1 per month” to “1 or more servings per day.” We combined responses for the four questions and generated a count of weekly total second-trimester fish servings. We used second-trimester diet as the measure of fish exposure because the timing of intake assessed by this questionnaire best overlapped with the timing of mercury exposure assessed by the maternal hair length sampled.
During the entire study period from April 1999 through February 2003, 2,128 participants delivered a live infant. We began collecting maternal hair samples in February 2002 and continued through the end of the study. During this period, 409 study participants delivered, of whom 302 were approached for collection of a hair sample during the hospitalization for delivery. The others were not approached mainly because they were hospitalized over a weekend only or because a trained research assistant was not available to meet them during their hospital stay. Thirty-two women were ineligible (hair too short or in braids), and 211 of the remaining 270 (78%) consented to provide a hair sample. Of these, 135 mother–infant pairs had complete information on both maternal second-trimester diet and 6-month infant cognitive testing and thus constitute the sample for the present analysis.
In the hospital after delivery, research assistants collected a sample of approximately 50–100 strands of hair from the mother’s occipital scalp, tied the hair at the proximal end, and stored it in a paper envelope at room temperature. We also asked about known exposure to mercury, for example, from occupational exposure. We did not collect information about dental amalgams or recent dental work.
Hair mercury assay.
We analyzed the proximal 3-cm length of the maternal hair specimen for total mercury content. This hair length represents growth during approximately months 6–8 of pregnancy, because the hair produced in the month before sampling remains under the scalp (National Research Council 2000). All samples were handled in a class 100 clean hood. We precleaned plastic and glassware by soaking them in 10% HNO3 for 24 hr, and then rinsing them several times with deionized water. Hair samples were sonicated for 15 min in approximately 10 mL of 1% Triton X-100 solution in precleaned 50-mL Pyrex beakers. After sonication, samples were rinsed several times with distilled deionized water and dried at 60°C for 24 hr.
We performed mercury assays using the Direct Mercury Analyzer 80 (Milestone Inc., Monroe, CT). This automatic mercury analyzer requires no sample digestion or pretreatment. The cleaned sample of hair was weighed into a nickel boat, thermally decomposed, and amalgamated, and then the released mercury was measured by atomic absorption spectroscopy at 253.7 nm as a function of mercury concentration. Samples were analyzed by using a matrix-matched calibration curve created with different weights of certified reference material GBW 09101 (human hair; Shanghai Institute of Nuclear Research, Academia Sinica, China) containing 2.16 ppm mercury.
Quality control steps included daily calibration with verification of a high- and a low-concentration standard for each working range, a procedural blank, and certified reference material NIES CRM-13 (human hair; National Institute for Environmental Studies, Ibaraki, Japan) as the standard. Mercury recovery was 90–110%, with > 95% precision.
Cognitive testing.
When the infants reached approximately 6 months of age, we performed cognitive testing using the visual recognition memory (VRM) paradigm. All subjects were first tested for visual acuity and had results within the normal range (Teller et al. 1986). We performed tests at the child’s home or in a research clinic. The infant was seated on the mother’s lap, with the mother’s view of the test stimuli shielded to minimize any possible influence on the infant’s performance.
Trained test administrators presented the infant with two identical photographs of an infant’s face, at a standardized distance. The habituation trials were repeated, with no maximum number of presentations, until the infant became habituated to this stimulus. In the testing phase, the infant was presented with the previously seen photo simultaneously with a novel photo of another infant’s face. Using a laptop computer, test administrators tracked the amount of time that the infant looked at each stimulus. The computer then calculated a novelty preference, the percentage of the total test time that the infant spent looking at the novel stimulus. The test administrator also recorded her confidence that the test was performed without distractions or any other concerns that might influence results. We excluded results for which the administrator did not have confidence in the test performance.
Each infant had two test trials, with the positions of the two faces alternated; the final score represents the average of the two trials. This test reflects the infant’s ability to encode a stimulus into memory, to recognize that stimulus, and to look preferentially at a novel stimulus (McCall and Carriger 1993), and is correlated with later IQ (Rose and Feldman 1995; Rose et al. 1992).
Statistical analysis.
We assessed factors associated with VRM score using linear regression. We first performed individual bivariate analyses with each maternal and child characteristic. For multivariable analyses, we included as independent predictors of VRM score both maternal hair mercury obtained at delivery (hereafter referred to as “mercury level”), and maternal second-trimester weekly intake of combined fish and seafood (hereafter referred to as “fish intake”), as well as the covariates maternal age (continuous), race/ethnicity (white vs. nonwhite), education (college graduate vs. not), and marital status (married or cohabiting vs. not), and infant sex, gestational age at birth (continuous), birth weight for gestational age (continuous), breast-feeding duration (continuous), and age at cognitive testing (continuous).
We studied both mercury levels and fish intake as continuous predictors. In addition, we dichotomized fish intake and mercury exposure based on public health recommendations. Federal advisories have recommended that pregnant women should consume two or fewer weekly fish meals (U.S. Department of Health and Human Services 2004). A hair methylmercury level of 1.2 ppm has been recommended as the U.S. reference dose (National Research Council 2000). Most hair mercury is in the methyl form, and total rather than methylmercury in hair has been recommended as a biomarker for mercury exposure (Davidson et al. 1998). We explored associations with VRM score using these dichotomized measures.
All final models met standard assumptions for linear regression. We did not see any evidence of influential outliers. We performed all analyses using SAS (version 8.2; SAS Institute Inc., Cary NC).
Results
Women in this cohort were predominantly white (82%), married or living with a partner (92%), and well educated (Table 1). The great majority of participants (94%) breast-fed their children, and half continued to breast-feed until at least 6 months postpartum. During pregnancy, 106 women (79%) consumed some alcohol, although only 25 (19%) continued consuming alcohol after learning they were pregnant; median first- and second-trimester alcohol consumption was 1.0 g/day (25th percentile 0.4, 75th percentile 3.0). Ten women (7%) reported smoking during pregnancy, and one reported illicit drug use. Use of these substances was not related to maternal hair mercury or to fish consumption. Six infants were born preterm (< 37 weeks), and three were small for gestational age (< 10th percentile).
Women consumed an average of 1.2 servings per week of combined tuna, dark meat, white meat, and shellfish (range, 0–5.5 servings/week) in the second trimester of pregnancy. The mean maternal hair mercury level was 0.55 ppm (micrograms per gram), with a range of 0.02–2.38. The geometric mean hair mercury level was 0.45 ppm. Fourteen participants (10%) had mercury levels > 1.2 ppm. No participants reported occupational exposure to mercury.
Total maternal fish intake was moderately associated with hair mercury content (Spearman r = 0.47). For each additional weekly total fish serving, mercury was 0.17 ppm higher [95% confidence interval (CI), 0.10 to 0.24]. Consumption of each group of fish was also correlated with hair mercury, with Spearman correlation coefficients with hair mercury ranging from r = 0.43 for canned tuna to r = 0.23 for white meat fish. No other maternal and infant sociodemographic characteristics, including breast-feeding, were associated with fish consumption or mercury levels (data not shown).
The mean VRM score (percent novelty preference) was 59.8 (range, 10.8–92.5). Infants were tested between 5.5 and 8.4 months of age (mean, 6.5 months). VRM score did not differ by maternal and infant characteristics (Table 1), including maternal use of cigarettes or alcohol during pregnancy. On bivariate analyses, maternal fish intake was positively associated with VRM score, and mercury was negatively associated with VRM score, although CIs did not exclude zero (Table 2). After adjusting for maternal and infant characteristics, each additional weekly fish serving was associated with a VRM score that was 2.8 (95% CI, 0.2 to 5.4) points higher (Table 2). In the similarly adjusted analysis, CIs for mercury did not exclude zero, although the estimate suggested that increasing mercury was associated with reduced VRM score.
To determine the independent effect of each on offspring cognition, we included both fish consumption and mercury level simultaneously in the linear regression model. In this multivariate model, both fish intake (direct) and mercury (inverse) were significantly and more strongly associated with infant cognition (Table 2). We further included as covariates maternal age, race/ethnicity, education and marital status, and infant sex, gestational age, birth weight for gestational age, breast-feeding duration, and age at cognitive testing. After adjustment, an increase of 1 ppm in maternal hair mercury was associated with a decrement in VRM score of 7.5 (95% CI, –13.7 to –1.2) points. For each additional weekly fish serving consumed by the mother, offspring VRM score was 4.0 (95% CI, 1.3 to 6.7) points higher on similarly adjusted analysis. Exclusion of participants with extreme values for mercury, fish intake, or VRM score did not markedly change results. Results were similar among nonsmokers only, and among nondrinkers only (data not shown).
We also examined multivariate associations with VRM score according to recommended thresholds for fish intake and mercury exposure. After adjustment for participant characteristics and mercury (continuous), the nine participants (7%) who consumed more than two weekly fish servings had infants with VRM scores that were 12.0 (95% CI, –0.1 to 24.1) points higher than those who consumed two or fewer weekly servings. Offspring of the 14 mothers (10%) with hair mercury > 1.2 ppm had VRM scores 9.3 (95% CI, –19.3 to 0.8) points lower than those with hair mercury ≤1.2 ppm, after adjustment for participant characteristics and fish intake (continuous).
We next investigated whether an interaction existed between low fish consumption and high mercury levels. Because of small numbers, we were not able to perform a multivariate analysis including both dichotomized variables. However, unadjusted VRM scores appeared highest among infants of mothers with high fish intake and low mercury levels, whereas scores appeared lowest in infants of mothers with low fish intake and high mercury (Table 3). We lacked power to demonstrate a statistically significant interaction in this unadjusted analysis.
Discussion
These results support findings from some studies that higher mercury exposure in pregnancy is associated with lower offspring cognitive scores, even at these relatively low levels of exposure. In addition, higher maternal fish intake was associated with higher mercury levels. However, higher maternal fish consumption was associated with better infant cognition. This benefit appeared greatest among infants whose mothers consumed more fish but had lower mercury levels.
The conflicting results from the two large-scale longitudinal studies in the Faroe and Seychelles Islands have led to disagreement about whether moderate mercury exposure from frequent seafood consumption may harm offspring development. The mean maternal hair mercury levels in the Faroe Islands (4.3 ppm) (Grandjean et al. 1997) and Seychelles Islands (6.8 ppm) (Davidson et al. 1998) studies were much higher than in our cohort (0.55 ppm), which had levels similar to other pregnant (Morrissette et al. 2004; Stern et al. 2001) and nonpregnant (Centers for Disease Control and Prevention 2001) U.S. populations. Some data suggest that no threshold exists for adverse neuropsychologic effects from methylmercury exposure (Rice 2004). A recent study in the United Kingdom did not show any adverse association of low levels of umbilical cord tissue mercury (median, 0.01 ppm) with child development (Daniels et al. 2004). However, in the U.K. study, development at 15 months of age was assessed by parental self-report, which is not likely to be as sensitive to the adverse effects of mercury.
Nevertheless, because high-dose organic mercury is known to harm the developing fetus, and because fish contain organic mercury, advisory bodies in the United States (National Research Council 2000; U.S. Department of Health and Human Services 2004), Canada (Health Canada 2002), and the United Kingdom (Committee on Toxicity 2004) have recommended that pregnant women limit their fish consumption. However, fish also contain nutrients such as iron, vitamin E, selenium, and long-chain n-3 polyunsaturated fatty acids that may benefit brain development (Clarkson and Strain 2003; National Research Council 2000; Neuringer et al. 1994). Little information has been available about the balance of risk and benefit for fish consumption.
It may seem contradictory that, on the one hand, fish intake raises mercury levels and higher mercury levels lead to worse cognition but, on the other hand, higher fish consumption is associated with better cognition. The most likely explanation is that the benefit is conferred by consuming fish types with the combination of relatively little mercury and high amounts of beneficial nutrients. This explanation is supported by results from multivariable models in the present analysis, in which adjustment for mercury strengthened the observed positive association of fish intake and cognition. Similarly, in the stratified analysis, we observed the highest cognitive scores among offspring of mothers with higher fish intake but lower mercury levels.
The fish questions in the Project Viva food frequency questionnaire were designed to estimate intake of fatty acids, not mercury. Because this questionnaire does not assess intake of individual fish types, but rather groups of fish, we cannot report associations for specific types of fish. Mercury levels vary among different fish species. In general, white meat fish such as cod and haddock tend to have lower mercury levels but also lower levels of long-chain n-3 fatty acids, whereas dark meat fish, such as swordfish, mackerel, and other large long-lived predatory fish, tend to contain both more mercury and more n-3 fatty acids. Because mercury and n-3 fatty acids often travel together, it may be difficult to isolate the opposing influences of the two on child cognition. Small fatty fish such as sardines and canned light tuna (vs. albacore tuna) may contain relatively more fatty acids with less mercury. Future studies incorporating more detailed dietary information may help advise women about specific fish species that are better or worse for their children’s cognition.
Resolving this issue remains important because women may indiscriminately reduce fish consumption in response to concerns about mercury exposure, perhaps substituting fish with other, less healthful foods. In a previous study, we demonstrated that a different subset of pregnant women enrolled earlier in our cohort reduced consumption not only of dark meat fish, which are likely to have higher mercury levels, but also of canned tuna and white meat fish, which tend to have lower mercury levels, after dissemination of a 2001 U.S. federal mercury advisory (Oken et al. 2003a). A recently updated federal advisory reiterated health warnings while encouraging women to consume up to two seafood meals per week (U.S. Department of Health and Human Services 2004), but it is unclear to what extent women understand the details of the health message or simply hear that seafood contains mercury and therefore is harmful.
In Project Viva, we assessed infant cognition using the VRM protocol. This test of cognitive function has many advantages for studies of prenatal exposures: It can be performed early in infancy and assesses cognition isolated from motor function. VRM tests have predicted IQ in childhood and early adolescence as strongly as other standardized tests of infant development (e.g., the Bayley Scale of Infant Development) (Rose and Feldman 1995; Rose et al. 1992), with correlations with intelligence in later childhood ranging from 0.44 to 0.66 (Laucht et al. 1994). However, similar to all tests of infant cognition, the VRM is most strongly correlated with later IQ when mental development is impaired, whereas the relationship is less strong when cognition is within the normal range. The VRM, which uses visual preference rather than motor skills to assess cognitive ability, might be particularly sensitive to the benefits of fish consumption: The marine fatty acid docosahexaenoic acid is an essential component of the retina and promotes infant vision (SanGiovanni et al. 2000). VRM testing has been used to demonstrate associations of infant cognition with infant intake of marine n-3 fatty acids (O’Connor et al. 2001; Uauy et al. 2001) as well as prenatal exposure to environmental pollutants such as lead and polychlorinated biphenyls (Darvill et al. 2000; Emory et al. 2003). Inverse associations of methylmercury exposure and novelty preference scores have been seen among non-human primates (Gunderson et al. 1986, 1988); however, other studies among human infants, most notably in the Seychelles Islands cohort, have not demonstrated an association of prenatal mercury exposure with VRM results (Darvill et al. 2000; Myers et al. 1995). Future cohort follow-up will help determine whether fish and mercury are associated with intelligence later in childhood.
Results should be generalized with some caution because our study population contained a high proportion of educated, white, and well-off mothers, all of whom lived in one area of the United States. Even within this sample, however, the main exposures and outcome were similar to those of other populations. For example, levels of hair mercury were similar to those in women of childbearing age in the most recent National Health and Nutrition Examination Survey (90th percentile, 1.4 ppm), which is nationally representative of the U.S. population (Centers for Disease Control and Prevention 2001). The mean novelty preference score of 59.8 in our population was similar to that observed among children enrolled in the Seychelles Child Development Study (60.3) (Myers et al. 1995) and among U.S. children living in New York State (59.8) (Darvill et al. 2000).
As with all observational studies, it is possible that we did not completely adjust for potentially confounding factors. For example, we did not collect information on parental IQ or home environmental stimulation, because Project Viva includes many outcomes of interest in addition to cognition, and we wished to reduce participant burden. Both of these factors might be associated with fish intake and have been associated with child cognition in prior studies, although usually only beginning in the later half of the second year of life. We also did not measure exposure to persistent organic pollutants such as polychlorinated biphenyls, although we would expect that accounting for the adverse effect of these toxicants would strengthen the observed benefit of seafood. We assessed mercury exposure using total maternal hair mercury, which is a recommended biomarker for estimating the methylmercury dose received by the offspring’s brain (National Research Council 2000) and strongly associated with both maternal blood and cord blood organic mercury (Morrissette et al. 2004). More than 80% of the mercury in hair is in the methyl form; in addition, organic mercury can be partly transformed to inorganic mercury, so the concentration of total mercury more accurately represents the mercury entering the hair follicle from the blood stream (Davidson et al. 1998; Myers et al. 2003). Some mercury may have derived from the inorganic mercury in dental amalgams. If this were true, however, the beneficial effect of fish on cognition then would be even stronger than what we observed. Our sample was too small to address whether there are subgroups within the population that are more susceptible to mercury exposure because of genetic, nutritional, or social factors that may sensitize the developing brain to neurotoxicants.
Results from the present study among a U.S. cohort with moderate fish intake suggest that maternal fish consumption during pregnancy may benefit offspring cognition in infancy but that exposure to higher levels of mercury has adverse effects on child cognition. These findings, based on a relatively small group of women, merit further investigation and verification in other populations consuming moderate amounts of seafood. Meanwhile, we recommend that women continue to consume fish during pregnancy but seek out varieties with lower levels of mercury.
We appreciate the invaluable contributions provided to this project by the staff and participants of Project Viva.
This project was supported by grants from the National Institutes of Health (HD34568, HL68041, HD44807, ES00002, P01ES012874), Harvard Medical School, and the Harvard Pilgrim Health Care Foundation.
Table 1 Participant characteristics and their unadjusted associations with cognition in 6-month-old infants, assessed by VRM testing: results from 135 mother–infant pairs enrolled in Project Viva.
Maternal characteristics Subjects (%) VRM score (95% CI) p-Value
Age (years) 0.80
< 30 16 60 (51 to 68)
30–34 53 61 (57 to 65)
≥35 31 59 (52 to 67)
Race/ethnicity 0.81
White 82 60 (56 to 63)
Nonwhite 18 61 (54 to 67)
Marital status 0.79
Married or cohabitating 92 60 (57 to 63)
Divorced or single 8 61 (51 to 70)
Education level 0.25
College or graduate degree 80 59 (56 to 62)
< College graduate 20 63 (58 to 68)
Sex 0.56
Female 51 59 (55 to 62)
Male 49 61 (56 to 66)
Gestation length (weeks) 0.73
< 37 4 65 (50 to 79)
37–38 22 58 (51 to 65)
39–40 54 61 (57 to 65)
≥41 19 58 (50 to 65)
Birth weight for gestational age 0.76
Small (< 10th percentile) 2 53 (28 to 78)
Appropriate 85 60 (57 to 63)
Large (> 90th percentile) 13 59 (49 to 69)
Breast-feeding duration (months) 0.23
< 2 19 54 (47 to 62)
2–4 23 60 (53 to 67)
≥5 58 61 (57 to 66)
Infant age at testing (months) 0.37
< 7 80 60 (57 to 64)
≥7 20 57 (51 to 64)
Table 2 Associations of maternal second-trimester fish consumption and maternal hair mercury at delivery with infant cognition at 6 months (VRM score): results from six linear regression models among 135 mother–infant pairs in Project Viva.
Change in VRM score [% novelty preference (95% CI)]
Model Effect per weekly fish serving Effect per ppm maternal hair mercury
Fish only 2.5 (–0.01 to 5.0) —
Fish and participant characteristicsa 2.8 (0.2 to 5.4) —
Mercury only — –4.6 (–10.3 to 1.1)
Mercury and participant characteristicsa — –4.0 (–10.0 to 2.0)
Fish and mercury 3.9 (1.2 to 6.5) –8.1 (–14.1 to –2.0)
Fish, mercury, and participant characteristicsa 4.0 (1.3 to 6.7) –7.5 (–13.7 to –1.2)
a Participant characteristics adjusted for include maternal age (continuous), race/ethnicity (white vs. nonwhite), education (college graduate vs. not), marital status (married or cohabiting vs. not), and infant sex, gestational age at birth (continuous), birth weight for gestational age (continuous), breast-feeding duration (continuous), and age at cognitive testing (continuous).
Table 3 Mean cognitive (VRM) scores (% novelty preference) among offspring of mothers with high or low second-trimester fish intake and high or low hair mercury levels at delivery.
Hair mercury
Weekly fish intake ≤1.2 ppm > 1.2 ppm
> 2 servings 72 (n = 7) 55 (n = 2)
≤2 servings 60 (n = 114) 53 (n = 12)
==== Refs
References
Centers for Disease Control and Prevention 2001 Blood and hair mercury levels in young children and women of childbearing age—United States, 1999 JAMA 285 1436 1437 11280332
Chapman L Chan HM 2000 The influence of nutrition on methyl mercury intoxication Environ Health Perspect 108 suppl 1 29 56 10698722
Clarkson TW Strain JJ 2003 Nutritional factors may modify the toxic action of methyl mercury in fish-eating populations J Nutr 133 1539S 1543S 12730461
Committee on Toxicity 2004. Advice on Fish Consumption: Benefits and Risks. London:The Stationery Office. Available: http://www.foodstandards.gov.uk/multimedia/pdfs/cotstatementmercuryfish.PDF [accessed 20 July 2004].
Daniels JL Longnecker MP Rowland AS Golding J 2004 Fish intake during pregnancy and early cognitive development of offspring Epidemiology 15 394 402 15232398
Darvill T Lonky E Reihman J Stewart P Pagano J 2000 Prenatal exposure to PCBs and infant performance on the Fagan test of infant intelligence Neurotoxicology 21 1029 1038 11233749
Davidson PW Myers GJ Cox C Axtell C Shamlaye C Sloane-Reeves J 1998 Effects of prenatal and postnatal methylmercury exposure from fish consumption on neurodevelopment: outcomes at 66 months of age in the Seychelles Child Development Study JAMA 280 701 707 9728641
Emory E Ansari Z Pattillo R Archibold E Chevalier J 2003 Maternal blood lead effects on infant intelligence at age 7 months Am J Obstet Gynecol 188 S26 S32 12712133
Fawzi WW Rifas-Shiman SL Rich-Edwards JW Willett WC Gillman MW 2004 Calibration of a semi-quantitative food frequency questionnaire in early pregnancy Ann Epidemiol 14 754 762 15519898
Gillman MW Rich-Edwards JW Rifas-Shiman SL Lieberman ES Kleinman KP Lipshultz SE 2004 Maternal age and other predictors of newborn blood pressure J Pediatr 144 240 245 14760269
Grandjean P Weihe P White RF Debes F Araki S Yokoyama K 1997 Cognitive deficit in 7-year-old children with prenatal exposure to methylmercury Neurotoxicol Teratol 19 417 428 9392777
Gunderson VM Grant KS Burbacher TM Fagan JF III Mottet NK 1986 The effect of low-level prenatal methylmercury exposure on visual recognition memory in infant crab-eating macaques Child Dev 57 1076 1083 3757602
Gunderson VM Grant-Webster KS Burbacher TM Mottet NK 1988 Visual recognition memory deficits in methylmercury-exposed Macaca fascicularis infants Neurotoxicol Teratol 10 373 379 3226381
Health Canada 2002. Information on Mercury Levels in Fish. 29 May 2002. Ottawa, Ontario, Canada:Health Canada. Available: http://www.hc-sc.gc.ca/english/protection/warnings/2002/2002_41e.htm [accessed 20 July 2004].
Laucht M Esser G Schmidt MH 1994 Contrasting infant predictors of later cognitive functioning J Child Psychol Psychiatry 35 649 662 7518825
Lauritzen L Hansen HS Jorgensen MH Michaelsen KF 2001 The essentiality of long chain n-3 fatty acids in relation to development and function of the brain and retina Prog Lipid Res 40 1 94 11137568
McCall RB Carriger MS 1993 A meta-analysis of infant habituation and recognition memory performance as predictors of later IQ Child Dev 64 57 79 8436038
Morrissette J Takser L St-Amour G Smargiassi A Lafond J Mergler D 2004 Temporal variation of blood and hair mercury levels in pregnancy in relation to fish consumption history in a population living along the St. Lawrence River Environ Res 95 363 374 15220070
Myers GJ Davidson PW 2000 Does methylmercury have a role in causing developmental disabilities in children? Environ Health Perspect 108 suppl 3 413 420 10852838
Myers GJ Davidson PW Cox C Shamlaye CF Palumbo D Cernichiari E 2003 Prenatal methylmercury exposure from ocean fish consumption in the Seychelles Child Development Study Lancet 361 1686 1692 12767734
Myers GJ Marsh DO Davidson PW Cox C Shamlaye CF Tanner M 1995 Main neurodevelopmental study of Seychellois children following in utero exposure to methylmercury from a maternal fish diet: outcome at six months Neurotoxicology 16 653 664 8714870
National Research Council 2000. Toxicological Effects of Methylmercury. Washington, DC:National Academy Press.
Neuringer M Reisbick S Janowsky J 1994 The role of n-3 fatty acids in visual and cognitive development: current evidence and methods of assessment J Pediatr 125 S39 S47 7965452
O’Connor DL Hall R Adamkin D Auestad N Castillo M Connor WE 2001 Growth and development in preterm infants fed long-chain polyunsaturated fatty acids: a prospective, randomized controlled trial Pediatrics 108 359 371 11483801
Oken E Kleinman KP Berland WE Simon SR Rich-Edwards JW Gillman MW 2003a Decline in fish consumption among pregnant women after a national mercury advisory Obstet Gynecol 102 346 351 12907111
Oken E Kleinman KP Olsen SF Rich-Edwards JW Gillman MW 2004 Associations of seafood and elongated n-3 fatty acid intake with fetal growth and length of gestation: results from a US pregnancy cohort Am J Epidemiol 160 774 783 15466500
Oken E Kleinman KP Rich-Edwards J Gillman MW 2003b A nearly continuous measure of birth weight for gestational age using a United States national reference BMC Pediatr 3 6 12848901
Olsen SF Grandjean P Weihe P Videro T 1993 Frequency of seafood intake in pregnancy as a determinant of birth weight: evidence for a dose dependent relationship J Epidemiol Community Health 47 436 440 8120495
Rice D Barone S Jr 2000 Critical periods of vulnerability for the developing nervous system: evidence from humans and animal models Environ Health Perspect 108 suppl 3 511 533 10852851
Rice DC 2004 The US EPA reference dose for methylmercury: sources of uncertainty Environ Res 95 406 413 15220074
Rose SA Feldman JF 1995 Prediction of IQ and specific cognitive abilities at 11 years from infancy measures Dev Psychol 31 685 696
Rose SA Feldman JF Wallace IF 1992 Infant information processing in relation to six-year cognitive outcomes Child Dev 63 1126 1141 1446544
SanGiovanni JP Parra-Cabrera S Colditz GA Berkey CS Dwyer JT 2000 Meta-analysis of dietary essential fatty acids and long-chain polyunsaturated fatty acids as they relate to visual resolution acuity in healthy preterm infants Pediatrics 105 1292 1298 10835071
Stern AH Gochfeld M Weisel C Burger J 2001 Mercury and methylmercury exposure in the New Jersey pregnant population Arch Environ Health 56 4 10 11256856
Teller DY McDonald MA Preston K Sebris SL Dobson V 1986 Assessment of visual acuity in infants and children: the acuity card procedure Dev Med Child Neurol 28 779 789 3817317
Uauy R Hoffman DR Peirano P Birch DG Birch EE 2001 Essential fatty acids in visual and brain development Lipids 36 885 895 11724460
U.S. Department of Health and Human Services 2004. What you need to know about mercury in fish and shellfish. March 2004. Washington, DC:U.S. Department of Health and Human Services. Available: http://www.cfsan.fda.gov/~dms/admehg3.html [accessed 20 July 2004].
World Medical Association 1997 World Medical Association declaration of Helsinki. Recommendations guiding physicians in biomedical research involving human subjects JAMA 277 925 926 9062334
|
16203250
|
PMC1281283
|
CC0
|
2021-01-04 23:41:29
|
no
|
Environ Health Perspect. 2005 Oct 26; 113(10):1376-1380
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.8041
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences 10.1289/ehp.7856ehp0113-00138116203251ResearchChildren's HealthTransport of Methylmercury and Inorganic Mercury to the Fetus and Breast-Fed Infant Björnberg Karolin Ask 1Vahter Marie 1Berglund Birgitta 2Niklasson Boel 3Blennow Mats 2Sandborgh-Englund Gunilla 41 Division of Metals and Health, Institute of Environmental Medicine, Karolinska Institutet, Stockholm, Sweden2 Division of Pediatrics and3 Department of Obstetrics and Gynecology, Karolinska University Hospital, Huddinge, Sweden4 Division of Dental Biomaterials, Institute of Odontology, Karolinska Institutet, Huddinge, SwedenAddress correspondence to G. Sandborgh-Englund, Division of Dental Biomaterials, Institute of Odontology, Karolinska Institutet, Box 4064, SE-141 04 Huddinge, Sweden. Telephone: 46-8-524-881-03. Fax: 46-8-711-83-43. E-mail:
[email protected] authors declare they have no competing financial interests.
10 2005 15 6 2005 113 10 1381 1385 15 12 2004 14 6 2005 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright. It is well established that methylmercury (MeHg) and mercury vapor pass the placenta, but little is known about infant exposure via breast milk. We measured MeHg and inorganic mercury (I-Hg) in blood of Swedish mothers (n = 20) and their infants, as well as total mercury (T-Hg) in breast milk up to 13 weeks postpartum. Infant blood MeHg was highly associated with maternal blood MeHg at delivery, although more than twice as high. Infant MeHg decreased markedly until 13 weeks of age. Infant blood I-Hg was associated with, and about as high as, maternal blood I-Hg at delivery. Infant I-Hg decreased until 13 weeks. In breast milk, T-Hg decreased significantly from day 4 to 6 weeks after delivery but remained unchanged thereafter. At 13 weeks, T-Hg in breast milk was associated with infant MeHg but not with maternal MeHg. Conversely, T-Hg in breast milk was associated with maternal I-Hg but not with infant I-Hg. From the findings of the present study in which the exposure to both MeHg and I-Hg was low, we conclude that the exposure to both forms of mercury is higher before birth than during the breast-feeding period, and that MeHg seems to contribute more than I-Hg to infant exposure postnatally via breast milk.
breast-feedingbreast milkhumaninfant exposureinorganic mercurymethylmercurypregnancy
==== Body
People are exposed to methylmercury (MeHg) mainly through consumption of predatory fish species, and to inorganic mercury (I-Hg) mainly through release of mercury vapor (Hg0) from amalgam fillings. Both MeHg and I-Hg are neurotoxic, especially for the developing brain [National Research Council (NRC) 2000], and it is well documented that both MeHg and Hg0 readily pass the placenta [Ask et al. 2002; NRC 2000; Vahter et al. 2000; World Health Organization (WHO) 1991]. In general, MeHg concentrations in cord blood are almost twice those in maternal blood (Sakamoto et al. 2004; Stern and Smith 2003; Vahter et al. 2000), probably due to transport via the neutral amino acid carrier (Kajiwara et al. 1996). Blood I-Hg concentrations are about the same in mother and newborn (Vahter et al. 2000).
The brain is sensitive to chemical assaults also postnatally (Rice and Barone 2000) but little is known about infant mercury exposure via breast milk. Breast-feeding has many benefits, and breast milk is the best source of nutrition for infants (Gartner et al. 1997; Oddy 2002; Pronczuk et al. 2004). However, it may also be a source of environmental contaminants (Anderson and Wolff 2000; Dorea 2004). Findings of decreasing concentrations of I-Hg in maternal blood during breast-feeding indicate that I-Hg is excreted in breast milk (Vahter et al. 2000). This is supported by reported correlations between total mercury (T-Hg) in milk and plasma, believed to contain mainly I-Hg (Skerfving 1988), and between I-Hg in milk and I-Hg in whole blood (Oskarsson et al. 1996). Also, T-Hg in milk correlated with number of amalgam fillings, but not with fish consumption (Oskarsson et al. 1996). However, another study found correlations to both (Drexler and Schaller 1998). Both human and animal studies indicate that the infant may be exposed to MeHg via breast milk (Grandjean et al. 1994; Nordenhall et al. 1998; Sundberg et al. 1991). However, to what extent I-Hg and MeHg in breast milk are taken up by the child is not known. By speciating the blood concentrations of MeHg and I-Hg at birth and during breast-feeding in relation to the concentrations in maternal blood and breast milk, we aimed in the present study to clarify the transport of MeHg and I-Hg during breast-feeding and infant exposure through breast milk.
Materials and Methods
In 2001 we recruited 20 women at delivery at the Huddinge University Hospital, using a convenience sample. The midwives involved in the project informed women registering at the delivery clinic about the study. No particular exclusions were made, but emergency cases were not included. The participation was voluntary, and only a few of the approached women chose not to participate. There were no incentives. Reason for nonparticipation was the inconvenience of sampling.
We obtained information about smoking and alcohol consumption from the antenatal care centers. Four women reported that they did consume alcohol, although seldom, and two women reported smoking, one of them quitting during pregnancy. The women had normal and healthy pregnancies. However, one woman developed late preeclampsia (week 39), and one had premature delivery (week 35). Two women delivered by cesarean section. One infant was born with cleft palate, and one with functional ileal obstruction. The general health and growth of all infants were good at the last sampling occasion.
We collected blood samples from the mothers (antecubital vein) and infants at the time of delivery and at approximately 4 days (in connection with the phenylketonuria test; range, 3–6 days) and at 13 weeks (on average, 94 days; range, 86–115 days) after delivery. Infant blood was collected from the umbilical artery at delivery, the femoral vein at 4 days, and the hand vein at 13 weeks. The mothers collected breast milk samples at approximately 4 days (colostrum; range, 2–5 days), 6 weeks (46 days; range, 34–59 days), and 13 weeks (88 days; range, 77–101 days) after delivery. Because of the known variability in breast milk composition, the women were asked to collect three different samples (> 5 mL) at 6 and 13 weeks, respectively; to register the time when the sample was taken; and to specify whether the sample was collected at the beginning, middle, or end of the feeding. A total of 15 women collected samples on all three occasions.
At delivery, the women filled out a questionnaire concerning fish consumption, vaccinations, and dental care during the preceding 6 months. At the 13-week appointment, the women completed a similar questionnaire that also included information about breast-feeding and the use of infant formula. Six women reported use of infant formula in combination with breast-feeding at 13 weeks. In addition, a dentist recorded the number of amalgam-filled surfaces. We obtained informed consent from the women, and the Ethics Committee at Karolinska Institutet approved the study.
We analyzed T-Hg and I-Hg in blood by alkaline solubilization/reduction and cold-vapor atomic fluorescence spectrophotometry (CVAFS; Merlin, PSA 10.023; P.S. Analytical Ltd., Orpington, Kent, UK) as previously described (Vahter et al. 2000). The concentration of organic mercury was calculated as the difference between T-Hg and I-Hg. We assume that essentially all of the organic mercury in blood was in the form of MeHg, because the only other known exposure sources of organic mercury compounds in Sweden are a few vaccines containing thimerosal (a preservative containing ethylmercury). None of the women had received any vaccines during the pertinent period, and none of the infants had received vaccines containing thimerosal. The use of thimerosal in vaccines for infants and children in Sweden stopped in the early 1990s. In breast milk, we analyzed T-Hg by CVAFS (Merlin, PSA 10.003; P.S. Analytical Ltd.) (Sandborgh-Englund et al. 1998) after acid microwave digestion. Breast milk samples of 1.0 mL were mixed with 1.5 mL concentrated HNO3 (suprapure) and digested in a microwave oven (model MDS-2000; CEM-Innovators Microwave Technology, Matthews, NC, USA). We were not able to speciate mercury in breast milk, although we tried different methods. The main problem was the solubilization step. Digestion of breast milk samples in a microwave oven enabled us to determine T-Hg. Because the concentrations were very low, we made no further attempts to speciate mercury in milk. All utensils used were checked to be free from mercury contamination.
The limit of detection (LOD; 3 × SD of the reagent blanks) in blood varied between 0.04 and 0.09 μg/L for T-Hg and between 0.03 and 0.08 μg/L for I-Hg. The LOD for T-Hg in breast milk varied between 0.03 and 0.07 μg/L. One-third (n = 40) of the blood samples (from 10 of the women and 15 of the infants) had I-Hg concentrations below the LOD, and two breast milk samples (from two different women) had T-Hg concentrations below the LOD.
Results of the analytical quality control are presented in Table 1. There are no recommended values for I-Hg in the Seronorm reference blood samples (Nycomed Co., Oslo, Norway), but the values obtained for I-Hg were in good agreement with our previous analytical runs of the same Seronorm sample (Seronorm 404107: 0.53 ± 0.06 μg I-Hg/L, n = 21; Seronorm 404108: 6.2 ± 0.59 μg I-Hg/L, n = 27). Repeated analysis of a MeHg standard containing 0.4 μg Hg/L during blood analysis gave a recovery of 99 ± 5%. Repeated analysis of standard solutions containing 1.25 and 2.5 ng T-Hg during breast milk analysis gave a recovery of 100 ± 4%. Seven breast milk samples from one woman, sampled at different occasions, were replicated in three separate analyses. The coefficient of variation varied between 2 and 16%.
Statistics.
The mercury concentrations were not normally distributed (skewness and Shapiro-Wilk tests of normality). To test for correlations between variables, we used Pearson correlation (r) whenever the requirements for normally distributed residuals were met; otherwise, we used Spearman correlation (rS). We used Wilcoxon and Kruskal-Wallis nonparametric tests to test for differences between variables and groups.
For evaluation of variation in mercury concentrations over time, we log-transformed the data to be able to use analysis of variance (ANOVA) for repeated measures with one within-subject factor (time) with three levels. To test for differences in mercury concentrations between specific time points, we used contrast analysis.
Statistical analyses were conducted mainly with SPSS (version 12.0.1 for Windows; SPSS Inc., Chicago, IL, USA). The influence of sampling in relation to feeding (beginning, middle, or end) was evaluated using the Mixed procedure in SAS System 8.2 (SAS Institute Inc., Cary, NC, USA). We computed two-way repeated-measures ANOVA with feeding (beginning, middle, and end) and week (6 and 13) as the within-subjects variables. Statistical significance level was set to p < 0.05.
Although values below LOD have larger uncertainty than do those above, analytical results below LOD were not rejected or changed so that distributions would not be distorted. In the evaluation of correlations between T-Hg in breast milk and the different forms of mercury in blood, we used mean values of all breast milk samples from each woman taken at 6 weeks and 13 weeks, respectively.
Results
A summary of study characteristics is presented in Table 2. Minor dental treatment was reported by seven women during pregnancy and two women during breast-feeding. They did not have higher blood mercury concentrations compared with those who did not have dental treatment. The blood mercury concentration of children given infant formula in combination with breast-feeding (n = 6) did not differ from those exclusively breast-fed. The women’s total fish intake was similar during pregnancy as during breast-feeding (p = 0.25). No woman reported intake of freshwater fish during pregnancy or breast-feeding. Information on consumption of predatory marine fish was not explicitly asked for.
As shown in Figure 1, maternal blood MeHg increased from delivery (median, 0.45 μg/L; range, 0.24–1.5 μg/L) to 13 weeks postpartum (median, 0.60 μg/L; range, 0.20–1.6 μg/L; p = 0.01). MeHg in maternal blood was associated with MeHg in cord blood (median, 0.99 μg/L; range, 0.52–3.8 μg/L; Figure 2) and in infant blood at 4 days (median, 1.1 μg/L; range, 0.62–4.4 μg/L; r = 0.95; p < 0.001), although the cord and infant blood concentrations were more than twice as high as those in maternal blood. Infant blood MeHg decreased (p < 0.001) from 4 days to 13 weeks after birth (median, 0.38 μg/L; range, 0.10–1.1 μg/L; Figure 1). We did not find any significant associations between reported fish consumption and maternal or infant blood MeHg.
As shown in Figure 3, maternal blood I-Hg was about the same at delivery (median, 0.09 μg/L; range, 0.03–0.75 μg/L) as at 13 weeks postpartum (p = 0.78). Infant blood I-Hg (median, 0.09 μg/L; range, 0.02–0.34 μg/L) was similar to maternal concentrations at birth and was significantly associated with maternal blood I-Hg both at birth (Figure 2) and at 4 days (rS = 0.51; p = 0.02). Infant blood I-Hg decreased (p = 0.001) until 13 weeks of age (median, 0.05 μg/L; range, 0–0.13 μg/L). Maternal blood I-Hg correlated significantly with the number of amalgam-filled surfaces at delivery (rS = 0.55; p = 0.01).
As shown in Figure 4, T-Hg in breast milk at 13 weeks correlated significantly to maternal blood I-Hg but not to infant blood I-Hg. Conversely, T-Hg in breast milk at 13 weeks correlated significantly to infant blood MeHg (Figure 5) but not to maternal blood MeHg. T-Hg in breast milk (Figure 6) decreased significantly (p < 0.001) from day 4 (colostrum; median, 0.29 μg/L; range, 0.06–2.1 μg/L) to 6 weeks postpartum (mature milk; median, 0.14 μg/L; range, 0.07–0.37 μg/L) but remained unchanged thereafter. T-Hg in milk increased significantly with time during each feeding session at 6 weeks (n = 15; p < 0.001). The median concentrations were 0.12 μg/L (range, 0.04–0.31 μg/L) in the first milk that was pumped out, 0.15 μg/L (range, 0.07–0.32 μg/L) about halfway through the feeding session, and 0.18 μg/L (range, 0.07–0.49 μg/L) at the end. This change was less pronounced at 13 weeks.
T-Hg in breast milk correlated with the number of amalgam-filled surfaces at 4 days (rS = 0.49; p = 0.04) and 6 weeks (rS = 0.44; p = 0.05). We did not find any significant associations between T-Hg in breast milk and reported fish consumption. Parity did not affect the concentrations of T-Hg in breast milk.
Discussion
The present study clearly demonstrates that infant exposure to MeHg and I-Hg via breast-feeding is low compared with late fetal exposure. For both MeHg and I-Hg, the concentrations in maternal blood and cord blood at delivery were highly associated, which is in accordance with previous studies (Stern and Smith 2003; Vahter et al. 2000). Because the women of the present study had few amalgam fillings and a limited consumption of fish, especially fish species potentially high in MeHg, they also had low mercury exposures. Apparently, their compliance with the recommendation that pregnant and breast-feeding women avoid consumption of certain fish species (Livsmedelsverket 2004) was good.
Exposure to MeHg and I-Hg in breast-fed infants is largely unknown. Animal studies suggest that both MeHg and I-Hg are transported from plasma to breast milk bound to serum albumin, with I-Hg also bound to casein (Sundberg et al. 1999). Besides the binding to albumin and casein in breast milk, the transport is also affected by the distribution in maternal blood. About 65% of I-Hg but only about 10% of MeHg in whole blood is present in plasma (Kershaw et al. 1980) and thus available for transport to breast milk.
We found no significant association between MeHg in maternal blood and mercury in breast milk, which is in contrast to a recent Japanese study (Sakamoto et al. 2002). Probably this was due to the lower MeHg exposure and higher I-Hg exposure (as Hg0 via amalgam fillings) in the present study. Unfortunately, we were not able to speciate mercury in breast milk. Based on the reported milk-to-plasma ratios of about 0.2 for MeHg (Sakamoto et al. 2002; Sundberg et al. 1998) and 0.6–1.0 for I-Hg (Oskarsson et al. 1996; Sundberg et al. 1998), it seems that I-Hg is more easily transported to breast milk than is MeHg. Also, the association between I-Hg in maternal blood and mercury in breast milk indicated transport of I-Hg into milk. The absorption of Hg2+ in the gastrointestinal tract is known to be low (Sandborgh-Englund et al. in press; WHO 1991), but it has been suggested to be higher in infants than in adults (Clarkson 1992). However, we found that infant blood I-Hg concentrations declined from birth to 13 weeks of age (Figure 3) and there was no association between mercury in breast milk and I-Hg in infant blood. Thus, our result raises the question as to what extent I-Hg in milk is absorbed from the gastrointestinal tract in the infant.
On the other hand, we found that mercury in breast milk correlated with MeHg in infant blood at 13 weeks. This might be explained by the fact that the small amount of MeHg that passes from maternal plasma to breast milk is almost completely taken up by the infant [about 95% is absorbed in the gastrointestinal tract (WHO 1990)]. In this way, there is a small exposure of the breast-fed infant to MeHg via breast milk. Still, infant blood MeHg decreased markedly until 13 weeks of age (Figure 1). The decline, which is in line with previous findings (Sakamoto et al. 2002; Sandborgh-Englund et al. 2001), can be explained partly by the rapid increase in body weight, and partly by a decrease in hematocrit (Ciba-Geigy 1984) due to the exchange from fetal to adult hemoglobin. It is generally believed, based on results from animal studies, that excretion of MeHg (which takes place mainly via feces) by the breast-fed newborn is limited, because demethylating bacteria in the gastrointestinal tract become first established after weaning (Rowland et al. 1983). Unless it is demethylated, MeHg is reabsorbed via the enterohepatic circulation. With limited excretion of MeHg in the infant, one would not expect a marked decline in blood MeHg concentrations during the first 3 months of life as found in the present study. Studies of the total postnatal excretion of MeHg are needed to confirm that the prenatal accumulated MeHg actually may be excreted during the first postnatal weeks.
Both volume and composition of breast milk change over time, during a feeding session, during the course of the day, and during the breast-feeding period (Mitoulas et al. 2002). In the present study we found T-Hg levels approximately twice as high in colostrum as in mature breast milk. Probably the higher concentration of albumin in colostrum compared with that in mature milk (Neville et al. 1983) enables transport of both forms of mercury. However, the increasing volume of breast milk during the breast-feeding period (Kunz et al. 1999) contributes to a dilution of the concentrations of mercury in mature milk. Interestingly, we also found an increase in mercury concentrations in breast milk during the same feeding, which could be explained by an increase in casein concentrations during the feeding (Hytten 1954).
Given the findings of the present study, we conclude that the child’s exposure to MeHg and I-Hg is much higher before birth than during breast-feeding. Exposure of infants to MeHg may also occur postnatally through breast milk, depending on maternal plasma MeHg concentrations. This underlines the importance of dietary recommendations to pregnant and breast-feeding women not to eat MeHg-contaminated fish. However, because the present study was carried out in a small group with low exposure to both MeHg and I-Hg, further studies of the different forms of mercury in breast milk and infant uptake are warranted in populations with higher exposure (Barbosa and Dórea 1998).
We express our gratitude to the participating women and to B. Palm for skillful laboratory assistance. We also acknowledge E. Berg for support with statistical analysis and J. Holmén for editing the manuscript.
The study was performed by financial support from the Swedish Medical Research Council (grant 13450-02).
Figure 1 The concentrations of MeHg in maternal and infant blood at delivery and 4 days and 13 weeks postpartum (n = 20). Box length illustrates the interquartile range; line is the median. Circles are cases with values 1.5–3 box lengths from the upper or lower edge of the box, and asterisks are cases with values > 3 box lengths from the upper or lower edge of the box.
Figure 2 The associations between concentrations in cord blood and maternal blood for MeHg (r = 0.95; p < 0.001) and I-Hg (rS = 0.77; p < 0.001).
Figure 3 The concentrations of I-Hg in maternal and infant blood at delivery and 4 days and 13 weeks postpartum (n = 20). Box length illustrates the interquartile range; line is the median. Circles are cases with values 1.5–3 box lengths from the upper or lower edge of the box, and asterisks are cases with values > 3 box lengths from the upper or lower edge of the box.
Figure 4 The associations between T-Hg in breast milk and I-Hg in maternal blood (A; rS = 0.61; p = 0.006) and infant blood (B; rS = 0.17; p = 0.50) at 13 weeks after birth.
Figure 5 The associations between T-Hg in breast milk and MeHg in maternal blood (A; rS = 0.26; p = 0.28) and infant blood (B; rS = 0.55; p = 0.01) at 13 weeks after birth.
Figure 6 The concentrations of T-Hg in breast milk at 4 days (colostrum; n = 19), 6 weeks (n = 20), and 13 weeks postpartum (n = 19). Box length illustrates the interquartile range; line is the median. Circles are cases with values 1.5–3 box lengths from the upper or lower edge of the box, and asterisks are cases with values > 3 box lengths from the upper or lower edge of the box.
Table 1 Results from analytical quality control of T-Hg and I-Hg in blood and T-Hg in breast milk (μg/L).
Quality control Species Recommended value Obtained value (n = 6) Coefficient of variation (%)
Breast milk
SRM-1549 T-Hg 0.3 ± 0.2 0.3 ± 0.1 25
Blood
Seronorm T-Hg 2.2–3.3 2.4 ± 0.2 8.6
404107 I-Hg — 0.55 ± 0.03 5.0
Seronorm T-Hg 6.7–8.4 8.0 ± 0.3 3.1
404108 I-Hg — 6.5 ± 0.2 2.6
Abbreviations: —, no data; SRM, standard reference material (National Institute of Standards and Technology, Gaithersburg, MD, USA).
Table 2 Subject characteristics (n = 20).
Characteristic Mean Minimum Maximum
Age (years) 31 24 37
Body mass index 26 21 36
Weight gain during pregnancy (kg) 15 –1 24
Parity (n) 1 0 3
Gestational length (weeks) 39 35 43
Fish intake, pregnancy (times/month) 2.0 0 8
Fish intake, breast-feeding (times/month) 2.4 0 8
Amalgam-filled surfaces (n) 5 0 24
Infant birth weight (g) 3,666 2,720 4,505
==== Refs
References
Anderson HA Wolff MS 2000 Environmental contaminants in human milk J Expo Anal Care Environ Epidemiol 10 755 760
Ask K Åkesson A Berglund M Vahter M 2002 Inorganic mercury and methylmercury in placentas of Swedish women Environ Health Perspect 110 523 526 12003757
Barbosa AC Dorea JG 1998 Indices of mercury contamination during breast feeding in the Amazon basin Environ Toxicol Pharmacol 6 71 79 21781883
Ciba-Geigy 1984. Geigy Scientific Tables, Vol 3. Physical Chemistry, Composition of Blood, Hematology, Somatometric Data. Basel, Switzerland:Ciba-Geigy.
Clarkson T 1992. The uptake and disposition of inhaled mercury vapor. In: Potential Biological Consequences of Mercury Released from Dental Amalgam, Proceedings from a Conference (Bergman B, Boström H, Larsson K, Löe H, eds). Stockholm:Swedish Medical Research Council, 59–75.
Dorea JG 2004 Mercury and lead during breast-feeding Br J Nutr 92 21 40 15230985
Drexler H Schaller KH 1998 The mercury concentration in breast milk resulting from amalgam fillings and dietary habits Environ Res 77 124 129 9600805
Gartner LM Black LS Eaton AP Lawrence RA Naylor AJ Neifert ME 1997 Breastfeeding and the use of human milk Pediatrics 100 1035 1039 9411381
Grandjean P Jorgensen PJ Weihe P 1994 Human milk as a source of methylmercury exposure in infants Environ Health Perspect 102 74 77 9719671
Hytten FE 1954 Clinical and chemical studies in human lactation Br Med J 23 175 182 13106514
Kajiwara Y Yasutake A Adachi T Hirayama K 1996 Methyl-mercury transport across the placenta via neutral amino acid carrier Arch Toxicol 70 310 314 8852703
Kershaw TG Clarkson TW Dhahir PH 1980 The relationship between blood levels and dose of methylmercury in man Arch Environ Health 35 28 36 7189107
Kunz C Rodriguez-Palermo M Korletzko B Jensen R 1999 Nutritional and biochemical properties of human milk, part I: general aspects, proteins and carbohydrates Clin Perinatol 26 307 333 10394490
Livsmedelsverket (Swedish National Food Administration) 2004. Kostråd om Fiskkonsumtion [in Swedish] (Dietary recommendations concerning fish consumption). Uppsala, Sweden:Livsmedelsverket. Available: http://www.slv.se/templates/SLV_Page____11374.aspx [accessed 1 December 2004].
Mitoulas LR Kent JC Cox DB Owens RA Sherriff JL Hartmann PE 2002 Variation in fat, lactose and protein in human milk over 24 h and throughout the first year of lactation Br J Nutr 88 29 37 12117425
Neville MC, Allen JC, Watters C 1983. The mechanism of milk secretion. In: Lactation, Physiology, Nutrition, and Breast-feeding (Neville MC, Neifert MR, eds). New York:Plenum Press, 49–102.
Nordenhall K Dock L Vahter M 1998 Cross-fostering study of methyl mercury retention, demethylation and excretion in the neonatal hamster Pharmacol Toxicol 82 132 136 9553991
NRC (National Research Council) 2000. Toxicological Effects of Methylmercury. Washington DC:National Academy Press.
Oddy WH 2002 The impact of breastmilk on infant and child health Breastfeed Rev 10 5 18 12592775
Oskarsson A Schutz A Skerfving S Hallen IP Lagerkvist BJ 1996 Total and inorganic mercury in breast milk and blood in relation to fish consumption and amalgam fillings in lactating women Arch Environ Health 51 234 241 8687245
Pronczuk J Moy G Vallenas C 2004 Breast milk: an optimal food Environ Health Perspect 112 A722 A723 15345351
Rice D Barone SJ 2000 Critical periods of vulnerability for the developing nervous system: evidence from humans and animal models Environ Health Perspect 108 suppl 3 511 533 10852851
Rowland I Robinson R Doherty R Landry T 1983. Are developmental changes in methylmercury metabolism and excretion mediated by the intestinal microflora. In: Reproductive and Developmental Toxicity of Metals (Clarkson TW, Nordberg GF, Sager PR, eds). New York:Plenum Press, 745–758.
Sakamoto M Kubota M Liu XJ Murata K Nakai K Satoh H 2004 Maternal and fetal mercury and n-3 polyunsaturated fatty acids as a risk and benefit of fish consumption to fetus Environ Sci Technol 38 3860 3863 15298193
Sakamoto M Kubota M Matsumoto S Nakano A Akagi H 2002 Declining risk of methylmercury exposure to infants during lactation Environ Res 90 185 189 12477463
Sandborgh-Englund G Ask K Belfrage E Ekstrand J 2001 Mercury exposure in utero and during infancy J Toxicol Environ Health A 63 317 320 11471864
Sandborgh-Englund G Bjorkhem I Bjorkman L Valtersson C 1998 Determination of low levels of total mercury in blood and plasma by cold vapour atomic fluorescence spectrometry Scand J Clin Lab Invest 58 155 160 9587168
Sandborgh-Englund G Einarsson C Sandström M Ekstrand J In press. Gastrointestinal absorption of metallic mercury. Arch Environ Health.
Skerfving S 1988 Mercury in women exposed to methylmercury through fish consumption, and in their newborn babies and breast-milk Bull Environ Contam Toxicol 41 475 482 3224165
Stern AH Smith AE 2003 An assessment of the cord blood:maternal blood methylmercury ratio: implications for risk assessment Environ Health Perspect 111 1465 1470 12948885
Sundberg J Ersson B Lonnerdal B Oskarsson A 1999 Protein binding of mercury in milk and plasma from mice and man—a comparison between methylmercury and inorganic mercury Toxicology 137 169 184 10522497
Sundberg J Jonsson S Karlsson MO Hallen IP Oskarsson A 1998 Kinetics of methylmercury and inorganic mercury in lactating and nonlactating mice Toxicol Appl Pharmacol 151 319 329 9707508
Sundberg J Oskarsson A Albanus L 1991 Methylmercury exposure during lactation—milk concentration and tissue uptake of mercury in the neonatal rat Bull Environ Contam Toxicol 46 255 262 2018872
Vahter M Åkesson A Lind B Björs U Schütz A Berglund M 2000 Longitudinal study of methylmercury and inorganic mercury in blood and urine of pregnant and lactating women, as well as in umbilical cord blood Environ Res 84 186 194 11068932
WHO 1990. Methylmercury. Environmental Health Criteria 101. Geneva:International Programme on Chemical Safety, World Health Organization.
WHO 1991. Inorganic Mercury. Environmental Health Criteria 118. Geneva:International Programme on Chemical Safety, World Health Organization.
|
16203251
|
PMC1281284
|
CC0
|
2021-01-04 23:41:29
|
no
|
Environ Health Perspect. 2005 Oct 15; 113(10):1381-1385
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.7856
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences 10.1289/ehp.7401ehp0113-00138616203252ResearchChildren's HealthIncreased Levels of 8-Hydroxy-2′-Deoxyguanosine Attributable to Carcinogenic Metal Exposure among Schoolchildren Wong Ruey-Hong Kuo Chung-Yih Hsu Ming-Lin Wang Tsun-Yen Chang Pi-I Wu Tsung-Hsun Huang Shuai Department of Public Health, College of Health Care and Management, Chung Shan Medical University, Taichung, TaiwanAddress correspondence to R.-H. Wong, Department of Public Health, College of Health Care and Management, Chung Shan Medical University, 110 Chien-Kuo North Rd., Section 1, Taichung, Taiwan 40242. Telphone: 886-4-24730022, ext. 11792. Fax: 886-4-23248179. E-mail:
[email protected] authors declare they have no competing financial interests.
10 2005 27 5 2005 113 10 1386 1390 9 7 2004 26 5 2005 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright. Arsenic, chromium, and nickel are reported in several epidemiologic studies to be associated with lung cancer. However, the health effects of arsenic, chromium, and nickel exposures are equivocal for children. Therefore, we performed a cross-sectional study to investigate possible associations between the internal concentrations of arsenic, chromium, and nickel and the level of oxidative stress to DNA in children. We measured urinary levels of arsenic, chromium, and nickel for 142 nonsmoking children using atomic absorption spectrometry. As a biomarker for oxidative stress, urinary 8-hydroxy-2′-deoxyguanosine (8-OHdG) levels were analyzed with an enzyme-linked immunosorbent assay kit. The median urinary 8-OHdG level for our subjects was 11.7 ng/mg creatinine. No obvious relationship between the levels of urinary nickel and 8-OHdG was found. Multiple linear regression analysis showed that children with higher urinary chromium had greater urinary 8-OHdG than did those with lower urinary chromium. Similarly, subjects with higher urinary arsenic had greater urinary 8-OHdG than did those with lower urinary arsenic. Furthermore, children with both high urinary arsenic and high urinary chromium had the highest 8-OHdG levels (mean ± SE, 16.0 ± 1.3; vs. low arsenic/low chromium, p < 0.01) in urine, followed by those with low arsenic/high chromium (13.7 ± 1.6; vs. low arsenic/low chromium, p = 0.25), high arsenic/low chromium (12.9 ± 1.6 vs. low arsenic/low chromium, p = 0.52), and low arsenic/low chromium (11.5 ± 1.3); the trend was significant (p < 0.001). Thus, environmental carcinogenic metal exposure to chromium and arsenic may play an important role in oxidative DNA damage to children.
arsenicchildrenchromium8-hydroxy-2′-deoxyguanosinenickel
==== Body
Potential hazardous pollutants from industrial sources such as thermal power plants are often emitted to our living environments, where they possibly expose adults and children to heavy metals through inhalation and ingestion of contaminated soil and dust. Earlier literature pointed out that increased use of coal for power production will lead to increased release of metals into the environment (Sabbioni et al. 1984). Furthermore, high quantities of arsenic, chromium, and nickel are detected in milled coal and ash of coal-fired power plants (Goodarzi and Huggins 2001), and these three metals are also reportedly associated with human lung cancers in several occupational epidemiologic studies (Chen and Chen 2002; Droste et al. 1999; Grimsrud et al. 2002). However, the health effects of arsenic, chromium, and nickel exposure are especially equivocal for children. Children are considered to be a population susceptible to adverse health effects induced by air pollutants (Nicolai 1999). Previous studies of metal exposure to children in Taiwan focused primarily on lead in occupational sites (Wang et al. 2002) and Chinese herbal medicine (Cheng et al. 1998), whereas effects of other environmental contaminants such as arsenic, chromium, and nickel on children’s health have been largely ignored.
The carcinogenic potential of arsenic, chromium, and nickel compounds is well established for humans and experimental animals (Hayes 1997). However, the molecular damage formation after exposure to metals is still not well understood. One mechanism proposed frequently is an increase in oxidative DNA lesions attributable to metal exposure, mediated by increased generation of highly reactive oxygen species (ROS) (Dally and Hartwig 1997; Kasprzak 1991; Kasprzak et al. 1997). Oxidative DNA lesions are supposed to play important roles in various diseases including cancer and premature aging (Beckman and Ames 1998; Cerutti 1994; Grisham 1994; Jenner 1994; Witztum 1994). Among the diverse oxidative DNA lesions, 8-hydroxy-2′-deoxyguanosine (8-OHdG) is one of the most abundant base modifications and has attracted special attention because it is premutagenic, causing G-to-T transversions (Cheng et al. 1992); thus, the presence of 8-OHdG may lead to mutagenesis. Moreover, the repair process for 8-OHdG–inflicted damage results in excised 8-OHdG adduct being excreted into the urine (Marnett 2000; Shigenaga et al. 1989). Because of easy collection, urinary 8-OHdG is thus regarded as a suitable biomarker of oxidative stress (Toraason 1999; Wong et al. 2003).
To investigate possible associations between the incorporated internal concentrations of arsenic, chromium, and nickel and the level of oxidative stress in schoolchildren, we performed a cross-sectional study in Taiwan.
Materials and Methods
Study areas and subject selection.
The study subjects were nonsmoking fifth-grade pupils (10–12 years of age) who in January 2003 were attending three elementary schools from three different towns of Taichung County, Taiwan. Each selected primary school was close to, and within 1 km of, an ambient air quality monitoring station. Of these schools, Longgang elementary school is also adjacent to the Taichung Thermal Power Plant on the southern side of Taichung harbor, with eight coal-fired generation units in operation to accommodate a total installed capacity of 4,688 MW. The coal consumption of this power plant is approximately 1.28 million tons per year. Coal is stored at and delivered from three coal yards (storage capacity, 310, 0.52, and 0.42 million tons, respectively) near the power plant. Thus, it is possible that coal particles are emitted into the atmosphere from these three coal yards. Additionally, there were no major roads or factories within the Longgang elementary school district. The remaining schools, Shalach and Shuntian, are located in suburban communities and are on the northeastern upwind side approximately 8 and 18 km of the Taichung Thermal Power Plant, respectively. In addition, these two school districts are intersected by major trunk roads.
All subjects who participated in the medical surveillance process underwent physical examinations conducted by a qualified pediatrician. All participating schoolchildren voluntarily entered the study after informed written consent was obtained from the children’s parents. Parents of schoolchildren completed the questionnaire before collecting children’s urine samples. The questionnaire was divided into the following parts: demographic data of the children, respiratory symptoms and diseases of the children, smoking habits and occupation of the parents, and possible sources of indoor air pollution such as household smoking, pet feeding, incense burning all day, and home dampness. According to the responses to the questionnaires, asthma was assessed by the question “has your child had wheezing in the chest accompanied by dyspnea that had ever been given the diagnosis of asthma by a physician during the past 12 months?” Similarly, children who had been diagnosed as having allergic rhinitis by a physician were considered to have a history of allergic rhinitis. Of these study subjects, 16 schoolchildren were excluded from the study because of incomplete urine samples or residence far from the school. Finally, a total of 142 subjects without any disease history except respiratory diseases participated in this study.
Urine collection.
In our study period, spot morning urine samples were collected in polypropylene specimen containers. The decision to use first morning voids rather than 24-hr collections was based on the report by Thompson et al. (1999), which indicated that 24-hr average urinary levels were not statistically different in values from first voids. Moreover, it is difficult to collect 24-hr urine samples and first voids of morning urine from every subject. Finally, most morning urine specimens we obtained were first spot and small numbers of them were second spot specimens. Immediately after the collection, urine samples were stored at −20°C until used for analysis.
Measurement of metal levels in urine.
Metal levels in urine samples including arsenic, chromium, and nickel were measured using atomic absorption spectrometry with a graphite furnace (model 4110ZL; Perkin-Elmer, Norwalk, CT, USA) technique and Zeeman background correction. All analytical glassware and plasticware purchased were of low-metal grade and were further cleaned with diluted nitric acid before use. Initially, all the frozen samples were thawed and aliquotted at room temperature. A solution of Triton X-100 (0.1%, wt/vol) was prepared in nitric acid (0.2%, vol/vol). Subsequently, the 18-mL urine samples were diluted with 2 mL of prepared Triton X-100 solution (9:1) and stored at −20°C until required for analysis. Urine test portion and aqueous standards were injected at 20 μL using the autosamplers in the furnace. We used a mixture containing palladium plus magnesium nitrate, and magnesium nitrate as chemical modifiers for the determination of arsenic and chromium in urine, respectively. No matrix modifier was used for the determination of nickel. For analyses of urinary metals, we checked the accuracy of the instrumental methods and the analytical procedure by using reference solutions (standard reference material 12111, normal-range metals urine toxicology control; UTAK Laboratories, Valencia, CA, USA), which were run before every batch of samples. The correlation coefficients for each of the values of the standard curves were all > 0.990. The mean recovery rates ranged from 90 to 105%, and coefficients of variation for reproducibility were all < 10%. Metal concentrations in urine were corrected for each individual according to their urinary creatinine values, and urinary samples were analyzed blind to the status of the individuals for the presence of metals.
Determination of urinary 8-OHdG levels.
Before examination, urine samples were centrifuged at 2,000 × g for 10 min to remove any suspended cell debris. The supernatants were used for the determination of 8-OHdG levels using a competitive enzyme-linked immunosorbent assay kit (ELISA; Japan Institute for the Control of Aging, Fukuroi, Japan). The determination range was 0.5–200 ng/mL. The 8-OHdG monoclonal antibody N45.1 and urine sample were loaded at 50 μL onto a microtiter plate that was coated with 8-OHdG, and incubated at 37°C for 60 min, in accordance with the instructions of the manufacturer. After the wells were washed three times, the antibodies that remained bound to the 8-OHdG in the sample were further bound with the horseradish peroxidase–conjugated secondary antibody, followed by incubation at 37°C for 60 min. The wells were again washed three times. Subsequently, a substrate containing 3,3′,5,5′-tetramethylbenzidine was added, and the wells were incubated at room temperature for 15 min, resulting in the development of color intensity proportional to the amount of antibody bound to the plate. The color reaction was terminated by the addition of stop solution (1 M phosphoric acid), and the absorbance was measured using a computer-controlled spectrophotometric plate reader at a wavelength of 450 nm. The concentration of 8-OHdG in the test samples was interpolated from a standard curve drawn with the assistance of logarithmic transformation. Urinary 8-OHdG levels were subsequently adjusted by urinary creatinine levels.
Statistical analysis.
Because of the positively skewed distribution of the urinary arsenic, chromium, and nickel levels, we used nonparametric testing to test the differences of urinary metal levels among our study children at three different elementary schools. Similarly, because of the positively skewed distribution of the urinary 8-OHdG levels, we used nonparametric testing to test the differences of urinary 8-OHdG level for each variable. Median values of urinary arsenic, chromium, and nickel levels were used as cutoff. Subsequently, we developed a multiple linear regression analysis to adjust for significant covariate identified in the univariate analysis to evaluate potential differences in urinary 8-OHdG. We also computed regression coefficients and their SEs and calculated least-square means to predict the adjusted 8-OHdG levels for children with different urinary metal contents.
Results
In total, 142 children (74 boys and 68 girls) were involved in this study. Their ages ranged from 10 to 12 years (mean age, 11.2 years). All study subjects lived within the limits of the Taichung harbor area and lived near their schools. More than half of the parents of study subjects had achieved greater than a senior high school education (58.5% for fathers, 51.4% for mothers). Almost half (49.3%) the parents were smokers. Half the parents were industrial workers. Possible sources of indoor air pollution such as pet feeding were reported in 19.0% of children’s homes, incense burning all day was reported in 40.8% of homes, and home dampness was reported in 20.4% of homes. In addition, prevalences of asthma and allergic rhinitis among study participants were 7.0 and 31.0%, respectively.
The median urinary metal levels were 6.4 μg/L for arsenic, 1.9 μg/L for chromium, and 3.4 μg/L for nickel (Table 1). The creatinine-adjusted median levels were 7.7 μg/g for arsenic, 2.0 μg/g for chromium, and 4.1 μg/g for nickel. Among the 142 urine samples, 19 urine samples were below the detection limit of 0.1 μg/L urine for arsenic, 19 samples were below the detection limit of 0.7 μg/L for chromium, and 16 samples were below the detection limit of 0.8 μg/L for nickel. Especially, the study children at Longgang elementary school had significantly higher urinary levels of arsenic, chromium, and nickel than did those at Shalach and Shuntian elementary schools (p-values ≤ 0.01, Kruskal-Wallis test).
Median urinary 8-OHdG level for the study subjects was 11.7 ng/mg creatinine (range, 0.4–59.7 ng/mg creatinine) (Table 2). Children at Longgang elementary school had higher 8-OHdG levels than did those at Shalach and Shuntian elementary schools (p < 0.01, Kruskal-Wallis test). Children of mothers who had greater than a senior high school education had significantly lower 8-OHdG levels than did children of mothers who had less than a senior high school education (p = 0.04, Wilcoxon rank sum test). Children whose parents smoked at home also had significantly higher 8-OHdG levels than did children whose parents did not smoke at home (p = 0.07). Children with allergic rhinitis had significantly lower 8-OHdG levels than did those without allergic rhinitis (p = 0.02). However, sex (p = 0.94), paternal education (p = 0.70), parental occupation (p = 0.11), pet feeding (p = 0.92), incense burning at home all day (p = 0.18), home dampness (p = 0.40), and asthma history (p = 0.51) were not associated with increased urinary 8-OHdG levels.
Children with urinary arsenic levels > 7.7 μg/g creatinine (median, n = 71) had higher 8-OHdG levels than did those with urinary arsenic levels < 7.7 μg/g creatinine (n = 71, p = 0.09). Children with urinary chromium > 2.2 μg/g creatinine (median, n = 71) also had higher 8-OHdG levels than did those with urinary chromium levels < 2.2 μg/g creatinine (n = 71, p = 0.06). However, children with urinary nickel levels > 4.1 μg/g creatinine (median, n = 71) did not have higher 8-OHdG levels than those with urinary nickel levels < 4.1 μg/g creatinine (n = 71, p = 0.62).
Our univariate analysis showed that only elementary school, maternal education, parental smoking status, history of allergic rhinitis, and urinary creatinine-adjusted concentrations of arsenic and chromium were obviously associated with elevated levels of urinary 8-OHdG (p-values < 0.10). Therefore, we performed a multiple linear regression model for urinary 8-OHdG level as a function of maternal education, parental smoking status, history of allergic rhinitis, and urinary creatinine-adjusted concentrations of arsenic and chromium [general linear model (GLM); Table 3]. We excluded elementary school in this model because the variables of elementary school and urinary arsenic and chromium levels had high colinearity. Urinary 8-OHdG level was positively associated with maternal educational level below senior high school (p = 0.05) and was negatively associated with history of allergic rhinitis (p = 0.05). However, parental smoking status (p = 0.47) appeared not to influence the concentrations of urinary 8-OHdG for individuals when examining the data using GLM analysis. Interestingly, a mean difference of 1.9 ng/mg creatinine for urinary 8-OHdG was noted for children with high urinary arsenic levels compared with those with low urinary arsenic levels (p = 0.18). Children with high urinary chromium levels also had a mean difference of 3.0 ng/mg creatinine for urinary 8-OHdG compared with those with low urinary chromium levels (p = 0.04). Furthermore, in this model, the partial R2 value was 17.5% for urinary arsenic and 21.5% for urinary chromium.
Subsequently, we performed a least-squares mean analysis to assess the urinary 8-OHdG levels of children with combination of urinary arsenic and chromium adjusted for maternal education and history of allergic rhinitis. Children with both low urinary arsenic and low urinary chromium levels had the lowest urinary 8-OHdG mean levels of 11.4 ng/mg creatinine (n = 44; Figure 1), whereas those with both high urinary arsenic and high urinary chromium levels had the highest urinary 8-OHdG mean levels of 16.2 ng/mg creatinine (n = 43; vs. low arsenic/low chromium, p < 0.01). Those with both low urinary arsenic and high urinary chromium levels (13.7 ng/mg creatinine, n = 28; vs. low arsenic/low chromium, p = 0.25) and those with both high urinary arsenic and low urinary chromium levels (12.7 ng/mg creatinine, n = 27; vs. low arsenic/low chromium, p = 0.52) had a moderately increased 8-OHdG mean levels. This trend in urinary 8-OHdG levels was statistically significant (p = 0.01, GLM). Furthermore, the difference in urinary 8-OHdG levels between the combination high urinary arsenic and chromium and combination low urinary arsenic and chromium (4.8 ng/mg creatinine) was greater than the summation of differences of urinary 8-OHdG levels between the combination low urinary arsenic and high urinary chromium and combination low urinary arsenic and low urinary chromium (2.3 ng/mg creatinine), and combined high urinary arsenic and low urinary chromium with combined low urinary arsenic and low urinary chromium (1.3 ng/mg creatinine).
Discussion
Attacks on DNA by ROS frequently result in oxidative DNA damage. 8-OHdG is a modified base that occurs in DNA because of attack by hydroxyl radicals. Because it is premutagenic, causing G-to-T transversions (Cheng et al. 1992), the presence of 8-OHdG may lead to mutagenesis. The possibility of 8-OHdG arising from oxidation of deoxyguanosine has been also proposed (Shigenaga et al. 1989), and the result of this deoxyguanosine oxidation does not occur in DNA, so this 8-OHdG has no mutagenic potential. Thus, urinary 8-OHdG is commonly considered a biomarker of oxidative stress, reflecting its repair from DNA. Nonetheless, urinary 8-OHdG has not been used in previous studies to detect the effects of environmental carcinogenic metal exposure in children. In our study, the median value (11.7 ng/mg creatinine) of urinary 8-OHdG for our participants was similar to that of a previous study for normal English children (10.0 ng/mg creatinine) (Drury et al. 1998).
A major fraction of arsenic, chromium, and nickel absorbed by humans appears to be eliminated relatively quickly and mainly via urine. The biologic half-life for these metals has been estimated to be between 1 and 3 days (Hwang et al. 1997; Paustenbach et al. 1997; Sunderman 1993; Vahter 1994). Thus, these metal concentrations in urine samples are determined as important short-term exposure biomarkers and have been used in many epidemiologic studies (Moore et al. 1997; Smith-Sivertsen et al. 1998; Stern et al. 1998). In the present study, we found statistically significant relationships between the urinary concentrations of chromium and arsenic and the level of DNA oxidative stress. The lack of correlation between exposure to nickel and DNA oxidative stress could be attributable to a low biologically relevant dose in the study population. Several studies have demonstrated that arsenic and chromium cause oxidative DNA damage to cultured cells (Kessel et al. 2002; Yuann et al. 1999). Previous epidemiologic studies have also shown increased 8-OHdG levels in humans exposed to arsenic (Matsui et al. 1999) or chromium (Kuo et al. 2003). Transition metals are commonly thought to produce ROS such as hydroxyl radicals that can directly damage cellular DNA. The other mechanism is indirect oxidative DNA damage due to inflammation caused by metal exposure (Donaldson et al. 2002). Some metals may stimulate the defense systems of the body so that they react against the toxic damage to produce cytokines (Carter et al. 1997; Donaldson et al. 2002). Several cytokines cause production of large amounts of ROS. Some propose that ROS generated in inflamed tissues can cause injury to target cells and also damage DNA, which contributes to carcinogenic processes (Chazotte-Aubert et al. 1999; Eiserich et al. 1998).
Environmentally relevant metals seldom occur alone. Little is known about the exact mechanism of carcinogenesis of two or more metals when they are present together. It is generally assumed that the concept of additivity is operative on low-level exposures to chemical mixtures (Hartwig and Schwerdtle 2002). It is particularly interesting that we observed a synergistic effect for combined arsenic and chromium exposure on DNA oxidative stress in the present study (Figure 1). It may be that other types of cellular damage are caused by metal exposure, which also contributes to their carcinogenic potentials. There is accumulating evidence that metals including arsenic and chromium can interfere with distinct steps of diverse DNA repair systems (Hartwig and Schwerdtle 2002). Thus, oxidative DNA lesions are not only induced by metals at biologically relevant concentrations, but their extent may also be enhanced indirectly by impaired repair. Further studies are required to clarify these findings.
The amount of the modified base in cellular DNA excreted into urine should represent the average rate of DNA damage in the whole body (Cooke et al. 2000). Thus, it is possible that the levels of oxidative DNA damage are reflective of different active diseases, especially active inflammation (Wong et al. 2003). Besides, urinary levels of any oxidative lesion rely on efficient renal excretion of the damage products, so renal impairment can therefore affect urinary 8-OHdG levels (Akagi et al. 2003). In our study, urinary creatinine levels were used to correct for variation in urine concentration. In addition, no medical histories were reported by our participants except asthma and allergic rhinitis. There is ample evidence indicating that allergic disorders, such as asthma and rhinitis, are mediated by oxidative stress (Bowler and Crapo 2002). In our study, we did not observe a significant association between asthma and urinary 8-OHdG level in children. On the contrary, children with allergic rhinitis had significantly lower 8-OHdG than did those without. One could interpret this finding as an effect rather than a cause; that is, children with past allergies or past episodes of respiratory symptoms have had previous medical care, and their parents may have been urged to improve their environment to alleviate the symptoms. We also found that maternal education level, used as a proxy for socioeconomic status, was significantly related to children’s oxidative DNA damage. Maternal education may convey information that influences the patterns of potential metal exposure as well as health care for children. In addition, we also observed that children whose parents smoked had higher 8-OHdG expression, although it was not significant in our multiple linear regression model. Because cigarette smoke contains ROS, the association between cigarette smoking and urinary 8-OHdG has been previously reported (Loft and Poulsen 1996). However, the association between 8-OHdG and children exposed to environmental tobacco smoke has not been investigated. Oxidative damage occurs rapidly after exposure, and this damage can be repaired rapidly. Because Taiwanese children may not have regular chances to be exposed to tobacco smoke from their family members, the effects of passive smoking on 8-OHdG in children are less likely to appear in the model.
Children spend their most of their time indoors. It is therefore important to consider the effects that exposure to indoor air pollutants may have on children’s oxidative DNA damage. House dust and fungi are the major indoor pollutants in our subtropical area (Li et al. 1994; Wang et al. 1999). In addition, burning Chinese incense releases polycyclic aromatic hydrocarbons (Lung et al. 2003), which may increase cellular oxidative stress (Wu et al. 2003). Our data provide no evidence for an association between the levels of indoor environmental factors such as pet feeding, incense burning, and home dampness. In our study, these indicators were self-reported and therefore were subjective and could have resulted in misclassification of exposure that might reduce the observed associations.
When the balance between pro-oxidant and antioxidant processes is shifted in favor of the former, increased 8-OHdG would be generated from further DNA oxidation and ring opening followed by rearrangements (Cadet et al. 2003). However, the role of pro-oxidant and antioxidant on 8-OHdG in our study children was not identified. In addition, the 8-OHdG monoclonal antibody used in our ELISA assay has similar binding affinity for the oxidized free base 8-hydroxyguanine, and the oxidized nucleoside 8-hydroxyguanosine (Yin et al. 1995). Thus, the possibility of overestimation of urinary 8-OHdG levels by our ELISA assay cannot be ruled out, and this bias would be likely to attenuate the observed association if it was nondifferential.
In our study, children’s exposure to arsenic and chromium was associated with increased generation of subsequent 8-OHdG. However, the role of other carcinogenic metals on oxidative DNA damage also requires further study. For the future, a longitudinal rather than a cross-sectional study should be conducted to ascertain the possible association between carcinogenic metal exposure and oxidative DNA lesions. A longitudinal study that includes a relevant number of environmentally exposed participants offers an advantage for studying dose–effect relationships over time with repeated measurements.
This study was supported by the Bureau of Environmental Protection of Taichung County, Taiwan, Republic of China.
Figure 1 Adjusted urinary 8-OHdG level (ng/mg creatinine) by urinary arsenic and urinary chromium concentrations. Values shown are mean ± SE. Cut points were determined according to medians (arsenic, 7.7 μg/g creatinine; chromium, 2.0 μg/g creatinine) of urinary creatinine-adjusted levels among all subjects.
Table 1 Concentrations of arsenic, chromium, and nickel in the study population.
Elementary school
All (n = 142)
Variable Longgang (n = 49) Shalach (n = 45) Shuntian (n = 48) Mean ± SE Geometric mean Median Maximum
Arsenic (μg/L urine) 16.2 ± 2.6** 8.7 ± 0.9 3.7 ± 0.6 9.1 ± 1.1 3.9 6.4 86.4
Arsenic (μg/g urinary creatinine) 21.3 ± 4.3** 10.0 ± 1.2 4.7 ± 0.7 12.1 ± 1.6 5.0 7.7 149.3
Chromium (μg/L urine) 2.6 ± 0.2** 1.9 ± 0.1 1.3 ± 0.1 1.9 ± 0.1 1.7 1.9 6.6
Chromium (μg/g urinary creatinine) 3.7 ± 0.5** 2.2 ± 0.2 2.0 ± 0.2 2.7 ± 0.2 2.2 2.0 21.4
Nickel (μg/L urine) 5.4 ± 0.5* 3.9 ± 0.3 2.8 ± 0.2 4.0 ± 0.2 3.2 3.4 14.5
Nickel (μg/g urinary creatinine) 7.2 ± 1.1** 4.6 ± 0.5 4.1 ± 0.5 5.4 ± 0.4 4.0 4.1 47.8
All values shown are mean ± SE.
* p = 0.01
** p < 0.01, Kruskal-Wallis test.
Table 2 Urinary 8-OHdG (ng/mg) creatinine stratified by different variables.
Variable No. Mean ± SE Median (range)
Total 142 13.6 ± 0.7 11.7 (0.4–59.7)
Elementary school
Longgang 49 18.8 ± 1.5 18.8 (3.9–59.7)**
Shalach 45 9.8 ± 0.7 9.9 (0.4–22.1)
Shuntian 48 11.8 ± 0.9 11.6 (0.7–37.3)
Sex
Boys 74 13.2 ± 0.9 12.0 (0.4–39.7)
Girls 68 14.0 ± 1.1 11.3 (2.5–59.7)
Paternal education (years)
> senior high school (≥ 12) 83 12.8 ± 0.7 11.4 (2.5–33.0)
< senior high school (< 12) 59 14.7 ± 1.4 12.0 (0.4–59.7)
Maternal education (years)
> senior high school (≥ 12) 73 11.8 ± 0.7 10.8 (0.8–31.3)*
< senior high school (< 12) 69 15.5 ± 1.2 13.0 (0.4–59.7)
Parental occupation
Industry 72 14.8 ± 1.1 12.8 (0.7–59.7)
Nonindustry 70 12.4 ± 0.9 10.6 (0.4–33.0)
Parental smoking status
Yes 70 14.3 ± 0.9 12.8 (0.7–37.3)#
No 72 12.9 ± 1.1 10.6 (0.4–59.7)
Possible indoor pollutants
Pet feeding
Yes 27 13.3 ± 1.6 11.1 (3.2–33.1)
No 115 13.7 ± 0.8 12.0 (0.4–59.7)
Incense burning at home all day
Yes 58 14.5 ± 1.1 11.1 (0.4–59.7)
No 84 12.9 ± 0.9 13.5 (0.8–39.7)
Home dampness
Yes 29 15.3 ± 2.1 12.0 (2.5–59.7)
No 113 13.1 ± 0.7 11.4 (0.4–39.7)
Personal medical histories
Asthma
Yes 10 12.2 ± 2.3 10.1 (5.2–25.1)
No 132 13.7 ± 0.8 11.9 (0.4–59.7)
Allergic rhinitis
Yes 44 11.0 ± 0.9 9.9 (0.4–25.1)*
No 98 14.8 ± 0.9 12.3 (0.7–59.7)
Adjusted urinary arsenica
High (≥ 7.7 μg/g creatinine) 71 14.8 ± 1.1 12.4 (0.4–59.7)#
Low (< 7.7 μg/g creatinine) 71 12.4 ± 0.9 11.1 (0.7–39.7)
Adjusted urinary chromiuma
High (≥ 2.0 μg/g creatinine) 71 15.5 ± 1.3 12.4 (0.4–59.7)#
Low (< 2.0 μg/g creatinine) 71 11.6 ± 0.6 11.1 (3.1–24.5)
Adjusted urinary nickela
High (≥ 4.1 μg/g creatinine) 71 13.8 ± 1.2 11.4 (0.4–59.7)
Low (< 4.1 μg/g creatinine) 71 13.3 ± 0.8 12.3 (0.7–39.7)
a Cut points were determined according to medians of urinary creatinine-adjusted levels among all subjects.
* 0.01 < p < 0.05;
** p < 0.01;
# 0.05 < p < 0.10.
Table 3 Multiple linear regression model between the level of urinary 8-OHdG and incorporated concentration of arsenic and chromium (n = 142).
Component model Model parameter (SE) Significance (p-value) Explained variance (%)
Maternal education (< senior high school vs. > senior high school) 2.8 (1.4) 0.05 34.7
Parental smoking status (yes vs. no) 1.0 (1.4) 0.47 2.2
Allergic rhinitis (yes vs. no) −3.0 (1.6) 0.05 24.1
Adjusted urinary arsenic (high vs. low)a 1.9 (1.4) 0.18 17.5
Adjusted urinary chromium (high vs. low)a 3.0 (1.4) 0.04 21.5
a Cut points were determined according to medians (arsenic, 7.7 μg/g creatinine; chromium, 2.0 μg/g creatinine) of urinary creatinine-adjusted levels among all subjects.
==== Refs
References
Akagi S Nagake Y Kasahara J Sarai A Kihara T Morimoto H 2003 Significance of 8-hydroxy-2′-deoxyguanosine levels in patients with chronic renal failure Nephrology 8 192 195 15012720
Beckman KB Ames BN 1998 The free radical theory of aging matures Physiol Rev 78 547 581 9562038
Bowler RP Crapo JD 2002 Oxidative stress in allergic respiratory diseases J Allergy Clin Immunol 110 349 356 12209079
Cadet J Douki T Gasparutto D Ravanat JL 2003 Oxidative damage to DNA: formation, measurement and biochemical features Mutat Res 531 5 23 14637244
Carter JD Ghio AJ Samet JM Devlin RB 1997 Cytokine production by human airway epithelial cells after exposure to an air pollution particle is metal-dependent Toxicol Appl Pharmacol 146 180 188 9344885
Cerutti PA 1994 Oxy-radicals and cancer Lancet 344 862 863 7916406
Chazotte-Aubert L Oikawa S Gilibert I Bianchini F Kawanishi S Ohshima H 1999 Cytotoxicity and site-specific DNA damage induced by nitroxyl anion (NO− ) in the presence of hydrogen peroxide. Implications for various pathophysiological conditions J Biol Chem 274 20909 20915 10409635
Chen W Chen J 2002 Nested case-control study of lung cancer in four Chinese tin mines Occup Environ Med 59 113 118 11850554
Cheng KC Cahill DS Kasai H Nishimura S Loeb LA 1992 8-Hydroxyguanine, an abundant form of oxidative DNA damage, causes G–T and A–C substitutions J Biol Chem 267 166 172 1730583
Cheng TJ Wong RH Lin YP Hwang YH Horng JJ Wang JD 1998 Chinese herbal medicine, sibship, and blood lead in children Occup Environ Med 55 573 576 9849547
Cooke MS Evans MD Herbert KE Lunec J 2000 Urinary 8-oxo-2′-deoxyguanosine—source, significance and supplements Free Radic Res 32 381 397 10766407
Dally H Hartwig A 1997 Induction and repair inhibition of oxidative DNA damage by nickel(II) and cadmium(II) in mammalian cells Carcinogenesis 18 1021 1026 9163690
Donaldson K Brown D Clouter A Duffin R MacNee W Renwick L 2002 The pulmonary toxicology of ultra-fine particles J Aerosol Med 15 213 220 12184871
Droste JH Weyler JJ Van Meerbeeck JP Vermeire PA van Sprundel MP 1999 Occupational risk factors of lung cancer: a hospital based case-control study Occup Environ Med 56 322 327 10472306
Drury JA Jeffers G Cooke RW 1998 Urinary 8-hydroxy-deoxyguanosine in infants and children Free Radic Res 28 423 428 9684987
Eiserich JP Hristova M Cross CE Jones AD Freeman BA Halliwell B 1998 Formation of nitric oxide-derived inflammatory oxidants by myeloperoxidase in neutrophils Nature 391 393 397 9450756
Goodarzi F Huggins FE 2001 Monitoring the species of arsenic, chromium and nickel in milled coal, bottom ash and fly ash from a pulverized coal-fired power plant in western Canada J Environ Monit 3 1 6 11253001
Grimsrud TK Berge SR Haldorsen T Andersen A 2002 Exposure to different forms of nickel and risk of lung cancer Am J Epidemiol 156 1123 1132 12480657
Grisham MB 1994 Oxidants and free radicals in inflammatory bowel disease Lancet 344 859 861 7916405
Hartwig A Schwerdtle T 2002 Interactions by carcinogenic metal compounds with DNA repair processes: toxicological implications Toxicol Lett 127 47 54 12052640
Hayes RB 1997 The carcinogenicity of metals in humans Cancer Causes Control 8 371 385 9498900
Hwang YH Bornschein RL Grote J Menrath W Roda S 1997 Urinary arsenic excretion as a biomarker of arsenic exposure in children Arch Environ Health 52 139 147 9124875
Jenner P 1994 Oxidative damage in neurodegenerative disease Lancet 344 796 798 7916079
Kasprzak KS 1991 The role of oxidative damage in metal carcinogenicity Chem Res Toxicol 4 604 615 1807443
Kasprzak KS Jaruga P Zastawny TH North SL Riggs CW Olinski R 1997 Oxidative DNA base damage and its repair in kidneys and livers of nickel(II)-treated male F344 rats Carcinogenesis 18 271 277 9054618
Kessel M Liu SX Xu A Santella R Hei TK 2002 Arsenic induces oxidative DNA damage in mammalian cells Mol Cell Biochem 234 301 308 12162448
Kuo HW Chang SF Wu KY Wu FY 2003 Chromium (VI) induced oxidative damage to DNA: increase of urinary 8-hydroxydeoxyguanosine concentrations (8-OHdG) among electroplating workers Occup Environ Med 60 590 594 12883020
Li CS Wan GH Hsieh KH Chua KY Lin RH 1994 Seasonal variation of house dust mite allergen (Der pI) in a subtropical climate J Allergy Clin Immunol 94 131 134 8027492
Loft S Poulsen HE 1996 Cancer risk and oxidative DNA damage in man J Mol Med 74 297 312 8862511
Lung SC Kao MC Hu SC 2003 Contribution of incense burning to indoor PM10 and particle-bound polycyclic aromatic hydrocarbons under two ventilation conditions Indoor Air 13 194 199 12756013
Marnett LJ 2000 Oxyradicals and DNA damage Carcinogenesis 21 361 370 10688856
Matsui M Nishigori C Toyokuni S Takada J Akaboshi M Ishikawa M 1999 The role of oxidative DNA damage in human arsenic carcinogenesis: detection of 8-hydroxy-2′-deoxyguanosine in arsenic-related Bowen’s disease J Invest Dermatol 113 26 31 10417614
Moore LE Smith AH Hopenhayn-Rich C Biggs ML Kalman DA Smith MT 1997 Micronuclei in exfoliated bladder cells among individuals chronically exposed to arsenic in drinking water Cancer Epidemiol Biomarkers Prev 6 31 36 8993795
Nicolai T 1999 Air pollution and respiratory disease in children: what is the clinically relevant impact? Pediatr Pulmonol Suppl 18 9 13 10093083
Paustenbach DJ Panko JM Fredrick MM Finley BL Proctor DM 1997 Urinary chromium as a biological marker of environmental exposure: what are the limitations? Regul Toxicol Pharmacol 26 S23 S34 9380834
Sabbioni E Goetz L Bignoli G 1984 Health and environmental implications of trace metals released from coal-fired power plants: an assessment study of the situation in the European Community Sci Total Environ 40 141 154 6523129
Shigenaga MK Gimeno CJ Ames BN 1989 Urinary 8-hydroxy-2′-deoxyguanosine as a biological marker of in vivo oxidative DNA damage Proc Natl Acad Sci USA 86 9697 9701 2602371
Smith-Sivertsen T Tchachtchine V Lund E Bykov V Thomassen Y Norseth T 1998 Urinary nickel excretion in populations living in the proximity of two Russian nickel refineries: a Norwegian–Russian population-based study Environ Health Perspect 106 503 511 9681979
Stern AH Fagliano JA Savrin JE Freeman NC Lioy PJ 1998 The association of chromium in household dust with urinary chromium in residences adjacent to chromate production waste sites Environ Health Perspect 106 833 839 9831544
Sunderman FW Jr 1993 Biological monitoring of nickel in humans Scand J Work Environ Health 19 34 38 8159970
Thompson HJ Heimendinger J Haegele A Sedlacek SM Gillette C O’Neill C 1999 Effect of increased vegetable and fruit consumption on markers of oxidative cellular damage Carcinogenesis 20 2261 2266 10590217
Toraason M 1999 8-Hydroxydeoxyguanosine as a biomarker of workplace exposures Biomarkers 4 3 26 23898791
Vahter M 1994 What are the chemical forms of arsenic in urine, and what can they tell us about exposure? Clin Chem 40 679 680 8174236
Wang CL Chuang HY Ho CK Yang CY Tsai JL Wu TS 2002 Relationship between blood lead concentrations and learning achievement among primary school children in Taiwan Environ Res 89 12 18 12051780
Wang TN Ko YC Chao YY Huang CC Lin RS 1999 Association between indoor and outdoor air pollution and adolescent asthma from 1995 to 1996 in Taiwan Environ Res 81 239 247 10585020
Witztum JL 1994 The oxidation hypothesis of atherosclerosis Lancet 344 793 795 7916078
Wong RH Yeh CY Hsueh YM Wang JD Lei YC Cheng TJ 2003 Association of hepatitis virus infection, alcohol consumption and plasma vitamin A levels with urinary 8-hydroxydeoxyguanosine in chemical workers Mutat Res 535 181 186 12581536
Wu MT Pan CH Huang YL Tsai PJ Chen CJ Wu TN 2003 Urinary excretion of 8-hydroxy-2-deoxyguanosine and 1-hydroxypyrene in coke-oven workers Environ Mol Mutagen 42 98 105 12929122
Yin B Whyatt RM Perera FP Randall MC Cooper TB Santella RM 1995 Determination of 8-hydroxydeoxyguanosine by an immunoaffinity chromatography-monoclonal antibody-based ELISA Free Radic Biol Med 18 1023 1032 7628728
Yuann JM Liu KJ Hamilton JW Wetterhahn KE 1999 In vivo effects of ascorbate and glutathione on the uptake of chromium, formation of chromium(V), chromium-DNA binding and 8-hydroxy-2′-deoxyguanosine in liver and kidney of osteogenic disorder shionogi rats following treatment with chromium(VI) Carcinogenesis 20 1267 1275 10383900
|
16203252
|
PMC1281285
|
CC0
|
2021-01-04 23:41:29
|
no
|
Environ Health Perspect. 2005 Oct 27; 113(10):1386-1390
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.7401
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences 10.1289/ehp.7610ehp0113-00139116203253ResearchChildren's HealthAir Pollution and Lymphocyte Phenotype Proportions in Cord Blood Hertz-Picciotto Irva 1Herr Caroline E.W. 12Yap Poh-Sin 1Dostál Miroslav 3Shumway Robert H. 4Ashwood Paul 5Lipsett Michael 6Joad Jesse P. 7Pinkerton Kent E. 8Šrám Radim J. 31 Department of Public Health Sciences, University of California, Davis, California, USA2 Institute of Hygiene and Environmental Medicine, University of Giessen, Germany3 Laboratory of Genetic Ecotoxicology, Institute of Experimental Medicine, Academy of Sciences of the Czech Republic, and Health Institute of Central Bohemia, Prague, Czech Republic4 Department of Statistics and5 Department of Rheumatology, Allergy, and Clinical Immunology, University of California, Davis, California, USA6 Department of Epidemiology and Biostatistics, University of California, San Francisco, California, USA7 Department of Pediatrics, and8 Center for Health and the Environment, University of California, Davis, California, USAAddress correspondence to I. Hertz-Picciotto, Department of Public Health Sciences, Division of Epidemiology, TB #168, University of California Davis, Davis, CA 95616 USA. Telephone: (530) 752-3025. Fax: (530) 752-3239. E-mail:
[email protected] authors declare they have no competing financial interests.
10 2005 15 6 2005 113 10 1391 1398 27 9 2004 14 6 2005 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright. Effects of air pollution on morbidity and mortality may be mediated by alterations in immune competence. In this study we examined short-term associations of air pollution exposures with lymphocyte immunophenotypes in cord blood among 1,397 deliveries in two districts of the Czech Republic. We measured fine particulate matter < 2.5 μm in diameter (PM2.5) and 12 polycyclic aromatic hydrocarbons (PAHs) in 24-hr samples collected by versatile air pollution samplers. Cord blood samples were analyzed using a FACSort flow cytometer to determine phenotypes of CD3+ T-lymphocytes and their subsets CD4+ and CD8+, CD19+ B-lymphocytes, and natural killer cells. The mothers were interviewed regarding sociodemographic and lifestyle factors, and medical records were abstracted for obstetric, labor and delivery characteristics. During the period 1994 to 1998, the mean daily ambient concentration of PM2.5 was 24.8 μg/m3 and that of PAHs was 63.5 ng/m3. In multiple linear regression models adjusted for temperature, season, and other covariates, average PAH or PM2.5 levels during the 14 days before birth were associated with decreases in T-lymphocyte phenotype fractions (i.e., CD3+ CD4+, and CD8+), and a clear increase in the B-lymphocyte (CD19+) fraction. For a 100-ng/m3 increase in PAHs, which represented approximately two standard deviations, the percentage decrease was −3.3% [95% confidence interval (CI), −5.6 to −1.0%] for CD3+, −3.1% (95% CI, −4.9 to −1.3%) for CD4+, and −1.0% (95% CI, −1.8 to −0.2%) for CD8+ cells. The corresponding increase in the CD19+ cell proportion was 1.7% (95% CI, 0.4 to 3.0%). Associations were similar but slightly weaker for PM2.5. Ambient air pollution may influence the relative distribution of lymphocyte immunophenotypes of the fetus.
air pollutionB-cellcord bloodimmune statusimmunologylymphocytesnatural killer cellsPAHPM10pregnancyT-cell
==== Body
Early life is a potentially susceptible period for pollution-induced perturbation of the respiratory system (Braun-Fahrländer et al. 1997), DNA (Binková et al. 1995), and possibly the immune system (Hertz-Picciotto et al. 2002). Disturbances during this developmental period may result in transient or irreversible long-term effects and may also provide models for examining the mechanisms by which air pollution may affect the entire population.
With regard to prenatal exposures, current evidence from several countries is compatible with small adverse effects of particulate matter (PM), polycyclic aromatic hydrocarbons (PAHs), and/or other air pollutants on fetal growth, infant mortality, and duration of pregnancy (Dejmek et al. 1999, 2000; Glinianaia et al. 2004; Jedrychowski et al. 2004; Perera et al. 2003; Šrám et al. 1999; Wang et al. 1997; Wilhelm and Ritz 2003). Ambient PAH exposure has also been linked to somatic mutations in newborns (Perera et al. 2002) and possibly to heritable genetic changes (Somers et al. 2004). Benzo[a]pyrene (BaP)–DNA adducts were found to be substantially higher in cord blood compared with maternal blood samples from New York City residents (Perera et al. 2004).
A recent European study reported associations of PM with absolute numbers of B-cells, CD4+ and CD8+ T-cells, and natural killer (NK) cells in schoolchildren, after adjustment for numerous factors (Leonardi et al. 2000). In earlier work, we observed lower percentages of T-lymphocyte and higher NK cell percentages among residents of a polluted community, and among deliveries in winter, when air pollutant levels are highest (Hertz-Picciotto et al. 2002). Ambient PM pollution causes oxidative stress (Li et al. 2003) and may influence allergic (Diaz-Sanchez et al. 2003), immunologic (Burchiel et al. 2004), and systemic inflammatory responses (Li et al. 2003; van Eeden and Hogg 2002).
A further rationale for concern about ambient air pollution is the similarity with constituents of cigarette smoke. Evidence of early-life vulnerability to cigarette smoke is seen in reduced physical height; higher incidence of lower respiratory infections, asthma and wheeze, middle-ear disease, and sudden infant death syndrome (SIDS) (Courage 2002; DiFranza et al. 2004); stronger antigen-induced lymphoproliferation in cord blood (Devereux et al. 2002); and significantly higher cord blood mononuclear cell production of Th2 cytokines (interleukin-13) in response to stimulation with antigen (Noakes et al. 2003).
This report extends earlier work on births in the Czech Republic. In contrast to earlier analyses of long-term ambient air pollution, in which we focused on the comparison of births from a highly versus a less polluted district (Hertz-Picciotto et al. 2002), in the present project we examine temporal variability in exposures in the days immediately before birth. In particular, we examine associations between neonatal immunophenotypes and specific pollutants not previously examined for developmental immunotoxicity—namely, PM2.5 (fine PM < 2.5 μm in aerodynamic diameter) and PAHs—in a large study population.
Materials and Methods
Subject enrollment and data collection.
From May 1994 through March 1999, women who delivered in the districts of Teplice or Prachatice were asked to participate in the Pregnancy Outcome Study (Dejmek et al. 2000). Of approximately 8,500 births in the two districts during the study period, nurses trained in research methods recruited close to 90% of mothers (n = 7,465) during their hospital stays and obtained written informed consent. While in the hospital, mothers completed self-administered questionnaires regarding reproductive histories, medical conditions and medications, smoking, alcohol and other lifestyle factors, and occupational information. The nurses then reviewed the mothers’ responses for ambiguity and completeness.
From this cohort, a subset of 1,476 mother–infant pairs was recruited into the Immune Biomarker Study, for which maternal and cord blood samples were obtained at delivery (Hertz-Picciotto et al. 2002). No exclusions were made, and a stratified random sample was obtained as follows: Nurses were instructed to enroll all preterm and low-birth-weight infants, as well as a systematic, one in five random sample of other births. However, sampling of the full-term, normal-birth-weight deliveries was increased when a meteorologic inversion occurred in January 1996 and was subsequently maintained through the end of the study. Friday or weekend deliveries were not enrolled because of the requirement for flow cytometry within 24 hr of sample collection.The overall sampling fraction was 20%. The refusal rate was 5%.
We abstracted data on pregnancy, labor, delivery, and the neonate from medical records—for example, birth weight, gestational age, date and time of birth, and medications used during each stage of labor. Home heating and cooking sources and family history of allergy were taken from a follow-up questionnaire administered to parents when children were 3 or 4.5 years of age. For the analysis below, we excluded data from a) mothers with incomplete sociodemographic or labor data (n = 36), b) infants delivered by cesarean section (n = 42), and c) one exceptionally long labor (> 24 hr), because labor itself is reported to influence lymphocyte distribution (Pittard et al. 1989), leaving 1,397 mother–infant pairs. This study was approved by the Ethical Committee of the Health Institute of Central Bohemia, Prague, Czech Republic, and the Institutional Review Board at the University of California, Davis.
Exposure assessment.
In January 1992, an air-quality field measurement program was initiated by the Czech Ministry of Environment and carried out by the District Institutes of Hygiene, with assistance from the U.S. Environmental Protection Agency (Pinto et al. 1998). Two monitoring sites were established, one in Teplice, Northern Bohemia, and one in Prachatice, Southern Bohemia. Measurements of PM2.5, PM < 10 μm in diameter (PM10), and PAHs were performed daily in November–March, every third day in April–June and September–October, and every sixth day in July and August, when air pollution is lowest. Daily measurements of sulfur dioxide and nitrogen oxides were conducted year-round and were used in the imputation of PM2.5 and PAHs for nonsampled days.
We assessed fine and coarse particle concentrations, as well as PAH concentrations, from samples collected by the versatile air pollution sampler described by Pinto et al. (1998). This device is a modified dichotomous sampler, with a size-selective inlet of 10 μm aerodynamic diameter. A virtual impactor separates the airflow into two channels that collect fine particles (PM2.5) and a third that collects coarse particles (PM2.5–10). We determined the mass PM2.5 gravimetrically from a sample collected on a Teflon filter pack in the first fine-particle channel.
In the second fine-particle channel, a polyurethane foam (PUF) trap located downstream of a quartz filter collects PAHs that have either evaporated from the quartz filter or were originally in the gas phase. PAH concentrations are obtained by extraction of the PUF trap and the quartz filters, followed by high-performance liquid chromatography analysis. We measured 12 compounds and used their sum (total PAH) for analysis: phenanthrene, anthracene, fluoranthene, pyrene, benzo[a]anthracene, chrysene, benzo[b]fluoranthene, benzo[k]fluoranthene, BaP, dibenzo[a,h]anthracene, benzo[g,h,i]perylene, and indeno[c,d]pyrene. We calculated a BaP equivalent exposure using potency factors according to Petry at al. (1996). Seven PAHs potentially carcinogenic to humans were evaluated separately as the sum of c-PAHs: chrysene, benzo[a]anthracene, benzo[b]fluoranthene, benzo[k]fluoranthene, BaP, dibenz[a,h]-anthracene, and indeno[1,2,3-c,d]pyrene. This sum had been associated with intrauterine growth retardation in a prior analysis of the cohort (Dejmek et al. 2000).
For periods when daily PM2.5 and PAHs were not measured, between 3 and 5% of the values had to be imputed. The time-series nature of the data dictated an imputation procedure that replaced missing values with estimates that took into account the predictive capabilities of nearby neighbors in time as well as relations with other pollution series. To minimize the effects of potential outliers, the log-transformed values were imputed and then transformed back to determine imputed pollution values. The underlying pollution series were assumed to satisfy a first-order vector autoregressive model, with Kalman filters and smoothers (Shumway and Stoffer 2000) used for imputation. More specifically, the imputed values are always the conditional expectations of the unobserved values given the observed series. The general methodology follows that described in Little and Rubin (2002).
Because not all PAHs were detected each time measurements were made, we assumed that the log-transformed values were independent and normally distributed to complete the 12 dimensional component vectors. We imputed the missing observations on days when PAHs were measured (about 5.2% for Teplice and 3.8% for Prachatice) using conditional expectations obtained during the estimation process, following a commonly used procedure for estimating parameters with missing data (Little and Rubin 2002).
Laboratory methods: flow cytometry.
Venous cord blood sampled immediately after labor was collected into heparinized Vacutainers (10-mL Vacuette; Greiner, Kremsmuenster, Austria). The samples were stored at 4°C in polystyrene boxes and transported for analysis in coolers. Samples arriving at the laboratory later than 24 hr after delivery were discarded; all others were analyzed on arrival. Lymphocytes in lysed whole blood were immunophenotyped using a FACSort flow cytometer, Simulset software, and a Simultest IMK lymphocyte kit of monoclonal antibodies (Becton Dickinson Immunocytometry Systems, San Jose, California, USA).
The following lymphocyte subsets were determined: CD3+ T-lymphocytes, CD3−CD19+ B-lymphocytes, and CD3−CD16+/CD56+ NK cells. Subsets of the CD3+ cells were also ascertained, including CD4+ and CD8+ cells, sometimes referred to as T-helper and T-suppressor cells, respectively. Problems with contamination of cord blood lymphocytes with nucleated red blood cells were solved by a lysed whole blood method (Harris et al. 1994). The Simulset software provides the three-part differential of leukocytes in the gate, based on CD45+CD14+ staining. Correspondence of the percentage of identified lymphocytes (sum of T + B + NK cells) to the percentage of lymphocytes in the gate was used to control the quality of staining. The proportions of the lymphocyte subsets were converted to percentages of the sum of T, B, and NK lymphocytes.
Data handling and statistical methods.
We entered abstracted medical record data and maternal questionnaire information into electronic files in Prague at the Laboratory of Genetic Ecotoxicology (Institute of Experimental Medicine, Academy of Sciences, Czech Republic). One of us (M.D.) then reviewed the medical records a final time to correct errors. Outliers, implausible values, and inconsistencies across variables were identified and resolved. Multiple files were merged into an analysis file that included a) 24-hr average concentrations of air pollutants; b) daily mean temperature measurements; c) maternal, family, and household characteristics; d) data from medical records on pregnancy, labor, and delivery; and e) lymphocyte results.
To quantify the relationships between measurements of air pollutants and immune parameters, multiple linear regression models were fit separately for PM2.5 and total PAHs. For each of these, we examined the lymphocyte percentages in relation to exposures for five averaging periods before the date of cord blood collection: 3-, 7-, 14-, 30-, and 45-day intervals before birth. The strongest and most precise associations were with 7-day and 14-day intervals; for brevity, we present only the 14-day results.
Covariates of concern were identified from the literature, the conceptual model we developed a priori, and an empirical screen of the variables available. Empirically, we screened for predictivity using variables previously associated (p < 0.15) with lymphocyte distributions in cord blood (Hertz-Picciotto et al. 2002), including season, length of labor, parity, number of previous stillbirths, medication during delivery, working status of mother, maternal education, and exposure to active and secondhand smoke. In the present analysis, we also examined family history of allergy, self-reports of workplace exposure to dust during pregnancy, and self-reported maternal chronic or severe respiratory diseases during pregnancy.
Meteorologic variables are potential confounders because of strong associations with air pollution and lymphocytes (Afoke et al. 1993; Levi et al. 1988). Therefore, we explored five averaging periods of ambient temperature (3-, 7-, 14-, 30-, and 45-day intervals) for multivariable models. We show results adjusted for 3-day average temperature before birth to account for short-term and 45-day average temperature to represent potential longer term associations. In addition, seasonal and circadian rhythms are characteristics of both air pollutants and lymphocytes. We adjusted for time of day of delivery, as well as season, where summer (June–August) is the reference, with three binary variables representing winter (December–February), spring (March–May), and fall (September–November).
We evaluated four possible effect modifiers: home heating source, cigarette smoking, ethnicity, and low birth weight or prematurity. Coal or wood heating or cooking can markedly increase residential indoor exposure to PAHs (Siwinska et al. 1999), and because the effect of ambient air pollution might differ among those already exposed to high background exposures in their homes, we introduced an interaction term in the models. For similar reasons, we evaluated effect modification from cigarette smoking, either by the mother or by other members of the family. We also addressed possible heterogeneity by ethnicity—that is, whether associations with ambient air pollutants differed between newborns of Gypsy (Rom) ethnicity and those of European origin. As reported previously, frequency of smoking was greater among the Roma, parity was higher, and infants were more likely to be low birth weight. Finally, infants born prematurely or who were small for their gestational age were hypothesized to represent a potentially susceptible subgroup.
The initial model included average PAH level during the 14 days before birth and all variables on the candidate list that were appropriate for adjustment. Subsequently, predictor variables were removed if they did not predict the outcome with adequate precision (i.e., they were eliminated if p > 0.15), did not block a backdoor path in the directed acyclic graph (Cole and Hernan 2002), and resulted in changes < 15% in the estimated coefficient for PAHs. Once the predictive or confounder covariates were determined, they were used in subsequent models exploring different averaging periods or alternative pollutant mixes (i.e., BaP equivalents or carcinogenic PAHs). To ensure that confounders specific to the other pollutant groups were not missed, the final models for PM2.5, BaP equivalents, and carcinogenic PAHs were expanded to examine potential confounding (by the change-inestimate criterion) from any previously removed variables.
All models were adjusted using SUDAAN statistical software (version 8.2004; Research Triangle Institute, Research Triangle Park, NC, USA) for the sampling design, namely, stratified sampling without replacement in strata defined by three variables (district, preterm or low birth weight vs. full term and normal birth weight, and year of birth). Results were expressed as predicted changes in lymphocyte distribution for an increase of 100 ng/m3 in PAH concentration, and of 25 μg/m3 in PM2.5. These increments were close to two standard deviations of the distributions of 14-day averages of the two pollutants in this study: 48 ng/m3 for total PAHs and 13 μg/m3 for PM2.5. We chose to use absolute increments, rather than interquartile ranges, to facilitate comparison across studies.
Results
Table 1 compares the 1,397 deliveries in the immunity study that met our inclusion criteria with the full cohort from which they were sampled. The sample in the immunity study consisted of proportionally fewer births from Teplice (28% vs. 39% from Prachatice), more mothers of low parity, and, because of the sampling strategy, more winter/spring and 1996–1998 births and more preterm and low-birth-weight infants. The sample did not differ from the full cohort with respect to maternal or paternal education, smoking, maternal age, or ethnicity.
Mean daily averages and mean 14-day averages were nearly identical both within districts and averaged across districts (e.g., in Teplice the daily mean PAH concentration was 68.7 ng/m3, whereas the 14-day mean was 68.9 ng/m3). PAHs were about 10% lower in Prachatice. However, Prachatice PAH concentrations were higher in the winters of 1994–1995 and 1995–1996 (Figure 1). The interdistrict differences are more pronounced for PM2.5: The mean 14-day average over the study period was about 1.5 times higher in Teplice (30.1 μg/m3) than in Prachatice (19.8 μg/m3). These PM2.5 differences are fairly consistent throughout the study period, with close tracking of seasonal peaks in the two districts. Correlations varied by district (Table 2). Temperature showed strong negative correlations with both PAHs (r = −0.7 to −0.9) and PM2.5 (r = −0.4 to −0.6). The correlation of 14-day average PAHs with PM2.5 was 0.6 in Prachatice and 0.9 in Teplice. When PAHs from different averaging periods were compared, the correlations declined from almost 1 (3-day vs. 7-day average) to 0.9 (3-day vs. 45-day average); for PM2.5, these correlations ranged from a high of 0.9 to 0.5 (data not shown).
Adjusted for 3-day temperature and season, total PAH exposure during the 14 days before birth was associated with reduced T-lymphocyte fractions CD3+, CD4+, and CD8+, and an increase in the B-lymphocyte fraction (CD19+) (Figure 2). For a 100-ng/m3 increase in PAHs, the percentage decrease was −3.3% [95% confidence interval (CI), −5.6 to −1.0%] for CD3+, −3.1% (95% CI, −4.9 to −1.3%) for CD4+, and −1.0% (95% CI, −1.8 to −0.2%) for CD8+ cells. The corresponding increase in the CD19+ cells was +1.7% (95% CI, 0.4 to 3.0%).
These findings were robust to the parameterization of temperature and season; the strongest findings for all fractions, except CD19+, were observed when adjustment was made for season alone. In the models adjusted for season, an increase of the NK-cell fraction was also seen (+2.5%; 95% CI, 0.2 to 4.7%); additional adjustment for 3-day temperature reduced the magnitude of associations, but most remained significant (Table 3).
These findings were robust to the parameterization of temperature and season; the strongest findings for all fractions, except CD19+, were observed when adjustment was made for season alone. In the models adjusted for season, an increase of the NK-cell fraction was also seen (+2.5%; 95% CI, 0.2 to 4.7%); additional adjustment for 3-day temperature reduced the magnitude of associations, but most remained significant (Table 3).
When the alternative PAH averaging periods (3, 7, 30, and 45 days) were used, the results usually yielded lower precision, although the overall patterns were fairly similar (data not shown); for models using 7-day averages, the results were very close to those from the 14-day models—that is, none of the associations with lymphocyte subsets changed by > 15%.
Models using PM2.5 concentrations showed similar results for CD4+, CD3+, and CD19+, whereas associations for CD8+ disappeared. Associations were very much attenuated or not present in models of BaP equivalents or carcinogenic PAHs (data not shown).
Temperature (either 3- or 45-day average before birth) was a strong predictor of all lymphocyte phenotype subsets except B-cells (CD19+). When 3-day temperature was included in the model, fall season was strongly related to increased CD3+ (+3.7%; 95% CI, 1.0 to 6.4%) and CD4+ cell fractions (+3.7%; 95% CI, 1.4 to 6.1%) and a lower CD19+ cell fraction (−2.6%; 95% CI, −4.1 to −1.0%). Spring was associated with lower CD19+ (−1.9%; 95% CI, −3.6 to −0.1%) and higher NK (+2.8%; 95% CI, 2.7 to 2.9%) cell fractions.
Other variables predictive of one or more lymphocyte immunophenotypes were district and year of birth, duration of labor, medications administered during labor, number of previous pregnancies, maternal education, time of day of delivery, and workplace exposure to dust. Family history of allergy and maternal chronic respiratory diseases during pregnancy were not associated with lymphocyte phenotype fractions in cord blood and were therefore not included in the final models.
We observed no difference in the impact of any of the air pollutants between the Roma infants and infants born to mothers of eastern European descent (data not shown). Among newborns from homes using coal or wood for cooking or heating, PAHs were associated with greater decreases in fractions of CD3+ and CD4+ and greater increases in the percentage of NK cells (Figure 3). The pollutant-associated decreases of CD3+, CD4+, and CD8+ percentages were greater in newborns exposed to cigarette smoke from the mother or others around her versus newborns not exposed to cigarette smoke, whereas an increase in the CD19+ percentage was observed only among births to nonsmokers. PAH-associated increases in NK cell fractions were largest in homes of coal users and smokers. For PM2.5, the only demonstrable heterogeneity was observed for neonates in homes using coal, with greater reductions in CD3+ and CD4+ percentages and an elevated percentage of NK cells, a pattern similar to that observed in relation to ambient PAHs.
Preterm or low-birth-weight infants were not at higher risk than were normal neonates for pollution-associated changes in lymphocyte proportions (Figure 3). However, in this subset, there appeared to be an interaction between ambient PAH exposure and maternal exposure (active or passive) to cigarette smoke (not shown in figure). Specifically, preterm/low-birth-weight infants born to mothers reporting active or second-hand cigarette smoke exposure had cord blood samples with a decreased proportion of cells with CD3+ surface antigens (−9.7%; 95% CI, −15.0 to −4.4%), whereas cells with CD19+ were elevated by 3.8% (95% CI, 0.3 to 7.3%) and NK cells were elevated by 4.5% (95% CI, 0.6 to 8.4%), for an increase of 100 ng PAH/m3, compared with preterm/low-birth-weight infants born to mothers not exposed to cigarette smoke.
Discussion
In this study population, ambient concentrations of PAHs and PM2.5 during the last 2 weeks of gestation were associated with decreases in the percentages of T-lymphocytes in cord blood. These associations were stronger for the percentage of CD4+ cells than for the percentage of CD8+ subsets. However, partly because CD4+ are relatively more numerous, the resulting effect of air pollution on the CD4+:CD8+ ratio was essentially null. The association with CD8+ cells was more marked for PAHs, compared with PM2.5. Accompanying these T-cell decreases, the percentage of B-cells (CD19+) increased. Because these outcomes represent distributions of immune phenotype subsets relative to each other, if the percentage of one subset goes down, the percentage of at least one other will increase. It would have been informative to measure absolute lymphocyte counts and functional immune parameters; however, because the quality of these measurements degrades with increasing storage time and transportation, such measurements were deemed not feasible in this project.
The possibility that redistributions of lymphocyte phenotypes may alter susceptibility to infections or inflammatory diseases in otherwise healthy persons, particularly small children, is worthy of investigation in its own right. The longitudinal life-course changes in lymphocyte populations, both absolute and relative, are discussed by Schultz et al. (2000), who concluded that the developmental patterns in absolute counts were different from those in relative distributions, and that “discordance between the absolute and relative size of lymphocyte subpopulations emphasizes the consideration of both variables in the assessment of lymphocyte maturation.” Published data suggest that the percentages of many of the immunophenotypic subpopulations differ for cord versus adult blood (Zhao et al. 2002) but that differences in absolute counts are not always accompanied by differences in percentages (D’Arena et al. 1998). Percentages of some lymphocyte subtypes change from the fetal to neonatal period (Zhao et al. 2002). Given that immunologic development is intimately connected to the interactions between the organism and the environment via antigenic challenge, specific chemical exposures could influence these relationships (e.g., exposure to diesel particles can modify the host response to allergen) (Diaz-Sanchez et al. 2003). Whether alterations in developmental patterns of lymphocytes have clinical implications remains to be established. To shed light on the larger picture, we are addressing, in other work, immunoglobulin production of the neonate and relationships between relative lymphocyte distributions at birth and subsequent early childhood morbidity.
The direction of associations was generally unchanged, whether adjusted for short-term (3 days) or longer-term (45 days) average temperature, season, or both season and short-term temperature, although the precision and magnitude varied. The associations with CD19+ were the most consistent across pollutant metrics (Table 3).
Few studies have examined the relationship of air pollution to pediatric immune status. Leonardi et al. (2000) examined absolute levels of specific lymphocytes in schoolchildren, whereas we assessed percentages of lymphocytes in newborns; hence, our results cannot be compared. Similarly, a cross-sectional study of schoolchildren in two industrial cities suggested a relationship of air pollution with changes in several immune cell fractions, but this report was based on ecologic, not individual-level, data and did not control for any confounders (Skachkova et al. 2001). Overall, the proportions of T-, B-, and NK lymphocytes in cord blood in our study were not markedly different from those reported in a small Italian study (D’Arena et al. 1998) and a large Mexican sample (Garcia et al. 1995). Different laboratory methods and our inclusion of a population-based sample, with no exclusions, can explain any discrepancies.
Consistent with previous findings in this project (Hertz-Picciotto et al. 2002) and other studies, the length of labor, delivery medication, number of previous pregnancies, maternal education, and time of day of delivery (Levi et al. 1988) were all related to lymphocyte outcomes at birth. Maternal chronic or severe respiratory diseases during pregnancy did not predict lymphocyte distributions at birth, possibly because these self-reports did not include information on the period in gestation during which they occurred. Although our previous findings using district of residence as a surrogate for chronic exposure might have been confounded by unrecognized interdistrict differences (Hertz-Picciotto et al. 2002), this problem was eliminated by analysis of short-term temporal variations in PM2.5 and PAHs before birth, in which district and other confounders were controlled. The PM2.5 fraction was selected because it generally displays less spatial heterogeneity than do coarse particles. With only one monitor in each district, fine particles would be subject to less misclassification error. Nevertheless, some error is unavoidable, especially for families living at some distance from the monitors.
Participants in the immune biomarker study were randomly sampled as described above, and the refusal rate was < 5%, supporting the validity of the results based on this sample. Although the distributions of several variables were significantly different in the study sample compared with the full cohort (e.g., the proportion of low-birth-weight infants was 7% in the study sample and 5% in the full cohort), these differences are not likely to have affected either the internal or external validity of the results, because the variables were controlled in the analysis through adjustment for the sampling design or by multivariate adjustment or both. These adjustments combined with the use of sampling weights also provided generalizability to the population that participated in the full cohort study—about 90% of the population in the two districts. Because 14 days before birth could represent different developmental periods for low-birth-weight and preterm newborns, we also conducted a sensitivity analysis in which these infants were excluded and found little change in the results (data not shown).
We furthermore believe our results could be generalized outside the Czech Republic, given that the exposures are similar to those in other countries. Concentrations of PAHs in this investigation were in a range similar to what has been reported from some U.S., European, and Asian cities. For instance, for BaP, our geometric mean was 1.4 ng/m3, which was comparable to levels observed in Pavia (Italy) in 1996 and in Taipei in 1995 and 1996, 1.2 and 1.7 ng/m3, respectively (Naumova et al. 2002).
In the Central European Study of Air Quality and Respiratory Health (CESAR, conducted in Bulgaria, Hungary, Czech Republic, Slovak Republic, Poland, and Romania) annual PM2.5 means were between 29 and 67 μg/m3 (Leonardi et al. 2000). Although the annual average concentration in our study was lower (25 μg/m3), the winter mean was 38 μg/m3, which is higher than the winter means of 18 of the 21 cities analyzed in the European Community Respiratory Health Survey study (Hazenkamp-von Arx et al. 2003). Most U.S. cities are reported to have lower annual PM2.5 means, although comparable means occurred in Los Angeles (25 μg/m3) and Riverside (29 μg/m3), California, in 2000 (Pinto et al. 2004).
The less than daily monitoring during 7 months of the year made it necessary to impute pollution data. Given the schedule of every third day of measurement for 5 months of the year, and every sixth day for 2 months, the probability that missing data are related to levels of PM2.5 or PAH, after conditioning on daily measurements for SO2 and NOx and on the existing data for PM10, PM2.5, and PAH, appears to be extremely small. Thus, the rationale for the assumption of “missing-at-random” is strong.
Confounding due to meteorology and seasonal lymphocyte patterns was accounted for by including 3- or 45-day averages of temperature or season, or both season and temperature in the model. Short-term and longer term time spans of antenatal ambient temperatures were related to a similar extent to T-lymphocyte fractions. Except for CD19+, inclusion of either temperature interval (3- or 45-day average before birth) resulted in less precise estimates of the pollution effects. Season alone does not appear to have this effect; winter by itself does not have any impact once temperature is included. Possibly, our adjustment for temperature could induce bias, because it is highly correlated (Table 2) with air pollution, particularly within season; however, the direction of such bias is difficult to deduce.
The B-lymphocyte fraction was not related to ambient temperature but was related to both spring and fall seasons. This suggests that the seasonal associations must be due to other variables, perhaps respiratory infections during fall or pollen in spring. Seasonal variations in immunologic parameters occur in healthy children (Afoke et al. 1993) and are hypothesized to represent adaptive responses to climatic variability and other environmental factors.
We also found that maternal active smoking and/or exposure to second-hand smoke predicted cord blood lymphocyte distributions in preterm low-birth-weight newborns (decreased T-lymphocytes and increased B-lymphocytes). We did not find published results on lymphocyte phenotype fractions in cord blood in relation to cigarette smoking. Studies in adults consistently suggest a rise in either the percentage or absolute count of CD3+ or CD4+ cells in association with cigarette smoke exposure (Santagostino et al. 1999; Schaberg et al. 1997; Tollerud et al. 1989). One of these studies also reports a higher B-cell fraction related to smoking (Santagostino et al. 1999), which is concordant with our finding.
The pattern in relation to cigarette smoke in our data was similar to what we observed for lymphocyte distributions in relation to ambient air pollution, that is, PAHs. Whether this similarity is due to PAHs or other constituents common to tobacco smoke and ambient air pollution is unclear. Furthermore, lymphocyte changes associated with ambient PAHs in all but the B-cell fractions were greater in births where the mother or others around her smoked (Figure 3). This association of PAHs beyond the effect of smoke exposure on lymphocyte phenotype fractions is consistent with an impact of these aromatic compounds over a wide range of exposure levels.
The subgroup of children most susceptible to lymphocyte effects from air pollution appears to be those from homes heated by coal. This may be a result of high exposures to PAHs from this source.
Distinct from our earlier work suggesting that chronic, long-term exposure to high ambient air pollution may influence immune development (Hertz-Picciott et al. 2002), these results demonstrate short-term associations with fine particles and PAHs. Adjustment for temperature does not eliminate these associations. A comparison of the associations with PAHs and with PM2.5 indicates a number of stronger findings for the former. Reductions in CD3+ and CD4+ fractions were larger for PAHs, and no association was observed between PM2.5 and CD8+ fractions. Because the increments used for PAHs and PM2.5 represented approximately a change of two standard deviations for each, these results can be compared.
Our measurements included both the semivolatile and particulate-bound fractions of PAHs. Given this detailed characterization of PAHs and the associations observed here, chemical composition may play a key role in the immune-mediated effects of air pollution. Others have demonstrated that fetal exposures to ambient PAHs occur trans-placentally and can result in mutations in cord blood lymphocytes (Perera et al. 2002). Moreover, PAHs have been demonstrated to exhibit immunomodulatory properties (Nel et al. 2001).
The striking finding of this study is that exposure to ambient PM2.5 and PAHs in late pregnancy is associated with statistically significant changes in the distributions of lymphocyte phenotypes in cord blood. Although the biologic relevance of this finding is not entirely clear, the observation of note is that the fetal immune system may be altered by maternal exposure to these environmental pollutants. Further research on critical windows of vulnerability throughout gestation is warranted, as well as study of whether such changes persist beyond birth and/or are associated with adverse health effects. We have recently completed a follow-up study of these children, which will give the opportunity to relate the changes observed in T-cell and B-cell fractions at birth with subsequent morbidity.
This work was supported in part by the Czech Ministry of Environment (Teplice Program), the U.S. Environmental Protection Agency (CR 820076), the U.S. Agency for International Development, the Commission of the European Community (PHARE II, EC/HEA 18/CZ), Health Effects Institute, and National Institute of Environmental Health Sciences 1P01-ES11269.
Figure 1 Time series plots for the two districts in the study. (A) PAHs. (B) PM2.5.
Figure 2 Associations between air pollutants and lymphocyte fractions: Percent changes in cord blood lymphocyte distributions, with 95% CIs, associated with increases of 100 ng/m3 PAH (A) or 25 μg/m3 PM2.5 (B) during the 14 days before birth. All models adjusted for district, year of birth, time of day of delivery, labor medication and duration, number of previous pregnancies, and maternal education and smoking.
Figure 3 Effect modification of associations between air pollutants and lymphocyte fractions by size and maturity at birth, exposure to cigarette smoke and home coal use: Percent changes in cord blood lymphocyte distributions, with 95% CIs, associated with increases of 100 ng/m3 PAH (A) or 25 μg/m3 PM2.5 (B) during the 14 days before birth, in various subsets of the study population. All models adjusted for season, average temperature in the three days before birth, district, year of birth, time of day of delivery, labor medication and duration, number of previous pregnancies, and maternal education and smoking.
Table 1 Comparison of characteristics at birth in full cohort versus subset with lymphocytes in cord blood: deliveries in 1994–1999, Prachatice and Teplice, Czech Republic.
Characteristic Immunity [n = 1,397; no. (%)] Full cohort [n = 7,502; no. (%)]
District*
Prachatice 548 (39) 2,144 (29)
Teplice 849 (61) 5,358 (71)
Season of birth*
Winter 380 (27) 1,826 (24)
Spring 397 (28) 2,017 (27)
Summer 300 (21) 1,929 (26)
Fall 320 (23) 1,730 (23)
Year of birth*
1994 106 (8) 1,313 (18)
1995 181 (13) 1,606 (21)
1996 300 (22) 1,420 (19)
1997 367 (26) 1,419 (19)
1998 374 (27) 1,394 (19)
1999 69 (5) 350 (5)
Delivery hour
0600–1159 hr 383 (28) 2,082 (28)
1200–1759 hr 402 (29) 2,101 (28)
1800–2359 hr 316 (23) 1,668 (22)
0000–0559 hr 285 (20) 1,583 (21)
Missing 11 (1) 68 (1)
Sex
Male 724 (52) 3,856 (51)
Female 673 (48) 3,643 (49)
Missing 0 (0) 3 (0)
Birth weight (g)*
≤ 2,500 102 (7) 365 (5)
> 2,500 1,294 (93) 7,132 (95)
Missing 1 (0) 5 (0)
Weeks of gestation at birth*
< 37 98 (7) 339 (5)
≥ 37 1,299 (93) 7,163 (96)
Mother’s age at delivery (years)
< 20 171 (12) 934 (13)
20–24.9 638 (46) 3,274 (44)
25–29.9 395 (28) 2,120 (28)
30–34.9 142 (10) 856 (11)
≥ 35 51 (4) 313 (4)
Missing 0 (0) 5 (0)
Ethnicity of mother
European 1,228 (88) 6,556 (87)
Romani 155 (11) 856 (11)
Other 9 (1) 70 (1)
Don’t know 2 (0) 14 (0)
Missing 3 (0) 6 (0)
No. of live births (parity)*
0–1 651 (47) 2,813 (38)
2 519 (37) 2,306 (31)
≥ 3 227 (16) 2,348 (31)
Missing 0 (0) 35 (0)
Mother’s education
Did not complete primary school 18 (1) 120 (2)
Primary school 293 (2) 1,587 (21)
Secondary school 602 (43) 3,205 (43)
Secondary school with leaving exam 400 (29) 2,092 (28)
Student 6 (0) 36 (0)
University 78 (6) 404 (5)
Missing 0 (0) 58 (1)
No. of cigarettes/day smoked by mother before pregnancy
None 880 (63) 4,584 (61)
1–10 286 (21) 1,593 (21)
11–20 186 (13) 1,033 (14)
≥ 21 20 (1) 130 (2)
Missing 25 (2) 162 (2)
No. of cigarettes/day smoked by father during pregnancy
None 627 (45) 3,197 (43)
1–10 312 (22) 1,670 (22)
11–20 365 (26) 2,004 (27)
≥ 21 42 (3) 310 (4)
Missing 51 (4) 321 (4)
Father’s education
Did not complete primary school 10 (1) 53 (1)
Primary school 207 (15) 1,250 (17)
Secondary school 722 (52) 3,757 (50)
Secondary school with leaving exam 305 (22) 1,593 (21)
Student 5 (0) 28 (0)
University 91 (7) 443 (6)
Missing 57 (4) 378 (5)
* Chi-squared, p < 0.05.
Table 2 Means and Spearman correlations for pollutants and temperature, April 1994 through March 1999 (n = 1,796 days).
Arithmetic mean 14-Day PAH 14-Day PM2.5 3-Day temperature 45-Day temperature
Prachatice
14-Day PAH (ng/m3) 62.77 1.00 0.56 −0.67 −0.70
14-Day PM2.5 (ng/m3) 18.49 1.00 −0.45 −0.47
3-Day temperature (°C) 7.58 1.00 0.83
45-Day temperature (°C) 7.61 1.00
Teplice
14-Day PAH (ng/m3) 70.00 1.00 0.79 −0.80 −0.75
14-Day PM2.5 (ng/m3) 28.80 1.00 −0.54 −0.58
3-Day temperature (°C) 9.54 1.00 0.86
45-Day temperature (°C) 9.52 1.00
All correlations are significant at p < 0.0001.
Table 3 Adjusteda percent changes in cord lymphocyte outcome for increments of 100 ng/m3 PAH and 25 μg/m3 PM2.5 average during 14 days before birth: impact of adjustment for meteorologic variables.
Models adjusted for the following meteorologic variables
45-Day average temperature
3-Day average temperature
Season
3-Day average temperature + season
Lymphocyte Air pollution Percent change (95% CI) p-Value Percent change (95% CI) p-Value Percent change (95% CI) p-Value Percent change (95% CI) p-Value
CD3 100 ng/m3 PAHs −2.43 (−4.74 to −0.12) 0.04 −2.74 (−4.81 to −0.66) 0.01 −4.33 (−6.44 to −2.22) < 0.0005 −3.26 (−5.55 to −0.98) 0.01
25 μg/m3 PM2.5 −1.95 (−4.00 to 0.10) 0.06 −2.34 (−4.29 to −0.39) 0.02 −3.12 (−5.05 to −1.19) < 0.0005 −1.86 (−3.81 to 0.09) 0.06
CD4 100 ng/m3 PAHs −2.33 (−4.12 to −0.53) 0.01 −2.68 (−4.34 to −1.03) < 0.0005 −3.86 (−5.54 to −2.18) < 0.0005 −3.12 (−4.94 to −1.29) 0.0005
25 μg/m3 PM2.5 −2.37 (−4.04 to −0.69) 0.01 −2.71 (−4.35 to −1.07) < 0.0005 −3.18 (−4.79 to −1.56) < 0.0005 −2.27 (−3.93 to −0.60) 0.01
CD8 100 ng/m3 PAHs −1.01 (−1.85 to −0.18) 0.02 −0.80 (−1.58 to −0.03) 0.04 −1.23 (−1.96 to −0.51) < 0.0005 −1.01 (−1.82 to −0.19) 0.02
25 μg/m3 PM2.5 −0.16 (−1.08 to 0.75) 0.73 −0.07 (−0.93 to 0.80) 0.88 −0.39 (−1.21 to 0.43) 0.35 −0.02 (−0.89 to 0.85) 0.96
CD4:CD8 100 ng/m3 PAHs 0.21 (−0.08 to 0.49) 0.15 0.16 (−0.19 to 0.32) 0.27 0.07 (−0.19 to 0.32) 0.61 0.11 (−0.18 to 0.40) 0.46
25 μg/m3 PM2.5 0.09 (−0.20 to 0.37) 0.55 0.05 (−0.25 to 0.35) 0.73 0.03 (−0.24 to 0.29) 0.83 0.06 (−0.24 to 0.36) 0.72
CD19 100 ng/m3 PAHs 2.11 (0.74 to 3.45) 0.003 2.34 (1.04 to 3.63) < 0.0005 1.87 (0.65 to 3.08) 0.003 1.69 (0.36 to 3.01) 0.01
25 μg/m3 PM2.5 1.77 (0.56 to 2.98) < 0.0005 1.96 (0.79 to 3.12) < 0.0005 1.81 (0.70 to 2.92) < 0.0005 1.55 (0.39 to 2.71) 0.01
NK 100 ng/m3 PAHs 0.27 (−2.18 to 2.71) 0.83 0.39 (−1.77 to 2.55) 0.72 2.45 (0.22 to 4.68) 0.03 1.54 (−0.83 to 3.92) 0.20
25 μg/m3 PM2.5 0.13 (−1.76 to 2.02) 0.89 0.37 (−1.39 to 2.13) 0.68 1.28 (−0.54 to 3.10) 0.17 0.27 (−1.53 to 2.06) 0.77
a Adjusted for district, year and time of birth, labor medication and duration, number of previous pregnancies, maternal education, and maternal active smoking or exposure to second-hand smoke.
==== Refs
References
Afoke AO Eeg-Olofsson O Hed J Kjellman NI Lindblom B Ludvigsson J 1993 Seasonal variation and sex differences of circulating macrophages, immunoglobulins and lymphocytes in healthy school children Scand J Immunol 37 209 215 8094576
Binková B Lewtas J Míšková I Leníček J Šrám R 1995 DNA adducts and personal air monitoring of carcinogenic polycyclic aromatic hydrocarbons in an environmentally exposed population Carcinogenesis 16 1037 1046 7767962
Braun-Fahrländer C Vuille J Sennhauser F Neu U Kunzle T Grize L 1997 Respiratory health and long-term exposure to air pollutants in Swiss schoolchildren. SCARPOL Team. Swiss Study on Childhood Allergy and Respiratory Symptoms with Respect to Air Pollution, Climate and Pollen Am J Respir Crit Care Med 155 1042 1049 9116984
Burchiel SW Lauer FT McDonald JD Reed MD 2004 Systemic immunotoxicity in AJ mice following 6-month whole body inhalation exposure to diesel exhaust Toxicol Appl Pharmacol 196 337 345 15094304
Cole SR Hernan MA 2002 Int J Epidemiol 31 163 165 11914314
Courage CM 2002. Environmental tobacco smoke. In: Children’s Health and Environment: A Review of Evidence. A Joint Report from the European Environment Agency and the WHO Regional Office for Europe (Tamburlini G, von Ehrenstein OS, Bertollini R, eds). Copenhagen:European Environment Agency and the World Health Organization Regional Office for Europe, 79–98.
D’Arena G Musto P Cascavilla N Di Giorgio G Fusilli S Zendoli F 1998 Flow cytometric characterization of human umbilical cord blood lymphocytes: immunophenotypic features Haematologica 83 197 203 9573672
Dejmek J Selevan S Beneš I Solansky I Šrám R 1999 Fetal growth and maternal exposure to particulate matter during pregnancy Environ Health Perspect 107 475 480 10339448
Dejmek J Solansky I Benes I Lenicek J Šrám RJ 2000 The impact of polycyclic aromatic hydrocarbons and fine particles on pregnancy outcome Environ Health Perspect 108 1159 1164 11133396
Devereux G Barker RN Seaton A 2002 Antenatal determinants of neonatal immune responses to allergens Clin Exp Allergy 32 43 50 12002736
Diaz-Sanchez D Proietti L Polosa R 2003 Diesel fumes and the rising prevalence of atopy: an urban legend? Curr Allergy Asthma Rep 3 146 152 12562554
DiFranza JR Aligne CA Weitzman M 2004 Prenatal and post-natal environmental tobacco smoke exposure and children’s health Pediatrics 113 1007 1015 15060193
Garcia LA Garcia GA Martinez PD Langle AE Fernandez PS Vazquez TL 1995 Reference values of T-lymphocyte subsets in umbilical cord blood Rev Alerg Mex 42 60 63 7552186
Glinianaia SV Rankin J Bell R Pless-Mulloli T Howel D 2004 Particulate air pollution and fetal health: a systematic review of the epidemiologic evidence Epidemiology 15 36 45 14712145
Harris D Schumacher M Rychlik S Booth A Acevedo A Rubinstein P 1994 Collection, separation and cryop-reservation of umbilical cord blood for use in transplantation Bone Marrow Transplant 13 135 143 8205082
Hazenkamp-von Arx ME Gotschi Fellmann T Oglesby L Ackermann-Liebrich U Gislason T Heinrich J 2003 PM2.5 assessment in 21 European study centers of ECRHS II: method and first winter results J Air Waste Manag Assoc 53 617 628 12774995
Hertz-Picciotto I Dostal M Dejmek J Selevan SG Wegienka G Šrám RJ 2002 Air pollution and distributions of lymphocyte immunophenotypes in cord and maternal blood at delivery Epidemiology 13 172 183 11880758
Jedrychowski W Bendkowska I Flak E Penar A Jacek R Kaim I 2004 Estimated risk for altered fetal growth resulting from exposure to fine particles during pregnancy: an epidemiologic prospective cohort study in Poland Environ Health Perspect 112 1398 1402 15471732
Leonardi GS Houthuijs D Steerenberg PA Fletcher T Armstrong B Antova T 2000 Immune biomarkers in relation to exposure to particulate matter: a cross-sectional survey in 17 cities of Central Europe Inhal Toxicol 12 suppl 4 1 14 12881884
Levi FA Canon C Touitou Y Reinberg A Mathe G 1988 Seasonal modulation of the circadian time structure of circulating T and natural killer lymphocyte subsets from healthy subjects J Clin Invest 81 407 413 3257498
Li N Sioutas C Cho A Schmitz D Misra C Sempf J 2003 Ultrafine particulate pollutants induce oxidative stress and mitochondrial damage Environ Health Perspect 111 455 460 12676598
Little RJA Rubin DB 2002. Statistical Analysis with Missing Data. 2nd ed. New York:Wiley Interscience, 248–251.
Naumova YY Eisenreich SJ Turpin BJ Weisel CP Morandi MT Colome SD 2002 Polycyclic aromatic hydrocarbons in the indoor and outdoor air of three cities in the U.S Environ Sci Technol 36 2552 2559 12099449
Nel AE Diaz-Sanchez D Li N 2001 The role of particulate pollutants in pulmonary inflammation and asthma: evidence for the involvement of organic chemicals and oxidative stress Curr Opin Pulm Med 7 20 26 11140402
Noakes PS Holt PG Prescott SL 2003 Maternal smoking in pregnancy alters neonatal cytokine responses Allergy 58 1053 1058 14510725
Perera F Hemminki K Jedrychowski W Whyatt R Campbell U Hsu Y 2002 In utero DNA damage from environmental pollution is associated with somatic gene mutation in newborns Cancer Epidemiol Biomarkers Prev 11 1134 1137 12376523
Perera FP Rauh V Tsai WY Kinney P Camann D Barr D 2003 Effects of transplacental exposure to environmental pollutants on birth outcomes in a multiethnic population Environ Health Perspect 111 201 205 12573906
Perera FP Tang D Yi-Hsuan T Cru LA Borjas M Bernert T 2004 Antimony: biomarkers in maternal and newborn blood indicate heightened fetal susceptibility to procarcinogenic DNA damage Environ Health Perspect 112 1133 1136 15238289
Petry T Schmid P Schlatter CH 1996 The use of toxic equivalency factors in assessing occupational and airborne mixtures of polycyclic aromatic hydrocarbons (PAHs) Chemosphere 32 639 648 8867146
Pinto JP Lefohn AS Shadwick DS 2004 Spatial variability of PM2.5 in urban areas in the United States J Air Waste Manag Assoc 54 440 449 15115373
Pinto JP Stevens RK Willis RD Kellogg R Mamane Y Nowak J 1998 Czech air quality monitoring and receptor modelling study Environ Sci Technol 32 843 854
Pittard W Schleich D Geddes K Sorensen R 1989 Newborn lymphocyte subpopulations: the influence of labor Am J Obstet Gynecol 160 151 154 2521424
Santagostino A Garbaccio G Pistorio A Bolis V Camisasca G Pagliaro P 1999 An Italian national multicenter study for the definition of reference ranges for normal values of peripheral blood lymphocyte subsets in healthy adults Haematologica 84 499 504 10366792
Schaberg T Theilacker C Nitschke OT Lode H 1997 Lymphocyte subsets in peripheral blood and smoking habits Lung 75 387 394 9330248
Schultz C Reiss I Bucsky P Gopel W Gembruch U Ziesenitz S 2000 Maturational changes of lymphocyte surface antigens in human blood: comparison between fetuses, neonates and adults Biol Neonate 78 2 77 82 10970998
Shumway RH Stoffer DS 2000. Time Series Analysis and Its Applications. New York:Springer-Verlag, 329–333.
Siwinska E Mielzynska D Bubak A Smolik E 1999 The effect of coal stoves and environmental tobacco smoke on the level of urinary 1-hydroxypyrene Mutat Res 445 147 153 10575425
Skachkova MA Skachkov MV Smolyagin AI Boev VM Vereshchagin NN Mikhailova IV 2001 Immune and interferon status in schoolchildren living in cities with different anthropogenic load Bull Exp Biol Med 131 371 373 11550029
Somers CM McCarry BE Malek F Quinn JS 2004 Reduction of particulate air pollution lowers the risk of heritable mutations in mice Science 304 1008 1010 15143280
Šrám RJ Binková B Rössner P Rubeš J Topinka J Dejmek J 1999 Adverse reproductive outcomes from exposure to environmental mutagens Mutat Res 428 203 215 10517994
Tollerud DJ Clark JW Brown LM Neuland CY Mann DL Pankiw-Trost LK 1989 The effects of cigarette smoking on T cell subsets. A population-based survey of healthy Caucasians Am Rev Respir Dis 139 1446 1451 2786361
van Eeden SF Hogg JC 2002 Systemic inflammatory response induced by particulate matter air pollution: the importance of bone-marrow stimulation J Toxicol Environ Health A 65 1597 1613 12396870
Wang X Ding H Ryan L Xu X 1997 Association between air pollution and low birthweight: a community-based study Environ Health Perspect 105 514 520 9222137
Wilhelm M Ritz B 2003 Residential proximity to traffic and adverse birth outcomes in Los Angeles County, California, 1994–1996 Environ Health Perspect 111 207 216 12573907
Zhao Y Dai ZP Lv P Gao XM 2002 Phenotypic and functional analysis of human T lymphocytes in early second- and third-trimester fetuses Clin Exp Immunol 129 302 308 12165087
|
16203253
|
PMC1281286
|
CC0
|
2021-01-04 23:41:28
|
no
|
Environ Health Perspect. 2005 Oct 15; 113(10):1391-1398
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.7610
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences 10.1289/ehp.7809ehp0113-00139916203254ResearchChildren's HealthPhthalates in Indoor Dust and Their Association with Building Characteristics Bornehag Carl-Gustaf 123Lundgren Björn 1Weschler Charles J. 24Sigsgaard Torben 5Hagerhed-Engman Linda 1Sundell Jan 21 Swedish National Testing and Research Institute, Borås, Sweden2 International Centre for Indoor Environment and Technology, Technical University of Denmark, Lyngby, Denmark3 Department of Public Health Sciences, Karlstad University, Karlstad, Sweden4 Environmental and Occupational Health Sciences Institute, University of Medicine and Dentistry of New Jersey/Robert Wood Johnson Medical School and Rutgers University, Piscataway, New Jersey, USA5 Department of Environmental and Occupational Medicine, Aarhus University, Aarhus, DenmarkAddress correspondence to C.-G. Bornehag, Public Health Sciences, Karlstad University, 651 88 Karlstad, Sweden. Telephone: 46-54-700-25-40. Fax: 46-54-700-22-20. E-mail:
[email protected] authors declare they have no competing financial interests.
10 2005 1 6 2005 113 10 1399 1404 1 12 2004 1 6 2005 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright. In a recent study of 198 Swedish children with persistent allergic symptoms and 202 controls without such symptoms, we reported associations between the symptoms and the concentrations of n-butyl benzyl phthalate (BBzP) and di(2-ethylhexyl) phthalate (DEHP) in dust taken from the childrens’ bedrooms. In the present study we examined associations between the concentrations of different phthalate esters in the dust from these bedrooms and various characteristics of the home. The study focused on BBzP and DEHP because these were the phthalates associated with health complaints. Associations have been examined using parametric and nonparametric tests as well as multiple logistic regression. For both BBzP and DEHP, we found associations between their dust concentrations and the amount of polyvinyl chloride (PVC) used as flooring and wall material in the home. Furthermore, high concentrations of BBzP (above median) were associated with self-reported water leakage in the home, and high concentrations of DEHP were associated with buildings constructed before 1960. Other associations, as well as absence of associations, are reported. Both BBzP and DEHP were found in buildings with neither PVC flooring nor wall covering, consistent with the numerous additional plasticized materials that are anticipated to be present in a typical home. The building characteristics examined in this study cannot serve as complete proxies for these quite varied sources. However, the associations reported here can help identify homes where phthalate concentrations are likely to be elevated and can aid in developing mitigation strategies.
BBzPbuilding characteristicsDEHPDnBPhomesPVC flooringsources
==== Body
For almost a quarter-century, phthalate esters have been recognized as major indoor pollutants (Clausen et al. 2003; Fromme et al. 2004; Rudel et al. 2003; Wensing et al. 2005; Weschler 1980, 1984). This reflects their widespread use, primarily as plasticizers, in products ranging from polyvinyl chloride (PVC) flooring to vinyl toys. Worldwide phthalate production has been estimated to exceed 3.5 million tons/year (Cadogan and Howick 1996). Different phthalate esters have different chemical and physical properties and, consequently, have different uses. Di(2-ethylhexyl) phthalate (DEHP) accounts for roughly 50% of overall phthalate production, although this percentage has been decreasing in recent years. Most of the current DEHP production is used in PVC products, including PVC flooring, where it typically constitutes 30% of PVC by weight [Cadogan and Howick 1996; Kavlock et al. 2002b; National Toxicology Program (NTP) 2003]. The production of n-butyl benzyl phthalate (BBzP) and di-n-butyl phthalate (DnBP) is about one-tenth that of DEHP. BBzP is also used as a plasticizer for PVC flooring, as well as for vinyl tile, carpet tiles, and artificial leather and in certain adhesives (Kavlock et al. 2002a). DnBP is used in latex adhesives, as a plasticizer in cellulose plastics, as a solvent for certain dyes, and, to a lesser extent than DEHP, as a plasticizer in PVC (Kavlock et al. 2002c).
Health concerns related to phthalate ester exposures have focused primarily on cancer and reproductive effects (Kavlock et al. 2002a, 2002b, 2002c; NTP 2003). However, phthalate exposures have also been postulated to have a role in the pathogenesis of asthma (Oie et al. 1997), and plasticized indoor materials have been associated with the development of bronchial obstruction in young children (Jaakkola et al. 1999). We recently reported an association between asthma and allergies in children and phthalate concentrations in dust collected from the children’s bedrooms (Bornehag et al. 2004b). The geometric mean concentrations of BBzP were higher in dust from rooms of children with rhinitis compared with controls (0.237 vs. 0.157 mg/g dust, p = 0.001) and of children with eczema compared with controls (0.224 vs. 0.157 mg/g dust, p = 0.001). Regarding DEHP, dust from rooms of children with asthma had a higher geometric mean concentration compared with that of controls (0.966 vs. 0.741 mg/g dust, p = 0.022). For these associations, a dose–response relationship was supported by trend analyses (p < 0.05) when the phthalate concentrations in dust were divided into quartiles. DnBP was not associated with doctor-diagnosed disease. Related to these findings, various di- and monophthalate esters have been shown to have an adjuvant effect in a mouse model (Larsen et al. 2001a, 2001b, 2002, 2003), to enhance the production of interleukin-4 in mouse T-cells (Lee et al. 2004), and to potentiate the response of allergic effector cells (Glue et al. 2005).
The aim of the present study was to examine associations between the concentration of phthalates in dust from Swedish homes and selected building characteristics.
Materials and Methods
Selection of buildings.
The study is based on 390 homes that participated in the nested case–control study of 400 children in Sweden (Bornehag et al. 2004a). The cases and controls were selected from phase 1 of the Dampness in Buildings and Health (DBH) study. This was a cross-sectional questionnaire study soliciting health and environmental information regarding all 14,077 children 1–6 years of age in the county of Värmland, Sweden; responses were obtained for 10,852 children (Bornehag et al. 2003, 2005a).
The selection criteria for the cases in DBH phase 2 were, in the initial questionnaire, reports of at least two symptoms of “wheezing during last 12 months without a cold,” “rhinitis during last 12 months without a cold,” and “eczema during last 12 months.” In the follow-up questionnaire 1.5 years later, cases had to report at least two of three possible symptoms. Inclusion criteria for the controls were no symptoms in the first questionnaire and no symptoms in the follow-up questionnaire. Both cases and controls must not have rebuilt their homes because of moisture problems, and not have changed residence since the first questionnaire. This process ultimately yielded 198 cases and 202 controls, living in 390 homes.
Factors associated with participation in the study included a greater number of health problems in the case families. Furthermore, in both case and control families, participation was associated with more health-conscious lifestyle factors such as nonsmoking parents and cotton diapers for the child. Higher socioeconomic status, as a selection factor, was indicated by a higher participation among families living in single-family houses compared with multifamily houses, and higher participation among families with two parents living in the home compared with single parent homes (Bornehag et al., unpublished data).
Building investigations.
There were 10 pairs of siblings among the 400 children; hence, they lived in 390 buildings. Between October 2001 and April 2002, six professional inspectors performed visual inspections and indoor air quality assessments, including dust sampling, in the homes. The inspectors were blinded to case–control status of the children living in the homes. During these investigations, a checklist was followed regarding factors such as the type of building, building construction, building materials, type of ventilation, and mold and moisture problems.
For each residence, 1-week average ventilation rates of both the whole home and the bedroom of the index child were measured using a passive tracer gas method (Nordtest 1997).
Phthalates in dust.
Samples of dust from 390 homes were collected from moldings and shelves in the children’s bedroom. All dust was sampled during heating season from October 2001 to April 2002. The dust was collected on 90-mm membrane filters made of pure cellulose in holders made of styrene-acrylonitrile polymer mounted on a sampler made of polypropylene (VacuuMark disposable nozzle; Petersen Bach, Bjerringbro, Denmark) connected to a vacuum cleaner. The filters were first packed in aluminum foil and then in a polyethylene bag and stored in a refrigerator for 2–3 days. The filter was weighed before and after sampling under controlled conditions. Before weighing, the filter samples were conditioned at 23°C and 50% relative humidity.
From the 390 homes there were 9 missing samples, 13 samples with errors in the laboratory analysis, and 6 samples with a negative dust weight. Consequently, there were 362 valid samples. Only filters with a net increase in weight of ≥ 25 mg were included in the present analysis; 346 of the 362 dust samples met this criterion.
The dust samples were extracted in pre-cleaned 10-mL glass vials for 30 min using 2 mL dichloromethane. This procedure was repeated, and the two extracts were then combined and transferred to 3-mL autosampler vials. Aliquots from these vials were injected into either a gas chromatograph/mass selective detector for phthalate identification or a gas chromatograph/flame ionization detector for quantitation. The dust concentrations (mg/g dust) were determined for six phthalates: diethyl phthalate (DEP), diisobutyl phthalate (DIBP), DnBP, BBzP, DEHP, and diisononyl phthalate (DINP). For further details regarding chemical analyses, see Bornehag et al. (2004b).
Statistical method.
We performed analyses of potential associations between concentrations of phthalates in dust and building characteristics using nonparametric tests (Mann-Whitney U-test). Log-transformed, normally distributed concentrations (where concentrations below the detection limit have been excluded) were tested with parametric tests (t-test) and Pearson correlation (r). The analyses were considered statistically significant when p < 0.05. The concentrations are reported as medians, as arithmetic means, and as geometric means with 95% confidence intervals (CIs). The CIs were calculated with a back-transform of mean log ± 2 × SE.
We used multiple logistic regression (backward elimination) for analyzing associations between a high phthalate concentration in dust (above median concentration) and building characteristics: PVC as flooring material in the child’s bedroom (no, yes), type of building (single-family house, multifamily houses), construction period (before 1960before 1960–1983, after 1983), and ventilation rate (in quartiles). Data on water leakage in the home during the previous 3 years was collected in the DBH phase 1 questionnaire, 18–24 months before the exposure measurements were conducted.
The study was approved by the ethics committee in Orebro, Sweden.
Results
Descriptions of the 390 homes included in this case–control study are presented in Table 1. The buildings were primarily single-family houses, and almost 50% of these buildings were constructed before 1960. PVC flooring was the most commonly used flooring material, followed by wood flooring and laminate. There was little difference in the frequency of PVC use between single-family houses and chain houses, but PVC was more commonly used in multifamily houses compared with either of these.
Table 2 lists the phthalate concentrations in dust collected from 346 children’s bedrooms; these were the dust samples that met the criteria for reliable analyses (see “Materials and Methods”). The most frequently identified phthalate was DEHP, which was found in nearly all samples; DnBP was found in 89%, and BBzP was found in 79% of the samples. DEHP also had the highest average concentration in the dust, with a median concentration of 0.77 mg/g dust. All other phthalates were detected at median concentrations below 0.2 mg/g dust. DnBP, BBzP, and DEHP were not highly correlated with each other (r 2 < 0.35).
Surface materials.
The distribution of surface materials on floors and walls in the bedrooms of cases and controls are presented in Table 3. Significantly more PVC and less wood flooring were found among the cases. This difference is due partly to selection bias. However, the earlier reported association between phthalates in dust and asthma/allergic symptoms among children is not a consequence of either selection bias or active avoidance of specific flooring materials because of allergic disease in the family (Bornehag et al. 2005c). In the instance of vinyl as a wall material, no difference was found between cases and controls, and no selection bias was found. Additionally, more painted wallpaper and less painted glass fiber wallpaper were found among the cases.
As shown in the last three columns of Table 2, the median concentrations of BBzP and DEHP in dust were significantly higher in bedrooms with PVC flooring compared with other flooring materials. In the case of the other four identified phthalates, there were no significant differences between bedrooms with and without PVC flooring. The more rooms with PVC flooring in the home, the higher the geometric mean dust concentrations of both DEHP and BBzP. The particular characteristics of the groups are as follows: group I, no PVC in the child’s bedroom and no PVC in other rooms (parent’s bedroom, living room, kitchen, and hall); II, no PVC in child’s bedroom and PVC in at least one of the other rooms; III, PVC in the child’s bedroom and no PVC in other rooms; IV, PVC in the child’s bedroom and PVC in at least one of the other rooms; V, PVC in the child’s bedroom and PVC in all other rooms. This is illustrated in Figure 1. The association between PVC flooring and the concentration of phthalates in the dust was stronger for BBzP than for DEHP.
The data in Table 2 and Figure 1 also illustrate that PVC flooring is not the only source of BBzP and DEHP in the dust. When there is no PVC flooring in the bedroom, the median amount of DEHP in the dust is 0.7 mg/g; when there is no PVC flooring anywhere in the house, the median amount of DEHP in the dust is 0.55 mg/g. Hence, there is a large background concentration of DEHP to which the DEHP from PVC flooring is contributing. The background concentration for BBzP is not as large. When there is no PVC flooring in the bedroom, its dust concentration is 0.089 mg/g; when there is no PVC flooring anywhere, its dust concentration is comparable (0.100 mg/g).
Of the 26 homes with vinyl on walls in the child’s bedroom, 12 had PVC as flooring material in the same room. Homes with vinyl on the wall in the child’s bedroom had a higher concentration of DEHP in the dust compared with bedrooms that had other types of wall coverings [1.24 mg/g dust (n = 26; 95% CI, 0.79–1.96) vs. 0.74 mg/g dust (n = 319; 95% CI, 0.67–0.83), p = 0.009 by t-test]. There was no significant difference with wall coverings for BBzP. The highest concentration of DEHP was found in bedrooms with a combination of PVC on the floor and vinyl on the walls (Figure 2).
Type of building and construction period.
The concentrations of DnBP, BBzP, and DEHP were higher in multifamily houses than in single-family houses, but the differences did not reach significance. Neither were there any significant differences in phthalate concentrations between buildings from different construction periods (i.e., before 1960i.e., before 1961–1983, and after 1983). However, when including only homes with PVC as flooring material in the child’s bedroom, the geometric mean concentrations of DEHP and BBzP were significantly higher in buildings erected before 1960 [DEHP: 1.25 mg/g dust (n = 72; 95% CI, 0.97–1.61); BBzP: 0.25 mg/g dust (n = 60; 95% CI, 0.19–0.33)] compared with buildings constructed after 1983 [DEHP: 0.79 mg/g dust (n = 32; 95% CI, 0.61–1.03); BBzP: 0.15 mg/g dust (n = 32; 95% CI, 0.11–0.20); both p < 0.05 by t-test].
Type of foundation.
Different types of foundation may produce different moisture loads in a building. Moisture from the ground and/or construction materials such as concrete may have an impact on PVC flooring via various degradation processes (e.g., hydrolysis of phthalate plasticizers). Data on the type of foundation were available only for single-family houses. In such buildings, a significantly higher geometric mean dust concentration of BBzP was found in buildings with a concrete slab on the ground as the foundation [0.20 mg/g dust (n = 72; 95% CI, 0.16–0.26)] compared with buildings with a basement [0.13 mg/g dust (n = 90; 95% CI, 0.11–0.16); p < 0.01 by t-test]. Furthermore, buildings with a concrete slab on the ground had a higher geometric mean concentration of BBzP compared with those with a crawl space; however, the difference did not reach significance (p = 0.077 by t-test).
Ventilation.
There was no association between the geometric mean concentration of BBzP and the mean ventilation rate (during a week) in the child’s bedroom, but the geometric mean concentration of DEHP was higher in buildings with higher ventilation rates. No association was found between the type of ventilation system and the concentration of phthalates in dust. There was no association between phthalate concentrations in dust and the relative humidity or the temperature in the child’s bedroom.
Self-reported water leakage.
Homes with self-reported water leakage during the preceding 3 years had higher geometric mean concentrations of BBzP and DEHP in dust than did buildings without such reports [BBzP: 0.19 mg/g dust (n = 67; 95% CI, 0.16–0.24) vs. 0.15 mg/g dust (n = 202; 95% CI, 0.13–0.17), p = 0.049 by t-test; DEHP: 0.93 mg/g dust (n = 78; 95% CI, 0.77–1.13) vs. 0.75 mg/g dust (n = 242; 95% CI, 0.66–0.85), p = 0.084 by t-test]. When the analysis included buildings with only PVC as the flooring material in the child’s bedroom, the association became somewhat stronger (by t-test: BBzP, p = 0.012; DEHP, p = 0.062).
Multivariate analyses.
Table 4 displays associations between building characteristics and the dust concentrations of BBzP or DEHP as determined by multiple logistic regression models. (Data on type of foundation were not included in the models because such data were available only for single-family houses.) In these analyses the dependent variable (i.e., the concentration of phthalate in the dust) was divided into two groups: low, below the median concentration, and high, above the median concentration. In a backward stepwise logistic regression, high BBzP concentration was associated with PVC flooring and, to a lesser degree, with self-reported water leakage during the previous 3 years. Elevated DEHP concentration was associated with PVC flooring and with home construction before 1960. In the univariate multiple logistic regression, ventilation rate was associated with DEHP in dust. However, in the adjusted model such an association disappeared. Neither type of building nor vinyl wall covering was included in the final models. When type of foundation was included in the analyses (data available only for single-family houses), the associations in Table 4 remained.
Discussion
Measured concentrations.
The concentrations of phthalate esters are somewhat higher in our study than in some of the other studies (Table 5). This may reflect more frequent use of PVC flooring in Sweden than in other countries. Additionally, we suspect that dust samples collected by filter methods contain smaller dust particles than those obtained from vacuum cleaner bags; for semivolatile organic compounds associated with the dust via sorption processes, this would mean higher dust concentrations for filter samples compared with samples from vacuum cleaner bags. The surface from which the dust is collected can also influence the resulting chemical constituents of the dust. The highest median DEHP concentration in Table 5 is for samples collected from flooring in schools (Clausen et al. 2003). In a subsequent study, Clausen et al. (2004) presented results that indicate direct transfer of DEHP from PVC flooring to dust in contact with the PVC flooring. Finally, the method of extraction and analysis can also influence the measured concentrations.
Associated building characteristics.
High concentrations (above median) of BBzP and DEHP in dust were associated with PVC flooring; however, BBzP was more strongly associated with PVC than was DEHP. Furthermore, BBzP was associated with self-reported water leakage, and DEHP was, to a lesser degree, associated with construction before 1960.
PVC flooring appears to be a source for both BBzP and DEHP in settled dust. The more rooms with PVC, the higher the concentration of these phthalates in dust. However, for both the phthalates, there is a “background” concentration (geometric means: DEHP, 0.5 mg/g dust; BBzP, 0.1 mg/g dust) in buildings with no PVC flooring (except for the bathroom). This is consistent with other known sources for phthalates in indoor dust.
Vinyl materials on walls were associated with a higher concentration of DEHP, but not BBzP, in dust based on the univariate analysis. However, the association disappeared in the multivariate model. This could reflect the few rooms with vinyl on walls and the fact that most of the bedrooms with vinyl on walls had PVC as flooring materials. Emission of DEHP from vinyl materials has been shown in other studies (Afshari et al. 2004; Fujii et al. 2003).
The correlation between DnBP, BBzP, and DEHP was not high, which implies that PVC materials can be plasticized with one or more of these phthalates, but that it is not routinely plasticized with a fixed ratio of these.
Ventilation rate.
In crude analysis, there was an association between a high DEHP concentration in dust and a higher mean ventilation rate in the child’s bedroom. However, in the adjusted analysis, such an association disappeared, probably because of confounding mechanisms; for example, a higher ventilation rate is associated with an earlier construction period as well as several other building-related factors, as described elsewhere (Bornehag et al. 2005b).
Construction period.
Buildings constructed before 1960 were found to have higher concentrations of DEHP than buildings from later periods. Such a finding could be due to a larger content of DEHP in older flooring materials (PVC), but there was no correlation between the concentration of different phthalates in dust and the age of the PVC flooring (data not shown). However, the Swedish Chemicals Inspectorate (KemI) reports that the total consumption of DEHP has decreased in Sweden over the past years (KemI 2004).
Water leakage and change of flooring materials.
In the multiple regression analyses, water leakage during the previous 3 years was associated with an elevated concentration of BBzP in the dust. It should be stressed that the data regarding water leakage was self-reported by the parents, and that there was an 18- to 24-month interval between reports of water leakage and the exposure measurements. The association could be due to degradation of PVC floors caused by moisture/water and, in some cases, highly basic (high pH) moist concrete surfaces. On the other hand, reports of water damage may be a proxy for renovations in which old flooring materials have been replaced by new materials. Thus, there are several possible explanations regarding the association between BBzP concentration in dust and water leakage.
In this study we focused on only two indoor sources of phthalates, PVC flooring and vinyl wall covering. A typical home contains numerous other materials that are plasticized with phthalates. Examples include furniture covered with synthetic leather, vinyl raincoats, vinyl notebook covers, toys and sports equipment made of PVC, vinyl lampshades, vinyl garment bags, PVC containers, and PVC insulation on telephone, television, and computer cables. The building characteristics examined in this study cannot be proxies for these quite varied sources. However, the associations reported in this study can help to estimate, without chemical analyses, whether high or low BBzP and DEHP levels can be anticipated in a home’s dust.
Conclusions
The main finding from this study is that the concentrations of BBzP and DEHP in dust are associated with the amount of PVC/vinyl used as flooring and wall material in the home, but that there are also many other sources of these phthalates. Although PVC flooring and vinyl on walls do not fully explain the concentration of phthalates in dust, occurrences of such materials are associated with higher concentrations of DEHP and BBzP in dust indoors. There are also associations between the concentration of BBzP in bedroom dust and water leakage in the previous 3 years, as well as higher levels of DEHP in bedroom dust and buildings constructed before 1960. The reason for the association between high BBzP concentration and self-reported water leakage is not obvious. The finding that DEHP was higher for buildings erected before 1960 could reflect higher fractional concentrations in older products or higher emission rates as products degrade.
The study was supported by the Swedish Research Council for Environment, Agricultural Sciences and Spatial Planning (Formas), Swedish Asthma and Allergy Association’s Research Foundation, the Swedish Foundation for Health Care Sciences and Allergy Research, and the European Council for Plasticisers and Intermediates.
Figure 1 Geometric mean concentration (95% CI) of (A) DEHP and (B) BBzP in surface dust (mg/g dust) in homes with different combinations of flooring material.
Figure 2 Geometric mean concentration (95% CI) of DEHP in surface dust (mg/g dust) in homes with different combinations of flooring (PVC vs. no PVC) and wall materials (vinyl vs. no vinyl).
Table 1 Description of the 390 homes in the case–control study.
No. of buildings with different characteristics (%)
Building characteristicsa Single-family houses Chain houses Multifamily houses Total
No. of buildings in the study 323 (82.8) 23 (5.9) 44 (11.3) 390 (100)
Flooring material in child’s bedroom
PVC 167 (52.0) 12 (52.2) 32 (72.7) 211 (54.4)
Wood/parquet 108 (33.6) 7 (30.4) 5 (11.4) 120 (30.9)
Laminate 34 (10.6) 2 (8.7) 3 (6.8) 39 (10.1)
Linoleum 8 (2.5) 1 (4.3) 4 (9.1) 13 (3.4)
Wall-to-wall carpet 3 (0.9) 1 (4.3) 0 (0) 4 (1.0)
Other 1 (0.3) 0 (0) 0 (0) 3 (0.8)
Wall material in child’s bedroom
Wallpaper 230 (71.2) 12 (52.2) 39 (88.6) 281 (72.0)
Painted wallpaper 41 (12.7) 5 (21.7) 4 (9.1) 50 (12.8)
Painted glass fiber 27 (8.3) 4 (17.4) 0 (0) 31 (7.9)
Vinyl 29 (9.0) 4 (17.4) 4 (9.1) 37 (9.5)
Wood 11 (3.4) 0 (0) 0 (0) 11 (2.8)
Textile 1 (0.3) 0 (0) 0 (0) 1 (0.2)
Construction period
Before 1940 101 (31.1) 1 (4.3) 7 (15.9) 109 (27.9)
1940–1960 58 (18.0) 2 (8.7) 10 (22.7) 70 (17.9)
1961–1970 34 (10.5) 4 (17.4) 13 (29.5) 51 (13.1)
1971–1976 46 (14.2) 6 (26.1) 3 (6.8) 55 (14.1)
1977–1983 48 (14.9) 2 (8.7) 1 (2.3) 51 (13.1)
1984–1993 29 (9.0) 6 (26.1) 7 (15.9) 42 (10.8)
After 1993 7 (2.2) 2 (8.7) 3 (6.8) 12 (3.1)
Ventilation system
Natural including kitchen fan 233 (74.4) 6 (28.6) 10 (22.7) 249 (65.9)
Mechanical exhaust 51 (16.3) 11 (52.4) 30 (68.2) 92 (24.3)
Mechanical exhaust and supply 29 (9.3) 4 (19.0) 4 (9.1) 37 (9.8)
Self-reported water leakageb
Yes, during previous 3 years 68 (21.5) 8 (34.8) 7 (16.7) 83 (21.7)
No 222 (70.0) 13 (56.5) 28 (66.7) 263 (68.8)
Don’t know 27 (8.5) 2 (8.7) 7 (16.7) 36 (9.4)
a Data from inspections of the buildings in DBH phase 2 except for flooding, which was collected in the first questionnaire in DBH phase 1.
b Data from questionnaire investigation in DBH phase 1, which was collected 18 months before the exposure measurements were conducted.
Table 2 Concentrations (mg/g dust) for different phthalates in settled dust from 346 bedrooms.
All samples (n = 346)
Type of flooringa (median mg/g dust)
Phthalate Above detection limitb [n (%)] Mean Median Min–Max 95th percentile No PVC (n = 157) PVC (n = 187) p-Valuec
DEP 32 (9.2) 0.031 0.000 0.000–2.425 0.115 0.000 0.000 0.241
DINP 173 (50.0) 0.639 0.041 0.000–40.667 1.930 0.000 0.082 0.394
DIBP 188 (54.3) 0.097 0.045 0.000–3.810 0.311 0.042 0.050 0.120
BBzP 272 (78.6) 0.319 0.135 0.000–45.549 0.599 0.089 0.192 < 0.001
DnBP 308 (89.0) 0.226 0.150 0.000–5.446 0.568 0.133 0.159 0.138
DEHP 343 (99.1) 1.310 0.770 0.000–40.459 4.069 0.700 0.868 0.001
Abbreviations: Max, maximum; Min, minimum.
a Type of flooring in the child’s bedroom.
b Number of samples with a concentration greater than the detection limits (0.040 mg/g dust).
c Mann-Whitney U-test regarding differences in phthalate concentration between bedrooms with and without PVC as flooring material.
Table 3 Frequency of surface materials in the child’s bedroom (floors and walls) among cases and controls [n (%)].
Surface material Cases Controls
Flooring material
PVC 118 (59.6) 97 (48.8)
Wood 47 (23.7) 76 (38.0)
Laminate 24 (12.1) 18 (9.0)
Linoleum 6 (3.0) 7 (3.5)
Wall-to-wall carpet 2 (1.0) 2 (1.0)
Other 1 (0.5) 0 (0)
Wall material
Wallpaper 143 (72.2) 142 (70.2)
Painted wallpaper 35 (17.7) 19 (9.4)
Painted glass fiber 12 (6.0) 21 (10.4)
Vinyl 19 (9.6) 20 (9.9)
Wood 4 (2.0) 8 (3.9)
Textile 0 (0) 1 (0.5)
Table 4 Association between concentration of phthalates in dust (> median) and building characteristics.
Odds ratio (95% CI)a
Factor No. BBzPb DEHPc
PVC as flooring
No 138 1.0 1.0
Yes 165 3.85 (2.37–6.24) 1.85 (1.15–2.98)
Vinyl as wall material
No 282 1.0 1.0
Yes 21 NS NS
Type of building
Single-family house 277 1.0 1.0
Multifamily house 26 NS NS
Construction period
Before 1960 144 NS 2.30 (1.17–4.52)
1960–1983 110 NS 1.09 (0.55–2.18)
After 1983 49 1.0 1.0
Ventilation rate in child’s bedroom
1st quartile 74 NS NS
2nd quartile 79 NS NS
3rd quartile 80 NS NS
4th quartile 70 1.0 1.0
Water leakage during previous 3 years
No 227 1.0 1.0
Yes 76 1.84 (1.05–3.22) NS
a Backward conditional logistic regression in two different models. Only significant variables included in the final model; variables with no significant contribution to the model have been eliminated (NS).
b Model 1: Dependent variable BBzP coded as 1 ≤median concentration and 2 > median concentration.
c Model 2: Dependent variable DEHP coded as 1 ≤median concentration and 2 > median concentration.
Table 5 Measurements of the concentration of phthalates in dust in different countries.
DEHP (μg/g dust)
BBzP (μg/g dust)
DnBP (μg/g dust)
Study Country No. 50tha 95tha 50tha 95tha 50tha 95tha Sampling technique
Present study Sweden 346 770 4,069 135 599 150 568 Surface dust above floor (filter)b
Pohner et al. 1997 Germany 272 450 2,000 — — — — “Fine dust”?
Oie et al. 1997 Norway 38 640c — 110c — 100c — Surface dust (filter)d
Butte et al. 2001 Germany 286 740 2,600 49 320 49 240 Vacuum cleaner bags
Becker et al. 2002 Germany 199 416 1,190 15 207 42 160 Vacuum cleaner bags
Clausen et al. 2003 Denmark 23 858 2,595 — — — — Floor dust (cyclone/glass bottle)
Rudel et al. 2003 USA 120 340 854e 45 277e 20 44e Surface dust (filter)d
Kersten and Reich 2003 Germany 65 600 1,600 19 230 47 180 Vacuum cleaner bags
Fromme et al. 2004 Germany 30 703 1,540 30 218 56 130 Vacuum cleaner bags
Becker et al. 2004 Germany 252 515 1,840 — — — — Vacuum cleaner bags
a 50th, 95th: 50th and 95th percentiles.
b Multiple surfaces excluding floors.
c Mean concentration.
d Multiple surfaces including floors.
e 90% percentile.
==== Refs
References
Afshari A Gunnarsen L Clausen PA Hansen V 2004 Emission of phthalates from PVC and other materials Indoor Air 14 120 128 15009418
Becker K Seiwert M Angerer J Heger W Koch HM Nagorka R 2004 DEHP metabolites in urine of children and DEHP in house dust Int J Hyg Environ Health 207 409 417 15575555
Becker K Seiwert M Kaus S Krause C Schulz C Seifert B 2002. German Environmental Survey 1998 (GerES III): pesticides and other pollutants in house dust. In: Proceedings of the 9th International Conference on Indoor Air Quality and Climate, 30 June–5 July, Monterey, California (Levin H, ed). Santa Cruz, CA:Indoor Air 2002: 883–887.
Bornehag CG Sundell J Hagerhed L Janson S 2003 Pet-keeping in early childhood and airway, nose and skin symptoms later in life Allergy 58 939 944 12911425
Bornehag CG Sundell J Hagerhed-Engman L Sigsggard T Janson S Aberg N 2005a “Dampness” at home and its association with airway, nose and skin symptoms among 10 851 preschool children in Sweden: a cross sectional study Indoor Air 15 suppl 10 48 55 10.1111/j.1600–0668.2005.00306.x.15926944
Bornehag CG Sundell J Sigsgaard T 2004a Dampness in buildings and health (DBH). Report from an on-going epidemiological investigation on the association between indoor environmental factors and health effects among children in Sweden Indoor Air 14 59 66 15330773
Bornehag CG Sundell J Sigsgaard T Hagerhed-Engman L 2005b Association between ventilation rates in 390 Swedish homes and allergic symptoms in children Indoor Air 15 275 280 11.1111/j.1600–0668.2005.00372.x [Online 15 June 2005].15982274
Bornehag CG Sundell J Weschler CJ Sigsgaard T 2005c Potential selection bias [Letter] Environ Health Perspect 113 A152 A153 15743700
Bornehag CG Sundell J Weschler CJ Sigsgaard T Lundgren B Hasselgren M 2004b The association between asthma and allergic symptoms in children and phthalates in house dust: a nested case-control study Environ Health Perspect 112 1393 1397 15471731
Butte W Hoffmann W Hostrup O Schmidt A Walker G 2001 Endocrine disrupting chemicals in house dust: results of a representative monitoring Gefahrstoffe Reinhaltung Der Luft 61 19 23
Cadogan DF Howick CJ 1996. Plasticizers. In: Kirk-Othmer Encyclopedia of Chemical Technology. New York:John Wiley and Sons, 258–290.
Clausen PA Hansen LG Gunnarsen L Afshari A Wolkoff P 2004 Emission of di-2-ethylhexyl phthalate from PVC flooring into air and uptake in dust: emission and sorption experiments in FLEC and CLIMPAQ Environ Sci Technol 38 2531 2537 15180047
Clausen PA Lindeberg Bille RL Nilsson T Hansen V Svensmark B Bowadt S 2003 Simultaneous extraction of di(2-ethylhexyl) phthalate and nonionic surfactants from house dust. Concentrations in floor dust from 15 Danish schools J Chromatogr A 986 179 190 12597625
Fromme H Lahrz T Piloty M Gebhart H Oddoy A Ruden H 2004 Occurrence of phthalates and musk fragrances in indoor air and dust from apartments and kindergartens in Berlin (Germany) Indoor Air 14 188 195 15104786
Fujii M Shinohara N Lim A Otake T Kumagai K Yanagisawa Y 2003 A study on emission of phthalate esters from plastic materials using a passive flux sampler Atmos Environ 37 5495 5504
Glue C Platzer MH Larsen ST Nielsen GD Skov PS Poulsen LK 2005 Phthalates potentiate the response of allergic effector cells Basic Clin Pharmacol Toxicol 96 140 142 15679477
Jaakkola JJ Oie L Nafstad P Botten G Samuelsen SO Magnus P 1999 Interior surface materials in the home and the development of bronchial obstruction in young children in Oslo, Norway Am J Public Health 89 188 192 9949747
Kavlock R Boekelheide K Chapin R Cunningham M Faustman E Foster P 2002a NTP Center for the Evaluation of Risks to Human Reproduction: phthalates expert panel report on the reproductive and developmental toxicity of butyl benzyl phthalate Reprod Toxicol 16 453 487 12406492
Kavlock R Boekelheide K Chapin R Cunningham M Faustman E Foster P 2002b NTP Center for the Evaluation of Risks to Human Reproduction: phthalates expert panel report on the reproductive and developmental toxicity of di(2-ethylhexyl) phthalate Reprod Toxicol 16 529 653 12406494
Kavlock R Boekelheide K Chapin R Cunningham M Faustman E Foster P 2002c NTP Center for the Evaluation of Risks to Human Reproduction: phthalates expert panel report on the reproductive and developmental toxicity of di-n -butyl phthalate Reprod Toxicol 16 489 527 12406493
KemI (Swedish Chemicals Inspectorate) 2004. Use of Phthalates in Sweden 1996–2003. Available: http://www.kemi.se/templates/Page____1857.aspx [accessed 25 August 2004].
Kersten W Reich T 2003 Schwerfluchtige organische umweltchemikalien in Hamburger hausstäben [in German] Reinhaltung der Luft 63 85 91
Larsen ST Hansen JS Thygesen P Begtrup M Poulsen OM Nielsen GD 2001a Adjuvant and immuno-suppressive effect of six monophthalates in a subcutaneous injection model with BALB/c mice Toxicology 169 37 51 11696408
Larsen ST Lund RM Damgard Nielsen G Thygesen P Poulsen OM 2001b Di-(2-ethylhexyl) phthalate possesses an adjuvant effect in a subcutaneous injection model with BALB/c mice Toxicol Lett 125 11 18 11701218
Larsen ST Lund RM Nielsen GD Thygesen P Poulsen OM 2002 Adjuvant effect of di-n -butyl-, di-n -octyl-, di-iso-nonyl- and di-iso-decyl phthalate in a subcutaneous injection model using BALB/c mice Pharmacol Toxicol 91 264 272 12570034
Larsen ST Lund RM Thygesen P Poulsen OM Nielsen GD 2003 Investigation of the adjuvant and immunosuppressive effects of benzyl butyl phthalate, phthalic acid and benzyl alcohol in a murine injection model Food Chem Toxicol 41 439 446 12504176
Lee MH Park J Chung SW Kang BY Kim SH Kim TS 2004 Enhacement of interleukin-4 production in activated CD4+ T cells by diphthalate plasticizers via increased NF-AT binding activity Int Arch Allergy Immunol 134 213 222 15178890
NTP 2003. 10th Report on Carcinogens. Research Triangle Park, NC:National Toxicology Program.
Nordtest 1997. Ventilation: Local Mean Age of Air—Homogenous Emission Techniques. Nordtest method NT VVS 118. Espoo, Finland:Nordtest.
Oie L Hersoug LG Madsen JO 1997 Residential exposure to plasticizers and its possible role in the pathogenesis of asthma Environ Health Perspect 105 972 978 9374082
Pohner A Simrock S Thumulla J Weber S Wirkner T 1997 Hintergrundbelastung des hausstaubes von privathauhalten mit mittel- und schwerfluchtigen organischen schadstoffen [in German] Umwelt Gesundheit 2 1 64
Rudel RA Camann DE Spengler JD Korn LR Brody JG 2003 Phthalates, alkylphenols, pesticides, polybrominated diphenyl ethers, and other endocrine-disrupting compounds in indoor air and dust Environ Sci Technol 37 4543 4553 14594359
Wensing M Uhde E Salthammer T 2005 Plastics additives in the indoor environment—flame retardants and plasticizers Sci Total Environ 339 19 40 15740755
Weschler CJ 1980 Characterization of selected organics in size-fractionated indoor aerosols Environ Sci Technol 14 428 431 22288628
Weschler CJ 1984 Indoor-outdoor relationships for nonpolar organic constituents or aerosol particles Environ Sci Technol 18 648 652
|
16203254
|
PMC1281287
|
CC0
|
2021-01-04 23:41:27
|
no
|
Environ Health Perspect. 2005 Oct 1; 113(10):1399-1404
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.7809
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences 10.1289/ehp.7844ehp0113-00140516203255ResearchChildren's HealthFungal Levels in the Home and Allergic Rhinitis by 5 Years of Age Stark Paul C. 1Celedón Juan C. 2Chew Ginger L. 3Ryan Louise M. 4Burge Harriet A. 5Muilenberg Michael L. 5Gold Diane R. 21 Biostatistics Research Center, Institute for Clinical Research and Health Policy Studies, Department of Medicine, Tufts–New England Medical Center, Boston, Massachusetts, USA2 Channing Laboratory, Department of Medicine, Brigham and Women’s Hospital and the Harvard Medical School, Boston, Massachusetts, USA3 Department of Environmental Health Sciences, Mailman School of Public Health, Columbia University, New York, New York, USA4 Department of Biostatistics, and5 Department of Environmental Health, Harvard University School of Public Health, Boston, Massachusetts, USAAddress correspondence to P.C. Stark, Biostatistics Research Center, Institute for Clinical Research and Health Policy Studies, Tufts–New England Medical Center, 750 Washington St., Tufts-NEMC #63, Boston, MA 02111 USA. Telephone: (617) 636-5457. Fax: (617) 636-5560. E-mail:
[email protected] authors declare they have no competing financial interests.
10 2005 20 5 2005 113 10 1405 1409 10 12 2004 19 5 2005 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright. Studies have repeatedly demonstrated that sensitization to fungi, such as Alternaria, is strongly associated with allergic rhinitis and asthma in children. However, the role of exposure to fungi in the development of childhood allergic rhinitis is poorly understood. In a prospective birth cohort of 405 children of asthmatic/allergic parents from metropolitan Boston, Massachusetts, we examined in-home high fungal concentrations (> 90th percentile) measured once within the first 3 months of life as predictors of doctor-diagnosed allergic rhinitis in the first 5 years of life. In multivariate Cox regression analyses, predictors of allergic rhinitis included high levels of dust-borne Aspergillus [hazard ratio (HR) = 3.27; 95% confidence interval (CI), 1.50–7.14], Aureobasidium (HR = 3.04; 95% CI, 1.33–6.93), and yeasts (HR = 2.67; 95% CI, 1.26–5.66). The factors controlled for in these analyses included water damage or mild or mildew in the building during the first year of the child’s life, any lower respiratory tract infection in the first year, male sex, African-American race, fall date of birth, and maternal IgE to Alternaria > 0.35 U/mL. Dust-borne Alternaria and non-sporulating and total fungi were also predictors of allergic rhinitis in models excluding other fungi but adjusting for all of the potential confounders listed above. High measured fungal concentrations and reports of water damage, mold, or mildew in homes may predispose children with a family history of asthma or allergy to the development of allergic rhinitis.
allergic rhinitisfungimoldrespiratory health effectswater damage
==== Body
Studies have repeatedly demonstrated that sensitization to fungi, such as Alternaria, is strongly associated with allergic rhinitis and asthma in children (Arshad et al. 2001; Downs et al. 2001; Halonen et al. 1997; Nolles et al. 2001). However, the role of exposure to fungi, measured in environmental samples, in the development of childhood allergic rhinitis is poorly understood. Allergic rhinitis affects an estimated 20–40 million people in the United States alone, and the incidence is increasing. Allergic rhinitis has been associated with snoring, sleep apnea, and sleep disturbances in children (Camhi et al. 2000; Corbo et al. 2001; Scharf and Cohen 1998). The prevalence of symptoms of allergic rhinoconjunctivitis ranged between 15 and 25% among children 13–14 years of age who lived in the United States or the United Kingdom (International Study of Asthma and Allergies in Childhood 1998). A recent study documented that 93% of adolescents with allergic asthma also have allergic rhinitis. The diagnosis of allergic rhinitis preceded asthma in up to 64% of patients with both diseases (Kapsali et al. 1997). Because allergic rhinitis is a common disease that lessens quality of life and is associated with significant morbidity, determining environmental exposures associated with the development of allergic rhinitis is of great public health importance.
In a prospective birth-cohort study of children with parental history of asthma or allergies, we assessed environmental fungal exposure (culturable fungi in house air and dust samples) and evaluated whether high fungal levels were independently associated with doctor-diagnosed allergic rhinitis in the first 5 years of life.
Materials and Methods
Study protocol.
Participants were part of a metropolitan Boston, Massachusetts, prospective birth-cohort study designed to examine relationships between amounts of indoor allergens, including fungi, and the development of allergy and asthma. The study was approved by the institutional review board of Brigham and Women’s Hospital. With the consent of the primary caregiver, screening and recruitment of the families, which we have previously described (Gold et al. 1999), was conducted between September 1994 and June 1996. Monday through Friday, all mothers who delivered at Brigham and Women’s Hospital, a large Boston hospital, were approached for screening within 24–48 hr after delivery. Eligibility criteria included residence inside route 128 (a highway encircling the Boston metropolitan area); maternal age ≥18 years; history of hay fever, asthma, or allergies in either parent; and maternal ability to speak English or Spanish. Families were not screened if the newborn was hospitalized in the intensive care unit, was of gestational age < 36 weeks, or had a congenital anomaly. One month after birth, a questionnaire regarding the child’s health was administered by telephone to families who had initially expressed interest in participating in the study and who fit the inclusion criteria. At 1 month, parents who expressed plans to move or lack of interest in the study were also excluded. Of the 1,405 families screened, 906 were excluded from the study before the initial home visit. Reasons for exclusion were reluctance to participate in a longitudinal study (51%), plans to move within 1 year (39%), early loss to follow-up (9%), and other (1%). After written informed consent was obtained, a home visit was made when the child was 2–3 months of age, and a questionnaire regarding health outcomes, home characteristics, environmental exposures, smoking, and demographics was administered by trained research assistants. Every 2 months, beginning when the child was 2 months old and continuing until the second birthday, a telephone questionnaire was administered to the child’s primary caretaker. After that, health outcomes were ascertained semiannually.
At the initial home visit, indoor air samples were collected from each home using a Burkard culture plate sampler (Burkard Manufacturing Co. Ltd., Hertfordshire, UK) operated at 45 L/min [calculated cut point, diameter allowing 50% (D50) = 2 μm] that collected particles onto DG18 agar (a glucose medium containing dichloran and 18% glycerol) in 90-mm Petri dishes. Sequential duplicate 1-min air samples were collected in the bedroom 1–1.5 m above the area of the floor demarcated for dust collection. After sampling, the Petri plates were returned to the laboratory on the same day for incubation. The sampler sieve plate was cleaned with an isopropanol swab after each home visit.
After air samples were collected, 2 m2 of the floor surrounding the newborn’s bed was vacuumed for 5 min, using a Eureka Mighty-Mite canister vacuum cleaner (Eureka Co., Bloomington, IN) modified to collect dust in a 19- × 90-mm cellulose extraction thimble (Whatman International Ltd., Herefordshire, UK). In cases where both a smooth floor and a rug were present, 2.5 min was devoted to sampling the rug and 2.5 min was spent vacuuming the smooth floor surface. After sampling, the thimble was sealed in a plastic bag and returned to the laboratory for sifting (425-μm mesh sieve) and weighing.
A weighed aliquot of fine dust from the dust sampling was suspended (~ 25 mg/mL) and serially diluted (e.g., full strength, 1:10, and 1:100) in an aqueous solution of 0.02% Tween. Duplicates of each dilution were spread-plated on DG18 culture media (Verhoeff et al. 1994). After a mean ± SD incubation period of 27.9 ± 14.5 days, we identified fungal colonies on both air- and dust-sample plates to genus (subgenus in the case of Aspergillus isolates) using standard mycologic criteria (Ellis 1971; von Arx 1970). We adjusted the number of colonies recovered on the air sample plates for possible multiple impactions using the positive-hole correction equation (Andersen 1958). The calculated concentrations of dust-borne fungi were the number of colony-forming units (cfu) per gram of dust, and those of airborne fungi were cfu per cubic meter of air. We prioritized the use of limited amounts of dust collected from a home so that measurement of allergens for cockroach, cat, and dust mites could be performed first (Chew et al. 1998), followed by analyses of fungi and endotoxin, respectively. If there was not enough dust for allergen analyses, then the fungal analyses were performed first. Thus, when there was > 200 mg or < 25 mg dust collected, fungal concentrations were determined. There were 84 children eliminated because the amount of dust collected was between 25 and 200 mg.
Fungal data.
In previous analyses of these data (Stark et al. 2003), we found that it was useful to create indices indicating high levels of fungi in the home. The variable was coded 1 if that home produced “high” levels of that fungus, and 0 otherwise. We define “high” as greater than the 90th percentile for each taxon. Based on Akaike’s Information Criterion (Hastie and Tibshirani 1990) and –2 log likelihood, models with indices for high fungal levels fit the data better than do models treating the fungal concentrations as continuous or natural logarithm–transformed data.
Definition of allergic rhinitis and predictor variables.
We assessed allergic rhinitis from 12 to 60 months of age by asking “Has your child had allergic rhinitis or hay fever diagnosed by a doctor?” Children without 12-month follow-up (n = 10) were excluded, leaving 405 children with > 12 months of health outcome data, as well as fungal data.
Definitions of other predictor variables.
Sociodemographic, familial, and perinatal variables considered for inclusion in the multivariate analysis of the relation between indoor fungal exposure and allergic rhinitis included sex, season of birth [defined as winter (December–February), spring (March–May), summer (June–August), or fall (September–November)], race/ethnicity of the child according to parental reporting (Litonjua et al. 1998), total annual family income (< $30,000, $30,000–50,000, and > $50,000), active maternal history of asthma (physician-diagnosed asthma and wheeze in the past year), maternal sensitization to Alternaria [specific immunoglobulin E (IgE) > 0.35 U/mL], maternal smoking during pregnancy, and presence of older siblings in the household.
Additional variables considered for inclusion in the multivariate analysis included report of any of the following in the child’s first year of life: any physician-diagnosed lower respiratory tract infections (pneumonia, croup, bronchitis, or bronchiolitis), as determined by bimonthly questions to the child’s primary caregiver (Celedon et al. 1999); visible water damage; mold or mildew inside the home; and report of water damage in the basement, presence of concrete floors in the baby’s room, use of a dehumidifier, and use of an air conditioner. A composite variable representing the presence of either water damage or visible mold or mildew in the home in the child’s first year of life was also examined.
Statistical methods.
All final models reported used survival analysis (Cox proportional hazards regression) to explore time to first report of doctor-diagnosed allergic rhinitis or hay fever. We performed initial exploratory analyses of univariate associations between allergic rhinitis and potential predictor variables using logistic regression, chi-square analyses, and Cox regression. We performed all statistical analyses using SAS statistical software (version 8.2, SAS Institute Inc., Cary, NC) and S-Plus (version 6.1, Insightful Corporation, Seattle, WA). Multivariate models were constructed that included significant (p ≤0.05) fungal taxa, significant independent predictors of allergic rhinitis, and variables that may confound the relationship between the fungal levels and allergic rhinitis. The proportional hazards assumption was validated in final models.
Lasso approach.
Multivariate models were independently validated using the lasso approach for model shrinkage described by Tibshirani (1997). This technique prevents the tendency to overfit a model by constraining the effect that the variables will have in the model. This is one approach used to combat the multiple-comparison hazards that arise when so many different taxa of fungi need to be considered.
Results
Of the 405 children included in this analysis, 52 (12.8%) were diagnosed with allergic rhinitis or hay fever by a doctor before or at 5 years of age.
Relationship between cohort characteristics and allergic rhinitis.
Table 1 shows the relationship between cohort characteristics and doctor-diagnosed allergic rhinitis by 5 years of age. Factors associated with an increased risk of allergic rhinitis included African-American ethnicity, being born between September and November, maternal sensitization to Alternaria (IgE to Alternaria > 0.35 U/mL), and having at least one lower respiratory tract illness in the first year of life. Male sex and water damage or mold or mildew in the building of residence within the first year of life were marginally associated with doctor-diagnosed allergic rhinitis by 5 years of age.
In Table 2, we report the number of houses from which each taxon was cultured, and the value corresponding to the 90th percentile. Penicillium was the most common taxon sampled from the air, but Cladosporium, Penicillium, and nonsporulating fungi had the highest levels recovered from the air. From the dust samples, Aspergillus was most commonly recovered sporulating taxon, followed by Cladosporium.
High levels of dust-borne Aureobasidium, Aspergillus, Alternaria, yeasts, and nonsporulating fungi as well as high levels of total fungi were independently associated with the development of doctor-diagnosed allergic rhinitis by 5 years of age in analyses controlling for factors significantly associated with allergic rhinitis (Table 3). These factors included water damage or mold or mildew in the building during the first year of the child’s life, any lower respiratory tract illness in the first year, male sex, African-American race, birth date between September and November, and maternal IgE to Alternaria > 0.35 U/mL.
In a multivariate adjusted proportional hazards regression analysis looking simultaneously at the effects of high fungal levels and significant independent predictors (Table 4), dust-borne Aspergillus [hazard ratio (HR) = 3.27; 95% confidence interval (CI), 1.50–7.14], Aureobasidium (HR = 3.04; 95% CI, 1.33–6.93), and yeasts (HR = 2.67; 95% CI, 1.26–5.66) remained significantly associated with allergic rhinitis. Alternaria and Cladosporium were not significant in the multivariate model, because they are highly correlated with other fungi that were significant. Alternaria was significantly correlated with Aureobasidium (R = 0.33, p < 0.0001) to the point where its contribution to the model is not statistically significant after considering the effect of Aureobasidium. When Aureobasidium is omitted from the model (Table 4, model 3), Alternaria is borderline significant (HR = 2.07; 95% CI, 0.98–4.37). Alternaria and Cladosporium were highly correlated as well (R = 0.30, p < 0.0001). Having high levels of total dust-borne fungi was not significant in multivariate models that had the specific taxa listed above (p = 0.44). The magnitude and strength of these associations remained essentially unchanged when controlling for the presence of dust mite, cockroach, and cat allergens and endotoxin in the home, none of which had significant independent associations with allergic rhinitis by age 5 in multivariate models (data not shown).
Discussion
High dust-borne fungal levels of Alternaria, Aspergillus, Aureobasidium, and yeasts measured in the home in the first 3 months of life were associated with the subsequent development of doctor-diagnosed allergic rhinitis within the first 5 years of life, independent of water damage or mold or mildew in the home, race, sex, season of birth, lower respiratory tract infection in the first year of life, and maternal IgE to Alternaria > 0.35 U/mL. Home dampness has been shown to be related to allergic rhinitis exacerbations (Kilpelainen et al. 2001), but to our knowledge, the relation of either water damage or mold/mildew or high levels of specific fungal taxa to the development of allergic rhinitis in childhood has not previously been reported.
Although other studies have found associations between measures of total fungi and health outcomes (Gent et al. 2002), it is likely that, in our study, “total fungi” is a noisier estimate of relevant fungal exposures than are our other metrics of fungi. The measure of “total fungi” would combine fungi not likely to be related to rhinitis along with fungi that have been identified as being related to allergic sensitization. This could lead to less precise estimates of fungal heath effects and an attenuation of the relationship between fungi and allergic rhinitis.
Fungal exposure is complex. Fungi can contain allergens, irritants, toxins, and sometimes even potentially infectious units (Johnston 2000). (1→3)-β-d-Glucans—glucose polymers that are structural components of most fungal cell walls—are known to stimulate macrophages and neutrophils and have been shown to be good markers for the overall levels of fungal concentrations in floor dust (Chew et al. 2001). Beijer et al. (2002) found that an inhalation challenge to (1→3)-β-d-glucans has an effect on the inflammatory cells, which may be related to chronic exposure to household molds. Fungi are ubiquitous, and exposure is impossible to completely avoid. Often, there are no adverse effects from these exposures (Macher 2000), but at times exposure to fungi can directly and indirectly influence an individual’s health. Fungal exposure in sensitized individuals may result in histamine and IgE-mediated nasal inflammation (Skoner 2001). Chronic exposure to fungi may also cause chronic nasal symptoms that are not allergic in nature, but rather caused by irritants (Daisey et al. 2003; Dotterud et al. 1996). Symptoms of allergic rhinitis are similar to the chronic nasal symptoms often found with sick building syndrome. Given that fungi may play a role in sick building syndrome as well (Scheel et al. 2001), differentiation may be difficult.
In previous articles we have reported that endotoxin was a risk factor for wheeze (Park et al. 2001) yet had a protective effect on eczema in the first year of life (Phipatanakul et al. 2004). However, the associations of fungi with allergic rhinitis remained essentially unchanged when controlling for the presence of indoor allergens and endotoxin.
There is a large literature demonstrating the relationship between fungal sensitization and allergic rhinitis and asthma, as well as a growing literature linking fungal exposure to asthma and rhinitis exacerbation (Downs et al. 2001; Halonen et al. 1997) in sensitized children (Arshad et al. 2001; Nolles et al. 2001). However, the fact that children with allergic rhinitis are sensitized to fungi does not necessarily mean that fungal exposure will lead to allergy or allergic symptoms. This study suggests a prospective relationship between exposure and symptoms, although elevation of the individual fungal taxa may not lead directly to symptom development. For example, relative elevation of specific taxa such as Aspergillus, Aureobasidium, and yeasts may be a marker for overall fungal exposure. In models containing the individual taxa and our measure of total fungi, total fungi was not predictive of allergic rhinitis, suggesting that the combination of individual taxa was a better marker of clinically relevant fungal exposure.
Sensitization to Alternaria, Aspergillus, and Penicillium allergens have been suggested as being related to allergic rhinitis and asthma exacerbation in children (Clark et al. 1999; Halonen et al. 1997). Purified and characterized allergens have been prepared for Aspergillus fumigatus (Asp f 1 and Asp f 3) (Macher 2000), which has been shown to cause skin reactivity in patients with allergic rhinitis (Johnston 2000). However, A. fumigatus was sampled from only three houses. The most commonly sampled species in this cohort were members of the Aspergillus glaucus group (n = 275) and Aspergillus versicolor (n = 255). Although skin test material is commonly available for both, less is known about the allergenicity of these species. A. versicolor has been implicated in an outbreak of allergic respiratory disease (Jarvis and Morey 2001).
Cladosporium and Alternaria are known sources of allergens, including the proteins Alt a 1, Alt a 2, and Cla h 1 (Johnston 2000). Many cross-sectional studies have correlated Cladosporium and Alternaria sensitization with diagnoses of asthma, eczema, and rhinitis (Arshad et al. 2001; Downs et al. 2001; Halonen et al. 1997; Nolles et al. 2001).
Aureobasidium has been associated with hypersensitivity pneumonitis (Greinert et al. 2000), but its role in the development of allergic rhinitis is unknown. In this study, it may be a marker for mold in the home, rather than the actual source of the chronic nasal symptoms.
In their association with chronic nasal symptoms, yeasts may also be a marker for overall mold exposure, because although they have known associations with skin manifestations of allergy, nasal symptoms are less commonly documented. Yeasts are unicellular fungi that reproduce primarily by budding and, in culture, form pasty colonies similar to those of bacteria (Macher 2000). Several studies have shown that patients with atopic dermatitis and respiratory allergy often develop IgE antibodies against yeasts (Devos and Van der Valk 2000; Nittner-Marszalska et al. 2001; Tengvall Linder et al. 2000). One showed that patients with atopic dermatitis of the head and neck developed IgE antibodies against the yeast Pityrosporum ovale much more often than did their control group. Another found that the enzyme enolase from baker’s yeast (Saccharomyces cerevisiae) induces an immediate dermal allergic reaction in subjects with respiratory allergy and positive skin tests to Candida albicans and other fungi. A third group demonstrated that Pityrosporum orbiculare can induce an eczematous reaction in sensitized atopic dermatitis patients.
Potential study limitations.
This study does have some limitations. Potential limitations of our work include the relatively small sample size, along with the multiplicity of variables being considered. Our use of the Lasso approach provides an efficient and appealing method for handling multiplicity issues. The Lasso method is much less prone to missing real associations than more traditional adjustments for multiple comparisons, such as Bonferroni. Even so, the reality is that with only 52 cases, some important differences may have been missed. It is unlikely, however, that our analysis erroneously identified many false associations, if any. In addition, because of the collinearity of the fungal data and the paucity of events, it is difficult to assess the risk of each taxon controlling for the others. For example, Alternaria is so significantly correlated with Aureobasidium that its effect is masked when both are included in the model. Even still, in a combined multivariate model, high levels of Aspergillus, Aureobasidium, and yeasts were independently associated with an increased risk of doctor-diagnosed allergic rhinitis within the first 5 years of life. In addition, the air and dust samples were single, short-term samples. Therefore, it is not clear how representative the samples are of long-term household fungal levels. We used only a culture-based analysis to determine fungal concentrations, which would underestimate actual fungal spore exposures and does not account for other fungal agents. Also, the culture media used for both the air and dust samples, DG18, has high solute concentration, which limits the amount of available water. Thus, some strongly hydrophilic fungi (e.g., Stachybotrys sp., some yeasts) might be underrepresented. In addition, all relationships discussed in this article are related to exposures that occurred in the first home in which the child lived, and do not account for conditions in any subsequent homes or locations where the child may have spent significant time (e.g., a child care center). Moreover, without sampling performed throughout the study (Chew et al. 2001), it is impossible to determine whether high fungal levels early in life increase the likelihood of having doctor-diagnosed allergic rhinitis later in life, or whether the early sampling is representative of the environment later in life. In addition, the absence of definitive measurements of allergy, and the relatively early age of ascertainment of outcome in this study such that some children may yet develop allergic rhinitis in this cohort, may be responsible for the lack of an endotoxin effect on the risk of allergic rhinitis in this analysis. Further, parental report of allergic rhinitis or hay fever was not confirmed by chart review, but primary care physician documentation of these diagnoses may not be the gold standard because it is likely to vary by individual practice. In addition, it is possible that the parents did not understand or recall a diagnosis of allergic rhinitis. Lastly, without skin-prick testing, the possibility exists that nasal symptoms related to chronic irritation were misdiagnosed as allergic rhinitis by the physicians of some of the participating children. Even if fungal exposure in infancy leads to allergic rhinitis in some children and to chronic irritant nonallergic rhinitis in others, these findings still have important public health implications.
Conclusions
Recently, more attention has been given to the need for monitoring of fungal levels and active intervention, where necessary (Johnston 2000). Epidemiologic studies have found associations between report of home dampness and respiratory symptoms in early life, but most have not measured fungi directly and have not determined whether fungal effects can be distinguished from other exposures related to dampness that prospectively predict respiratory disease. Independent of early life report of home dampness, we found that specific dust-borne fungi measurable in the home in the first 3 months of a child’s life was associated with an increased risk of developing doctor-diagnosed allergic rhinitis by 5 years of age.
We thank the participating families for their enthusiastic participation. We also thank C. Rogers and D. Sredl for their efforts.
This study was supported by National Institutes of Health grant AI/EHS35786. P.C.S. was supported by the Yamaguchi Endowment.
Table 1 Cohort characteristics as predictors of allergic rhinitis.
Factor No. Percent with allergic rhinitis HR 95% CI
Race/ethnicity
White 309 12.0
African American 46 21.7 2.22* 1.10–4.47
Hispanic 18 5.6 0.53 0.07–3.82
Asian 26 15.4 1.37 0.49–3.85
Other 6 0 — —
Season of birth
Winter (December–February) 103 3.9
Spring (March–May) 112 12.5 3.33* 1.10–10.12
Summer (June–August) 87 13.8 3.55* 1.15–11.00
Fall (September–November) 103 21.4 5.76** 1.99–16.72
Sex of child
Female 195 9.7
Male 210 15.7 1.64 0.93–2.88
Maternal IgE to Alternaria > 0.35 U/mL
No 316 11.4
Yes 43 25.6 2.49** 1.27–4.89
Any LRI in year 1
No 297 10.8
Yes 108 18.5 1.80* 1.03–3.15
Water damage in year 1
No 263 11.0
Yes 135 17.0 1.59 0.92–2.74
Mold or mildew in year 1
No 246 11.8
Yes 152 15.1 1.28 0.74–2.22
Water damage or mold
No 169 10.1
Yes 235 14.9 1.46 0.82–2.61
Abbreviations: —, no data; LRI, lower respiratory infection.
* p < 0.05;
** p < 0.01.
Table 2 Description of fungi included in the analysis.a
Fungi No. > 0 90th percentile
Airborne (cfu/m3)
Aspergillus 253 100
Cladosporium 315 400
Nonsporulating 303 278
Penicillium 346 189
Yeasts 147 33
Total airborne 402 1,044
Dust-borne (cfu/g)
Alternaria 233 8,333
Aspergillus 367 33,333
Aureobasidium 322 24,000
Cladosporium 343 33,088
Coelomyces 93 2,941
Fusarium 49 400
Nonsporulating 373 38,889
Penicillium 362 26,087
Ulocladium 60 435
Wallemia 61 1,429
Yeasts 361 58,000
Zygomycetes 57 909
Total dust-borne 405 250,000
a Must have been sampled from the indoor environment and have a 90th percentile > 1 cfu.
Table 3 Results of survival analysis of allergic rhinitis.
Unadjusted
Adjusteda
High fungi level HR 95% CI HR 95% CI
Airborne (cfu/m3)
Aspergillus 1.09 0.44–2.75 1.10 0.43–2.80
Cladosporium 0.72 0.26–2.00 1.25 0.43–3.64
Nonsporulating 0.54 0.17–1.74 0.55 0.17–1.81
Penicillium 1.07 0.42–2.68 0.69 0.23–2.06
Yeasts 0.98 0.36–2.73 0.79 0.24–2.60
Total airborne 0.73 0.27–2.04 0.83 0.28–2.43
Dust-borne (cfu/g)
Alternaria 2.50** 1.28–4.86 2.34* 1.12–4.91
Aspergillus 2.47** 1.27–4.81 2.57* 1.22–5.40
Aureobasidium 2.77** 1.42–5.39 3.12** 1.50–6.50
Cladosporium 1.63 0.77–3.46 1.88 0.81–4.35
Coelomyces 1.18 0.51–2.77 0.93 0.36–2.38
Fusarium 1.25 0.54–2.94 1.81 0.76–4.34
Nonsporulating 2.21* 1.11–4.41 2.45* 1.15–5.22
Penicillium 1.34 0.57–3.13 1.51 0.63–3.64
Ulocladium 0.82 0.30–2.28 1.04 0.37–2.95
Wallemia 2.23* 1.12–4.45 1.73 0.80–3.75
Yeasts 2.04* 1.00–4.19 2.90** 1.37–6.09
Zygomycetes 0.80 0.29–2.21 0.87 0.31–2.44
Total dust-borne 2.56** 1.32–4.98 3.13** 1.51–6.47
a Controlling for water damage or mold or mildew in year 1, African-American ethnicity, maternal Alternaria IgE > 0.35 U/mL, sex, and birth date in fall.
* p < 0.05;
** p < 0.01.
Table 4 Multivariate Cox regression models of allergic rhinitis.
Model 1a Model 2b Model 3c
Factor RR 95% CI RR 95% CI RR 95% CI
Water damage or mold/mildew in year 1 1.66 0.88–3.15 1.77 0.94–3.34 1.66 0.87–3.17
Male sex 1.79 0.97–3.29 1.88 * 1.03–3.42 1.78 0.97–3.28
LRI ever in year 1 1.58 0.83–3.02 — — 1.62 0.84–3.12
Maternal IgE to Alternaria > 0.35 U/mol 3.07** 1.49–6.32 2.96** 1.44–6.09 3.21** 1.56–6.60
Race (black vs. others) 3.82** 1.73–8.44 3.79** 1.72–8.38 3.27** 1.51–7.08
Season of birth (Sep–Nov vs. others) 2.44** 1.35–4.42 2.46** 1.35–4.47 2.46** 1.36–4.48
Dust-borne Alternaria 1.40 0.61–3.23 1.52 0.67–3.44 2.07 0.98–4.37
Dust-borne Aspergillus 3.27** 1.50–7.14 2.93** 1.36–6.30 2.73* 1.27–5.87
Dust-borne Aureobasidium 3.04** 1.33–6.93 3.06** 1.35–6.91 — —
Dust-borne yeasts 2.67* 1.26–5.66 2.80** 1.33–5.93 2.52* 1.18–5.36
Abbreviations: —, not included; LRI, lower respiratory infection.
a Full multivariate model.
b Multivariate model omitting any lower respiratory infection in year 1.
c Multivariate model omitting dust-borne Aureobasidium.
* p < 0.05;
** p < 0.01.
==== Refs
References
Andersen AA 1958 New sampler for the collection, sizing and enumeration of viable airborne particles J Bacteriol 76 471 484 13598704
Arshad SH Tariq SM Matthews S Hakim E 2001 Sensitization to common allergens and its association with allergic disorders at age 4 years: a whole population birth cohort study Pediatrics 108 E33 E37 11483843
Beijer L Thorn J Rylander R 2002 Effects after inhalation of (1 → 3)-β-D-glucan and relation to mould exposure in the home Mediators Inflamm 11 3 149 153 12137243
Camhi SL Morgan WJ Pernisco N Quan SF 2000 Factors affecting sleep disturbances in children and adolescents Sleep Med 1 2 117 123 10767652
Celedon JC Litonjua AA Weiss ST Gold DR 1999 Day care attendance in the first year of life and illnesses of the upper and lower respiratory tract in children with a familial history of atopy Pediatrics 104 3 pt 1 495 500 10469775
Chew GL Burge HA Dockery DW Muilenberg ML Weiss ST Gold DR 1998 Limitations of a home characteristics questionnaire as a predictor of indoor allergen levels Am J Respir Crit Care Med 157 1536 1541 9603135
Chew GL Douwes J Doekes G Higgins KM Spithoven J van Strien R 2001 Fungal extracellular polysaccharides, and β(1 → 3) glucans, and culturable fungi in repeated sampling of house dust Indoor Air 11 3 171 178 11521501
Clark NM Brown RW Parker E Robbins TG Remick DG Philbert MA 1999 Childhood asthma Environ Health Perspect 107 suppl 3 421 429 10423388
Corbo GM Forastiere F Agabiti N Pistelli R Dell’Orco V Perucci CA 2001 Snoring in 9- to 15-year old children: risk factors and clinical relevance Pediatrics 108 5 1149 1154 11694695
Daisey JM Angell WJ Apte MG 2003 Indoor air quality, ventilation and health symptoms in schools: an analysis of existing information Indoor Air 13 53 64 12608926
Devos SA Van der Valk PG 2000 The relevance of skin prick testing for Pityrosporum ovale in patients with head and neck dermititis Allergy 55 11 1056 1058 11097316
Dotterud LR Vorland LH Falk ES 1996 Mould allergy in school children in relation to airborne fungi and residential characteristics in homes and schools in northern Norway Indoor Air 6 71 76
Downs SH Mitakakis TZ Marks GB Car NG Belousova EG Leuppi JD 2001 Clinical importance of Alternaria exposure in children Am J Respir Crit Care Med 164 3 455 459 11500349
Ellis MB 1971. Dematiaceous Hyphomycetes. Kew, Surry, UK:Commonwealth Mycological Institute.
Gent JF Ren P Belanger K Triche E Bracken MB Holford TR 2002 Levels of household mold associated with respiratory symptoms in the first year of life in a cohort at risk for asthma Environ Health Perspect 110 A781 A786 12460818
Gold DR Burge HA Carey V Milton DK Platts-Mills T Weiss ST 1999 Predictors of repeated wheeze in the first year of life Am J Respir Crit Care Med 160 227 236 10390405
Greinert U Lepp U Becker W 2000 Bird keeper’s lung without bird keeping Eur J Med Res 5 3 124 10756167
Halonen M Stern DA Wright AL Taussig LM Martinez FD 1997 Alternaria as a major allergen for asthma in children raised in a desert environment. Am J Respir Crit Care Med 155 1356 1361
Hastie TJ Tibshirani RJ 1990. Generalized Additive Models. London:Chapman and Hall.
International Study of Asthma and Allergies in Childhood (ISAAC) Steering Committee 1998 Worldwide variation in prevalence of symptoms of asthma allergic rhinoconjunctivitis, and atopic eczema: ISAAC Lancet 351 1225 1232 9643741
Jarvis JQ Morey PR 2001 Allergic respiratory disease and fungal remediation in a building in a subtropical climate Appl Occup Environ Hyg 16 3 380 388 11297052
Johnston RB ed. 2000. Clearing the Air: Asthma and Indoor Air Exposures. Washington, DC:National Academy Press.
Kapsali T Horowitz E Diemer F Togias A 1997 Rhinitis is ubiquitous in allergic asthmatics J Allergy Clin Immunol 99 S138
Kilpelainen M Terho EO Helenius H Koskenvuo M 2001 Home dampness, current allergic diseases, and respiratory infections among young adults Thorax 56 462 467 11359962
Litonjua AA Carey VJ Burge HA Weiss ST Gold DR 1998 Parental history and the risk of childhood asthma. Does mother confer more risk than father? Am J Respir Crit Care Med 158 1 176 181 9655726
Macher J ed. 2000. Bioaerosols: Assessment and Control. Cincinnati, OH:American Conference of Government Industrial Hygienists Bioaerosols Committee.
Nittner-Marszalska M Wojcicka-Kustrzeba I Bogacka E Patkowski J Dobek R 2001 Skin prick test response to enzyme enolase of the baker’s yeast (Saccharomyces cerevisiae ) in diagnosis of respiratory allergy Med Sci Monit 7 1 121 124 11208506
Nolles G Hoekstra MO Schouten JP Gerritsen J Kauffman HF 2001 Prevalence of immunoglobulin E for fungi in atopic children Clin Exp Allergy 31 1564 1570 11678856
Park JH Gold DR Spiegelman DL Burge HA Milton DK 2001 House dust endotoxin and wheeze in the first year of life Am J Respir Crit Care Med 163 322 328 11179100
Phipatanakul W Celedon JC Raby BA Litonjua AA Milton DK Sredl D 2004 Endotoxin exposure and eczema in the first year of life Pediatrics 114 13 18 15231902
Scharf MB Cohen AP 1998 Diagnostic and treatment implications of nasal obstruction in snoring and obstructive sleep apnea Ann Allergy Asthma Immunol 81 4 279 287 9809490
Scheel CM Rosing WC Farone AL 2001 Possible sources of sick building syndrome in a Tennessee middle school Arch Environ Health 56 5 413 417 11777022
Skoner DP 2001 Allergic rhinitis: definition, epidemiology, pathophysiology, detection, and diagnosis J Allergy Clin Immunol 108 S2 S8 11449200
Stark PC Burge HA Ryan LM Milton DK Gold DR 2003 Fungal levels in the home and lower respiratory tract illnesses in the first year of life Am J Respir Crit Care Med 168 232 237 12724122
Tengvall Linder M Johansson C Scheynius A Wahlgren C 2000 Positive atopy patch test reactions to Pityrosporum orbiculare in atopic dermatitis patients Clin Exp Allergy 30 1 122 131 10606939
Tibshirani R 1997 The Lasso method for variable selection in the Cox model Statist Med 16 385 395
Verhoeff AP van Reenen-Hoekstra ES Samson RA Brunekreef B van Wijnen JH 1994 Fungal propagules in house dust. I. Comparison of analytic methods and their value as estimators of potential exposure Allergy 49 7 533 539 7825720
von Arx JA 1970. The Genera of Fungi Sporulating in Pure Culture. Lehre, Germany:Verlag von J Cramer.
|
16203255
|
PMC1281288
|
CC0
|
2021-01-04 23:41:28
|
no
|
Environ Health Perspect. 2005 Oct 20; 113(10):1405-1409
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.7844
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences 10.1289/ehp.7702ehp0113-00141016203256ResearchChildren's HealthEffect of Maternal Smoking on Breast Milk Interleukin-1α, β-Endorphin, and Leptin Concentrations Zanardo Vincenzo 1Nicolussi Silvia 1Cavallin Stefania 1Trevisanuto Daniele 1Barbato Angelo 1Faggian Diego 2Favaro Flaviano 2Plebani Mario 21 Department of Pediatrics and2 Institute of Laboratory Medicine, Padua University School of Medicine, Padua, ItalyAddress correspondence to V. Zanardo, Department of Pediatrics, Padua University School of Medicine, Via Giustiniani 3, 35128 Padua, Italy. Telephone: 049-8213505. Fax: 049-8213509. E-mail:
[email protected] authors declare they have no competing financial interests.
10 2005 15 6 2005 113 10 1410 1413 29 10 2004 14 6 2005 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright. Tobacco smoke is immunotoxic, but the effect of smoking on the immunologic function of the mammary gland of mothers who smoke cigarettes (“smoker mothers”) has not been studied. Our objective was to test, in smoker mothers, the colostral and transitional milk concentrations of interleukin-(IL)1α. The immunomodulators β-endorphin and leptin were also tested. Pregnant women who self-identified as smokers (≥ 5 cigarettes per day through pregnancy) or nonsmokers were recruited for study participation. The study population included 42 smoker and 40 non-smoker nursing mothers, with otherwise uncomplicated gestation, delivery, and puerperium, who were breast-feeding ad libitum their healthy neonates. Colostrum was obtained on the third postpartum day at 0900 hr and transitional milk on the 10th postpartum day at 0900 hr. IL-1α concentrations were significantly reduced in the colostrum of smoker mothers compared with nonsmoker mothers (p < 0.01). Colostral β-endorphin and leptin concentrations were comparable. No significant differences were found between smoker and nonsmoker lactating mothers in transitional milk concentrations of IL-1α, β-endorphin, and leptin. Moreover, β-endorphin and leptin concentrations were significantly reduced in transitional milk samples compared with colostrum of both smoker and nonsmoker mothers (p < 0.05); also, IL-1α transitional milk concentrations were reduced compared with colostrum, but without any significance. This analysis shows that maternal smoking alters the colostral milk levels of the proinflammatory cytokine IL-1α. The altered postnatal provision of alternative source of the proinflammatory cytokine IL-1α adds understanding to how breast-feeding could be nonprotective against infections among the neonates nursed by smoker mothers.
β-endorphincigarette smokecolostruminterleukin-1αleptinmaternal smokingtransitional milk
==== Body
Evidence increasingly indicates that breast milk protects against gastrointestinal, respiratory tract, middle ear infections, asthma, and sudden infant death syndrome (SIDS) (Klonoff-Cohen et al. 1995; Kovar et al. 1984; Oddy et al. 2003). Several components of human milk have been postulated to confer this protective effect (Goldman et al. 1998). Breast-feeding might provide an immediate line of defense against infectious agents, which compensates directly for the immaturity of the newborn’s immune system and lowered ability to resist infection. Protection might also be achieved through specific and nonspecific factors in milk, including bioactive enzymes, hormones, growth factors, and immunologic agents that augment and stimulate the development of immature host defense. Thus, it is not clear which of the many components of this complex, changing biologic fluid account for the protective effect.
Passive smoking in the same room as the infant increases the risk for respiratory disease and SIDS (Gordon et al. 2002), whereas breast-feeding is protective for SIDS among nonsmokers, but not smokers, when adjusted for potential confounders (Klonoff-Cohen et al. 1995), raising the possibility that there are also immunologic changes in the breast milk. However, a clear negative effect of nursing by a smoker mother has not been demonstrated (Mitchell et al. 1993), and very little is known of the mechanisms by which smoke might account for the reduced protective effect of breast milk (Blizzard et al. 2003; Elliot et al. 2003).
Cigarette smoke is composed of > 5,000 chemicals, including approximately 70 carcinogens (Stedman 1968), very few of which have actually been assayed for immunosuppressive activity (Ouyang et al. 2000). Conflicting reports exist regarding the effects of nicotine on cytokine production, and little is known about the nature of the immunosuppressive compounds in cigarette smoke (Ouyang et al. 2000). Nevertheless, the major phenolic components of cigarette tar, hydroquinone and catechol, have been reported to suppress the production of interleukin-(IL)1β, IL-2, and interferon-γ (Kalf et al. 1996; Ouyang et al. 2000). Breast-feeding substantially increases absorption of nicotine compared with only environmental tobacco smoke when the mother smokes (Dahlström et al. 1990; Luck and Nau 1985; Schulte-Hobein et al. 1992). In smoker mothers, the milk:plasma concentration ratio of nicotine is 2.9, whereas that of the primary metabolite cotinine is 1.2 (Dahlström et al. 1990). Nicotine has received much attention because it is immunotoxic, triggering the immune system and altering the humoral and cellular immunity and the levels of certain cytokines and their receptors (Sayers and Drucker 1999; Sopori et al. 1998).
It has only recently been shown that human milk contains cytokines (Goldman et al. 1996). Cytokines are small soluble glycoproteins that act in an autocrine or paracrine manner by binding to specific cellular receptors, operating in networks, and orchestrating the immune system’s development and function. Early milk has an abundance of cytokines at a time when neonatal organ systems are immature, suggesting that these bioactive components of milk might be important in neonatal immunity development. Growth, differentiation, and immunoglobulin production by B-cells were among the first activities found in human milk and attributed to the presence of cytokines (Bentzen 1994).
IL-1α is one of the two forms, indistinguishable in their biologic activities, of IL-1, the prototypic proinflammatory cytokine, involved in the mechanisms underlying many infectious and noninfectious inflammatory diseases. Although IL-1 can up-regulate host defenses and function as an immunoadjuvant, it is a highly proinflammatory cytokine. The margin between clinical benefit and unacceptable toxicity in humans is exceedingly narrow. In contrast, agents that reduce the production and/or activity of IL-1 are likely to have an impact on clinical medicine. The synthesis, processing, secretion, and activity of IL-1 are tightly regulated events. For instance, the presence of an IL-1 receptor antagonist is an important mechanism through which IL-1 signaling is down-regulated, and therefore its activity is reduced (Dinarello 1998). IL-1 is probably the first cytokine that has been quantified in human milk (Munoz et al. 1990). IL-1β, but not IL-1α, was frequently investigated and found in colostrum and early milk samples from healthy lactating mothers. To the best of our knowledge, the effect of cigarette smoking on the immunologic function of the mammary gland has not been studied previously.
Our objective was to determine whether there is a relationship between IL 1-α levels in milk and smoking habits in nursing mothers, with the consequent implications on the development of the immune system. Other factors in milk considered for association with infection risk in infancy were the immunomodulators β-endorphin and leptin. β-Endorphin was assessed because it is produced by the epithelial cells (i.e., the primary cells in human milk) (Khachaturian et al. 1958), it exists at detectable levels in human milk (Zanardo et al. 2001), and it influences the behavior of other cells and tissues. Leptin, induced by lipopolysaccharide (LPS) and cytokines, participates in the host response to inflammation by modulating the host immune and cytokine responses after LPS (Faggioni et al. 1999). Thus, any effects of a given cytokine could be tested for specificity by comparison with the other immunologically active components.
Materials and Methods
Longitudinal and cross-sectional human milk samples were obtained from mothers hospitalized in regional tertiary maternity care at Padua University (Italy), from June to December 2002. Human milk was collected by mothers via commercial breast pumps, frozen immediately, and stored. Collection date and time were recorded, as were infants’ birth date, gestational age, birth weight, and sex and mothers’ demographic and anthropometrical characteristics [age, education, parity, body mass index (BMI)], diagnoses, pregnancy complications, and route of delivery (cesarean section vs. vaginal).
Before the infant’s birth, pregnant women were asked by experienced research midwives to participate in a prospective study. The women received verbal and written information about the aim and design of the study. Only healthy women ≥ 18 years of age, not taking anti-inflammatory medications at the time of enrollment, with otherwise uncomplicated gestation, delivery, and puerperium and who planned to obtain care for their newborns through rooming-in and exclusively breast-feeding their newborns were eligible. Allergy in mothers was not an exclusion criterion because of the findings that maternal atopy shows no relationship with cytokine level in milk (Bottcher et al. 2000). After the women accepted participation, their smoking history was recorded.
Of the 1,217 eligible participants, 25 of 26 self-identified as smokers (≥ 5 cigarettes per day through pregnancy until last trimester) were recruited for study participation. One was excluded from the final analysis because of maternal fever. Control participants included consecutive women without history of smoking and matched a smoking participant on the basis of overall inclusion criteria. A written informed consent was obtained from all participants.
Milk specimen collection and processing.
Breast milk samples were collected during the third postpartum day (before discharge) and on the 10th postpartum day, close to expression of the colostrum and transitional milk production phases. Collection was standardized to reduce bias and potential diurnal variability in cytokine measurements. Specimens were obtained within 1 hour of the first feeding in the morning, defined as 0800 hr to 0900 hr. Mothers were asked not to feed for 3 hr before collection. All mothers were able to provide milk at each sampling point. An aliquot of 3 mL was taken using a manual breast pump. Samples were temporarily stored at −70°C in sterile plastic tubes and used for assays.
Samples were thawed and centrifuged for 10 min at 1800 rpm at 4°C, after which the lipid layer and cellular elements were removed. The aqueous fraction was filtered (0.45-μm Acrodisc; Gelman Sciences, Ann Arbor, MI, USA) and was used for cytokine, β-endorphin, and leptin determinations, with the lipid layer frozen for other studies.
Milk IL1-α, β-endorphin, and leptin assays.
Filtered aqueous milk fractions were assayed for IL-1α (by CytElisa Human IL-1α ACCUCYTE; CytImmune Sciences Inc., Rockville, MD, USA), β-endorphin (beta-Endorphin 60 T kit RIA; Nichols Institute Diagnostic, San Juan Capistrano, CA, USA), and leptin (ACTIVE Human Leptin IRMA, DSL-23100; Diagnostic Systems Laboratories Inc., Webster, TX, USA) levels. All fractions were assayed undiluted. The coefficients of variation for intraassay and interassay, respectively, are the following: IL-1α, 8.3 and 11.2%; β-endorphin, 4.1 and 7.1%; and leptin, 3.7 and 6.6%.
Statistical analysis.
Unless otherwise specified, results are expressed as mean concentrations in nanograms per liter or micrograms per liter ± SE. The p = 0.05 significance level was used for the statistical analysis. We grouped human milk samples into third postpartum day smoker and nonsmoker mother groups, with 42 and 40 samples per group, respectively, and into 10th postpartum day smoker and non-smoker mother groups, with 42 and 40 samples per group, respectively.
Because the concentrations of cytokine IL-1, β-endorphin, and leptin were not normally distributed, we performed a paired analysis with the Mann-Whitney U-test. We used the Student t-test for the analysis of the other data.
Ethics.
The research protocol was approved by the Hospital Ethical Committee for Human Research of the University of Padua.
Results
We analyzed 164 milk samples, collected from 82 mothers, 42 smokers, and 40 nonsmokers. Measurable immunoreactivity levels were observed in most colostral (third day) and transitional (10th day) milk samples throughout the collection period (> 90%). The smoker mothers of healthy-term infants and control nonsmoker mothers of healthy-term infants who planned to breast-feed their infants were comparable in their anthropometric characteristics and those of the respective neonates. The birth weight of the newborn infants of smoker mothers was not significantly lower (Table 1).
IL-1α concentrations were significantly reduced in the colostrum of smoker compared with the nonsmoker control mothers. β-Endorphin and leptin colostrum concentrations were comparable (Table 2).
Moreover, β-endorphin concentrations were significantly reduced in transitional milk samples compared with colostrum of both smoker and nonsmoker mothers. Also, leptin concentrations were significantly reduced in transitional milk samples compared with colostrum of both smoker and nonsmoker group. IL-1α concentrations were lower in transitional milk than in colostrum samples of smoker and nonsmoker mothers, but without any significance (Table 2).
We found no significant differences between smoker and nonsmoker lactating mothers in transitional milk concentrations of IL-1α, β-endorphin, and leptin (Table 2). And we found no correlations between reduced IL-1α levels of the colostral milk samples and related immunomodulator β-endorphin and leptin levels.
Discussion
In this article, we present novel evidence suggesting that exposure to tobacco smoke during pregnancy may affect the development of the immunologic function of the mammary gland, significantly influencing colostral milk provision of the proinflammatory cytokine IL-1α for breast-feeding infants. By contrast, the concentration of the immunomodulatory factors β-endorphin and leptin were unaffected. In this context, the possibility that delivery modalities, parity, or BMI affects the levels of cytokines and immunomediators in the breast milk is open to speculation.
The identification of reduced IL-1α levels in the colostral milk of mothers exposed to tobacco smoke during pregnancy offers support to the hypothesis that some clinical consequences of smoking on infancy might be initiated and enhanced by altered levels of inflammatory cytokines (Froen et al. 2000; Sayers and Drucker 1999).
Early breast milk may be of particular importance in the development of the mammalian newborn innate and acquired immunity (Goldman et al. 1998). Colostrum has been reported to contain a large amount of cytokines, but little is known as to their concentrations and associations within breast milk (Laiho et al. 2003). IL-1 is probably the first cytokine that has been quantified in human milk (Garofalo and Goldman 1998; Munoz et al. 1990). The synthesis, processing, secretion, and activity of IL-1 are tightly regulated events. For instance, the presence of an IL-1 receptor antagonist is an important mechanism through which IL-1 signaling is down-regulated, and therefore its activity is reduced (Dinarello 1997). Of interest, IL-1β receptor antagonists have been measured in human breast milk (Buescher and Malinowska 1996). To what extent the biologic effect of the cytokine in milk is modulated by the presence of soluble receptors and other cytokine antagonist is open to speculation.
Smoking is a major risk factor for both infantile infections and SIDS. Such infections, both viral and bacterial, also increase the SIDS risk (Gordon et al. 2002; Guntheroth 1989). In experimental animals, nicotine has been shown to depress both the primary and secondary immune response of the lungs, lymph nodes, and spleen (Sayers and Drucker 1999). Interestingly, most studies have shown a decrease in proinflammatory cytokine production (IL-1α, IL-6, and tumor necrosis factor-α) by lung macrophages in smokers, accounting for the greater concentration of tobacco in the lungs (Froen et al. 2000; Sayers and Drucker 1999). These cells have proven very sensitive to both smoking history and the duration of the smoke-free period (Zeidel et al. 2002). Similarly, the greater concentration of tobacco components reported in the breast milk of smoker mothers could imply a different magnitude of immune impairment of the mammary gland macrophages (Klonoff-Cohen et al. 1995). The possibility that altered levels of cytokines and other biologic factors in breast milk modify the immune function in the nursed infants is a very attractive hypothesis (Jones and Warner 2000). Physiologic delays or imbalance in the production of immune factors can occur in mammals, and such delays increase the risk of infection and SIDS (Laiho et al. 2003).
Human milk is an important source of other bioactive substances, including hormones, growth factors, and immunologic factors such as cytokines, but the functional consequences of an overexpression or of a down-regulation of most milk immunomodulatory constituents in neonates are unknown. In our study, the levels of β-endorphin and leptin in breast milk were not related to smoking habits of nursing mothers. The lack of additive effects and the dissimilarity in response to tobacco smoke might indicate that these immunomodulatory molecules are involved in different pathways. In this context, it is interesting that nicotine can cause a release of β-endorphin in various brain regions, and that cytokines and endotoxin can cause a release of β-endorphin in different tissues. Moreover, in mice, leptin, induced by LPS and cytokines, participates in the host response to inflammation by modulating the host immune and cytokine responses (Faggioni et al. 1999).
There are certain limitations in our work. First, only colostral and transitional milk samples were collected. It is well established that milk composition differs with delivery modalities, increasing parturition, throughout the day, at each feeding, and with time. Generally, milk proteins that are produced and secreted in the mammary gland are expected to constitute the largest proportion of protein in human milk and to have more postpartum variation than do serum proteins that are passively transferred into milk (Londerdal and Atkinson 1995). Having multiple cytokine and immunomediators measures taken over time from breast-feeding mothers with a dose–response relationship between smoking levels and levels of agents would permit verification of the assumption underlying this analysis. In addition, the selection of IL-1, of leptin, and of β-endorphin, among the wide range of immunomodulatory factors present in breast milk and interconnecting the neuroendocrine and immune systems, has been made in the context of the present study mainly for convenience and not according to solid scientific reasons.
However, the study has several strengths. Our measure of cytokines in colostrum has been used in previous studies and is valid. Furthermore, most of the colostral and transitional milk samples (> 90%) in the study contained IL-1α, so this is a large enough group to be confident of the results and of measurable presence of IL-1α in the early milk. Nevertheless, it is premature to know how the findings will pertain to later infections, because most children of smoking mothers do not develop infections in the early perinatal time.
Conclusion
Our data support the suggestion that some clinical consequences of smoking might be initiated and enhanced by the production of inflammatory cytokines from colostrum. Reduced colostral IL-1α concentrations provide additional data to the delineation of the pathophysiologic mechanism that includes mammary gland immunologic dysfunction in the cascade of events that can lead to infections and SIDS. To our knowledge, this is the first study on the effects of smoking on the content of cytokines and immunomediators in human milk. Prospective studies are required to thoroughly assess the differences in the composition of breast milk from mothers with different smoking histories and the immunologic consequences of this for the offspring.
Correction
The name of co-author Angelo Barbato was added after publication online.
We thank M. Mead and F. Cavallin for their contributions to this article.
Table 1 Demographics, pregnancy, and birth outcomes characteristics of the study groups.
Characteristic Smokers (n = 42) Nonsmokers (n = 40)
Maternal data
Maternal age (years) 31.1 ± 0.9 31.2 ± 1.1
Maternal smoking (cigarettes/day) 3.2 ± 0.7
Maternal education (years) 10.5 ± 2 10.6 ± 3
Parity [primiparous (%)] 35.1 41.4
BMI (kg/m2) 25.3 ± 4.0 25.7 ± 4.2
Vaginosis (%) 2.3 2
Labor induction (%) 7.1 7.2
Duration of labor (hr) 8.9 ± 4.2 8.6 ± 4.3
Mode of delivery [vaginal (%)] 88 92
Birth data
Gestational age (weeks) 39.0 ± 0.3 40.0 ± 0.3
Birth weight (kg) 3.23 ± 0.14 3.48 ± 0.09
Sex (%male) 44 42
Apgar score
1st min 8.8 ± 0.4 9.2 ± 0.2
5th min 9.5 ± 0.1 9.8 ± 0.1
Data are presented as mean ± SD unless otherwise indicated.
Table 2 IL-1α, β-endorphin, and leptin concentrations in the colostrum and transitional milk of smoker compared with nonsmoker mothers.
Colostrum
Transitional milk
Smokera Nonsmokerb Smokerc Nonsmokerd p-Value
IL-1α (ng/L) 17.2 ± 4.0 38.4 ± 7.4 14.4 ± 5.2 21.7 ± 12.5 a vs. b*
β-Endorphin (ng/L) 353.5 ± 29.1 317.5 ± 27.6 152.8 ± 26.5 127.7 ± 2.3 a vs. c**
b vs. d**
Leptin (μg/L) 1.6 ± 0.3 1.5 ± 0.2 0.7 ± 0.1 0.5 ± 0.1 a vs. c*
b vs. d**
Data are presented as mean ± SE, or p-value from Mann-Whitney U-test.
* p < 0.05
** p < 0.01.
==== Refs
References
Bentzen K 1994 Cytokines and natural regulators of cytokines Immunol Lett 43 111 123 7737681
Blizzard L Pansonby AL Dwyer T Venn A Cochrane JA 2003 Parental smoking and infant respiratory infection: how important is not smoking in the same room with the baby? Am J Public Health 93 482 488 12604500
Bottcher MF Jenmalm MC Garofalo RP Biokstén B 2000 Cytokines in breast milk from allergic and non-allergic mothers Pediatr Res 47 157 162 10625097
Buescher ES Malinowska I 1996 Soluble receptors and cytokine antagonists in human milk Pediatr Res 40 839 844 8947960
Dahlström A Lundell B Curvall M Thapper L 1999 Nicotine and cotinine concentrations in the nursing mother and her infant Acta Paediatr Scand 79 142 147 2321475
Dinarello CA 1997 Interleukin-1 Cytokine Growth Factor Rev 8 253 365 9620641
Dinarello CA 1998 Interleukin-1, interleukin-1 receptors and interleukin-1 receptor antagonist Int Rev Immunol 16 467 499
Elliot JC Carroll NG James AL Robinson PJ 2003 Airway alveolar attachment points and exposure to cigarette smoking in utero Am J Resp Crit Care Med 167 45 49 12502475
Faggioni R Fantuzzi G Gabay C Moser A Dinarello CA Feingold KR 1999 Leptin deficiency enhances sensitivity to endotoxin-induced lethality Am J Physiol 276 36 42
Froen JF Akre H Stray-Pedersen B Saugstad OD 2000 Adverse effects of nicotine exposure and interleukin-1β on autoresuscitation after apnea in piglets: implication for sudden infant death syndrome Pediatrics 105 E52 10742373
Garofalo RP Goldman AS 1998 Cytokines, chemokines and colony-stimulating factors in human milk: the 1997 update Biol Neonate 74 134 142 9691155
Goldman AS Chheda S Garofalo R 1998 Evolution of immunologic functions of the mammary gland and the postnatal development of immunity Pediatr Res 43 155 162 9475278
Goldman AS Chheda S Garofalo R Schmalstieg FC 1996 Cytokines in human milk: properties and potential effects upon the mammary gland and the neonate J Mammary Gland Biol Neoplasia 1 251 258 10887499
Gordon AE El Ahmer OR Chan R Al Madani OM Braun JM Weir DM 2002 Why is smoking a risk factor for sudden infant death syndrome? Child Care Health Dev 28 23 25 12515434
Guntheroth WG 1989 Interleukin-1 as intermediary causing prolonged sleep apnea and SIDS during respiratory infections Med Hypotheses 28 121 123 2784531
Jones CA Warner JO 2000 Breast milk as an alternative source of cytokines for offspring Clin Exp Allergy 30 599 601 10792349
Kalf GF Renz JF Niculescu R 1996 p -Benzoquinone, a reactive metabolite of benzene, prevents the processing of preinterleukins-1α and -1β to active cytokines by inhibition of the processing enzymes, calpain, and interleukin-1β converting enzyme. Environ Health Perspect 104 1251 1256 9118901
Khachaturian H Lewis ME Tsou K 1958. β-Endorphin, β-MSH, ACTH and related peptides. In: Proteins and Hormones (Bjorklund A, Hokfelt T, eds). New York:Elsevier Science, 216–272.
Klonoff-Cohen HS Edelstein SL Lefkowitz ES Srinivasan IP Kaegi D Chang JC 1995 The effect of passive smoking and tobacco exposure through breast milk on sudden infant death syndrome JAMA 273 795 798 7861574
Kovar MG Serdula MK Marks JS Fraser DW 1984 Review of the epidemiologic evidence for an association between infant feeding and infant health Pediatrics 74 615 638 6384916
Laiho K Lampi AM Hamalainen M Moilanen E Piironen V Arvola T 2003 Breast milk fatty acid, eicosanoids and cytokines in mothers with and without allergic disease Pediatr Res 53 4 642 647 12612204
Londerdal B Atkinson S 1995. Human milk proteins. In: Handbook of Milk Composition (Jensen R, ed). San Diego:Academic Press, 351–368.
Luck W Nau H 1985 Nicotine and cotinine concentrations in serum and urine of infants exposed via passive smoking or milk from smoking mothers J Pediatr 107 816 820 4056988
Mitchell EA Ford RP Stewart AW Becroft DM Taylor BJ Thompson JM 1993 Smoking and the SIDS Pediatrics 91 893 896 8474808
Munoz C Enders S Van Der Meer J Schlesinger L Arevalo M Dinarello C 1990 Interleukin-1β in human colostrum Res Immunol 141 501 513
Oddy WH Halonen M Martinez FD Lohman IC Stern DA Kurzius-Spencer M 2003 TGF-β in human milk is associated with wheeze in infancy J Allergy Clin Immunol 112 723 728 14564350
Ouyang Y Virasch N Hao P Aubrey MT Mukerjee N Bierer BE 2000 Suppression of human IL-1β, IL-2, IFN-γ, and TNF-α production by cigarette smoke extracts J Allergy Clin Immunol 106 280 287 10932071
Sayers NM Drucker DB 1999 Animal models used to test the interactions between infectious agents and products of cigarette smoke implicated in sudden infant death syndrome FEMS Immunol Med Microbiol 25 115 123 10443499
Schulte-Hobein B Schwartz-Bickenbach D Abt S Plum C Nau H 1992 Cigarette smoke exposure and development of infants throughout the first year of life: influence of passive smoking and nursing on cotinine levels in breast milk and infant’s urine Acta Paediatr 81 550 557 1392372
Sopori ML Kozak W Savage SM Geng Y Soszynski D Kluger MJ 1998 Effect of nicotine on the immune system: possible regulation of immune response by central and peripheral mechanism Psychoendocrinology 23 189 204
Stedman RL 1968 The chemical composition of tobacco and tobacco smoke Chem Rev 68 153 207 4868017
Zanardo V Nicolussi S Giacomin C Faggian D Favaro F Plebani M 2001 Labor pain effects on colostral milk β-endorphin concentration of lactating mothers Biol Neonate 79 87 90 11223648
Zeidel A Beilin B Yardeni I Mayburd E Smirnov G Bessler H 2002 Immune response in asymptomatic smokers Acta Anaesthesiol Scand 46 959 964 12190796
|
16203256
|
PMC1281289
|
CC0
|
2021-01-04 23:41:27
|
no
|
Environ Health Perspect. 2005 Oct 15; 113(10):1410-1413
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.7702
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences 10.1289/ehp.7669ehp0113-00141416203257ResearchMini-MonographLessons Learned for the National Children’s Study from the National Institute of Environmental Health Sciences/U.S. Environmental Protection Agency Centers for Children’s Environmental Health and Disease Prevention Research Kimmel Carole A. 1Collman Gwen W. 2Fields Nigel 3Eskenazi Brenda 41 National Children’s Study Interagency Coordinating Committee, National Center for Environmental Assessment, Office of Research and Development, U.S. Environmental Protection Agency, Washington, DC, USA2 Division of Extramural Research and Training, National Institute of Environmental Health Sciences, National Institutes of Health, Department of Health and Human Services, Research Triangle Park, North Carolina, USA3 National Center for Environmental Research, Office of Research and Development, U.S. Environmental Protection Agency, Washington, DC, USA4 Center for Children’s Environmental Health Research, School of Public Health, University of California, Berkeley, California, USAAddress correspondence to C.A. Kimmel, consultant, National Children’s Study, National Institute of Child Health and Human Development, 6100 Executive Blvd., Suite 5C01, Bethesda, MD 20892-7510 USA. Telephone: (301) 435-3427. Fax: (301) 480-1222. E-mail:
[email protected];
[email protected] authors declare they have no competing financial interests.
10 2005 24 6 2005 113 10 1414 1418 12 10 2004 14 3 2005 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright. This mini-monograph was developed to highlight the experiences of the National Institute of Environmental Health Sciences (NIEHS)/U.S. Environmental Protection Agency (EPA) Centers for Children’s Environmental Health and Disease Prevention Research, focusing particularly on several areas of interest for the National Children’s Study. These include general methodologic issues for conducting longitudinal birth cohort studies and community-based participatory research and for measuring air pollution exposures, pesticide exposures, asthma, and neuro-behavioral toxicity. Rather than a detailed description of the studies in each of the centers, this series of articles is intended to provide information on the practicalities of conducting such intensive studies and the lessons learned. This explication of lessons learned provides an outstanding opportunity for the planners of the National Children’s Study to draw on past experiences that provide information on what has and has not worked when studying diverse multiracial and multi-ethnic groups of children with unique urban and rural exposures. The Children’s Centers have addressed and overcome many hurdles in their efforts to understand the link between environmental exposures and health outcomes as well as interactions between exposures and a variety of social and cultural factors. Some of the major lessons learned include the critical importance of long-term studies for assessing the full range of developmental consequences of environmental exposures, recognition of the unique challenges presented at different life stages for both outcome and exposure measurement, and the importance of ethical issues that must be dealt with in a changing medical and legal environment. It is hoped that these articles will be of value to others who are embarking on studies of children’s environmental health.
asthmaautismchildrenenvironmental healthNational Children’s StudyNIEHS/EPA Children’s Centersobesitypregnancy
==== Body
The series of six articles in this mini-monograph developed out of a desire to learn from the collective experiences of the Centers for Children’s Environmental Health and Disease Prevention Research (hereafter Children’s Centers) in a way that could be useful for the design and implementation of the National Children’s Study. The Children’s Centers are co-sponsored by the National Institute for Environmental Health Sciences (NIEHS) and U.S. Environmental Protection Agency (EPA) and conduct both observational studies of etiology and intervention studies. Several of the exposures and outcomes studied by the Children’s Centers are of interest for the National Children’s Study. This mini-monograph highlights the experiences of the Children’s Centers’ studies, and the articles are meant to serve as a primer of lessons learned for the National Children’s Study. The articles represent a synthesis of thoughts on several topics: methodologic issues for conducting longitudinal birth cohort studies (Eskenazi et al. 2005) and community-based participatory research (Israel et al. 2005) and issues in the measurement of air pollution (Gilliland et al. 2005), pesticide exposures (Fenske et al. 2005), asthma (Eggleston et al. 2005), and neurobehavioral toxicity (Dietrich et al. 2005). Summarizing the combined experiences of the Children’s Centers at this time afforded an opportunity to inform the planning and protocol development of the National Children’s Study and to provide information on what has worked and what has not worked when studying diverse multi-racial and multiethnic groups of children with unique urban and rural exposures.
The National Children’s Study
The idea for the National Children’s Study originated from the Developmental Disorders Workgroup of the President’s Task Force on Environmental Health Risks and Safety Risks to Children, established in 1997 as a result of Federal Executive Order 13045, signed by President Clinton on 21 April 1997. This Executive Order, titled “Protection of Children from Environmental Health Risks and Safety Risks” (Clinton 1997) required federal agencies involved in related activities to consider special environmental circumstances that could pose a health threat to children. In late 1999, the task force approved the exploration of the feasibility of the Longitudinal Cohort Study of Environmental Effects on Parents and Children (later renamed the National Children’s Study). Subsequently, the Children’s Health Act (2000), which Congress passed unanimously in October 2000, authorized the planning and implementation of this study. The Children’s Health Act charged the director of the National Institute of Child Health and Human Development/National Institutes of Health together with a consortium of federal agencies, including the U.S. EPA, the Centers for Disease Control and Prevention, and other appropriate agencies to plan and implement this national longitudinal study of environmental influences on children’s health and development. The NIEHS shortly thereafter became the fourth lead partner agency. The Children’s Health Act broadly defined the environment to include physical, chemical, biologic, and psychosocial factors. Further, the act called for investigation of basic mechanisms of developmental disorders and environmental factors, both health adverse and health promoting.
Plans for the National Children’s Study are to enroll pregnant women as early in pregnancy as possible and to enroll a subset of the cohort before conception. Current plans are to follow 100,000 births to adulthood (21 years of age), with collection of data on exposures and outcomes at points throughout the study. Two major goals of the study are to use the longitudinal design as a way to link exposures with outcomes that occur at different points in time (i.e., exposure may precede the resulting outcome by months or years) and to explore interactions among various factors, including genetic traits. The general design and administrative structure are described by the National Children’s Study Interagency Coordinating Committee (2003) and the Study Plan (National Children’s Study 2005).
Planning for this large study began in mid-2000 with the initiation of a number of methods development studies, particularly the development of less burdensome and less costly exposure assessment methods (U.S. EPA 2004a). Other efforts have included laboratory studies on noninvasive procedures for analysis of biomarkers of exposures and outcomes (Rockett et al. 2002), development of comparable measures that could be administered in children and laboratory animal models enabling further exploration of exposure–outcome links in animal studies (Sharbaugh et al. 2003), and field studies using focus groups to explore recruitment and retention issues. A number of reviews and white papers have been developed, including an evaluation of advanced technology for capturing data, a review of the leading hypotheses of the effects of environmental factors on children’s health outcomes, and a review of the leading sampling strategies for consideration in the study. In addition, a resource database on biomarkers of exposures and outcomes was developed (U.S. EPA 2004b). More recently, a number of workshops have been organized to explore and refine the measures that might be included in the protocol for the study. A complete listing of workshops, meetings, and reports can be found at the National Children’s Study website (2005).
Because of the research being conducted by the Children’s Centers, a good deal of expertise and specialized knowledge have already been developed that can be particularly valuable to the development of protocols for the National Children’s Study. As detailed below, much of the work of the original eight centers was on pesticide exposures and neurodevelopment, and air pollution and asthma. The four additional Children’s Centers are focusing on developmental disabilities, particularly autism spectrum disorders, and the impact of various environmental pollutants on learning and behavior. These and other exposures and outcomes have been established as priorities for the National Children’s Study (Table 1). Thus, the Children’s Centers have experience in several of the priority focus areas that should provide valuable input to the design of the National Children’s Study.
In addition to these priority focus areas, it is recognized that community participation and involvement are key to the success of the National Children’s Study. The study design is likely not to be strictly community based, but rather will involve multiple sites throughout the country with a complex multilevel sampling strategy for obtaining the most representative population possible for inclusion in the study. Perhaps the greatest lesson to be learned from the Children’s Centers is the myriad of logistical complexities to be considered in an intensive investigation of children’s exposure and health outcomes. The cumulative experience of the Children’s Centers in conducting community-based participatory research, particularly in underserved populations, will provide valuable information for future studies.
The NIEHS/U.S. EPA Centers for Children’s Environmental Health and Disease Prevention Research
The same federal executive order that spawned the National Children’s Study also led to development of the Children’s Centers. The NIEHS and U.S. EPA collaborated in 1998 to develop a research program that would bring together efforts to better understand the exposures of infants and young children, study the health effects of such exposures to better understand the mechanisms by which they work, and explore intervention strategies for reducing such exposures in a way that would provide evidence for practice. The program was funded in two phases; the first phase in 1998 funded eight centers, and four more centers were funded in 2001. In 2003, a new round of competition was completed and seven awards were made. Six of the pre-existing centers continue to date, and a new center was added. The motivation for this program and a summary of the first eight centers has been more fully discussed by Dearry et al. (1999) and O’Fallon et al. (2000). A full description of the Children’s Centers can be found on the NIEHS website (NIEHS 2005).
The purpose of the Children’s Centers program is 2-fold: first, to create local research environments that promote multidisciplinary interactions among basic, clinical, and behavioral scientists through university/community partnering in order to accelerate translation of basic research findings into clinical prevention or intervention strategies; and second, to support a coordinated nationwide network of scientists and community advocacy groups synergistically sharing their experiences to address relevant questions related to the role of environmental exposures in the health of children in order to enhance community-level capacity to identify and address environmental threats and prevention opportunities. The aims of establishing this national network are to foster communication, innovation, and excellence in children’s environmental health; to provide training opportunities for scientists and clinicians for future development of this field of study; and to broaden the national discussions between diverse groups of community advocates and organizers on common interests in protecting and nurturing healthy environments for children.
Each center is designed around a central theme focusing on important questions in understanding the role of exposures in one of the following health outcome areas: respiratory disease, childhood learning, and growth and development including developmental disabilities. Exposures to toxicants such as polychlorinated biphenyls (PCBs), mercury, lead, air pollution, allergens, agricultural and urban pesticides, second-hand smoke, and others are part of the Centers Program’s research priorities. All centers have multiple projects across scientific disciplines, from basic laboratory-based research and genetics to exposure assessment, epidemiology, and clinical trials. Methods have been developed, field tested, and implemented to detect and define health symptoms and outcomes; new exposure technology has been developed to assess environmental exposures and body burden in a diverse array of biospecimens; and creative approaches to reaching and retaining traditionally “hard to follow” socioeconomic groups have been implemented with community input.
Over the past 5 years and using a variety of methods, scientists from the NIEHS/U.S. EPA Children’s Centers Program have made a number of advances that would not have been possible without the establishment of a coordinated network of centers to foster multi- and interdisciplinary research targeted at understanding children’s environmental health risks and reducing them. Many of the results of these collaborations and interactions have implications for children’s environmental health and the National Children’s Study. For example, studies have shown that blood and urine specimens from pregnant women show measurable levels of pesticides, suggesting that the fetus is exposed to these chemicals during early development (Bradman et al. 2003; Whyatt et al. 2003); children in urban and rural environments are exposed to a complex mix of agricultural and household pesticides, environmental tobacco smoke, and polycyclic aromatic hydrocarbons and negative social factors that can affect their early growth and development (Berkowitz et al. 2004; Eskenazi et al. 2004; Perera et al. 2004; Rauh et al. 2004); and exposures to lead in the urban environment can have lifelong effects such as behavioral problems and criminal behavior in early adult life (Ris et al. 2004). Research on air pollution and asthma has broadened our understanding of the inflammatory process in the lung (Walters et al. 2002) such that the effects of air pollution can be seen in school-age children as increased exacerbation of asthma symptoms and increased days absent from school (Gilliland et al. 2003; McConnell et al. 2003). Intervention studies have been conducted to show that the household environment can be cleaned up in a way to significantly reduce allergens from dust mites and cockroaches that should reduce the incidence of asthma symptoms in children (Eggleston et al. 2004).
Community-based participatory research.
A requirement for every Children’s Center is the inclusion of one project that uses community-based participatory research methods. This type of methodology encourages full participation of the community in the design, implementation, evaluation, and translation of the research. Research ideas may begin with the concerns of the community. Community partners, scientists, and clinicians share knowledge of exposure, health effects, and prevention strategies. Many studies train and employ community members as study coordinators, interviewers, or environmental technicians. Community participation ensures the relevance of the research questions and appropriateness of the research strategies. Research results are disseminated back to the community on an ongoing basis through community advisory boards, newsletters, health fairs, and other educational activities.
Cohorts and study designs.
Most of the centers have established cohorts of children in which to study the dynamic relationship between exposures to environmental agents and health outcomes. In general, two distinct age groups are targeted: birth cohorts where pregnant women were enrolled and their offspring become study participants, and cohorts of school-age children enrolled either in school settings or in a medical care environment. The birth cohort studies seek to understand exposures during fetal development and health risks related to respiratory illness progression and neurodevelopmental effects, including motor, sensory, and cognitive deficits. Because asthma cannot be definitively diagnosed until ages 3–4, prospective follow-up of a group of young children provides for new opportunities as they age. The school-age cohort studies focus on asthma, and children are recruited through school classrooms, neighborhood health clinics, and other medical care settings. One large cohort study of school children in Los Angeles, California, that was started 10 years ago continues to follow the children and compare genetic factors from recently collected specimens with historical air pollution and medical data. Case-only or case–control designs are used in two other studies of children that focus on understanding the possible environmental causes of autism, a relatively rare disorder. Intervention/prevention studies include cohorts of children with disease or unique exposures that can be found in urban and rural settings. These studies are unique with regard to community participation, recruitment and retention, and dissemination of study results; Table 2 lists the types of studies and centers.
Government/Children’s Centers partnerships.
The NIEHS and U.S. EPA, as federal funding partners, continue to align their priorities and working relationships to manage and support this $140 million program. The agencies’ commitment to overcoming differences in their regulatory and research mandates is reflected in the broad success of the Centers Program’s impact on public policy and influence in several fields of public health. The federal partners share responsibility in both supporting the national network of researchers and sponsoring annual center meetings. Bringing center scientists and community members together on a consistent basis has been instrumental in the success of these programs. This created a stronger working relationship across the Children’s Centers than would have been fostered with individual programs working alone. At the inception of the program, many meetings were held to discuss definitions of health outcomes and ways to measure them, methodologies for exposure assessment, questionnaire items, and follow-up strategies with special attention to retention of study participants, cultural sensitivities, and engagement of community. Information was shared and protocols were designed for individual studies that strove for commonality. The goal was not to have standardized methods employed, but rather to see where collaborations could be built and methodologies shared.
The NIEHS/U.S. EPA Children’s Centers Program is now in its seventh year. The program has generated important scientific results and expanded our knowledge of exposures to young children and how they affect their health status. There is a wealth of knowledge about issues that pertain to conducting future studies in this field, especially the National Children’s Study. This mini-monograph is an attempt to describe in detail the lessons learned from these important groundbreaking studies.
Major Lessons Considered Important for Planning the National Children’s Study
Several major lessons from the Children’s Centers are important for consideration in planning the National Children’s Study. These and a number of others are discussed in detail in the articles in this mini-monograph (Dietrich et al. 2005; Eggleston et al. 2005; Eskenazi et al. 2005; Fenske et al. 2005; Gilliland et al. 2005; Israel et al. 2005).
First, long-term studies that follow participants into adolescence and early adulthood are considered essential to assess the full range of developmental consequences of exposure to environmental chemicals.
It is also important to identify a population with a wide range of exposure concentrations for those key pollutants hypothesized a priori to be of interest in order to evaluate the relationships between the distributions of multiple exposures and observed effects.
It is necessary to allow for population differences in literacy, language, and culture when establishing study procedures for recruitment and retention and in determining the type of information collected and the methods of collection.
Assessment tools need to balance measures both broad and narrow in scope. Questionnaires, neurodevelopmental instruments, and the like employed in these studies should include a core set to evaluate the entire cohort and additional segments for selected populations that may be unique based on their exposure or other attributes.
Exposure assessment should include a combination of environmental and personal measurements as well as data derived from questionnaires and from observational and ecologic data. The exposure assessment effort should take advantage of modeling approaches to provide estimates for the entire cohort. Targeted exposure studies in a selected sub-sample of study subjects may be useful for improving exposure assessment. The depth of assessments that can be realistically implemented will be restricted in populations that are widely dispersed geographically, have limited transportation, or lack trained personnel in the community.
Procedures for monitoring the quality and accuracy of data collection must be established and maintained not only for the collection and analysis of biologic or environmental specimens, but also for the assessment of questionnaire, developmental testing, and other health outcome data. Data safety and monitoring procedures must be in place.
Active and meaningful participation of the community is essential for determining the relevant research questions, enrolling and retaining the cohort in an intensive investigation over the long term, and contributing to translation of scientific principles and research results for communities and the public at large. This requires establishing trust and respecting differences in culture and knowledge of the community. Sufficient time and resources are necessary to develop community partnerships.
The ethical issues in a longitudinal birth cohort study are likely to become increasingly more complex in the changing medical and legal environment and must be carefully considered in designing research protocols and following the cohort. It is necessary to develop clear plans of referral when children with disease, developmental difficulties, or adverse social situations emerge.
Communication of risk to participants and the community and translation of research findings into interventions and policies are of utmost importance and should be part of the research plan and cost consideration.
Funding for a longitudinal birth cohort study must be adequate for the start-up period, continuous without gaps, and long term. Costs have often been underestimated because tracking and maintaining study participants is labor intensive.
The unique characteristics at each developmental stage from birth through adulthood must be considered. Every age presents special challenges in both outcome and exposure assessment.
Finally, the health and development of children are multifactorially determined. The greatest challenge is anticipating the data and specimens that will allow the questions of the future to be answered. This requires state-of-the-art tools for data collection and tracking participants, environmental and biologic specimen repositories, and anticipation of future human subject requirements in consent procedures.
The unique challenges faced by the Children’s Centers in studying diverse populations will be especially helpful for the National Children’s Study, which is intended to be a nationwide study representative of the many populations across the United States. Although the Children’s Centers have reported important findings from their individual studies, it is only by examining the collective experiences of the Children’s Centers in these lessons learned articles that we gain a better perspective of the potential challenges to be met in the many National Children’s Study sites.
We acknowledge the support of the National Children’s Study in the development of the articles in this mini-monograph. We also thank A. Branum (National Center for Health Statistics, Centers for Disease Control and Prevention) for assistance in developing the questions to frame the articles in this series, as well as several other members of the National Children’s Study Interagency Coordinating Committee and Program Office who kindly reviewed various drafts of the manuscripts.
The views expressed in this article are those of the authors and do not necessarily represent the views or policies of the U.S. Environmental Protection Agency.
This article is part of the mini-monograph “Lessons Learned from the National Institute of Environmental Health Sciences/U.S. Environmental Protection Agency Centers for Children’s Environmental Health and Disease Prevention Research for the National Children’s Study.”
Table 1 Priority health outcomes and exposure areas identified for the National Children’s Study.
Priority health and disease outcomes
Pregnancy outcomes
Neurodevelopment and behavior
Childhood injury
Asthma
Obesity and physical development
Priority environmental exposures and other factors
Physical exposures and environment
Chemical exposures
Biologic environment and genetics
Psychosocial environment and exposures
Table 2 Primary outcomes, exposures, and populations studied by the Centers for Children’s Environmental Health and Disease Prevention Research.
Study Outcomes or health concern Exposures or intervention Population
Birth cohorts
University of California, Berkeley Infant growth and development Pesticides, allergens Latinos, in an agricultural community, Salinas Valley, California
Columbia University Infant growth and development, asthma Pesticides, lead, smoke, PAHs Urban Latinos (Dominican), African Americans, NYC
Mt. Sinai Medical Center Infant growth and development, obesity Chlorpyrifos, endocrine-disrupting chemicals, built environment Urban Latinos, African Americans, East Harlem, NYC
University of Illinois Infant growth and development PCBs, mercury Hmong, Wisconsin
Cincinnati Children’s Hospital Infant growth and development Lead, pesticides, smoke Urban African Americans, Cincinnati, Ohio
School-age cohorts
Johns Hopkins University Asthma Allergens, air pollution African Americans, health care based, Baltimore, Maryland
University of Southern California Asthma Air pollution School children in 12 communities in Los Angeles, California
University of Michigan Asthma Air pollution, allergens African Americans, school based, Detroit, MI
University of Iowa Asthma RSV, endotoxin Rural Iowa
Case-only and case–control studies
University of California, Davis Autism Environmental and other risk factors Cases and controls identified throughout California, mentally retarded controls and healthy controls
University of Medicine and Dentistry of New Jersey Autism, regression Home chemical exposures Autistic children with and without regression
Intervention/prevention studies
University of California, Berkeley Agricultural pesticide exposure Worker cleanup in fields Migrant workers in Salinas Valley, California
Columbia University Household pesticides and allergens Integrated pest management African Americans and Latinos, North Manhattan, NYC
Mount Sinai Medical Center Chlorpyrifos use, cockroach allergen Integrated pest management African Americans and Latinos, East Harlem, NYC
University of Illinois Birth outcomes and growth Education about fish consumption (PCBs, mercury) Hmong, Wisconsin
Cincinnati Children’s Hospital Lead Household cleanup and remediation African Americans, inner-city Cincinnati, Ohio
University of Michigan Asthma symptoms Household cleanup African Americans and Latinos, Detroit, Michigan
Johns Hopkins University Asthma symptoms Household cleanup African Americans, Baltimore, Maryland
University of Iowa Asthma symptoms Cleanup, medical management, personalized care plan Rural Iowa
University of Southern California Asthma symptoms Household cleanup School children in 12 communities in Los Angeles, California
University of Washington Agricultural pesticide exposure Pesticide reduction strategies Agricultural workers, Yakima Valley, Washington
Abbreviations: NYC, New York City; PAHs, polyaromatic hydrocarbons; RSV, respiratory syncytial virus.
==== Refs
References
Berkowitz GS Wetmur JG Birman-Deych E Obel J Lapinski RH Godhold JH 2004 In utero pesticide exposure, maternal paraoxonase activity, and head circumference. Environ Health Perspect 112 388 391
Bradman A McKone T Barr DB Harnly ME Eskenazi B 2003 Cumulative organophosphate pesticide exposure and risk assessment among pregnant women living in an agricultural community: a case study from the CHAMACOS cohort Environ Health Perspect 111 1779 1782 14594631
Children’s Health Act of 2000 2000. Public Law 106–310.
Clinton WJ 1997 Executive Order 13045. Protection of Children from Environmental Health Risks and Safety Risks Fed Reg 62 19883 19888
Dearry A Collman GW Saint C Fields N Redd S 1999 Introduction: building a network of research in children’s environmental health Environ Health Perspect 107 suppl 3 391 392 10346987
Dietrich KN Eskenazi B Schantz S Yolton K Rauh VA Johnson CB 2005 Principles and practices of neuro-developmental assessment in children: lessons learned from the Centers for Children’s Environmental Health and Disease Prevention Research Environ Health Perspect 113 1437 1446 16203260
Eggleston PA Butz A Rand C Callahan K Curtin-Brosnan J Merriman B 2004 Home environmental treatment improves asthma in inner city children [Abstract] J Allergy Clin Immunol 113 S179
Eggleston PA Diette G Lipsett M Lewis T Tager I McConnell R 2005 Lessons learned for the study of childhood asthma from the Centers for Children’s Environmental Health and Disease Prevention Research Environ Health Perspect 113 1430 1436 16203259
Eskenazi B Harley K Bradman A Weltzien E Jewell N Barr D 2004 Association of in utero organophosphate pesticide exposure and fetal growth and length of gestation in an agricultural population Environ Health Perspect 112 1116 1124 10.1289/ehp.6789 15238287
Eskenazi B Gladstone EA Berkowitz GS Drew CH Faustman EM Holland NT 2005 Methodologic and logistic issues in conducting longitudinal birth cohort studies: lessons learned from the Centers for Children’s Environmental Health and Disease Prevention Research Environ Health Perspect 113 1419 1429 16203258
Fenske RA Bradman A Whyatt RM Wolff MS Barr DB 2005 Lessons learned for the assessment of children’s pesticide exposure: critical sampling and analytical issues for future studies Environ Health Perspect 113 1455 1462 16203262
Gilliland FD Berhane K Islam T Wenten M Rappaport E Avol E 2003 Environmental tocbacco smoke and absenteeism related to respiratory illness in school children Am J Epidemiol 157 861 869 12746237
Gilliland F Avol E Kinney P Jerrett M Dvonch T Lurmann F 2005 Air pollution exposure assessment for epidemiologic studies of pregnant women and children: lessons learned from the Centers for Children’s Environmental Health and Disease Prevention Research Environ Health Perspect 113 1447 1454 16203261
Israel BA Parker EA Rowe Z Salvatore A Minkler M López J 2005 Community-based participatory research: lessons learned from the Centers for Children’s Environmental Health and Disease Prevention Research Environ Health Perspect 113 1463 1471 16203263
McConnell R Berhane K Gilliland F Molitor J Thomas D Lurman F 2003 Prospective study of air pollution and bronchitic symptoms in children with asthma Am J Respir Crit Care Med 168 790 797 12893648
National Children’s Study 2005. Homepage. Available: http://www.nationalchildrensstudy.gov/[accessed 8 August 2005).
National Children’s Study Interagency Coordinating Committee 2003 The National Children’s Study of environmental effects on child health and development Environ Health Perspect 111 642 646 12676629
NIEHS 2005. Centers for Children’s Environmental Health and Disease Prevention Research. Research Triangle Park, NC:National Institute of Environmental Health Sciences. Available: http://www.niehs.nih.gov/translat/children/children.htm [accessed 28 February 2005].
O’Fallon LR Collman GW Dearry A 2000 The National Institute of Environmental Health Sciences’ research program on children’s environmental health J Expo Anal Environ Epidemiol 10 6 pt 2 630 637 11138655
Perera FP Rauh V Whyatt RM Tsai WY Bernet JT Tu YH 2004 Molecular evidence of an interaction between pre-natal environmental exposures and birth outcomes in a multiethnic population Environ Health Perspect 112 626 630 15064172
Rauh VA Whyatt RM Garfinkel R Andrews H Hoepner L Reyes A 2004 Developmental effects of exposure to environmental tobacco smoke and material hardship among inner-city children Neurotoxicol Teratol 26 373 385 15113599
Ris MD Dietrich KN Succop PA Berger OG Bornschein RL 2004 Early exposure to lead and neuropsychological outcome in adolescence J Int Neuropsychol Soc 10 261 270 15012846
Rockett JC Kavlock RJ Lambright CR Parks LG Schmid JE Wilson VS 2002 DNA arrays to monitor gene expression in rat blood and uterus following 17β-estradiol exposure: biomonitoring environmental effects using surrogate tissues Toxicol Sci 69 49 59 12215660
Sharbaugh C Viet SM Frazier A Mcmaster SB 2003 Comparable measures of cognitive function in human infants and laboratory animals to identify environmental health risks to children Environ Health Perspect 111 1630 1639 14527843
U.S. EPA 2004a. Demonstration of Low Cost, Low Burden, Exposure Monitoring Strategies for use in Longitudinal Cohort Studies. EPA/600/R-04/109. Research Triangle Park, NC:U.S. Environmental Protection Agency. Available: http://www.epa.gov/heasdweb/children/children.htm/[accessed 28 February 2005].
U.S. EPA 2004b. Biomarkers Database. Research Triangle Park, NC:U.S. Environmental Protection Agency, 26 July. Available: http://cfpub.epa.gov/ncea/[accessed 28 February 2005].
Walters DM Breysse PN Schofield B Wills-Karp M 2002 Complement factor 3 mediates particulate matter-induced airway hyperresponsiveness Am J Respir Cell Mol Biol 27 413 418 12356574
Whyatt RM Barr DB Camann DE Kinney PL Barr JR Andrews HF 2003 Contemporary-use pesticides in personal air samples during pregnancy and blood samples at delivery among urban minority mothers and newborns Environ Health Perspect 111 749 756 12727605
|
16203257
|
PMC1281290
|
CC0
|
2021-01-04 23:41:28
|
no
|
Environ Health Perspect. 2005 Oct 24; 113(10):1414-1418
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.7669
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences 10.1289/ehp.7670ehp0113-00141916203258ResearchMini-MonographMethodologic and Logistic Issues in Conducting Longitudinal Birth Cohort Studies: Lessons Learned from the Centers for Children’s Environmental Health and Disease Prevention Research Eskenazi Brenda 1Gladstone Eleanor A. 1Berkowitz Gertrud S. 2Drew Christina H. 3Faustman Elaine M. 3Holland Nina T. 1Lanphear Bruce 4Meisel Stefanie J. 2Perera Frederica P. 5Rauh Virginia A. 5Sweeney Anne 6Whyatt Robin M. 5Yolton Kimberly 41 Center for Children’s Environmental Health Research, School of Public Health, University of California, Berkeley, California, USA2 Department of Community and Preventive Medicine, Mount Sinai School of Medicine, New York, New York, USA3 Center for Child Environmental Health Risks Research, Department of Environmental Health, University of Washington, Seattle, Washington, USA4 Children’s Environmental Health Center, Children’s Hospital Medical Center and University of Cincinnati College of Medicine, Department of Environmental Health, Cincinnati, Ohio, USA5 Columbia Center for Children’s Environmental Health, Mailman School of Public Health, Columbia University, New York, New York6 The Friends Children’s Environmental Health Center, University of Illinois, Urbana-Champaign, Illinois, USAAddress correspondence to B. Eskenazi, Center for Children’s Environmental Health Research, School of Public Health, UC Berkeley, 2150 Shattuck Ave., Suite 600, Berkeley, CA 94720-7380 USA. Telephone: (510) 642-3496. Fax: (510) 642-9083. E-mail:
[email protected] authors declare they have no competing financial interests.
10 2005 24 6 2005 113 10 1419 1429 12 10 2004 10 5 2005 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright. In anticipation of the National Children’s Study, lessons can be learned from the smaller birth cohort studies conducted by five Centers for Children’s Environmental Health and Disease Prevention Research funded by the National Institute of Environmental Health Sciences and the U.S. Environmental Protection Agency. The populations studied are diverse in ethnicity and social class and reside in urban and rural environments. Although almost all of the centers chose to enroll participants through medical care facilities, they had to develop independent staffs and structures because of the overburdened medical care system. Some of the lessons learned by the centers include the importance of continuous funding, building community partnerships to conduct culturally appropriate research, hiring bilingual and bicultural staff from the community, prioritizing research goals, developing biorepositories to ensure future utility of samples, instituting quality control procedures for all aspects of specimen and data collection, maintaining frequent contact with study participants, ensuring ethical conduct of the research in a changing medical-legal climate, and communicating results in a timely and appropriate manner to participants and the wider community. All centers underestimated the necessary start-up time, staff, and costs in conducting these birth cohort studies. Despite the logistical complexity and added expenses, all centers emphasize the importance of studying the impact of environmental exposures on those children most at risk, those living in minority and low-income communities. These centers present barriers encountered, solutions found, and considerations for future research, with the hope that the lessons learned can help inform the planning and conduct of the National Children’s Study.
biologic samplesbiorepositorybirth cohortchildrenenvironmental healthethicsgrowthNational Children’s Studyneurodevelopmentpregnancy
==== Body
Longitudinal birth cohort studies provide a rich source of information about antecedents of disease that originate in pregnancy or childhood. There have been two previous major longitudinal birth cohort studies in the United States: the Child Health and Development Studies (van den Berg et al. 1988) and the National Collaborative Perinatal Project (Niswander and Gordon 1972). Since these studies were conducted, > 40 years ago, science and research infrastructures have changed dramatically in the United States. These changes include but are not limited to advances in bio-markers and molecular and cellular biology, the use of computers in tracking and statistics, the increased difficulty of integrating research into routine clinical practice, the diversity and health disparity of the population, and growing complexities of medical-legal-ethical issues. The standards for quality research have risen considerably and with them the logistical complexities in conducting such research.
The purpose of this article is to outline the methods used by the five Centers for Children’s Environmental Health and Disease Prevention Research (Children’s Centers) that are conducting birth cohort studies. Three centers (University of California, Berkeley, Columbia University, and Mount Sinai School of Medicine) began their studies in 1998, and two centers (University of Cincinnati and University of Illinois) in 2000. All but two centers (Columbia and Mount Sinai), which had prior funding, started enrollment approximately 1 year after funding. Table 1 provides an overview of the five centers (also see Kimmel et al. 2005), each of which recruited racially/ethnically diverse and often low-income populations (Appalachian, Dominican, Hmong, Laotian, Mexican, African American, and Puerto Rican). Our centers worked closely with our respective communities to develop partnerships, strengthen community infrastructure, build trust, and conduct culturally appropriate research (Israel et al. 2005). We present here some of the barriers we faced and our solutions, with the hope that the lessons learned from our experience will assist in the planning and conduct of the National Children’s Study (2005).
Recruitment and Enrollment
Table 2 shows the eligibility criteria and recruitment strategies for each of the studies. Four centers recruited pregnant women. Three centers enrolled women before their third trimester of pregnancy; one center (Columbia) enrolled women throughout pregnancy, and another (Illinois) recruited couples before and during pregnancy.
The center at Illinois attempted to screen all Hmong and Laotian families in the study catchment area for eligibility, using telephone directories as suggested by community leaders. Study workers contacted all people with Hmong or Laotian surnames listed in the directory to describe the study and determine eligibility. Home visits were scheduled with eligible, willing families and were repeated every 2 months. Pregnancy tests were performed at each visit, and couples were enrolled in the cohort study when the women became pregnant.
The other four centers recruited through multiple hospital or clinic sites. They attempted to recruit as many consecutive, eligible patients as possible. The center at Berkeley used clinic staff to screen women for eligibility; eligible women were shown a video about the study and, if interested, were referred to a study worker. Berkeley also enrolled fathers, with only half participating. The center at Cincinnati, with a Health Insurance Portability and Accountability Act (HIPAA 1996) waiver, received weekly information about new patients directly from the clinics. Eligible patients were sent a letter describing the study and a mail-in postcard to decline further contact; those who did not return the postcard were contacted by phone to set up an appointment. The study staff at Columbia recruited participants by approaching women in clinic waiting areas, and the Mount Sinai staff recruited women from a prenatal clinic and two private practices.
Eligibility and exclusion criteria varied. The study populations differed considerably with respect to race/ethnicity, socioeconomic status, and geographic setting (urban vs. rural). Only one center specifically recruited primiparas; the same center was unique in its enrollment of mothers < 18 years of age. Although enrolling minors was not a problem for this center, other centers chose not to include minors because of additional institutional review board (IRB) requirements. Three centers excluded infants from continued follow-up if prenatal specimens or data were not collected. One of these centers also excluded infants if they were high risk (< 32 weeks gestation, < 1,500 g, or having congenital malformations). Some centers determined eligibility by race (e.g., African American) or country of birth (e.g., Dominican), whereas others determined eligibility by language (e.g., Spanish or English speaking). A number of the centers required some stability in residence or in medical care (e.g., planning to deliver at the hospital where the study was based) to be eligible. One center required that the women had lived in the study area for at least a year before pregnancy and planned to remain for at least a year afterward.
Overcoming barriers to recruitment and enrollment.
Participation rates for the studies ranged from 25 to 64%. All centers gathered demographic information on all eligible women, permitting later comparison of participants with eligible nonparticipants.
The most important barriers to participation, especially for working women, were the time required for each visit and the length of the follow-up period. Centers that recruited patients from clinic waiting areas found that even short waiting periods, especially in private practice offices, were a barrier. The one center using clinic staff for recruitment found that they were already overburdened and had little time for recruitment. Some centers also found that women were reluctant to enroll without their husband’s approval.
Many of the populations of interest in children’s environmental health studies are economically disadvantaged, undereducated, non-English-speaking, and distrustful of Western medicine and research. Many centers found that hiring study staff familiar with or from the target population was necessary for successful recruitment. Recruitment by or at clinics known by the community to respect patient confidentiality was particularly successful. In addition, response rates were improved by allowing potential participants time to discuss the study with their families before enrollment.
Assessment Methods
The centers have used a variety of tools to gather information about their cohort study participants (Table 3). At multiple time points during and after pregnancy, mothers completed questionnaires that focused on demographics, medical history, and exposure information (Appendix 1). Illinois included prepregnancy baseline questionnaires and menstrual cycle tracking. Several centers also completed multiple home visits over the course of the study. Questionnaires and home visits were completed in intervals ranging from every 3 months to annually; home visits were usually conducted at the same time points as questionnaires. In addition, all centers conducted neurodevelopmental and growth assessments, and most collected information on medical conditions such as asthma.
Growth, Development, and Other Health Outcomes
Most centers conducted neurodevelopmental assessments and growth assessment at numerous age points after birth (Dietrich et al. 2005). Various standardized neurodevelopmental assessment tools were used to assess the neonate, infant, and child. Two centers plan to collect school-based evaluations such as report cards and teacher ratings of classroom behavior. All centers used standardized anthropometric measurement protocols to measure height, weight, and head circumference at each contact point, some taking multiple measurements to reduce measurement error. Most centers used questionnaires and medical record review to obtain prenatal and child health information for respiratory disease (Eggleston et al. 2005) and other outcomes. Medical records were either abstracted for complete information or to confirm reported conditions. One center received all prenatal and delivery information on computer-ready forms from the participating hospital.
Social Environment
As prescribed by the centers’ Request for Application, many of the participants in these studies were from marginalized, low-income communities. Hence, most centers assessed aspects of the children’s social environment (Appendix 1) expected to affect their health and development. These measurements were obtained from observation, face-to-face interviews, and/or direct child assessment. All centers gathered information about the home environment and household composition, including presence of the father. Almost all centers used the Home Observation for the Measurement of the Environment scale (Caldwell and Bradley 1984). Others included measures of maternal depression, social support, parenting and marital stress, and use of childcare services. Centers that included immigrant populations obtained information on immigration history and acculturation. Socioeconomic status was ascertained by all centers; besides measuring total income and income per person supported, a few centers determined overall material hardship, food security, and use of social services.
Physical Environment
All centers assessed housing quality via questionnaire. In addition, three centers conducted home visits (see Table 3). Two centers used Global Positioning System (GPS) devices to determine the proximity of the home to services, pesticide applications, and high-crime or traffic areas (Gilliland et al. 2005). To assess the condition of the housing stock, centers either modified a measure developed by the U.S. Department of Housing and Urban Development (Jacobs et al. 2002) or developed their own instruments, which included visual assessments for molds/mildew, deteriorated paint, safety hazards, leaks, roach/rodent infestations, and other factors (Appendix 2). Because there are few validated tools to assess exposure to indoor pollutants, creating these materials was challenging. Home inspections themselves were time-consuming and required extensive training, in some cases provided by county housing inspectors. Several centers opted to visit homes multiple times to reassess household exposures, which may vary by season (Yiin et al. 2000) or change when families relocate.
During the home inspections, centers collected ambient measurements and samples, including wall moisture levels, mattress and floor dust samples, and air samples (Table 4).
Collecting environmental measurements often required the purchase of expensive, specialized collection equipment (e.g., air monitors) and a delay between home assessments to allow for cleaning of equipment. Standard practices for interpreting ambient measurements are not yet fully developed; for example, for most contaminants, it is unclear whether house dust concentration (micrograms per gram of dust) or loading (micrograms of surface area) is a better predictor of children’s exposure or body burden.
Overcoming Barriers to Assessment
Delivery events.
In the immediate postpartum period, some centers conducted neonatal assessments and interviews with the mother and, in one center, with the father. In addition, all centers attempted to collect biologic specimens during this time period. Shortening post-delivery hospital stays in the United States left a limited window of opportunity to collect information and samples from mothers and neonates in the hospital. Although women remained in the hospital for 48 hr after delivery at most centers, discharge was occasionally earlier.
Although all centers anticipated quick notification of participants’ admission for delivery, this was often overlooked in the frenzy of labor and birth. All centers relied on both participating women and delivery ward staff for notification. At one center, mothers without home phones were initially given cell phones to call the research team, whereas at other centers women were given special tee shirts or socks to wear to the hospital so as to alert the delivery staff. Some centers provided lists of participants approaching their due date to the medical station, and many checked delivery ward logbooks on a daily basis. Despite these efforts, for most centers where enrollment has ended, a large proportion of women were not tracked at the time of delivery (> 25%). In addition, nighttime and weekend admissions required that center staff were available at all times, sometimes resulting in costly overtime hours. Those centers whose employees were already integrated into the clinical program had a somewhat easier time completing delivery events.
Cord blood samples were particularly difficult to obtain; collection rates at the five centers ranged from 40 to 85%. Most missed collections occurred when women’s delivery admissions were not reported to research staff, although additional samples were missed from high-risk children with emergency deliveries. At least one center collected data at a hospital that did not routinely collect cord blood samples; when it did, a method was used that could result in contamination with maternal blood. In another case, hospital staff were concerned about accidental needle sticks from the traditional venipuncture collection method; this center worked with the hospital to develop an acceptable alternative. The greatest collection rate was reported by the one center that involved physicians on the research team in collecting the samples.
Numerous difficulties were encountered in conducting neonatal assessments. Few tests are available to assess newborn behavior, and their predictive validity is not high. The Brazelton Neonatal Behavioral Assessment Scale requires substantial training, which is available in only a few locations (Lester and Tronick 2001). Trained evaluators who left projects could not be easily replaced, sometimes leaving gaps in cohort assessment.
Although most centers attempted to complete neonatal assessments during the post-delivery hospital stay, this was not always possible. Assessments could not occur too soon after delivery lest behavior be affected by delivery medications, and short hospital stays left little time to schedule assessments when the child was not sleeping or eating. In addition, it was difficult to find a quiet assessment room in some hospitals and interruptions by medical personnel were common. The effect of these obstacles was that assessments intended for the neonatal period were in many cases conducted several weeks after delivery; again, high-risk children who required extra neonatal care were the most difficult to assess in a timely manner. One center increased success with hospital assessments by conducting early morning assessments. Another center chose to assess the child twice, soon after delivery and again a month later.
Participant fatigue.
Recognizing that participation in a longitudinal study is demanding for families, all centers attempted to minimize inconvenience to participants. Centers aimed to optimize contact frequency such that attrition was prevented but participants were not overly bothered. Likewise, all centers designed their contacts to be as brief and efficient as possible. This was particularly challenging for neurodevelopmental assessments; every additional developmental domain assessed increased the risk that children would be excessively fatigued. Many centers found it necessary to narrow the focus of their research questions out of respect for participants’ time.
Centers used several strategies to prevent participant fatigue. Some centers used multiple workers to simultaneously collect information at each visit (e.g., separate assessments of mother and child). This sometimes required that staff were trained in multiple aspects of the study protocol, for example, phlebotomists who were trained to conduct interviews. Some centers found that participants preferred multiple short visits to one long visit, both for convenience and to prevent child fatigue. Weekend and evening assessments, although needed by working parents, pose a strain on staff. For lengthy and demanding assessments, centers sometimes provided snacks and childcare to participants.
Many centers found it critical to use flexible assessment tools that could accommodate necessary protocol changes. Shortened versions of in-person questionnaires were found useful for telephone interviews when participants were not available to meet face to face. Some centers found it helpful to organize neurobehavioral assessments with the most important items first, minimizing data loss when children fatigued. Most centers also developed qualitative assessments that allowed study staff to document participants’ level of fatigue, cooperation, and attention and to record any changes made to the usual study protocol.
Distractions.
Distractions during interviews and assessments posed a challenge for all centers. Conducting assessments in the home was nearly impossible, especially for participants in crowded living conditions; thus, the provision of a standardized testing facility was essential. Minimizing distractions to children during neurobehavioral assessments was particularly challenging. For children > 12 months of age, it was desirable to assess the child separately from the mother to reduce interference; this, however, required additional time for the tester to build rapport with the child. Siblings were also a source of distraction during assessments. Centers accommodated siblings by providing on-site childcare, giving reimbursements for off-site childcare, and/or using videos or games to busy these children; however, most agreed that on-site childcare with dedicated space and personnel would be preferable to these arrangements. Finally, centers helped minimize disruptions to child assessments by scheduling bathroom and snack breaks.
A related issue was participant privacy. For one center, fathers frequently wanted to be present at maternal interviews. Because of concerns that the mothers might not answer personal questions honestly in their presence, partners were not permitted to attend.
Quality control of assessments and interviews.
All centers emphasized the importance of proper staff training. However, although all centers considered pilot testing extremely important, most expressed frustration that time, cost, and the need for prior IRB approval often hindered adequate pilot testing with noncohort participants.
Most centers instituted clear quality control protocols, particularly for the neurodevelopmental assessments. These included direct observations or review of videotapes by the other evaluators and lead psychologists. However, insufficient time and resources caused many centers with clear quality control protocols to fall short of what they considered appropriate quality control (e.g., taping and reviewing ~10% of assessments). Some centers expressed concern about inter-rater differences and reliability issues even after extensive staff training.
Missed appointments.
Many centers had problems with missed appointments and late arrivals. To minimize the frequency of this drain on staff time, centers used a number of strategies. Some called participants several days or hours before to confirm an appointment; others tied research appointments to clinical appointments, which participants seemed more likely to keep. Staff flexibility was required to ensure that even missed assessments could be completed.
Lack of transportation.
Particularly for studies with low-income participants, transportation was a barrier to successfully completing assessments. A number of centers either paid for taxi services or reimbursed participants for alternate travel costs. One center transported participants to the office for an assessment after completing a home visit. Another center purchased and outfitted an RV that could be driven to participants’ homes and used as a roving assessment room. A number of centers purchased a car for the study to reduce mileage reimbursement costs and wear and tear on staff cars.
Issues of literacy, language, and culture.
Many centers have enrolled participants with limited education and low literacy. The sixth-to eighth-grade reading level that is standard for questionnaires was often too high for center participants. Wording and phrasing had to be simplified for all study documents, including consent forms, and most study instruments, including those designed for self-administration, had to be administered orally.
With very few tools validated on non-English speakers, centers have devoted considerable resources to translating materials. This posed unique challenges. Centers with Spanish-speaking participants have learned that Spanish-language instruments do not necessarily reflect the dialect spoken by participants. Languages like Hmong are largely oral, with a written form having only developed recently. Potentially embarrassing topics that evade translation—for example, specific birth control methods—must sometimes be described graphically. Few neurodevelopmental assessment tools exist in other languages (even in Spanish), and these are often only translated and not validated.
Centers also faced unexpected challenges related to the culture and acculturation of participants. Obstacles encountered by centers included participants not knowing or (with undocumented immigrants) not sharing their exact date of birth, being hesitant to provide biologic samples (because of concerns that those in possession of the sample have the power to hurt them), and reporting pregnancy relatively late in gestation (when the fetus was believed strong enough to withstand evil spirits). Focus groups with community members were instrumental in understanding these types of issues and planning the research accordingly.
Staffing issues.
Many centers have found that building trusting relationships with participants is best accomplished by hiring bilingual, bicultural staff who are from the local community and are assigned to follow particular families ideally from pregnancy through the child assessments. Although this is helpful in building trusting relationships, it can introduce systematic bias. Center studies require staff with a particular gift for engaging children and encouraging optimal performance. In addition, they must have an appropriate level of acculturation, bilingual fluency, education, and computer skills. Often, more in-depth training on data collection techniques is needed than when hiring from within the academic community.
Some center staff have found their work to be emotionally demanding because of the difficult circumstances of participants. In response, one center provided an opportunity for staff to meet with a social worker who specialized in Latino mental health issues. In addition, some staff members have been trained on community resources (e.g., food banks) and, in some cases, initiate contact for participants. To maintain interest in the research, some centers also provided ongoing staff enrichment opportunities, including monthly discussion groups on topics such as child abuse, housing code violations, and child development.
In all centers, the number of staff required to maintain a birth cohort; to conduct weekday, weeknight, and weekend assessments; and to complete quality control tasks was much larger than projected. Staff time and funds were taxed by the need for extensive training and the necessity of sending staff members in pairs to dangerous neighborhoods. Gaps in funding were extremely detrimental to centers in that valuable staff could not be maintained and new staff required time-consuming training.
Retention
Retention of participants has been a critical concern for all centers. In the three centers that have completed follow-up to age 2 years, attrition rates ranged from 15 to 26%. For all centers, participants lost to follow-up differed from continuing participants in some demographic characteristics, such as age, marital status, medical insurance status, and race or ethnicity.
The most common reason for attrition was the inability to locate participants, usually due to disconnected phones and/or frequent or unreported moves (the latter was particularly true for a center that enrolled primarily migrant farmworkers). Lost participants were reinstated in some studies if they returned to the area or resumed contact. Some centers excluded participants who repeatedly missed appointments. Other participants refused to continue because of fatigue, lack of interest, or a partner’s disapproval. In a few cases, attrition was due to infant deaths.
All centers have used incentives to improve retention rates. Incentive amounts per visit have ranged from $20 to $300 (averaging ~$50), with some centers increasing the amount over the course of enrollment. Most centers provide incentives in the form of gift certificates (e.g., to grocery stores) rather than cash so as to minimize security concerns. Several centers offer bonus incentives for certain activities, such as calling study staff when in labor, returning on a separate day to finish an assessment or provide an additional sample, or providing new contact information on moving.
Most centers also provide small gifts such as toys, baby blankets, and tote bags. One center held a raffle at the end of the 24-month visits for participants who remained in the study, and another center intends to have yearly raffles. Although some centers have been successful in soliciting donations of raffle or incentive items, incentives remain a major budget item for all. In addition, some centers have questioned whether certain types of incentives—for example, educational items—could serve as an intervention in families with few such resources.
Overcoming barriers to retention.
To improve retention, the centers used a variety of strategies, including sending letters when phones were disconnected, using mail-forwarding services, sending research staff to the last known address, and using contacts (family and friends) to get updated information on the participants or to pass a message along. Some centers have used Internet-based “reverse look-up” sites to obtain addresses for participants who consistently do not answer phone calls; sending a letter to the address has had some success. Frequent contacts with the participants by phone or mail have also helped to reduce attrition. Most centers contact participants every 3–6 months. These contacts include birthday cards, brief telephone interviews about the child’s health (e.g., respiratory disease or injuries), or simple “check-ins” with the family to remind them of the next phases of the study. Because of a gap in funding, one center had nearly 2 years between contacts. This lapse resulted in considerable attrition and required significant costs and personnel time to locate the families.
In addition to phone calls and mailings, centers used other techniques to maintain communication and retain participants. One center organized a health fair for participants. Another distributed photograph key chains reminding participants to call if they moved and inserted a new photo of the child at each visit to promote its use. Other centers provided magnetized business cards for families’ refrigerators or distributed staff pager and cell phone numbers to encourage communication. One center has permitted participants who have moved from the study area or desire limited participation to complete phone interviews only or allow medical record review. This center has also made weeklong summer trips to complete assessments with participants who have moved to other areas of the state.
Research Infrastructure
Data and Specimen Management Systems
Computerized databases are an essential component of all centers’ participant tracking systems. Center databases contain basic information about participants (e.g., date of birth), information about visit events (e.g., event type and date completed), and detailed information about biologic and environmental samples (e.g., date collected, number and volume of aliquots). Centers use these systems to generate periodic reports (e.g., projected events for the coming month and volume of stored samples) and to check the completeness of final data sets.
Specimen Repository
As previously reviewed, there are many issues to be considered with regard to laboratory specimens (Eskenazi et al. 2003; Holland et al. 2003; Schulte and Perera 1993). All centers collected a variety of biologic samples (Table 5) from participants and/or environmental samples (Table 4) from home environments. Collectively, the centers obtained urine, peripheral blood, cord blood, breast milk, meconium, vernix, saliva, hair, placental tissue, infant formula, indoor and outdoor air, and house dust. The centers have analyzed levels of numerous compounds in these biologic and environmental samples, such as pesticides, phthalates, mercury, lead, cotinine, polycyclic aromatic hydrocarbons (PAHs), PAH–DNA adducts, allergens, endotoxin, antioxidant micronutrients, cytokines, immunoglobulin E (IgE), cholinesterase, and thyroid hormones. Some centers are also analyzing biomarkers of susceptibility, for example, DNA polymorphisms. An important goal of each center was to maximize future use of stored samples. Most centers banked samples for future analyses, such as blood samples for later derivation of RNA and for genomics assays using high-throughput methods based on polymerase chain reaction, chip, and microarray technologies (Appendix 3 for banked samples and Table 5 for intended analyses).
To assure the quality of the specimens for current and future use, the centers developed protocols for collecting, shipping, processing, and banking samples. Pilot studies were conducted to determine the collection and storage conditions necessary for stability of certain compounds and their range of levels in the cohort. Study protocols included written instructions and standard operating procedures, methods for documentation of procedures using chain-of-custody forms and discrepancy reports, and databases to track the location and flow of samples. Protocols were developed for quality assurance and control procedures, for separating specimens into several aliquots to eliminate the need for repeated thawing and freezing, and for avoiding potential contamination of the specimen. As part of their quality control protocols, most centers included field blanks, spikes, and duplicates in their analytical batch of samples. Most centers created bar-coded labels for specimens. Labels included the participant’s unique, coded identifier, the sample type, and the aliquot. In some cases, pilot studies were conducted to determine whether labels would withstand shipping and laboratory conditions over time. All these protocols aimed to maximize the potential for future use of sometimes low-volume samples (e.g., child blood samples).
Examples of problems in sample collection.
Blood collection from children is a challenge. Most centers collected research blood samples at the same time as clinical samples. This helped to avoid participant concerns about taking blood from children and pregnant women, especially in certain cultural groups. Researchers found it helpful to consult with community physicians to determine the amount of blood collection that is both clinically and culturally acceptable to the target population.
Centers found that collecting breast milk samples soon after delivery, although most convenient for the research team, was challenging for mothers. For most, the milk supply had not yet fully developed, and some new mothers (particularly primiparas) found it difficult to provide samples with a breast pump. In addition, some mothers feared that milk was being taken away from the baby. Later collection of breast milk avoided some of these problems, but timing problems arose for other sample types as well. When sample collections could not occur during a scheduled visit, centers scheduled extra visits or made alternate plans. For example, when children could not provide a urine sample, one center gave parents the supplies and instructions to collect the sample at home and arranged to pick it up the next day.
Studies conducted in rural areas faced additional barriers to successful collection and processing of samples. Centers with rural study sites encountered limited laboratory facilities that were not adequately equipped to process samples (e.g., to separate whole blood into blood products). For these centers, it was necessary to transport samples over long distances, increasing costs. In locations where necessary goods and services (e.g., dry ice or courier services) were in short supply, it was also difficult to ensure the prompt stabilization of samples. Finally, some rural areas lacked skilled pediatric phlebotomists.
Ethical Issues
The Children’s Centers have found themselves operating in a time of increasing ethical complexity. Particularly since the implementation of HIPAA, it has become more time-consuming to obtain participants’ informed consent. Concerns about potential lawsuits have increased and been exacerbated by the Grimes vs. Kennedy Krieger case (Mastroianni and Kahn 2002). Finally, centers struggle with conflicting ethical issues, such as deciding when the health and safety of a child takes precedence over a promise of confidentiality.
Consent and assent.
Longitudinal studies demand lengthy and complex consent forms. Ensuring that participants are well informed has been challenging for the centers and has required the allocation of adequate time to inform participants about the study and to review the consent form. For centers using medical records, the completion of HIPAA subject authorization forms adds time to the consent process. Centers’ consent forms differ in level of complexity and in time needed to complete them.
Centers have found it important to inculcate in staff an understanding that consent is an ongoing process. Instead of training staff to simply procure participant signatures, centers have trained staff to solicit and answer participants’ questions so that they can make informed decisions.
All centers recognized the importance of writing consent forms at a reading level understandable to all. Some centers wrote consent forms at an eighth-grade level, whereas others felt that even sixth-grade level was too high to assure comprehension. In addition to providing consent forms in multiple languages, some centers read consents aloud to participants to ensure that everyone, including participants who are embarrassed to admit their low literacy level, fully understood the information. Some centers solicited feedback from community partners, community board members, and community-based staff (in addition to the IRB) to help ensure that appropriate language was used. The centers’ experiences suggest that the language and style of a consent form in one community may not be appropriate in others.
Some studies used additional measures to enhance understanding of consent forms. Several centers used timetables and schedules to communicate study procedures or provided lists outlining the important items on the consent. One developed a short checklist to verify that participants understood the key aspects of the study. Two centers divided consent between two documents, one covering enrollment through delivery, and one covering the period after birth. This decreased the amount of complex information that participants had to digest at each visit, and gave participants an opportunity to re-evaluate their participation at a midway point. However, some participants expressed frustration with the continuing requests, indicating they would prefer full disclosure of the protocol up front.
Centers gave careful thought to who must consent to participate at each stage of the research. In all cases, pregnant women or mothers were asked to consent to their own participation and that of her child. However, once children reached a certain age (generally 5–9 years), child assent was usually also required by the IRB, posing new challenges for the centers. Centers needed to clarify for themselves and for their staff the difference between encouraging a child to try a new task and coercing him or her to do so. Some centers also needed to consider consent procedures in cases when the mother no longer had custody of the child (either officially or unofficially). Finally, centers that conducted home visits considered whether it was adequate for the mother to consent to a visit in a home shared with other families. In some such cases, centers skipped home visits to these participants or limited the visit to the portion of the home in which the participating mother and child lived.
Banked samples and informed consent.
Many centers have banked samples for future studies. This process requires special consideration, in that participants must be informed about and consent to future uses of these samples. Several centers’ consent forms allowed participants the option of either not having samples banked or not allowing future analysis of samples for unrelated studies. At least one center has needed IRB reapproval for each new analysis of banked samples. The center at the University of Washington has participated in a consortium formed by the Centers for Disease Control and Prevention (CDC) to develop informed consent approaches for integrating genetic variation studies into population-based research (Beskow et al. 2001); the group developed an informed consent template (CDC 2001).
Confidentiality and consideration of children at risk.
All centers strove to protect the identity and personal information of all participants. Some centers found it challenging, however, to maintain confidentiality in small or close-knit communities, especially when the research staff was hired from the local community. Most centers instructed staff on when to remove themselves from assessments with participants they know personally and on how to interact with participants when they meet in other settings.
Centers were also vigilant to ensure confidentiality within computerized databases. Centers modeled their own data management systems around guidance provided by their IRBs. All computerized files were password protected with knowledge of passwords restricted to a small number of staff, and the number of computer or paper files containing both the participant study number and identifying information (e.g., name) was limited. In complex studies with multiple contacts, centers found it necessary to work with both the IRB and the research staff to identify the types of linked information necessary for day-to-day operations and to provide that information with the least possible risk to participants. Centers kept files linking study numbers with participant name separate from event and sample tracking databases, linking them briefly only when necessary (e.g., to generate mailing lists of participants missing a particular event).
Certificates of confidentiality, which protect identifiable research information from forced disclosure, including in the case of legal action, are an important component in protecting participant confidentiality. However, many centers faced or anticipated facing circumstances in which they would need to break the promise of confidentiality without participant consent, for example, in cases where child abuse, severe depression, drug use or traffic in the home, and other potentially dangerous conditions were observed. Some centers have elected not to report housing code violations that do not pose an immediate threat to the child, because there is no clear legal mandate or options for the families, and because participants may fear eviction or reprisals from landlords. Centers developed clear protocols that included provisions for referral or reporting, including lists for center staff of what constituted an immediate housing threat. Staff were trained on human subject’s protection requirements and child abuse and neglect reporting issues, including mandatory or discretionary reporting protocols. Because some variation exists in state laws regarding mandated reporting of child abuse, the centers’ experiences suggest that special care should be taken in planning a nationwide study to ensure that protocols are in compliance with both the specific state laws and IRB requirements. Disclosure of such requirements (e.g., the need to report child abuse) was incorporated into consent forms, despite concern that it would repel some participants.
Centers have also developed protocols on intervening in cases of clear developmental delays or undiagnosed physical health problems. Most protocols include timely screening of developmental assessments and questionnaires to ensure prompt referral or treatment. Another aspect of these protocols is the centers’ designated cutoff score for developmental tests (e.g., > 2–3 standard deviations below the mean), children scoring below which are referred with parental permission for further evaluation or intervention. To ensure adequate follow-up of identified children, centers found it useful to identify local resources beforehand; the number of such resources, of course, varied by community. Centers were also required to report some exposure measures, such as lead results, to public health authorities when they exceeded certain action levels.
Communication
Communicating study results is a key step in any research project. In addition to publishing results in scientific journals, centers sought to share findings with participants and community members. Centers elicited the guidance of community collaborators to decide when and how to disseminate results, including how to craft messages that would be clearly understood by and of interest to the community. In some cases, communities expected interventions and actions that were outside the scope of the research; to prevent false expectations, the centers agree it is important to communicate the purposes and limitations of the research beforehand.
Timing of results communication.
Particularly in longitudinal studies with distant visit points that employ new or experimental laboratory methods, there is often a long delay between when data collection begins and when results are communicated to participants and communities. To ensure themselves adequate time to analyze and interpret results without causing undue frustration in participants, most centers found it beneficial to disclose all anticipated delays during the informed consent process.
Many centers have made it a priority to disclose findings to participants and/or community advisory boards before their publication in journal or newspaper articles. This disclosure has been an important step in building trust between researchers, participants, and communities. Community members resent hearing findings for the first time from the media.
Communication tools.
Dissemination strategies developed in collaboration with community advisory boards have included newsletters, fact sheets, pamphlets, press releases in local papers, pay-stub inserts, radio programs (particularly useful in rural areas), town hall meetings, and Internet sites. One center has a monthly radio program in which they report study progress and present a health and safety message. Investigators at all centers share their findings with other scientists and the public via presentations at national conferences, publications in peer-reviewed scientific journals, lectures at colleges and universities, and presentations at community meetings. Ideally, centers would also like to communicate results to children. The center at the University of Washington, based on results from their study of pesticide exposure in children, has created coloring books and curricula to educate preschool and school-age children on how to prevent such exposure.
Specialized tools are often needed for studies that target low-literacy or non-English-speaking communities. Many centers publish information in more than one language, and some centers have attempted to develop pictorial rather than verbal messages.
Group-versus individual-level results.
Perhaps the biggest communication issue that the centers have faced has been whether to provide individual-level results, particularly for measures of exposure or internal dose. The argument in favor of providing such results is that participants have the right to know; the counterargument is that participants may be unnecessarily alarmed by results with no interpretable meaning. Generally, results with a clear clinical implication (e.g., blood lead levels) have been reported to participants, whereas results without clear clinical implications (e.g., urinary pesticide metabolite levels) have not been returned. One center, however, on the basis of community advisory board input, has decided to offer participants the option of requesting their individual pesticide levels. That center is currently in the process of developing materials to provide these results and will work closely with community health care providers when clinical questions arise.
Regardless of whether group or individual-level results are returned, the centers agree it is important to provide to participants a context for these results. Providing a comparison, either to other study participants or nationwide data, has been particularly helpful. In communicating results, centers aim to clearly describe their implications for health and well-being; when these implications are not known (as in the case of pesticides), centers state this honestly (Faustman et al. 2000).
Conclusion: Lessons Learned
In anticipation of the National Children’s Study, lessons can be learned from the birth cohort studies being conducted by the National Institute of Environmental Health Sciences/U.S. Environmental Protection Agency Centers for Children’s Environmental Health and Disease Prevention Research. The National Children’s Study shares a mission with these centers—to understand the environmental causes of developmental diseases—and thus can benefit from the lessons learned during their implementation. Collectively, the centers offer the following advice gleaned from 7 years of research:
Building community infrastructure and trust is essential, especially in populations that are difficult to reach. Formative research, including focus groups with community members, is useful in understanding cultural barriers.
Research goals must be prioritized. Participant fatigue limits the length of the research protocol. Hence, the research needs of a large team of investigators need to be negotiated and prioritized. These research priorities may vary by geographic location.
A research study should be self-sufficient and rely minimally on clinical staff. Tagging a research protocol onto a clinical visit or hospitalization is usually not feasible. Medical care facilities, in particular those that treat low-income populations, are already overburdened.
Research protocols must be flexible and allow for variations in levels of care and medical protocols in hospital and clinics. Usually, only large urban areas have academic tertiary care facilities. If multiple facilities in both rural and urban areas are involved in the research, protocols must be flexible.
Long-term, continuous employment of high-quality and flexible research staff is imperative for the success of the study. Retaining high-quality staff over the course of the study is the key to project success. All centers greatly underestimated the staffing needed to enroll, evaluate, and maintain the birth cohort.
It is essential to find space that is without distraction and convenient to the families. This is difficult to find in medical care settings or in the home. Providing participant transportation was costly, but it was essential for keeping centers’ families enrolled.
Research protocols should be piloted and documented, and quality control protocols should be developed and enforced. Although the development of quality control protocols is standard for laboratory research, frequent checks and close oversight are also necessary for neurodevelopmental assessment and interviewing. All centers felt that the time and resources necessary for this effort were underestimated.
It may be necessary to allow for variations in the type of information collected and the methods of collection to allow for differences in literacy, language, and culture across disparate populations participating in a national study.
Biologic and environmental specimens should be carefully collected, processed, and banked in multiple aliquots. Specimen collection may need to vary by site to accommodate cultural concerns and logistical differences. There should be foresight in the funding of a biorepository for future generations of researchers.
Longitudinal cohort studies must closely track participants. Tracking families may be difficult, especially with mobile populations. Efforts should be made to maintain frequent contact with participants and to collect the contact information of people who will know their whereabouts. Again, the quality of the staff and their relationship to the participants is essential to retaining the cohort.
It is necessary to allow adequate time to obtain informed consent. Obtaining informed consent for low-literacy and immigrant populations may require additional steps. Consents should be written to allow materials and specimens to be used for future studies.
The complex ethical issues involved in conducting a birth cohort study, especially in low-income populations, must be carefully considered. When institutional IRBs differ, deference should be given to an IRB familiar with the culture of the population. Given the changing research climate, observational studies without a component of prevention or intervention may be perceived as unethical in the future.
Study results must be communicated to participants and lay and scientific communities in a timely and sensitive manner. A communication plan needs to be developed with community partners. The cost of regular communication with the community must be factored into the research plan.
If multiple centers are involved in the research, it is essential that there is close and regular communication among them aimed at problem solving and assuring similar methodologies. Resources should allow for frequent and ongoing communication.
Funding for a longitudinal birth cohort study must be adequate for the start-up period and continuous, without gaps, through the long term. Opportunities to evaluate specific developmental milestones may be lost when there are funding lapses. Long breaks between participant contacts can greatly increase attrition rates of valuable populations. The necessary start-up time, including time for formative research, was greatly underestimated by all centers.
On average, the centers have allocated at least $500,000 per year to their birth cohort studies. Given that the average sample size for these birth cohorts is 500, this would translate into a cost of about $500 million for the first 5 years of study of the 100,000-person birth cohort proposed for the National Children’s Study. This does not include additional costs such as for coordination among centers and long-term storage of specimens.
Despite numerous logistical challenges in collecting longitudinal birth cohort data, the Centers for Children’s Environmental Health and Disease Prevention Research have been successful in enrolling and maintaining a variety of populations, including from minority and low-income communities. Although the challenges of longitudinal data collection may be greatest in communities with the poorest and most marginalized populations, we maintain that it is crucial to include these diverse populations from both rural and urban environments to understand the health of those children at greatest risk for environmental hazards (Metzger et al. 1995; Pirkle et al. 1996; Sarpong et al. 1996; Whyatt et al. 2002). We hope that the lessons learned from the Centers for Children’s Environmental Health and Disease Prevention Research can help to inform the National Children’s Study.
This article is part of the mini-monograph “Lessons Learned from the National Institute of Environmental Health Sciences/U.S. Environmental Protection Agency Centers for Children’s Environmental Health and Disease Prevention Research for the National Children’s Study.”
We acknowledge the help of S. Miles and K. Kogut in preparing tables and editing the manuscript. We gratefully acknowledge all the dedicated staff, students, and community partners at each of our centers. Most of all, we thank the participants of our studies, without whom these centers would not be possible.
This research was supported by grants for the Centers for Children’s Environmental Health and Disease Prevention Research from the National Institute for Environmental Health Sciences (grants ES011263, ES011261, ES009600, ES009601, ES009584, ES009605) and the U.S. Environmental Protection Agency (grants R829390, R829389, R827027, R826886, R827039, R826709). We received additional funding for this article from the National Children’s Study.
Table 1 Study overview.
Component University of Illinois University of California, Berkeley Columbia University Mount Sinai Medical Center Cincinnati Children’s Hospital
Study design Prepregnancy/pregnancy cohort Pregnancy/birth cohort Pregnancy/birth cohort Pregnancy/birth cohort Pregnancy/birth cohort
No. of subjects enrolled 164 couplesa (goal = 400) 601 women 556 womena (goal = 730) 479 women 300 womena (goal = 400)
Characterization of the population Hmong and Laotian couples of childbearing age Low-income Mexican and Mexican-American women in agricultural community Dominican and African-American women Public and private prenatal patients Public and private prenatal patients
Location Green Bay and Appleton, Wisconsin Salinas Valley, California New York City (Harlem, Washington Heights, South Bronx) New York City (East Harlem) Cincinnati, Ohio
Major exposures PCBs, methylmercury Pesticides, allergens, metals PM, DEP, PAH, ETS (cotinine), pesticides, allergens, metals Pesticides, PCBs, metals Metals, PCBs, pesticides, tobacco smoke (cotinine), alcohol
Major outcomes Growth, hearing, neurodevelopment/behavior Growth, neurodevelopment/behavior, biomarkers, asthma/respiratory disease Growth, neurodevelopment/behavior, biomarkers, asthma/respiratory disease Growth, neurodevelopment/behavior, biomarkers Growth, neurodevelopment/behavior, hearing, asthma/respiratory disease
Abbreviations: DEP, diesel exhaust particulates; ETS, environmental tobacco smoke; PAH, polycyclic aromatic hydrocarbon; PCBs, polychlorinated biphenyls; PM, particulate matter.
a Still enrolling subjects.
Table 2 Eligibility criteria.a
Criterion University of Illinois University of California, Berkeley Columbia University Mount Sinai Medical Center Cincinnati Children’s Hospital
Age (years) Women 18–44, Men ≥ 18 ≥ 18 18–35 ≥ 16 ≥ 18
Language/ethnicity Hmong, Lao English, Spanish African-American, Dominican English, Spanish English
Gestational age (weeks) < 20 ≤ 26 13–19
Exclusions Will exclude infants after delivery if no prenatal samples None Diabetes
Hypertension
HIV
Smoker
Illicit drug use in last year
First prenatal visit > 20 weeks gestation
Infants after delivery if no prenatal interview, personal monitoring data, or delivery blood
Lived in study area for < 1 year before pregnancy Multiparas
Multiple gestations
Alcohol use
Illicit drug use
Infants after delivery if < 32 weeks gestation, < 1,500 g, birth defect, or no prenatal specimens Diabetes
Seizure disorder
HIV
Schizophrenia
Bipolar disorder
Thyroid disease
Living in mobile home
Home built before 1979
Prenatal care at participating clinic No Yes Yes Yes Yes
Planned delivery at participating hospital No Yes Yes Yes Yes
No plans to move < 1 year after delivery
Other Residents of Green Bay/Appleton MediCal eligible
Who determined eligibility? Research staff Clinic staff Research staff Research staff Research staff
a Blank fields indicate no relevant eligibility criteria for that center.
Table 3 Contact points and types.
Contact University of Illinois University of California, Berkeley Columbia University Mount Sinai Medical Center Cincinnati Children’s Hospital
Questionnaire Enrollment, every 2 months before pregnancy, monthly during pregnancy, delivery, child age 6, 9, 12 months Pregnancy (enrollment mean = 13 weeks), 3rd trimester, delivery (mother and father), child age 6, 12, 24, 42, 60,a 84a months Pregnancy (3rd trimester) child age 6, 12, 24, 36, 60, 72,a 84a months Pregnancy (3rd trimester), child age 12, 24, 48,a 72,a 84a months Pregnancy (enrollment mean = 20 weeks), child age 4 weeks, 12, 24 months; injury questionnaires every 3 months; sleep questionnaire every 6 months
Home walk-through Pregnancy, 6, 12, 24, 42, 60a months Pregnancy, 12, 36, 60 months Pregnancy, 12, 24 months
Neurodevelopment assessment Birth, 6, 9,12 months Birth, 6, 12, 24, 42, 60a months 6, 12, 24, 36, 60, 84a months Birth 12, 24, 48,a 72,a 84a months Birth, 4 weeks, 12, 24 months
School evaluation 84a months 96a months
Growth assessment Birth, 6, 9,12 months Birth, 6, 12, 24, 42, 60a months Birth, 6, 12, 24, 36, 60 months Birth, 12, 24, 48,a 72,a 84a months Birth, 4 weeks, 12, 24 months
Respiratory assessment 6, 12, 24, 42, 60,a 84a months (by questionnaire); spirometry at 60 months Every 3 months from birth to 24 months; every 6 months from 24 to 60 and to 84a months (by questionnaire); spirometry at 60 months
Incentives $20–35 in gift certificates per prepregnancy and pregnancy visits; amounts for child visits are to be determined; T-shirts, water bottles, fish measuring tapes, back-to-school packets; delivery gift baskets with baby T-shirts, socks, bottles; fishing supply raffles $20–60 in gift certificates per visit; car seat or stroller at delivery; hats, T-shirts, tote bags, toys; raffle after 24 and 60a months $50–300 in cash per visit, educational toys $50 in cash per visit, toys $25–100 in gift certificates per visit; tote bags, T-shirts, baby blankets, age-appropriate books
a Planned for second 5-year funding cycle.
Appendix 1 Questionnaire information collected.
Factors University of Illinois University of California, Berkeley Columbia University Mount Sinai Medical Center Cincinnati Children’s Hospital
Demographic information
Demographics X X X X X
Occupation X X X X X
Household income X X X X
Health and development
Reproductive history X X X X X
Medical history X X X X X
Medication use X X X X X
Child sleep X X
Breast-feeding/child diet X X X X X
Developmental milestones X X X
Respiratory symptoms, illness X X X X
Exposure assessment
Housing characteristics X X X X X
Pesticide exposure X X X X X
Allergen exposure X X X
House cleaning habits X X
Injury hazards X
Home remedies X X X X
ETS X X X X X
Household members X X X X X
Household pets X X X X
Fish consumption X X X X
Social factors
Social support X X X
Maternal depression X X X
Psychological distress X
Parenting stress X X
Marital conflict X
Life events X
Quality of life X
Neighborhood quality X X
Sense of control X
Neighborhood cohesion X
Family resources X
Material hardship X X X
Acculturation X X
Immigration history X X X
Childcare X X X X X
ETS, environmental tobacco smoke.
Appendix 2 Home visit information collected.
Characteristic University of California, Berkeley Columbia University Cincinnati Children’s Hospital
Age of housing X
GPS coordinates X X X
Type and condition of flooring X X
Cockroaches X X X
Rodents X X X
Mold/mildew X X X
Wall moisture X X X
Water damage X X X
Peeling paint X X X
Pets X X X
Proximity to agricultural fields X
Proximity to major streets X X
Pesticide use X X X
Pesticide storage X
Gas stove/gas heater X X X
Cleanliness X X
Safety of home environment X X
Lead hazards X
Table 4 Environmental samples.
Sample University of California, Berkeley Columbia University Cincinnati Children’s Hospital
House dust Pregnancy, 6, 12 months Pregnancy, 12, 36, 60 months Pregnancy, 12, 24 months
Lead X
Pesticides X X
Fungal spores/pollen X X
Allergens/endotoxin X X X
Vehicle dust 6 months (subset)
Pesticides X
Burkard air sampling, house Pregnancy, 6, 12 months
Fungal spores/pollen X
Burkard air sampling, area Ongoing 12 months
Water Pregnancy, 12, 24 months
Soil Pregnancy, 12, 24 months
Personal air sampling Pregnancy
Infant formula 1 month
PCBs X
Lead X
Pesticides X
Phthalates X
PCBs, polychlorinated biphenyls.
Table 5 Biologic samples—maternal, paternal, and child (and attendant analyses).
Sample University of Illinois University of California, Berkeley Columbia University Mount Sinai Medical Center Cincinnati Children’s Hospital
Maternal blood Enrollment, 1st, 2nd trimester, delivery, 6 weeks postpartum (pesticides, lead, other metal, PCBs, thyroid hormone) 2nd, 3rd trimester, delivery (pesticides, lead, PCBs, IgE, cholinesterase, genetic polymorphism, thyroid hormone) 1 day postpartum (pesticides, lead, mercury, tobacco, PCBs, IgE, DNA adducts, genetic polymorphism, micronutrients) 3rd trimester (pesticides, lead, PCBs, cholinesterase, genetic polymorphisms) 16, 26 weeks gestation, delivery (pesticides, lead, mercury, tobacco, genetic polymorphism)
Maternal urine Monthly during menstrual cycle (subset) (phthalates, hCG) Enrollment, 2nd, 3rd trimester, delivery 6 months postpartum (subset) (pesticides) 3rd trimester, every 2 weeks < 34 weeks gestation, delivery (subset) (pesticides) 3rd trimester (pesticides) 16, 26 weeks gestation, delivery (pesticides, phthalates)
Placental tissue Delivery
Breast milk Delivery, 6 months postpartum (pesticides) 1 month postpartum (PCBs, lead, pesticides, tobacco, phthalates)
Maternal saliva 16 weeks postpartum (pesticides)
Maternal hair Enrollment, 2nd trimester, 4 weeks postpartum (tobacco)
Paternal blood Enrollment (PCBs)
Paternal urine Delivery (pesticides)
Cord blood Delivery (pesticides, mercury, PCBs, chromosome damage) Delivery (pesticides, lead, IgE, cholinesterase, genetic polymorphisms) Delivery (pesticides, lead, mercury, tobacco, PCBs, IgE, DNA adducts, genetic polymorphisms, micronutrients) Delivery (lead, cholinesterase, genetic polymorphisms) Delivery (pesticides, lead, mercury, iron, tobacco, PCBs, genetic polymorphisms)
Child blood Planned (lead) 12, 24, 60a months (lead, IgE, cholinesterase, genetic polymorphisms, cytokines) 24, 36, 60 months (pesticides, tobacco, IgE, cytokines) 12, 24 months (pesticides, lead, mercury, iron, tobacco, PCBs, genetic polymorphisms)
Child urine 6, 12, 24, 42, 60a months (pesticides) 36, 60 months (pesticides) 12, 24 months (pesticides) 12, 24 months (pesticides, phthalates)
Child meconium Birth (pesticides) Birth (pesticides, lead, mercury, tobacco, PCBs, alcohol)
Child vernix Birth (pesticides, tobacco)
Child saliva 42, 60a months (pesticides)
Child hair 12, 24 months (tobacco)
Abbreviations: hCG, human chorionic gonadotropin; PCBs, polychlorinated biphenyls.
a Currently under way as part of second 5-year funding cycle.
Appendix 3 Banked blood samples.
University of California, Berkeley
Columbia University
Mount Sinai Medical Center
Cincinnati Children’s Hospital
Sample type C M C M C M C M
Whole blood X X X X X X X X
Serum X X X X X X
Clot X X
Plasma X X X X X X
Buffy coat X X X X X
Red blood cells X X X X X
Lymphocytes (cryopreserved) X X X
Blood smears X X
Cholinesterase (stabilized) X X
DNA X X X X X X X X
Other specimens X X X X X X
Abbreviations: C, child/cord blood; M, mother.
==== Refs
References
Beskow LM Burke W Merz JF Barr PA Terry S Penchaszadeh VB 2001 Informed consent for population-based research involving genetics JAMA 286 18 2315 2321 11710898
Caldwell BM Bradley RR 1984. Home Observation for the Measured Environment. Little Rock, AR:University of Arkansas at Little Rock.
CDC 2001. Informed Consent Template for Population-Based Research Involving Genetics. Atlanta, GA:Centers for Disease Control and Prevention. Available: http://www.cdc.gov/genomics/info/reports/policy/consent.htm [accessed 11 March 2005].
Dietrich KN Eskenazi B Schantz S Yolton K Rauh VA Johnson CB 2005 Principles and practices of neurodevelopmental assessment in children: lessons learned from the Centers for Children’s Environmental Health and Disease Prevention Research Environ Health Perspect 113 1437 1446 16203260
Eggleston PA Diette G Lipsett M Lewis T Tager I McConnell R 2005 Lessons learned for the study of childhood asthma from the Centers for Children’s Environmental Health and Disease Prevention Research Environ Health Perspect 113 1430 1436 16203259
Eskenazi B Bradman A Gladstone EA Jaramillo S Birch K Holland NT 2003 CHAMACOS, a longitudinal birth cohort study: lessons from the fields J Child Health 1 1 3 27
Faustman EM Silbernagel SM Fenske RA Burbacher TM Ponce RA 2000 Mechanisms underlying children’s susceptibility to environmental toxicants Environ Health Perspect 108 suppl 1 13 21 10698720
Gilliland F Avol E Kinney P Jerrett M Dvonch T Lurmann F 2005 Air pollution exposure assessment for epidemiologic studies of pregnant women and children: lessons learned from the Centers for Children’s Environmental Health and Disease Prevention Research Environ Health Perspect 113 1447 1454 16203261
HIPAA 1996. Health Insurance Portability and Accountability Act of 1996. Public Law 104–191.
Holland NT Smith MT Eskenazi B Bastaki M 2003 Biological sample collection and processing for molecular epidemiological studies Mutat Res 543 3 217 234 12787814
Israel BA Parker EA Rowe Z Salvatore A Minkler M López J 2005 Community-based participatory research: lessons learned from the Centers for Children’s Environmental Health and Disease Prevention Research Environ Health Perspect 113 1463 1471 16203263
Jacobs DE Clickner RP Zhou JY Viet SM Marker DA Rogers JW 2002 The prevalence of lead-based paint hazards in U.S. housing Environ Health Perspect 110 A599 A606 12361941
Kimmel CA Collman GW Fields N Eskenazi B 2005 Lessons learned for the National Children’s Study from the National Institute of Environmental Health Sciences/U.S. Environmental Protection Agency Centers for Children’s Environmental Health and Disease Prevention Research Environ Health Perspect 113 1414 1418 16203257
Lester BM Tronick E 2001. Behavioral assessment scales: the NICU Network Neurobehavioral Scale, the Neonatal Behavioral Assessment Scale, and the assessment of the preterm infant’s behavior. In: Biobehavioral Assessment of the Infant (Singer LT, Zeskind PS, eds). New York:Guilford Press, 363–380.
Mastroianni AC Kahn JP 2002 Risk and responsibility: ethics, Grimes v. Kennedy Krieger , and public health research involving children Am J Public Health 92 7 1073 1076 12084682
Metzger R Delgado JL Herrell R 1995 Environmental health and Hispanic children Environ Health Perspect 103 suppl 6 25 32 8549482
National Children’s Study 2005. Homepage. Available: http://www.nationalchildrensstudy.gov/ [accessed 8 August 2005).
Niswander KR Gordon M 1972. The women and their pregnancies; the Collaborative Perinatal Study of the National Institute of Neurological Diseases and Stroke. Washington, DC:National Institute of Health.
Pirkle JL Flegal KM Bernert JT Brody DJ Etzel RA Maurer KR 1996 Exposure of the US population to environmental tobacco smoke: the Third National Health and Nutrition Examination Survey, 1988 to 1991 JAMA 275 16 1233 1240 8601954
Sarpong SB Hamilton RG Eggleston PA Adkinson NF Jr 1996 Socioeconomic status and race as risk factors for cockroach allergen exposure and sensitization in children with asthma J Allergy Clin Immunol 97 6 1393 1401 8648037
Schulte PA Perera FP eds. 1993. Molecular Epidemiology: Principles and Practices. San Diego:Academic Press.
van den Berg BJ Christianson RE Oechsli FW 1988 The California Child Health and Development Studies of the School of Public Health, University of California at Berkeley Paediatr Perinat Epidemiol 2 3 265 282 3070486
Whyatt RM Camann DE Kinney PL Reyes A Ramirez J Dietrich J 2002 Residential pesticide use during pregnancy among a cohort of urban minority women Environ Health Perspect 110 507 514 12003754
Yiin LM Rhoads GG Lioy PJ 2000 Seasonal influences on childhood lead exposure Environ Health Perspect 108 177 182 10656860
|
16203258
|
PMC1281291
|
CC0
|
2021-01-04 23:41:28
|
no
|
Environ Health Perspect. 2005 Oct 24; 113(10):1419-1429
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.7670
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences 10.1289/ehp.7671ehp0113-00143016203259ResearchMini-MonographLessons Learned for the Study of Childhood Asthma from the Centers for Children’s Environmental Health and Disease Prevention Research Eggleston Peyton A. 1Diette Greg 1Lipsett Michael 2Lewis Toby 3Tager Ira 4McConnell Rob 5Chrischilles Elizabeth 6Lanphear Bruce 7Miller Rachel 8Krishnan Jerry 11 Johns Hopkins University School of Medicine, Baltimore, Maryland, USA2 University of California at San Francisco School of Medicine, San Francisco, California, USA3 University of Michigan School of Medicine, Ann Arbor, Michigan, USA4 University of California at Berkeley School of Public Health, Berkeley, California, USA5 University of Southern California School of Medicine, Los Angeles, California, USA6 University of Iowa College of Public Health, Ames, Iowa, USA7 University of Cincinnati School of Medicine, Cincinnati, Ohio, USA8 Columbia University College of Physicians and Surgeons, New York, New York, USAAddress correspondence to P.A. Eggleston, Department of Pediatrics, CMSC 1102, Johns Hopkins Hospital, 600 North Wolfe St., Baltimore, MD 21287 USA. Telephone: (410) 955-5883. Fax: (410) 955-0229. E-mail:
[email protected] authors declare they have no competing financial interests.
10 2005 24 6 2005 113 10 1430 1436 12 10 2004 23 6 2005 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright. The National Children’s Study will address, among other illnesses, the environmental causes of both incident asthma and exacerbations of asthma in children. Seven of the Centers for Children’s Environmental Health and Disease Prevention Research (Children’s Centers), funded by the National Institute of Environmental Health Sciences and the U.S. Environmental Protection Agency, conducted studies relating to asthma. The design of these studies was diverse and included cohorts, longitudinal studies of older children, and intervention trials involving asthmatic children. In addition to the general lessons provided regarding the conduct of clinical studies in both urban and rural populations, these studies provide important lessons regarding the successful conduct of community research addressing asthma. They demonstrate that it is necessary and feasible to conduct repeated evaluation of environmental exposures in the home to address environmental exposures relevant to asthma. The time and staff required were usually underestimated by the investigators, but through resourceful efforts, the studies were completed with a remarkably high completion rate. The definition of asthma and assessment of disease severity proved to be complex and required a combination of questionnaires, pulmonary function tests, and biologic samples for markers of immune response and disease activity. The definition of asthma was particularly problematic in younger children, who may exhibit typical asthma symptoms sporadically with respiratory infections without developing chronic asthma. Medications confounded the definition of asthma disease activity, and must be repeatedly and systematically estimated. Despite these many challenges, the Children’s Centers successfully conducted long-term studies of asthma.
asthmachildrenChildren’s Centersenvironmental healthNational Children’s Studypregnancy
==== Body
In this article we outline information from the Centers for Children’s Environmental Health and Disease Prevention Research (Children’s Centers) that have conducted studies related to asthma. We do not include information on sampling strategies because the goal of the monograph is to inform the National Children’s Study. Rather, we focus on major issues related to the identification of asthma, asthma-related symptoms and end points, relevant exposures, biologic markers, and follow-up requirements.
Studies of asthmatic children and their homes were conducted at seven Children’s Centers. The protocols for these studies are summarized in Table 1, and further information is provided in the introduction to this mini-monograph (Kimmel et al. 2005). The birth cohort studies conducted at two Children’s Centers are described more completely by Eskenazi et al. (2005); they are included here because both Children’s Centers address respiratory cross-sectional data and outcomes. In addition, three Children’s Centers conducted cohort studies with older children with established asthma, and the study at Johns Hopkins University (JHU) included a nonasthmatic control group as a comparator in initial observations. Four Children’s Centers conducted intervention trials, with diverse designs ranging from a randomized primary prevention trial in school-age children to formal randomized controlled clinical trials.
Recruitment of Participants
The studies that recruited children with asthma had key similarities (Table 1). Children were recruited whose parents reported a history of doctor-diagnosed asthma. Other children were recruited as controls. There were few exclusion criteria in most studies except for other respiratory diseases. Because all participants in these studies were minors, the investigators had to obtain informed consent from a parent or other guardian. When recruiting children > 7 years of age, assent was obtained from the child in addition to the consent obtained from a parent or guardian. In the two Children’s Centers that conducted birth cohort studies, mothers consented to obtain information regarding their own and their child’s medical and exposure histories (Eskenazi et al. 2005). In certain cases, the study setting included prespecified areas of a city bounded by ZIP codes or prespecified counties. In other cases, the study setting was defined by participants seeking care at a specific clinic or medical center regardless of where they lived.
Although three studies used local schools (University of Iowa, JHU intervention study, University of Michigan), most studies identified the sample from a health care setting such as health plans, emergency departments, and physician offices. In schools, the protocol was approved both by an institutional review board (IRB) and by a school system IRB. Informed consent was mailed to the parents to be signed, and followed by a mailed questionnaire in the case of Iowa and Michigan; in the JHU intervention study, the mailed consent was followed by a visit from a recruiter. For other studies, informing and recruiting potential participants included both passive and active methods. The passive approaches included posting study information publicly and providing information when a potential participant’s parent or guardian called. These were generally less efficient in generating interested families to participate and generally were considered a supplement to active methods. For the studies of children, most active recruitment consisted of generating patient lists from computerized databases of children who met entry criteria. Study information was then mailed directly to participants, and study staff telephoned sometime later to determine the family’s interest, to confirm eligibility, and to initiate the informed consent process.
Barriers to Recruitment
There were several barriers to recruitment of children to the studies. First among these was establishing the essential entry criterion for a diagnosis of asthma. This was a greater issue for those studies recruiting younger children (JHU longitudinal cohort study, Cincinnati) than in those focused on older children [JHU Intervention trial, University of Southern California (USC), Michigan, Iowa]. Older children were generally identified by a diagnosis being made by a physician and logged into the computerized database from which recruitment began. In some cases, children with typical asthma were entered into databases with related diagnoses such as recurrent bronchitis or reactive airway disease, which have slightly different codes in the International Classification of Diseases, 9th Revision [World Health Organization (WHO) 1978]. Study staff confirmed the diagnosis with screening questionnaires defining appropriate symptoms or with lung function tests. Younger children, who may have had asthmatic respiratory symptoms only during respiratory infections, were more difficult to qualify because they had less specific answers to respiratory questionnaires and could not perform lung function tests. In a birth cohort study, early wheeze may be recorded, with persistent wheeze or asthma being defined as an outcome later in life. However, entry criteria for the longitudinal cohort studies that recruited younger children used the same definition used with older children: doctor-diagnosed asthma. In a birth cohort study, cough or wheeze may be recorded as an outcome, with persistent wheeze or definable asthma used as an outcome later in life.
An important barrier was introduced by the need to preserve patient autonomy and privacy. IRBs have long required investigators to accommodate patients listed on health care databases who do not want to be contacted to participate in research studies. This has usually been accomplished by mailing an invitation letter containing a preaddressed and stamped postcard that families can return requesting that they not be called. In general, this has proven to be a minor barrier, excluding < 5% of identified families. The recently issued Health Insurance Portability and Accountability Act (HIPAA) regulations have introduced additional barriers by requiring that families give permission to use their health information even at preliminary stages of research (e.g., investigators having access to the health care database to obtain lists of children with a diagnosis of asthma). For example, HIPAA makes it illegal to create lists of persons with a specific diagnosis to whom to send an introductory letter and makes the approach used by most of these studies illegal. Various solutions have been developed, including having patients sign general waivers of HIPAA privacy rights as they enroll in a health care program or having their health care provider recommend them to the interviewer at the time of a health care encounter. Other approaches have included asking asthmatic patients or their parents who participate in health fairs, surveys, self-management programs, or other patient care activities to indicate in writing their willingness to participate in future research studies and creating a list of those who have so indicated. IRBs may approve a HIPAA waiver, but the investigators must justify why a waiver is necessary—that is, that the data being gathered are not sensitive, are not able to be linked with individual identifiers, or both are noninvasive and cannot be collected in another way. This requirement also involves clinic personnel, who are much less effective as recruiters that are trained, motivated study staff. Recruitment from school directories avoids these problems because health data are not used to identify potential subjects. One study that recruited from schools did so by conducting a mailed asthma screening survey of 10 school districts (87% response rate) to identify children with asthma (Chrischilles et al. 2004). Another reported a 78% response rate from 9,437 mailings (Lewis et al. 2004).
The two birth cohort studies (University of California at Berkeley, Columbia University) recruited pregnant mothers and thus faced the issue of participant identification to a smaller extent. At the time the studies were initiated, the HIPAA regulations had not been activated, so they did not face this issue. Were the studies to be conducted now, recruitment of identified pregnant mothers (with or without asthma) would still have to comply with the HIPAA requirements and, indeed, do so in the continued surveillance of the children.
A final barrier when dealing with families with lower socioeconomic status is obtaining a reliable means to contact persons on a list. Families move frequently (29% moved at least once during a 1-year observation) (Swartz et al. 2004), so the address and telephone numbers in a database may not be current. Additionally, as many as 25% of inner-city families may not have telephones (Wissow et al. 1988), and 52% may change their telephone number at least once during a year of follow-up (Swartz et al. 2004). This not only makes initial contact difficult but also interferes with follow-up.
Retention
As shown in Table 1, retention rates were generally > 80% in those studies that involved follow-up. A number of barriers had to be overcome to achieve these high rates, including frequent telephone number changes, residential address changes, and general reluctance in making follow-up appointments. The usual solutions to the lack of telephones or frequent changes of telephones and addresses are to maintain frequent contacts with participants (generally every 1–2 months), to ask families for secondary (and tertiary) alternative contact information, or to provide incentives to notify the study personnel in the event of a change in address or telephone number (Mitchell et al. 1997). Most study staff members were able to conduct home visits even when telephone contact strategies had proven unsuccessful. However, families who moved without providing new contact information interfered with this strategy as well as with follow-up home inspections that were part of several protocols. For some Children’s Centers, missed appointments were important barriers to retention. Reasons for missed appointments included inclement weather and illness among family members needed to help transport the child, but in many cases no explanation was provided. Some studies categorized repeated “no-shows” as having passively withdrawn from the study. This practical measure was necessary to preserve study resources. Children’s Centers that conducted the research in the homes of subjects (i.e., did not require participants to come to the study center) also faced appointment cancellations.
The Children’s Centers employed multiple strategies to maintain contact with participants and to encourage their continued participation. Commonly used incentives included cash, gift certificates, toys for the children, food stamps, and infant car seats. A wide variety of gifts were provided, including T-shirts, tote bags, hats, key chains, stuffed animals, games, and musical instruments. These incentives were provided either at follow-up visits or on successful completion of the study. Additional incentives included health-related devices such as peak flow meters, spacers for metered dose inhalers, and allergen-proof mattress and pillow covers. One Children’s Center provided up to $200 as reimbursement for electricity costs incurred during the study. The Columbia Children’s Center, faced with participants who moved to a new location (e.g., Florida), paid for air travel to do follow-up assessments, and in one case a research worker has flown to the Dominican Republic to obtain follow-up data. The Berkeley Children’s Center created a movable laboratory in a recreation vehicle and went on the road once a year to other areas in California to include participants who moved. Another common incentive was to provide reimbursement for travel expenses related to study participation.
In some cases, the incentives had no monetary value but focused on information provided to the family. In the JHU cohort study, participants learned about home environmental measurements, including allergens and pollutants; many of these families indicated that this was the principal incentive that attracted them to the study. All investigators felt that the most important factor in participant retention was the Children’s Centers’ staffs. Retention was highest when the staff members were able to create an empathetic bond with the participating families, who felt that the staff would try to help them not only with their child’s illness but also with other difficulties associated with their social and economic circumstances. Families frequently sought help with problems such as transportation, referrals, and problems of daily living and valued the “ear” provided by the staff regardless of whether the staff member could actually help with the issue.
Environmental Data
As shown in Table 2, all the Children’s Centers employed extensive questionnaires, collecting demographic, social, medical, and environmental exposure data similar to those collected in other studies. These data were collected repeatedly during longitudinal follow-up, although generally in abbreviated questionnaires.
Most Children’s Centers conducted ambient air sampling at one or more sites (Table 3). In addition, all Children’s Centers conducted home visits and in many cases conducted visits repeatedly (see Table 4). At these visits, the participants’ homes were formally inspected with checklists and families provided additional information through questionnaires. In most cases, settled dust samples were collected and were assayed for indoor allergens; concentrations of endotoxin and pesticides were also measured. Most Children’s Centers conducted repeated measures of environmental exposures, including inspection, settled dust sampling, and air sampling; completion rates for these evaluations ranged from 73 to 91%. The data were generally used to describe exposure at the times indicated and was compared with asthma morbidity at these times. Because the data were collected as several time points, time series analyses were conducted.
Biologic Samples
The number of studies that collected blood samples from children was remarkable (Table 5). These samples were generally used for radioallergosorbent tests (RAST) for specific immunoglobulin E (IgE) antibody to supplement or replace skin tests. Prick skin tests and RAST have been shown to correlate well for sensitivity to inhalant allergens (Wood et al. 1999) and food allergens (Sampson and Albergo 1984). When blood is saved for RAST, the investigator has the advantage of being able to test for additional sensitivities that were not considered at the time the study was conducted. The specific sensitivities determined in this manner were essential both to determine that a child was atopic and to detect specific IgE to the indoor allergens measured in home visits. Atopy, defined as a genetic predisposition to produce long-lived IgE antibody to environmental allergens in association with a constellation of chronic diseases including food allergy, eczema, allergic rhinitis, and asthma, is the most important risk factor for both incident asthma and asthma severity. The combination of specific IgE to environmental allergens and the presence of high concentrations of these allergens in the child’s home is the strongest known risk factor for asthma severity and morbidity (Rosenstreich et al. 1997).
In addition to blood samples, prick–puncture skin testing was usually done, and samples of meconium, urine, saliva, and hair were collected, usually to detect exposure to environmental tobacco smoke (ETS) and pesticides. What is most remarkable is that biologic samples were collected from children in most Children’s Centers on multiple occasions and that these collections did not have noticeable impact on participant retention rates (Table 1).
Asthma Disease Activity
All of these studies included questions regarding asthma disease activity as shown in Table 6. Assessing disease morbidity in epidemiologic studies of children with asthma involves the creation of a composite of symptoms, health care, and medication use from questionnaires, together with pulmonary function tests in older children (Kattan et al. 1997). Although there is a general consensus about asthma-associated symptoms, there is less agreement with regard to specific questions that provide the most appropriate assessment of these symptoms. Major symptoms include cough, wheeze, chest tightness, and dyspnea, but these are quite variable day to day and depend on exposures to specific stimuli or “triggers” such as respiratory viruses, allergens, or irritants such as ETS and air pollutants. It is now generally accepted that symptoms should be assessed over 2-week intervals and that questions be framed in terms of the number of days during which these symptoms are experienced rather than frequency per day or symptom intensity. The two most widely used questionnaires were developed by the American Thoracic Society (ATS) (Ferris 1978) and the Children’s Health Survey for Asthma (CHSA) (Asmussen et al. 1999).
Several scales have been created to synthesize the symptoms into a description of disease activity. The most commonly accepted comes from the National Institute of Health (NIH)’s National Asthma Education and Prevention Program (NAEPP) (National Heart, Lung, and Blood Institute 1997). This scale, summarized in Table 7, was originally intended to categorize severity in untreated asthmatics in clinical settings but has been used in epidemiologic studies, as well (Diette et al. 2001). More recently, as the use of daily medication for asthma has become more common, the NAEPP scale has been felt to more appropriately describe disease control rather than severity.
Some of the Children’s Centers included questions from the CHSA (Asmussen et al. 1999) or from the pediatric version of the ATS questionnaire. The JHU intervention study used the ATS questions as they had been adapted for the Children’s Asthma Management Program study (Szefler et al. 2000). The difference between these questionnaires is that the CHSA uses a Likert scale to define symptom frequency in the preceding 2 weeks, whereas the other questionnaires ask how many days the child had experienced individual symptoms; the reliability of the CHSA is better documented than is the case with other questionnaires. Because of the diurnal variation in asthma disease activity, symptoms occurring at night and during the day are always considered separately. Symptoms occurring with exercise or on days that the child does not have an upper respiratory tract infection are also recorded separately.
Other aspects of asthma morbidity also recorded in the Children’s Centers’ investigations included interference with a child’s activity (exercise, play, school, sleep), interference with parents’ activities, and acute events. The latter were generally defined as those requiring systemic corticosteroid use, unscheduled physician visits, visits to emergency rooms, or hospitalization. Because these were uncommon events, the recall period was generally longer, that is, 2 months or longer, in almost all Children’s Centers.
In the early 1990s, an international group created five core questions regarding childhood asthma to facilitate an international comparison of varying prevalence rates or severity indices. This scale from the International Study of Asthma and Allergy in Childhood (ISAAC) (Asher et al. 1995) has widespread support among epidemiologists. The ISAAC core asthma questions were commonly included in the asthma-specific questionnaires used in the Children’s Centers’ studies.
Juniper et al. (1996, 2000) developed a series of questions to assess the effect of symptoms on a child’s quality of life. These provide a composite assessment of disease activity and tend to correlate better with daily lung function measures than do symptom questionnaires. In addition, they have the advantage of providing a single summary number that has proven to be sensitive to change across time, an extremely valuable property in longitudinal cohort studies and interventions.
Pulmonary function tests are important measures of disease activity but correlate modestly with reported symptoms or other measures of morbidity. Daily measures usually are limited to peak expiratory flow rate (PEFR), although forced expiratory volume in 1 sec (FEV1) may also be measured by portable spirometers. Children, parents, and staff must be trained in the use of these devices; most preschool children given training can perform accurate PEFR measures. With proper coaching, 83% of children 3–6 years of age can complete technically acceptable and reproducible maneuvers (Eigen et al. 2001); the Children’s Center’s experience with technically acceptable data ranged from 52 to 99%.
Medication use both modifies symptoms and provides an independent measure of disease morbidity. Medications that are taken to reverse symptoms of obstruction [“relievers” is the term used in the NIH consensus guidelines (National Heart, Lung, and Blood Institute 1997)] are recorded as equivalent to symptoms during a day. Medications that are taken daily [“controllers” in the NIH consensus guidelines (National Heart, Lung, and Blood Institute 1997)] may modify symptoms, as may medications taken before exercise or other stimuli to prevent attacks. In this setting, it is appropriate to talk about disease “control” rather than “morbidity.” Validated scales have been published to describe disease control in adults (Nathan et al. 2004).
Diaries can potentially give more accurate records of disease activity, but retrieval and consistency are problematic. Diaries and periodic questionnaires generally correlate well (Gold et al. 1989), and the studies conducted in the Children’s Centers used repeated questionnaires to avoid the logistical problems of diary retrieval and data verification.
Interventions
Five of the Children’s Centers conducted interventions to test the efficacy of environmental control measures on improving indoor environmental exposures and asthma-related health. Strategies employed are summarized in Table 8. Although all Children’s Centers emphasized an environmental education program for families, they varied in the breadth and intensity of other components of the intervention programs. Some Children’s Centers focused on strategies targeting a few key triggers, whereas others chose a more comprehensive approach. In addition, some Children’s Centers chose to supplement home environmental strategies with education for families on asthma management, or education targeting physicians treating the asthmatic children. The relative benefits and challenges of these various strategies remain the subject of intense investigation.
Local Variations
Although childhood asthma is an issue of national and international significance, it is important to remember that it occurs within a local context. Cultural, social, and linguistic factors vary by location, as do systems of health care and community resources. Sources of environmental exposure, housing stock, and population behavior patterns relative to the exposure may vary tremendously. In addition, historical relationships between academia and local communities may range from strained to quite cooperative. Each of these factors influences the way asthma studies and interventions can be practically implemented in any given location. Some of the variation seen in study design between Children’s Centers is a direct result of variation in local priorities, circumstances, and resource constraints. What may work in one setting may not be feasible or relevant in another. Each of the Children’s Centers described here used a community-based participatory research approach (Israel et al. 2005) in which community partners contributed significantly to the research process. The Children’s Centers universally report that the involvement by community members enhanced their ability to accomplish their research goals. Multicenter studies, such as the proposed National Children’s Study, have an additional challenge of balancing the need for protocol uniformity across sites with the very real need to adjust to local contextual issues.
Conclusions: Lessons Learned
The experience gained in the studies conducted by the National Institute of Environmental Health Sciences/U.S. Environmental Protection Agency Children’s Centers for Children’s Environmental Health and Disease Prevention provide important lessons for the National Children’s Study. The lessons that specifically relate to asthma can be summarized as follows:
Asthma identification requires a combination of questionnaire and physiologic measures. Many validated questionnaires are available to record asthma symptoms in children; these differ in the terms used to describe symptoms and in the period of recall. To allow comparison with previously reported data and to allow data regarding children in the United States to be compared with those of children in other countries, it is advisable to include questions from the ISAAC, the ATS, and the CHSA in study questionnaires. In addition to historical information, objective measures such as spirometry, eosinophil counts in peripheral blood or secretions, or measures of specific IgE antibody are usually included in definitions of asthma.
Identification of asthma in preschool-age children is problematic. Many children wheeze or cough with respiratory infections and never wheeze when they are older. These episodes are not considered to be asthmatic, and current methods only modestly predict which infants with wheezing will later develop asthma. For this reason, it is appropriate to classify these episodes as recurrent wheezing illness rather than asthma and to reserve the definition of asthma for older children with more persistent symptoms.
Medication confounds the assessment of asthma symptoms and classification of disease severity. Short-acting β-adrenergic agonists (SABAs) will predictably improve acute asthma symptoms. In questionnaire histories, it is appropriate to equate the use of these medications with episodes of asthma; the fact that the SABAs are often used to prevent symptoms introduces uncertainty into this statement. Daily controller medications are used preventively and are currently used in more severe cases, so the use of these medications usually indicates more severe disease; however, the inconsistency with which these medications are prescribed lends considerable uncertainty to this statement.
Recruitment and data collection in health care settings require dedicated study staff. The Children’s Centers found that health care personnel in clinical settings could not be relied on either to recruit children into the studies or to collect outcome data. This added to the cost of recruiting in clinical settings and, in addition, added to the complexity of collecting health information from patients whose privacy was protected by HIPAA regulations. When accessed in a manner consistent with HIPAA guidelines, medical records can provide useful supplemental information, but in most clinics and hospitals the lack of standardized records makes this information less useful.
There are important longitudinal data to be gained from cohorts of older children. Longitudinal studies provide essential data regarding the sequence of exposure to environmental agents and incidence cases of asthma. Similarly, these studies provide important exposure response data with regard to the sequence of asthma episodes and environmental exposures in symptomatic asthma. Asthma is characteristically variable, so repeated measures are important. The frequency of asthmatic symptom recall and the variability of important environmental stimuli dictate how frequently these data must be recorded. For example, daily symptoms are best explained by environmental measures made during the same days. However, environmental samples have traditionally been collected for several (or many) days to accommodate analytic sensitivity, but paradoxically, it is difficult to attribute symptoms reported on some but not all days during which the measures were made. It also follows that sporadic events, such as hospitalizations, are difficult to associate with measures averaged over several months.
This article is part of the mini-monograph “Lessons Learned from the National Institute of Environmental Health Sciences/U.S. Environmental Protection Agency Centers for Children’s Environmental Health and Disease Prevention Research for the National Children’s Study.”
We acknowledge the support of the National Children’s Study in the development of the articles in this mini-monograph. We also gratefully acknowledge the outstanding, dedicated staff and co-investigators at each of the Children’s Centers, the community members who helped shape the studies, and the many families and children who participated.
This work was supported in part by the National Institute of Environmental Health Sciences (ES009606, ES09589, ES009605, ES009581, ES09607, ES011261, ES009600) and the U.S. Environmental Protection Agency (R826724, R826710, R826709, R826708, R826711, R829389, R827027).
Table 1 Overview of asthma-related studies and intervention.
Berkeley Columbia USC JHU Michigan CCH USC JHU Iowa Michigan
Study design BC BC LC LC LC RCT RCT RCT IT RCT
Sample size (n) 601a 861a 11,841 150 case, 150 control 298b 225 202 100 189 298b
Outcomes Respiratory symptoms, atopy, medications, health care use Respiratory symptoms, PFT, medications, health care use Asthma symptoms Asthma symptoms, medications, PEFR, health care use Asthma symptoms, PEFR Asthma symptoms, medications, health care use, child behavior Asthma symptoms, medications, health care use Asthma symptoms, medications, health care use, FEV1 Asthma symptoms, health care use, management behaviors Asthma symptoms, medications, health care use, PEFR, FEV1
Exposures Pesticides, ETS, dust allergens (home and ambient), dust endotoxin, social stressors PM, DEP, PAH, ETS, pesticides, dust, allergens, social stressors Air pollution, dust allergens, social stressors PM, NO2, O3 (home and ambient), dust allergens, ETS PM (ambient, home, personal), O3, ETS ETS, dust allergens, social stressors Dust allergens, social stressors PM, NO2, O3 (home and ambient), dust allergens, ETS Dust allergens, dust endotoxin, ETS Dust allergens, endotoxin, ETS, social stressors
Retention (%)c 86 90 78 83 67 96 77 93 76 77
Abbreviations: BC, birth cohort; CCH, Cincinnati Children’s Hospital; DEP, diesel exhaust particles; IT, intervention trial; LC, longitudinal cohort; NO2, nitrogen dioxide; O3, ozone; PAH, polycyclic aromatic hydrocarbon; PFT, pulmonary function tests; PM, particulate matter; RCT, randomized controlled clinical trial.
a Pregnant women.
b The same children were studied in the intervention trial and the longitudinal cohort in Michigan. The two studies were conducted simultaneously, with the longitudinal observational study extending follow-up after completion of the intervention.
c Time interval for retention varies across studies.
Table 2 Questionnaire information.
Characteristic Berkeley Columbia USC: LC JHU: LC Michigan: LC CCH USC: IT JHU: IT Iowa Michigan: IT
Collection schedule Pregnancy, birth, 6,12, 24, 42 months of age Pregnancy, every 3 months until age 24 months; every 6 months until 60 months of age Baseline, yearly Baseline, 3, 6 months Baseline, 3, 6, 9, 12, 15, 18, 21, 24, 27, 30 months Baseline, day 2, 3, 6, 9, 12, 12.5 months Baseline, 4, 8, 12, 16, 20 months Baseline, 3, 6, 9, 12 months Baseline, 12 months Baseline, 12 months
Demographics X X X X X X X X X X
Occupation X X X X X X
Housing characteristics X X X X X X X X X X
Pesticide exposure X X X X X
Allergen exposure X X X X X X X X X X
Cleaning habits X X X X X X X
Social support X X X X X X X X X
Maternal depression X X X X X X
Child diet X X X X X
Respiratory symptoms X X X X X X X X X X
Medication use X X X X X X X X X X
Home remedies X X X X
Smoking X X X X X X X
Medical history X X X X X X X X X X
Household income X X X X X X X X X X
Pets X X X X X X X X X X
Child care X X X X X X
Abbreviations: BC, birth cohort; CCH, Cincinnati Children’s Hospital; IT, intervention trial; LC, longitudinal cohort.
Table 3 Ambient air sampling.
Sample Berkeley Columbia USC: LC JHU: LC Michigan: LC CCH JHU: IT
Collection schedule Pregnancy, 6, 12 months and at community site Personal air during pregnancy, ambient and indoor at 12 months Continuous at community site Baseline, 6, 12 months Baseline, 3, 6, 9, 12, 15, 18, 21, 24, 27, 30 months 6, 12 months Baseline, 6, 12 months
PM10 X X X X
PM2.5 X X X X X
NO2 X X X
Ozone X X X X
Mold/pollen X X
Air nicotine X
Abbreviations: BC, birth cohort; CCH, Cincinnati Children’s Hospital; IT, intervention trial; LC, longitudinal cohort; PM, particulate matter; PM2.5, PM < 2.5 μm in diameter; PM10, PM < 10 μm in diameter. Samples were not collected in USC: IT, Iowa, and Michigan: IT.
Table 4 Home evaluation.
Characteristic Berkeley Columbia USC: LC JHU: LC Michigan: LC CCH USC: IT JHU: IT Iowa Michigan: IT
Collection schedule Pregnancy, 6, 12, 24, 42 months Pregnancy, 12, 24 months For subsample, one time Baseline, 3, 6 months Baseline, 12, 24 months Baseline, 6, 12 months Baseline, 4, 8, 12, 16, 20 months Baseline 6, 12 months Baseline, 12 months Baseline, 12 months
Age of house X X X X X X X
Proximity traffic X X X X X X X
Proximity fields X X
Flooring type, condition X X X X X X X
Cockroaches X X X X X X X X X
Rodents X X X X X X X X X
Mold X X X X X X X X X
Wall moisture X X X X X X X X X
Peeling paint X X X X X X
Water damage X X X X X X X X X
Pets X X X X X X X X X
Pesticide use X X X X X X
Gas stove/heater X X X X X X X X X
Cleanliness X X X X X X X
House dust X X X X X X X X X X
Allergens X X X X X X X X X X
Pesticide X X X X
Endotoxin X X X X X X
Home air Xa X Xb X X X
PM10 X Xb X
PM2.5 X X Xb X
NO2 X X X
Ozone X X
Allergen X X X
Endotoxin X X
Nicotine X X X X
Abbreviations: BC, birth cohort; CCH, Cincinnati Children’s Hospital; IT, intervention trial; LC, longitudinal cohort; PM, particulate matter; PM2.5, PM < 2.5 μm in diameter; PM10, PM < 10 μm in diameter.
a Mold spores documented.
b Evaluated at baseline and every 3 months.
Table 5 Biologic samples.
Sample Berkeley Columbia JHU: LC Michigan: LC CCH USC: IT JHU: IT Michigan: IT
Blood Maternal pregnancy and delivery; cord blood; 12, 24 months Cord blood, 24, 36, 60 months Baseline Baseline, 6, 12 months Baseline, 12 months
Urine Maternal pregnancy and delivery; 6, 12, 24, 42 months Maternal pregnancy, 36, 60 months 6 months Baseline
Meconium X
Saliva 42 months
Hair Baseline, 6, 12 months
Skin tests X X X X X X X
RAST X X X X
Abbreviations: BC, birth cohort; CCH, Cincinnati Children’s Hospital; IT, intervention trial; LC, longitudinal cohort. Samples were not collected in USC: LC and Iowa.
Table 6 Asthma disease activity.
Characteristic Berkeley Columbia USC-LC JHU-LC Michigan-LC CCH USC-IT JHU-IT Iowa Michigan-IT
Day symptoms X X X X X X X X X
Night symptoms X X X X X X X X X
Exercise symptoms X X X X X X X X X
Activity limited X X X X X X X X
ISAAC questionsa X X X X X X X X
Quality of lifea X X X X X X
Symptom medications X X X X X X X X X X
Control medications X X X X X X X X X X
Oral steroids X X X X X X X X X X
Recall periodb 6–12 months 3–6 months 1 year Daily 2 weeks, 3 months 8 weeks 2 weeks 2 weeks 2 weeks
ED visits X X X X X X X X X X
Hospitalization X X X X X X X X X X
Recall periodc 6–12 months 3 months 1 year 3 months Daily 3 months 4 months 3 months 2 months 3 months, 1 year
FEV1 X X X X X X X X
Daily FEV1 X X X
Daily PEFR X X X X X
Allergy history X X X X X X X X
Family history X X X X X X X X
Abbreviations: BC, birth cohort; CCH, Cincinnati Children’s Hospital; ED, emergency department; IT, intervention trial; LC, longitudinal cohort.
a One-year recall period for ISAAC, 1-week recall period for quality-of-life questions.
b Recall period for symptoms, medications.
c Recall period for emergency department, hospitalizations.
Table 7 NAEPP asthma activity classification.
Classification Days with symptomsa Nights with symptomsb PEFR, FEV1 (%)c
Mild intermittent 0–1 0–1 > 80
Mild persistent > 2 2 > 80
Moderate persistent Daily Weekly < 80
Severe persistent Constant > Weekly < 60
a Days during previous 2 weeks.
b Nights during previous month.
c Measures as a percentage of normal predicted for age, sex, and race.
Table 8 Interventions.
Cincinnati USC JHU Iowa Michigan
Pest control X X X X
Allergen-proof mattress covers X X X
HEPA air cleaners X
Vacuum cleaners with HEPA filters X
Cleaning supplies X X
Environmental education X X X X X
Asthma management education X X Xa
Smoking cessation X X X Xa
Physician education X
Social support and referrals X
a Provided only in a limited fashion.
==== Refs
References
Asher MI Anderson HR Beasley R Crane J Martinez F Mitchell EA 1995 International study of asthma and allergies in childhood (ISAAC): rationale and methods Eur Respir J 8 483 491 7789502
Asmussen L Olson LM Grant EN Fagan J Weiss KB 1999 Reliability and Validity of the Children’s Health Survey for Asthma Pediatrics 104 e71 Available: http://www.pediatrics.org/cgi/content/full/104/6/e71 [accessed 24 January 2005].10586005
Chrischilles E Ahrens R Kuehl A Kelly K Thorne P Burmeister L 2004 Asthma prevalence and morbidity among rural Iowa schoolchildren J Allergy Clin Immunol 113 66 71 14713909
Diette GB Skinner EA Nguyen TT Markson L Clark BD Wu AW 2001 Comparison of quality of care by specialist and generalist physicians as a usual source of asthma Pediatrics 108 432 437 11483811
Eigen H Bieler H Grant D Christoph K Terrill D Heilman DK 2001 Spirometric pulmonary function in healthy preschool children Am J Resp Crit Care Med 163 619 623 11254514
Eskenazi B Gladstone EA Berkowitz GS Drew CH Faustman EM Holland NT 2005 Methodologic and logistic issues in conducting longitudinal birth cohort studies: lessons learned from the Centers for Children’s Environmental Health and Disease Prevention Research Environ Health Perspect 113 1419 1429 16203258
Ferris BG 1978 Epidemiology standardization project. II. Recommended respiratory disease questionnaires for use with adults and children in epidemiologic research Am Rev Respir Dis 118 suppl 7 53 354446
Gold DR Weiss ST Tager IB Segal MR Speizer FE 1989 Comparison of questionnaires and diary methods in acute respiratory illness surveillance Am Rev Respir Dis 139 847 849 2923381
Israel BA Parker EA Rowe Z Salvatore A Minkler M López J 2005 Community-based participatory research: lessons learned from the Centers for Children’s Environmental Health and Disease Prevention Research Environ Health Perspect 113 1463 1471 16203263
Juniper EF Guyatt GH Feeny DH Ferrie PJ Griffith LE Townsend M 1996 Measuring quality of life in children with asthma Qual Life Res 5 35 46 8901365
Juniper EF O’Byrne PM Ferrrie PJ King DR Roberts JN 2000 Measuring asthma control: clinic questionnaire or daily diary Am J Respir Crit Care Med 162 1330 1334 11029340
Kattan M Mitchell H Eggleston P Gergen P Crain E Redline S 1997 Characteristics of inner-city children with asthma: the National Cooperative Inner-City Asthma Study Pediatr Pulmonol 24 253 262 9368259
Kimmel CA Collman GW Fields N Eskenazi B 2005 Lessons learned for the National Children’s Study from the National Institute of Environmental Health Sciences/U.S. Environmental Protection Agency Centers for Children’s Environmental Health and Disease Prevention Research Environ Health Perspect 113 1414 1418 16203257
Lewis TC Robins TG Joseph CLM Parker EA Israel BA Rowe Z 2004 Identification of gaps in the diagnosis and treatment of childhood asthma using a community-based participatory research approach J Urban Health 81 472 488 15273269
Mitchell H Senturia Y Gergen P Baker D Joseph C McNiff-Mortimer K 1997 Design and methods of the National Cooperative Inner-City Asthma Study Pediatr Pulmonol 24 237 252 9368258
Nathan RA Sorkness CA Kosinski M Schatz M Li JT Marcus P 2004 Development of the asthma control test: a survey of assessing asthma control J Allergy Clin Immunol 113 59 65 14713908
National Heart, Lung, and Blood Institute 1997. National Asthma Education and Prevention Program. Expert Panel Report 2: Guidelines for the Diagnosis and Management of Asthma. NIH Publication 97-4051. Bethesda, MD:National Heart, Lung, and Blood Institute.
Rosenstreich DL Eggleston PA Kattan M Baker D Slavin RG Gergen P for the National Cooperative Inner City Asthma Study 1997 The role of cockroach allergy and exposure to cockroach allergen in causing morbidity among inner-city children with asthma N Engl J Med 336 1356 1376 9134876
Sampson HA Albergo R 1984 Comparison of results of skin tests, RAST and double-blind, placebo-controlled food challenges in children with atopic dermatitis J Allergy Clin Immunol 74 26 33 6547461
Swartz LJ Callahan KA Butz AM Rand CS Kanchanaraksa S Diette GB 2004 Methods and issues in conducting a community-based environmental randomized trial Environ Res 95 156 165 15147921
Szefler SJ Weiss S Tonascia J for the Childhood Asthma Management Program Research Group 2000 Long-term effects of budesonide or nedocromil in children with asthma N Engl J Med 343 1054 1063 11027739
WHO 1978. International Classification of Diseases, 9th Revision. Geneva:World Health Organization.
Wissow LS Warshow Box J Baker D 1988 Case management and quality assurance of improved care of inner-city children with asthma Am J Dis Child 142 748 754 3289373
Wood RA Phipatanakul W Hamilton RG Eggleston PA 1999 A comparison of prick skin tests, intradermal skin tests, and RASTs in the diagnosis of cat allergy J Allergy Clin Immunol 103 1108 1114 10359893
|
16203259
|
PMC1281292
|
CC0
|
2021-01-04 23:41:28
|
no
|
Environ Health Perspect. 2005 Oct 24; 113(10):1430-1436
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.7671
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences 10.1289/ehp.7672ehp0113-00143716203260ResearchMini-MonographPrinciples and Practices of Neurodevelopmental Assessment in Children: Lessons Learned from the Centers for Children’s Environmental Health and Disease Prevention Research Dietrich Kim N. 1Eskenazi Brenda 2Schantz Susan 3Yolton Kimberly 4Rauh Virginia A. 5Johnson Caroline B. 6Alkon Abbey 7Canfield Richard L. 8Pessah Isaac N. 9Berman Robert F. 101 University of Cincinnati College of Medicine, Division of Epidemiology and Biostatistics, Department of Environmental Health, and the Cincinnati Children’s Environmental Health Center, Cincinnati, Ohio, USA2 Center for Children’s Environmental Health Research, School of Public Health, University of California at Berkeley, California, USA3 The Friends Children’s Environmental Health Center, University of Illinois, Urbana-Champaign, Illinois, USA4 Children’s Environmental Health Center, Cincinnati Children’s Hospital Medical Center and University of Cincinnati College of Medicine, Cincinnati, Ohio, USA5 Columbia Center for Children’s Environmental Health, Columbia University, New York, New York, USA6 Pediatric Neuropsychology Group, Berkeley, California, USA7 Department of Family Health Care Nursing, School of Nursing, University of California, San Francisco, California, USA8 Division of Nutritional Sciences, College of Human Ecology, Cornell University, Ithaca, New York, USA9 Department of Molecular Biosciences, and10 Department of Neurological Surgery, University of California at Davis, Davis, California, USAAddress correspondence to K.N. Dietrich, University of Cincinnati College of Medicine, Department of Environmental Health, Division of Epidemiology and Biostatistics, 3223 Eden Ave., Cincinnati, Ohio 45267-0056. Telephone: (513) 558-0531. Fax: (513) 558-4838. E-mail:
[email protected] authors declare they have no competing financial interests.
10 2005 24 6 2005 113 10 1437 1446 12 10 2004 14 3 2005 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright. Principles and practices of pediatric neurotoxicology are reviewed here with the purpose of guiding the design and execution of the planned National Children’s Study. The developing human central nervous system is the target organ most vulnerable to environmental chemicals. An investigation of the effects of environmental exposures on child development is a complex endeavor that requires consideration of numerous critical factors pertinent to a study’s concept, design, and execution. These include the timing of neurodevelopmental assessment, matters of biologic plausibility, site, child and population factors, data quality assurance and control, the selection of appropriate domains and measures of neurobehavior, and data safety and monitoring. Here we summarize instruments for the assessment of the neonate, infant, and child that are being employed in the Centers for Children’s Environmental Health and Disease Prevention Research, sponsored by the National Institute of Environmental Health Sciences and the U.S. Environmental Protection Agency, discuss neural and neurobiologic measures of development, and consider the promises of gene–environment studies. The vulnerability of the human central nervous system to environmental chemicals has been well established, but the contribution these exposures may make to problems such as attention deficit disorder, conduct problems, pervasive developmental disorder, or autism spectrum disorder remain uncertain. Large-scale studies such as the National Children’s Study may provide some important clues. The human neurodevelopmental phenotype will be most clearly represented in models that include environmental chemical exposures, the social milieu, and complex human genetic characteristics that we are just beginning to understand.
behaviordevelopmentNational Children’s Studyneurotoxicologystudy design
==== Body
The purpose of this article is to present lessons learned in neurodevelopmental assessment of children from the Centers for Children’s Environmental Health and Disease Prevention Research (Children’s Centers) and from previous research. These comments and recommendations are products of the authors’ experiences within their centers and their past works in the area of environmental neurotoxicology and teratology. We present both missteps and achievements in the hope that our collective experience can help guide in the planning and implementation of the National Children’s Study.
The Central Nervous System as a Critical Organ
A fundamental lesson learned from studies of the effects of environmental exposures to the fetus and child is that the developing brain is one of the organs in the human body most sensitive to damage. Functional manifestations ranging from frank mental retardation to milder learning disabilities are the most common class of birth defects (Lipkin 1991), although in most cases the specific etiology is unknown (Kallen 1988). Researchers have speculated that some cases of uncertain etiology may be due to environmental chemical exposures (Rees et al. 1990). The maturation of the central nervous system requires a more complex sequence of processes than any other structure, making this organ uniquely vulnerable to environmental influences (Rodier 1994, 2004).
National and international agencies have focused on the risks of early exposure to several major environmental contaminants such as lead, methylmercury, and polychlorinated biphenyls (PCBs) on neurodevelopment [Agency for Toxic Substances and Disease Registry (ATSDR) 2000; International Programme on Chemical Safety (IPCS) 1995; National Research Council 2000]. Although there is a paucity of data on the risks of lower-level pesticide exposures on human neurodevelopment, there are substantial animal model data and limited human developmental data that these poisons may affect intrauterine growth and have functional teratogenic properties as well (Guillette et al. 1998; Weiss 1997; Weiss et al. 2004).
Initial Considerations
Timing of neurodevelopmental assessment.
The vast and rapid growth of the child’s neurobehavioral repertoire from birth through adolescence means that functional expressions of earlier-damaged systems may not be present or accessible at any given moment in time. For example, finding no effects on infant behavior cannot be regarded as conclusive evidence that the toxic agent has had no impact. Negative findings in the preschool period are also inconclusive. A toxicant may damage higher cortical centers that are associated with neurocognitive processes that are not yet functional or only marginally functional in a preschooler. In older children, a wider and more differentiated range of abilities can be examined, scores on psychometric measures are more precise and reliable, and early academic performance and social functioning outside of the home environment can be evaluated. This provides the rationale for long-term prospective longitudinal studies of cohorts recruited prenatally or at birth as the optimal design to ascertain neurobehavioral deficits in relation to exposure to environmental chemicals (Krasnegor et al. 1994). The literature on lead provides a model example of early exposure associated with higher-order neuropsychological and behavioral dysfunctions in older children and adolescents (e.g., Bellinger et al. 1992; Burns et al. 1999; Needleman et al. 1990; Tong et al. 1996). Long-term neurodevelopmental consequences have also been observed in a cohort study of children exposed in utero to PCBs (Jacobson and Jacobson 1996, 2003). Evaluations at earlier ages can and have revealed changes in cognitive function, but these may be less robust and definitive (Dietrich and Bellinger 1994).
Biologic plausibility.
A number of factors determine the transplacental passage of environmental contaminants (Wilson 1977). Toxicants with low molecular weight, lipid affinity, nonpolarity, and low protein binding properties cross the placenta with ease (Slikker and Miller 1994). Unfortunately, most compounds possess one or more of these properties, cross the placenta, and enter the fetal circulation (Beckman and Brent 1999).
The immature blood–brain barrier of the fetus and young infant is more permeable to xenobiotics, and the fetus lacks drug-metabolizing detoxification capacities that are present postnatally (Rozman and Klaassen 1996). The chemical properties of certain toxicants also determine access to brain tissues. Lead imitates calcium ions and therefore crosses the blood–brain barrier with relative ease (Kerper and Hinkle 1997). In the blood stream, methylmercury combines with cysteine, forming a compound that is structurally similar to the essential amino acid methionine. This methylmercury–cysteine compound is actively transported into the endothelial cells in the blood–brain barrier on the methionine carrier and ultimately into the brain on a glutathione carrier (Kerper et al. 1992). The chlorobiphenyls and several classes of pesticides also gain access to brain tissues by virtue of their lipid solubility.
Toxicants linked with growth retardation and maturational delays in utero should be considered prime candidates for functional developmental toxicity. Any compound that retards intrauterine somatic growth should be examined as a potential neurobehavioral teratogen (Wilson 1977). Low-level intrauterine exposure to lead, PCBs, and organophosphate pesticides has been associated with lower birth weight, gestational maturity, or reduced head circumference in some prospective studies (e.g., Berkowitz et al. 2004; Dietrich et al. 1987; Eskenazi et al. 2004; Fein et al. 1984; McMichael et al. 1986; Whyatt et al. 2004).
Because of their potential disruption of central nervous system morphoregulation, hormonally active agents should also be considered as candidate neurodevelopmental toxicants. Hydroxylated metabolites of PCBs and related compounds may bind to human transthyretin, the only thyroid-hormone–binding protein synthesized in the brain. By binding to transthyretin, some hydroxylated PCBs may alter brain free thyroxine (T4) levels and interfere with central nervous system development and function (Cheek et al. 1999; Seegal 2000). Recent evidence also suggests that PCBs up-regulate several thyroid-hormone–responsive genes that are expressed during periods of brain development (Gauger et al. 2004).
Environmental chemicals with excitatory effects on neurons (prolonged depolarization) are potential developmental neurotoxicants, as well. Examples include nicotine, the organophosphate and carbamate pesticides (Slotkin 1999), and some metals, including lead (Bressler et al. 1999).
Population factors.
Population factors encompass a host of variables that affect a study’s design, plan of execution, and sensitivity to the potentially adverse neurodevelopmental consequences of exposure to environmental toxicants. Several sometimes highly intercorrelated co-factors are involved, such as socioeconomic status, ethnicity, nutrition, access to medical care and educational resources, language spoken in the home, and cultural milieu. These variables will need to be considered when devising recruitment and long-term retention strategies, and especially when selecting the most appropriate methods of neurodevelopmental assessment. The field of neurobehavioral testing has been at the center of this dilemma as psychologists have struggled with the problem of how to design tests that do not measure only culturally specific information that is quite familiar to children from certain groups or social backgrounds but less familiar to those from different backgrounds. Investigators can use tests that have been adapted for non-English-speaking children and their families. In the United States, this typically involves Spanish translations of extant instruments (e.g., intelligence or speech, language, and memory tests), but the inventory of instruments in Spanish is still quite low. Furthermore, different dialects within the U.S. Spanish-speaking population exist, thus adding another complication. For example, the Children’s Centers at the University of California at Berkeley, Columbia University, and Mount Sinai School of Medicine had Spanish-speaking populations from Puerto Rico, Mexico, and Dominican Republic with different regional dialects. In these cases, it is important to ask the caregiver what language is spoken in the home. All assessments of non-English–speaking children should be done by examiners who are bilingual, preferably with the language of concern (e.g., Spanish) as their native tongue. Piloting of previously translated tests in the population of interest is always essential to determine their suitability. Some nonverbal tests have been considered to be culturally neutral, such as the Raven’s Progressive Matrices (Raven 1998). However, even nonverbal tests may not be completely culturally neutral if the skills needed to complete the task are outside of the cultural experience of the population being evaluated.
One solution to this problem may lie in the use of so-called culture-neutral electrophysiologic, operant learning, or classical conditioning protocols. These types of tests also have the advantage that parallel animal tests that assess the same or similar functional domains are available. The drawback of using these “culture-fair” procedures is that the functional significance of any finding is sometimes difficult to calculate.
The degree of confounding can also vary greatly from one population to another and significantly affect a study’s ability to detect associations between toxicant exposures and various parameters of neurodevelopment. In some populations, the degree of confounding may be so great that, after statistical adjustment, the exposure variable(s) no longer accounts for any further unique (independent) variance in the neurobehavioral data. Lead is a model example where exposure is typically correlated with other suboptimal environmental and sociohereditary factors (e.g., Dietrich et al. 1991). In such cases, any additional attributable developmental risk, beyond that accounted for by confounders, is obscured. This raises the specter of type II error (Needleman and Bellinger 1986).
The problem of confounding is not limited to correlations between a chemical exposure and nonchemical covariables such as socioeconomic status, quality of child rearing, and parental intelligence, among others. In some cases, populations can be exposed to a mixture of compounds that are also intercorrelated. This presents a conceptual and biostatistical challenge when attempting to estimate the independent or combined additive or synergistic effects of multiple chemical exposures (Jacobson 2001). For example, critical reviews of the Faroe Islands study of methylmercury and child development have suggested that the effects attributed to intrauterine exposure to methylmercury might be the consequence of co-exposures to high levels of PCBs [Myers and Davidson 2000; Toxicology Excellence for Risk Assessment (TERA) 1999]. Most other investigations of neurotoxicants in humans have not measured a multitude of exposures but have examined the behavioral toxicity of single toxicants. With the advent of more sensitive biomarkers of environmental exposures, researchers such as those involved in the Children’s Centers and in the National Children’s Study will be facing a new challenge of developing both statistical and toxicologic methods for addressing the impact of multiple exposures. The importance of this has been shown in a recent study in animals demonstrating that the effect of the organophosphate chlorpyrifos was exacerbated by the pharmaceutical agent terbutaline commonly used to prevent preterm delivery (Rhodes et al. 2004).
The problem of confounding in environmental neuroepidemiology has led some to speculate about the advantages of studying chemical or drug exposures in lower risk populations for the purpose of reducing confounding and thus strengthening associations between measures of dose and disease (e.g., Bellinger 1995). However, restricting neurobehavioral studies of environmental toxicants to lower risk populations is not without scientific and ethical drawbacks. U.S. citizens at the lowest levels of social strata tend to suffer the greatest toxic burden. Furthermore, the opportunity to assess the interactions of toxicant exposures with other risk factors may be missed. The effects of prenatal and early postnatal exposure to developmental neurotoxicants may be more severe among the disadvantaged, where nutritional deficiencies, lack of adequate prenatal care, and suboptimal psychosocial environmental factors are likely to be more common.
Site factors.
Examiners are often compelled to conduct assessments under less than ideal conditions. Inadequate space, lighting, excess ambient noise, and other distractions can invalidate even the most carefully administered protocol. The developmental assessment of an infant or child should be regarded as a controlled experiment, just as in animal models involving behavioral measures.
For neonates and infants, it is especially critical that the clinic environment is conducive to a valid test. For the newborn, lighting, temperature, noise, and any other factors that affect the neonate’s state can determine the validity of the test. For an infant or child of any age, a comfortable, quiet, and well-lighted testing environment is a minimal requirement. Furniture, including chairs and table, should be appropriate for the test and developmental stage of the child. Test materials should be out of the child’s sight but still easily accessible to the examiner to avoid fumbling and other miscues. Infants and some preschool children will often need the security and support of a care-giver during the examination, so accommodations should be made to have another adult in the room, seated in an area where the subject will not be distracted.
In some studies the geographic dispersion of the population may make home or school place testing the only practical option. Indeed, testing sometimes must take place in multiple sites to accommodate families and prevent loss to follow-up. For example, studies in large rural communities (e.g., the Berkeley study of children in the Salinas Valley community) find it necessary to use a recreational vehicle in addition to the clinic site to reach families without any means of transportation and to avoid distractions within the home.
Child factors.
It would seem obvious that the infant or child should be in an appropriate state or physical condition for assessment. However, this factor is not always given the consideration it deserves. The child should be reasonably well when evaluated with no current infection that is likely to significantly affect performance such as a severe upper respiratory infection or acute otitis media. Any medications the child may be on should also be recorded. It is important to clarify this with the parent before the assessment begins. A child who is too ill to respond appropriately to the demands of the examination should be rescheduled.
Given that many neurodevelopmental testing procedures require normal sensory function, a vision and hearing screen should also precede the examination. In some studies, sensory functions may be core outcome variables. In this case, a more detailed assessment of vision (farsightedness, depth perception, color perception, acuity/contrast sensitivity) and hearing (tympanometry, pure-tone audiometry, central auditory processing) will be indicated. Assessment of audition can be conducted as early as 6 months using methods such as visual reinforcement play audiometry. The auditory brainstem response (ABR) can also be employed to noninvasively measure the electrophysiologic responses of the brainstem auditory pathways (Roizen 1996). The automated ABR is already a routine procedure for screening neonates for hearing problems in a large number of hospitals in the United States, although a program of universal screening has not been implemented [American Academy of Pediatrics (AAP) 1995]. Methods for assessing refractive errors and amblyopia in infants and young children with limited verbal abilities such as the Allen cards and Snellen chart can also be used. However, in the toddler and pre-schooler, observing the child’s response to visually engaging stimuli of various sizes can be just as valuable as a screening tool (Hoon 1996).
These neurosensory data can be used as core outcomes (i.e., if the toxicant in question is expected to affect vision or hearing), as effect modifiers (i.e., a toxicant’s effects on tasks involving visual or auditory processing may be more pronounced among youngsters with preexisting deficits in these areas), or as potential confounders. However, extreme care must be taken when using such data as confounding or control variables. If deficits in neurosensory function are an expression of exposure to the toxicant(s) in question, the resulting overcontrol may lead to a false negative finding or type II error.
Some way of rating the child’s behavior and affect during the test session is also desirable. The child’s response to examiner and test situation, attitude toward self and test performance, work habits and problem-solving style, motor functioning, visual and auditory acuity, oral communication, and mood are some of the factors that could potentially affect performance and should be noted. These data can be entertained as co-factors in outcome analyses, but this may lead to overcontrol. Indeed, behavioral disturbances that are rated by the examiner as limiting the validity of the examination can also be considered as legitimate behavioral sequelae of exposure.
Another factor that can affect a child’s performance is the presence of a parent. A parent provides physical and emotional support for an infant or toddler but can also be a source of disruptive influences on the child’s responses. Although the presence of a familiar caregiver is necessary and desirable for assessment of infants < 2 years of age, the parent or other caretaker should be instructed not to interfere with the presentation of stimulus materials, examiner instructions, or the child’s behaviors during the test. By the later preschool years, the parent should be encouraged to not be present during the examination. If a child absolutely needs the security of a familiar care-giver, the examiner should seat the adult companion outside of the child’s field of vision.
Quality assurance and quality control.
Analytical laboratories go to great lengths to assure the validity and reliability of their assessments of toxicants in environmental samples and biologic tissues with quality assurance and quality control protocols. Similar issues are germane to neurodevelopmental assessment. Results of individually administered behavioral assessments depend to a great degree on the interaction between the child and examiner. Therefore, it is imperative that the examiner’s contribution be equivalent for all children so that interchild differences in performance can be attributed to child characteristics rather than to some combination of child and examiner characteristics.
Although not always easy to achieve, the following recommendations apply to any longitudinal prospective study of the effects of neurodevelopmental toxicants. Ideally a single examiner should be used at each site or center. The examiner should be well seasoned and familiar with the population of interest. It is not necessary for a doctoral-level psychologist to administer most tests that are likely to be used in these studies (Brandt and van Gorp 1999). Individuals with baccalaureate or master degrees in psychology or related fields with additional training in the standardized administration and scoring of neuropsychological tests can examine children enrolled in a study, under professional supervision [American Academy of Clinical Neuropsychology (AACN) 1999]. The examiner should be blinded to the group membership or exposure status of the child. Children from high-, medium-, and low-level exposure groups (if known) should be tested in a randomly intermixed order. If more than one examiner is used, their comparability in training and technique should be explicitly checked and regularly monitored. In multicenter studies, examiner training should be standardized across sites and regular meetings, and conference calls should be arranged to discuss issues of administration and scoring as they arise.
Dietrich and Bellinger (1994) cite an example of the difficulties that can arise when the second and last guidelines are neglected. In a study of lead-exposed children, the average IQ scores assessed by two relatively inexperienced examiners differed by more than one standard deviation or 15 points (Gregory et al. 1976).
In long-term prospective studies, it is not always possible to rely on a single experienced examiner to test all children. Attrition of staff and requirements for backup psychometricians in the event of leave usually require that a research center employ the services of two or more examiners. Interexaminer differences can be minimized by having the same developmental neuropsychologist train them and by videotaping practice administration sessions to provide feedback to the trainee and assess the presence of any differences in adherence to standardized administration or style that may result in interexaminer variability and measurement error. Videotaping can also provide the opportunity to assess interexaminer reliability in scoring the protocols (Chandlee et al. 2002). In studies spanning several years, monitoring of interexaminer reliability and proficiency should be practiced at regular intervals. This is particularly critical for multicenter studies employing identical neurodevelopmental assessment protocols. Whether one or several examiners are active at a given site, it is imperative that their reliability and proficiency be monitored over the entire course of the study to prevent any drift away from standardized procedures for administration and scoring.
All studies should carefully document quality assurance and control procedures and report intertester and interscorer reliability coefficients where appropriate. As Bellinger (2002) has noted, “the expectations for reporting the reliability of these [neurobehavioral] measurements should be no different from the expectations for reporting exposure (i.e., bio-marker) measurements such as hair-mercury or blood-lead levels.”
Nevertheless, although the Children’s Centers instituted quality control protocols, particularly for the neurodevelopmental assessments, including direct observations or videotaping, insufficient time and resources made it difficult to fully meet the goals of these quality control programs. In our experience, the effort and cost associated with this are frequently underestimated. The National Children’s Study needs to make every effort not to fall short in this critical area.
Sensitivity and specificity of neurodevelopmental measures.
The terms “sensitivity” and “specificity” have two different meanings in the context of this article. In the evaluation of diagnostic tests, the sensitivity of a measure is defined as the proportion with the abnormality that the test classifies as abnormal (i.e., the proportion of true positives), whereas the “specificity” is the proportion of normal that the test classifies as normal (i.e., the proportion of true negatives). In the selection of neurodevelopmental measures for environmental studies, it is clearly advantageous to include tests that have the best possible prognostic value. This is particularly critical if neurobehavioral end points are incorporated into risk assessments conducted by regulatory bodies (Bellinger 2002). However, in this context we are also concerned with the selection of instruments that tap into neurodevelopmental domains that have been shown to be sensitive to (affected by) exposure to particular environmental toxicants. Of course, this assumes that some information already exists as to the effects of a given compound on the developing nervous system. Sometimes there are no prior human studies available to help dictate the domains that should be examined, and animal studies may not include domains that are unique to human cognition (e.g., language). In cases where there are multiple potential exposures with different mechanisms or loci of effect or when the exact exposures are not known or hypothesized, it may be necessary to use a broad spectrum of assessment tools. However, to the extent such information is available, selection of tests can be based on hypothesis-based inference. Under these circumstances, the focus shifts from “selecting tests” to the nomination of neurobehavioral domains to tap and subsequently to the employment of the best available measures (Bernstein 1994).
A wide range of neurobehavioral assessments has proven to be sensitive to lower level prenatal and early postnatal exposures to environmental contaminants, for example, lead, methylmercury, PCBs, environmental tobacco smoke, and other chemicals. However, the identification of behavioral phenotypes for specific agents has been an elusive goal. Some attempts have been made to differentiate the effects of alcohol exposure in utero from deficits in neurobehavior associated with PCBs and other toxicants [e.g., Jacobson 1998; National Institute on Drug Abuse (NIDA) 2000]. However, environmental studies have failed to clearly identify a “behavioral signature” for particular compounds. The best evidence for specific effects probably comes from the literature on PCBs, where deficits in cognitive processing involving visual discrimination and memory appear to be a somewhat consistent finding in exposed infants and preschool children (Darvill et al. 2000; Jacobson et al. 1992). However, toxicants such as lead, methylmercury, and even the chlorobiphenyls seem to affect a wide range of neurobehavioral outcomes without illustrating a convincing degree of consistency across domains from one study to another. The sample’s socioeconomic status; level, pattern, and timing of exposures; nutritional intake; general health; educational opportunities; and the particular instruments that were employed to examine neurodevelopment probably play an important role in between-study differences (Bellinger 1995; Schantz 1996). This could explain why the wide net provided by global and multiple-domain assessments of cognitive development such as IQ have proven to be so consistently sensitive across studies. Because these tests combine subscales that are representative of a broad number of underlying cognitive functions, they are likely to pick up exposure-related deficits across cohorts that differ in their functional expressions of toxicity. However, despite the sensitivity of these tests, their use in environmental studies has been sharply criticized by some reviewers. Critics of these apical measures note that they cannot readily yield information about the affected neural substrates or the complex executive and regulatory processes involved in learning, problem solving, and behavior (Bernstein 1994; Krasnegor et al. 1994; White et al. 1994). Furthermore, although global measures of intellectual functioning have proven to be sensitive measures of exposure to a number of environmental chemicals, this has not been the case for prenatal exposure to alcohol or drugs (NIDA 2000).
Data safety and monitoring.
There is an ethical responsibility that referral protocols be in place to deal with the needs of children who perform poorly in the course of their participation in the study. Criteria need to be established for referral before data collection begins. Because of the sensitivity of certain tests, only performance on some may result in referral. Also, because of the predictive validity of these tests, a more rigorous criterion may be used at older ages (e.g., scores three standard deviations below the mean up to 24 months and 2 standard deviations afterward). Referrals normally take place through the primary care provider with parental permission. All referral related contacts should be carefully documented and confidentiality closely guarded.
Neurodevelopmental Assessment in Environmental Studies: Initial Challenges
Constructing a battery of tests for an environmental study of developmental neurotoxicants can be a daunting task. First, the investigator is faced with choosing from a large number of potential end points that can be measured in a prospective study of any given agent or mixture of compounds to which the fetus or child may be exposed. Domains of interest in studies of this kind include overall neurologic status, sensorimotor skills, attention, memory, problem solving (executive functions, organization, and planning), visual–spatial and perceptual skills, speech and language abilities, behavioral problems, and adaptive skills as well as more global indices of intellectual attainment and academic achievement. Ideally, as stated above, the choice of neurobehavioral domains and the tests used to index them should be determined by what is known about the impact of a particular environmental agent on the development of specific cognitive, neuromotor, and behavioral features. However, as previously discussed, this is not always easily accomplished because of the lack of evidence for specific behavioral phenotypes. In this regard, animal studies have potential for identifying key functional domains that are affected by exposure and would be relevant to assess in human populations. For example, the Children’s Center at the University of Illinois is using studies in animals exposed to PCBs and methylmercury during gestation to identify specific aspects of cognitive, sensory, and motor functions that are affected by combined exposures. These data will be used to guide the assessment of children in a companion cohort study.
Ultimately, the best strategy is to develop a battery that includes both broad-based measures of cognitive and neuromotor status as well as more fine-grained assessments of specific skills. Narrow-band instruments should target domains that, based on the extant human and animal literature, are believed to be affected by exposure to the toxicant(s) in question (e.g., for hypothesis-based battery development from the perspective of childhood lead poisoning, see Dietrich et al. 2004; Rogan et al. 2001).
A second consideration in battery development is the amount of time that can be allocated to the assessment. The energy and attention spans of younger subjects are rapidly exhausted, calling for shorter periods of time for testing, more frequent breaks, and sometimes multiple appointments separated by days or weeks. Infants < 1 year of age can usually tolerate only about 45 min of standardized testing, whereas a 2-year-old can generally perform adequately over a period of roughly 75 min. For preschool and older children, breaks for a snack or just relaxing with the primary caregiver can fortify the child’s endurance. Furthermore, neurodevelopmental evaluations are often part of a larger protocol that includes collection of sociodemographic information from parents, a general medical evaluation of the child, and collection of biologic specimens such as blood, hair, and urine. Retention of children and their families in a long-term study can be compromised if the demands on their time and effort are excessive.
Unique Considerations for Assessment of the Neonate, Infant, and Child
The neonate.
The neonatal period extends from birth to roughly 1 month of age. Assessment of the neonate can be used to evaluate gestational maturity, neurologic status, and behavioral style and capacities. Neonatal assessment is a highly specialized area requiring continuous and careful attention to the infant’s state as the examination proceeds. Recognized infant states include several sleep states, drowsiness or transitional states, alert states, and crying. Reflexes and muscle tone, signs of stress, and alertness and orientation can all be affected by the newborn’s state.
Although neonatal neurodevelopmental assessments are not highly predictive of later functioning, performing a neurologic and neurobehavioral assessment after birth provides an assessment of the immediate effects of prenatal exposure before any influences of the extrauterine environment take place. A repeat exam (i.e., 2 or 4 weeks) allows assessment of neurobehavior after infants have stabilized. Inclusion of both assessment time points may provide insight into those effects of prenatal exposure that are transient and those that persist (Jacobson et al. 1984; Stewart et al. 2000). Neonatal assessment places special demands on quality control and site. Some instruments such as the Neonatal Behavioral Assessment Scale (NBAS) (Brazelton 1984; Brazelton and Nugent 1995) and the Neonatal Intensive Care Unit (NICU) Network Neurobehavioral Scale (Lester et al. 2004) require intensive training of examiners to master the techniques of administration and become reliable in scoring newborn responses. Training may demand travel to institutions with certification programs (for further details, see Brown Medical School Infant Development Center 2005). Each investigator will have to determine if these essential preconditions for neonatal neurobehavioral assessment can be met. The assessment of the neurologic and neurobehavioral status of the newborn will not usually be a practical option in studies involving populations that are widely dispersed geographically or living in communities lacking the required clinical facilities or trained and certified personnel.
The infant and toddler.
Infants and toddlers between 6 and 36 months of age may be the most challenging population for obtaining systematic and reliable behavioral data using paradigms that require the child to sustain attention, follow directions, and direct their cognitive effort toward a task that requires adherence to a somewhat rigid protocol. Although the cognitive capacities of older infants and toddlers are much advanced compared with their prelinguistic and prelocomotor days, directing that cognitive capacity at any particular time is often the most challenging part of the assessment. The clinical skills of the person examining the infant and toddler are of paramount importance for obtaining data that are valid and reliable.
Indices of neurodevelopment in infants and toddlers are less stable over time and, at least before 24 months, lack substantial predictive validity for later morbidity. This is partly because of the means by which infants are able to express their cognitive abilities (i.e., primarily through sensorimotor acts) and the lack of continuity in response modalities from infancy to older childhood and adolescence. However, neurobehavioral test scores in infancy retain strong concurrent validity. Bellinger (2002) suggests that scores on infant neurodevelopmental measures can be understood in a manner analogous to the neonatologist’s interpretation of birth weight. Except in very low-birth-weight infants, weight at birth is not predictive of later weight, although birth weight is a very informative index of a newborn’s general health status. Also, as with neonatal measures, the time between neurotoxicant exposure in utero and postnatally and the assessment of outcome is reduced. Thus, the influence of later intervening and potentially confounding factors is diminished along with increased strength and reduced bias in the estimate of the neurotoxicant’s contributions to development (Bellinger 2002). Furthermore, where neurotoxicants such as lead, PCBs, and methylmercury have predicted poorer performance in infants and preschoolers, forward studies have demonstrated that functional deficits in neurobehavior persist into later childhood (Dietrich 2000).
Older preschoolers and school-age children.
Because of the greater social maturity, autonomic stability, and endurance of the older preschool and school-age child, substantially more time can be devoted to a single assessment session. Also, owing to the tremendous growth in the range and clarity of a child’s response capabilities, assessments of preschoolers and older children generate a more differentiated picture of a subject’s developmental strengths and weaknesses. Functional impairments heretofore unnoticed may become apparent. To investigate the developmental effects associated with exposure to neurotoxicants in the fullness of time, studies should maintain follow-up at least until children attain school age. Indeed, recent studies of the long-term sequelae of early lead exposure would suggest follow-up to adolescence and young adulthood is necessary to reveal the full range of exposure related morbidities. There are a number of reasons for extending follow-up into late childhood and adolescence. Deficits still apparent at later ages are generally thought to be of greater practical significance because the predictive validity or prognostic value of later preschool and school-age performance is considerably greater than that of performance in infancy, thus providing a sounder basis for drawing inferences about the long-term effects of prenatal or early post-natal exposures.
Neurodevelopmental Assessment at the Children’s Centers
Neurodevelopmental assessment practices in the longitudinal birth cohort studies at the Children’s Centers were guided by many of the principles outlined above. Most centers conducted neurodevelopmental assessments at various ages after birth (see Table 1). For the first 5 years of funding, most centers conducted assessments at 12 months and 24 months of age.
A diverse group of standardized neurodevelopmental assessment tools were employed. There were no attempts to develop common protocols among the centers conducting neuroepidemiologic studies. Differences in tests and timing of assessment reflect variations in the toxicants under study, their hypothesized effects, and practical considerations. Four centers conducted neonatal assessments; three used the NBAS (Brazelton 1984; Brazelton and Nugent 1995), and one used the NICU Network Neurobehavioral Scale (Lester et al. 2004). The Bayley Scales of Infant Development–II (Bayley 1993), which provides standard scores for mental and psychomotor development, were used by all centers at 12 months of age and by most at 24 months. All centers adjusted the child’s chronologic age for prematurity on the Bayley scales until 24 months.
A variety of other assessment tools were used to measure domains including developmental milestones, language, and behavior. A few centers employed more experimental protocols, including assessment of visual recognition memory, the autonomic nervous system, and measures of learning that parallel some of those used in animal studies. The centers at Illinois and the University of Cincinnati also assess the child’s hearing.
The choice of neurodevelopmental tests was based first on the age of the child, followed by other considerations including tests employed in previous studies, the domains of behavior thought to be affected by the main toxicant(s) under study, the availability of tests in the language of the study population, time required to administer the examination, and the test’s suitability for administration by non-doctoral-level examiners. Some centers, especially those with low-income or bilingual study populations, found it necessary to select assessment tools with a lower end of functioning range, considerably below the ages of the children being assessed.
As discussed above, it can be difficult to determine the language of assessment for children who are raised in the United States in non-English-speaking households. All centers attempted to maximize the child’s performance on all tests. One center assessed bilingual children in English. Another center based the child’s assessment on what language was usually spoken in the home. In both cases, if the examiner noted language difficulties during the exam, the tasks were readministered in the other language.
The availability of assessment tools increases with the child’s age, enhancing the researcher’s ability to focus on specific developmental domains that may be affected by the exposure of interest. Because many of the chemicals examined by these centers have not been widely studied in children (e.g., pesticides, polycyclic aromatic hydrocarbons), the choice of the domains likely to be affected may be based solely on animal studies. Finding the human analog to animal behavior is challenging. However, as previously noted, the Illinois center is addressing this problem by conducting concurrent animal studies with exposures similar to those in their human population.
All centers used some quantitative measures of the parent’s cognitive functioning and quality of caregiving in the home to control for possible confounding. In centers dealing mainly with multicultural samples, tests of adult intellectual attainment with a minimum of verbal content were used.
Future Directions for Neurodevelopmental Assessments in the National Children’s Study
New technologies for assessment of the function and structure of the central nervous system may hold promise for advancing our understanding of the impact of environmental chemicals on neurodevelopment. Some of these are briefly discussed below.
Computer-based experimental measures for children.
Numerous attempts have been made to develop computer-administered batteries of tests for children exposed to environmental neurotoxicants. For example, adaptations of the Neurobehavioral Evaluation System (Baker et al. 1985) have been used in studies of children exposed to heavy metals in both the United States and abroad (Dahl et al. 1996). A computer-assisted comprehensive assessment for children enrolled in environmental studies has been developed by Anger and colleagues (Rohlman et al. 2003). The Behavioral Assessment and Research System has been designed for use in studies of children as young as 3 years of age and has been adapted for use with Hispanic populations (Rohlman et al. 2001). The entire battery consists of 11 computer-administered tests assessing sustained and selective attention, working memory, motivation, new learning, response speed, executive functions, and fine-motor coordination.
The use of computer-assisted tests has several advantages in that examiner effects are reduced and data collection and scoring are automated and objective. However, care must be exercised when applying these methods to populations that have little or no exposure to computers or similar kinds of automated systems. Another potential drawback of computer-assisted tasks is that they can minimize interaction with a friendly, supportive examiner, which can be essential in helping keep the younger child motivated to complete a lengthy battery of tests. If computer-assisted tests are used, it may be best to intersperse them with examiner-administered tasks.
Psychobiologic measures.
Studies of young children have included many psychosocial, environmental, and neurodevelopmental factors, but they rarely include individual difference measures of psychobiology. Recent studies show that children’s physiologic responses to different stressors are valid measures of individual psychobiology or autonomic reactivity (Alkon et al. 2003; Matthews et al. 1990). These psychobiology measures have been shown to affect how children interact with their family, adults, and peers at home and in school. In addition, children’s psychobiology is related to their physical and mental health (Boyce et al. 1998; Kagan et al. 1988; Porges et al. 1994). Autonomic reactivity is being studied as an outcome of organophosphate exposure prenatally and postnatally at the Berkeley Children’s Center. Autonomic dysregulation or an imbalance in the parasympathetic and sympathetic branches of the autonomic nervous system may be a sensitive indicator of acute and chronic pesticide exposure (Eskenazi et al. 1999). The Columbia center is using the orthostatic tilt test (Fifer et al. 1999) on a small subsample in the cohort. This is a noninvasive experimental paradigm used to identify infants with individual differences in autonomic regulation and diminished physiologic responses (as measured by heart rate) to blood pressure changes after postural adjustment.
Neurobiologically based markers of development.
Several environmental studies of children have used electrophysiologic techniques to assess the effects of neurotoxicants on central nervous system function. Visual as well as auditory evoked potentials have been examined and in many cases have been found to be sensitive to environmental chemical exposures. Another promising area for the future is the use of neuroradiologic techniques such as magnetic resonance imaging (MRI). MRI assessment of brain structure and function is beginning to be used in studies of developmental neurotoxicants. In the Cincinnati center, for example, adults exposed to high levels of lead in early childhood are being examined using volumetric, functional, and spectroscopic MRI methods. These methods provide data on exposure-related structural changes, brain activation in response to standardized verbal and visual–spatial problems, and brain biochemical processes.
The use of these methods assumes the availability of a clinical facility that can provide the needed equipment and trained personnel. For many centers, particularly those far from large metropolitan areas, this will normally not be an option.
Gene–environment interactions in neurodevelopment.
A promising area for future study is the interaction between certain genetic polymorphisms involved in neurotransmitter regulation and metabolism with environmental chemical exposures. For example, the Cincinnati center has examined the joint effects of a dopamine transporter (DAT) polymorphism and maternal prenatal smoking on childhood hyperactivity-impulsivity and inattentiveness (Kahn et al. 2003). Childhood hyperactivity–impulsivity and oppositional behaviors were associated with a DAT polymorphism, but only when the child also had in utero exposure to the products of maternal smoking. Cincinnati researchers have also identified a vitamin D receptor gene polymorphism that may be associated with greater absorption of environmental lead and thus an increased risk for neurodevelopmental toxicity (Haynes et al. 2003). The Cincinnati center is also investigating the relationship between early exposure to lead and adult criminality. The interactions of a history of early lead poisoning with four polymorphisms associated with dopamine and serotonin reception and/or transport and genes regulating monoamine oxidase A enzyme activity are being examined (Reif and Lesch 2003). Investigators at the University of California at Davis Children’s Center are investigating the interaction between specific GABAA receptor polymorphisms and autism spectrum disorder (ASD). It is well known that some of the most widely used and environmentally persistent pesticides block GABAA receptor chloride channels within the mammalian central nervous system, thereby producing hyperexcitability, tremors, and convulsions, depending on the level of exposure (Narahishi et al. 1998). Three genes that encode subunits of ionotropic GABAA receptors cluster on chromosome 15q11–13. Maternally inherited duplications of this chromosome account for approximately 1–3% of all cases of ASD (Veenstra-VanderWeele et al. 2003). The Children’s Centers at Mount Sinai, Berkeley, and the University of Washington are examining the relationship of the paraoxynase 1 (PON1) gene in modifying the associations between organophosphate pesticide exposure and neurodevelopment. Recent evidence for a relationship between genetic background and susceptibility to the neurobehavioral toxicity of neonatal exposure to thimerosal (ethylmercury) in mice (Hornig et al. 2004) further underscores the need to consider gene–environment interactions to identify susceptible populations.
Conclusions and Recommendations
Long-term studies that follow participants into adolescence and early adulthood are essential to assess the full range of neurodevelopmental consequences of exposure to environmental chemicals. The vast and rapid growth of human neurobehavioral capacities from birth through early adulthood means that the functional effects of earlier damage may not be fully expressed at any given moment in time.
Compounds targeted for investigation should be those that represent plausible neurodevelopmental hazards to fetal and postnatal central nervous system development. The compounds under investigation should represent a clear hazard to neurodevelopment on the basis of their biochemical properties. Most drugs and environmental chemicals cross the placenta with ease by virtue of their low molecular weight and many gain access to the developing central nervous system. This also requires some knowledge of the exposure levels that are likely to be encountered in the environments occupied by the samples of interest.
Population factors including ethnicity and language must be considered in the planning of study procedures for recruitment, retention, and selection of neurodevelopmental measures. Population factors can be particularly important if substantial confounding and multicultural issues are involved. These features become crucial when it comes to the choice of appropriate neurodevelopmental measures that will accurately reflect the cognitive and sensorimotor capacities of the population under investigation. Ideally, examiners familiar with the language and culture of the population under study should be employed.
The proposed work should be compatible with the physical and human resources at hand. Sites faced with populations that are widely dispersed geographically, having limited transportation, or lack of trained personnel will be restricted in the depth of neurodevelopmental assessments that can be realistically implemented.
Neuropsychologic batteries employed in these studies should include a balance of those with measures both broad and narrow in scope. Batteries too narrow in scope can easily miss deficits associated with environmental chemical exposures. On the other hand, fine-grained measures of more narrowly defined neurobehavioral domains can shed light on exposure-specific effects and brain–behavior relationships. If specific neurobehavioral domains are targeted for assessment, their selection should be guided by hypothesis-driven inference. An assessment battery that includes both broad- and narrow-band instruments will usually be optimal.
Procedures for monitoring the quality of data collection and scoring should be maintained throughout the life of the study. Monitoring of interexaminer reliability and proficiency should be practiced on a regular basis. In the event that multiple examiners are required at a single center, their training and examination techniques should be explicitly checked, monitored, and quantitatively recorded. In multicenter studies, training should be standardized across sites. This may require annual meetings and more frequent conference calls to discuss issues related to administrations and scoring of tests.
Data safety and monitoring procedures must be in place. As health care professionals as well as researchers, we have a duty to treat study participants with respect and concern. Referral of infants and children presenting with suspect neurologic or developmental signs or symptoms should be made through the primary care provider. In every case, consultation with the primary caregiver is mandatory.
The unique needs of study participants involved in neurodevelopmental assessment from birth through adulthood must be considered. Every age presents special challenges. Assessment of the neonate calls for extensive training of the examiner(s) and a highly controlled environment. Infants and toddlers present special challenges with regard to directing their attention to the demands of cognitive and sensorimotor tasks. Assessment of school-age children and adolescents provides an opportunity to examine a more differentiated picture of participants’ developmental strengths and weaknesses. Assessments of the older adolescent and young adult call for an approach that recognizes a participant’s legal rights as a research subject and sometimes requires special measures to assure subject protection and confidentiality of any data that may be obtained.
New approaches and technologies should be exploited in future studies. Computer-based assessment, psychobiologic measures, neural markers of central nervous system activity, and especially the promise of advances in human genetics have the potential to lead to substantial progress in our understanding of the effects of environmental chemicals on fetal and child development.
Final remarks.
A battery of tests that can be universally regarded as valid and reliable for evaluating the potential impact of neurotoxicants on the developing central nervous system does not exist. Specific approaches and tests have been reviewed and recommended based on the authors’ experiences and ongoing work in the Children’s Centers and other studies. The quandary faced by both seasoned investigators and researchers new to the area when planning a study was summarized by one experienced observer: “I cannot recommend any tests in this endeavor; many are appropriate. Overall strategy, a principled theoretical framework, and adequately specified domains are what count, not tests” (Bernstein 1994).
The vulnerability of the developing human central nervous system to environmental chemical compounds has been well established. The contribution of these exposures in utero or postnatally to the development of disorders of uncertain etiology such as attentional deficit hyperactivity disorder or pervasive developmental disorder/autism is not known at present. Large-scale human studies such as the National Children’s Study may provide some clues. In the final analysis, the human neurodevelopmental phenotype will be more fully revealed in large-scale studies that take account of environmental chemical influences, the social milieu, and complex human genetic characteristics that we are just beginning to understand (Hamer 2002).
This article is part of the mini-monograph “Lessons Learned from the National Institute of Environmental Health Sciences/U.S. Environmental Protection Agency Centers for Children’s Environmental Health and Disease Prevention Research for the National Children’s Study.”
This work was supported in part by the National Center for Environmental Assessment (ES011261, ES009605, ES011263, ES009600, ES011269, ES009584), the U.S. Environmental Protection Agency (R829389, R826709, R829390, R827027, R829388, R827039), and the National Children’s Study.
Table 1 Neurodevelopmental assessment in the Children’s Centers for the first 5-year funding cycle.
Assessment tool University of Illinois University of California, Berkeley Columbia University Mount Sinai Medical Center University of Cincinnati
Neonatal assessment tools
Brazelton (Brazelton and Nugent 1995) Birth Birth Birth
Ballard (Ballard et al. 1991) Birth
NICU Network Neurobehavioral Scale (Lester et al. 2004) Birth, 4 weeks
Infant and toddler assessment tools
Bayley scales (Bayley 1993) 6, 12 months 6, 12, 24 months 12, 24, 36 months 12, 24a months 12, 24 months
PLS (Zimmerman et al. 1992) 6, 12, 24 months
CBCL (Achenbach and Rescorla 2004) 12, 24 months
Denver (Frankenburg et al. 1992) 6 months
Fagan (Fagan et al. 1986) 6 months 6 months
Infant Behavior Questionnaire (Rothbart 1981) 12 months
Toddler Behavior Assessment Questionnaire (Goldsmith 1996) 24a months
Sleep questionnaire (adapted from Morrell 1999) 6, 12, 18, 23 months
Cognitive tasksb (Aguiar and Baillargeon 1999) 6, 9, 12 months
Child assessment tools
WPPSI-III (Wechsler 2002) 42c months 60 months
NEPSY (Korkman et al. 1998) 42c months
McCarthy (McCarthy 1974) 42c months
CBCL (Achenbach and Rescorla 2004) 42c months
Other child assessments
Auditory assessment Birth, 12 months
Autonomic Nervous System Assessment (Alkon et al. 2003) 6, 12, 42c months
HOME Scale (Caldwell and Bradley 1984) 6, 12, 24, 42c months 36 months 12, 24 months 12 months
Maternal intelligence
PPVT (Dunn and Dunn 1981)/TVIP (Dunn et al. 1986) 6 months Pregnancy
WASI (Wechsler 1999) 12 months
TONI-2 (Brown et al. 1990) 24 or 36 months
WAIS-III (Wechsler 1997) Enrollment (matrices and block design) 6 months (matrices)
Abbreviations: CBCL, Child Behavior Checklist; HOME, Home Observation for Measurement of the Environment; NEPSY, A Developmental Neuropsychological Assessment; PLS, Preschool Language Scale; PPVT, Peabody Picture Vocabulary Test; TONI-2, Test of Nonverbal Intelligence-2; TVIP, Test de Vocabulario en Imagenes Peabody; WAIS-III, Wechsler Adult Intelligence Scale-III; WASI, Wechsler Abbreviated Scales of Intelligence; WPPSI-III, Wechsler Preschool and Primary Scales of Intelligence-III.
a Children were brought in up to 5 years of age for the “24-month” assessment; if too old for the Bayley, they were assessed with the Batelle Developmental Inventory (Newborg et al. 1984).
b This consists of object retrieval tasks that examine working memory, executive control, and cognitive development.
c Currently underway as part of second 5-year funding cycle.
==== Refs
References
AACN 1999 American Academy of Clinical Neuropsychology policy on the use of non-doctoral-level personnel in conducting clinical neuropsychological evaluations [Editorial] Clin Psychol 13 385
AAP 1995 American Academy of Pediatrics Joint Committee on Infant Hearing 1994 position statement Pediatrics 95 152 156 7770297
Achenbach TM Rescorla LA 2004. The Achenbach System of Empirically Based Assessment (ASEBA) for ages 1.5 to 18 years. In: The Use of Psychological Testing for Treatment Planning and Outcomes Assessment (Maruish MR, ed). Mahwah, NJ:Lawrence Erlbaum Associates, 179–213.
Aguiar A Baillargeon R 1999 Perseveration and problem solving in infancy Adv Child Dev Behav 17 135 180 10884845
Alkon A Goldstein L Snider N Essex M Kupter W Boyce T 2003 Developmental and contextual influences on autonomic reactivity in young children Dev Psychobiol 42 64 78 12471637
ATSDR 2000. Toxicological Profile for Polychlorinated Biphenyls (PCBs). Atlanta, GA:Agency for Toxic Substances and Disease Registry.
Baker EL Letz RE Fidler AT Shalat S Pantamura D Lyndaom MA 1985 A computer-based neurobehavioral evaluation system for occupational and environmental epidemiology: methodology and validation studies Neurobehav Toxicolol Teratol 7 369 377
Ballard JL Khoury JC Wedig K Wang L Eilers-Walsman BL Lipp R 1991 New Ballard score expanded to include extremely premature infants J Pediatr 119 417 423 1880657
Bayley N 1993. Bayley Scales of Infant Development. 2nd ed. San Antonio, TX:Psychological Corporation.
Beckman DA Brent RL 1999. Basic principles of developmental toxicology. In: Medicine of the Fetus and Mother (Reece EA, Hobbin JC, eds). 2nd ed. Philadelphia:Lippincott-Raven, 281–288.
Bellinger DC 1995 Interpreting the literature on lead and child development: the neglected role of the “experimental system Neurotoxicol Teratol 17 201 212 7542726
Bellinger DC 2002 Perspectives on incorporating human neurobehavioral end points in risk assessments Risk Anal 22 487 498 12088228
Bellinger D Stiles K Needleman HL 1992 Low-level lead exposure, intelligence, and academic achievement: a long-term follow-up study Pediatrics 90 855 861 1437425
Berkowitz GS Wetmur JG Birman-Deych E Obel J Lapinski RH Godbold JH 2004 In utero pesticide exposure, maternal paraoxonase activity, and head circumference. Environ Health Perspect 112 388 391
Bernstein JH 1994 Assessment of developmental toxicity: neuropsychological batteries Environ Health Perspect 102 suppl 2 141 144 7523103
Boyce W Frank E Jensen P Kessler R Nelson C Steinberg L 1998 Social context in developmental psychopathology: recommendations for future research from the MacArthur Network on Psychopathology and Development Dev Psychopathol 10 143 164 9635219
Brandt J van Gorp W 1999 American Academy of Clinical Neuropsychology policy on the use of doctoral level personnel in conducting clinical neuropsychological evaluations [Editorial] J Clin Exp Neuropsychol 21 1
Brazelton T 1984: Neonatal Behavioral Assessment Scale. Philadelphia:JB Lippincott Co.
Brazelton TB Nugent JK 1995: Neonatal Behavioral Assessment Scale. 3rd ed. London:McKeith Press.
Bressler J Kim KA Chakraborti T Goldstein G 1999 Molecular mechanisms of lead neurotoxicity Neurochem Res 24 595 600 10227691
Brown L Sherbenou RJ Johnsen SK 1990. Test of Nonverbal Intelligence. 2nd ed. Austin, TX:PRO-ED.
Brown Medical School Infant Development Center 2005. Training and Education, NNNS Training Program. Available: http://www.infantdevelopment.org [accessed 21 March 2005].
Burns JM Baghurst PA Sawyer MG McMichael AJ Tong SL 1999 Lifetime low-level exposure to environmental lead and children’s emotional and behavioral development at ages 11–13 years. The Port Pirie Cohort Study Am J Epidemiol 149 740 749 10206624
Caldwell BM Bradley RR 1984. Home Observation for Measurement of the Environment. Little Rock, AK:University of Arkansas.
Chandlee J Heathfield LT Salganik M Damokosh A Radcliffe J 2002 Are we consistent in administering and scoring the Bayley Scales of Infant Development-II? J Psychoeduc Assess 20 183 200
Cheek AO Kow K Chen J McLachlan JA 1999 Potential mechanisms of thyroid disruption in humans: interaction of organochlorine compounds with thyroid receptor, transthyretin, and thyroid-binding globulin Environ Health Perspect 107 273 278 10090705
Dahl R White RF Weihe P Sorensen N Letz R Hudnell K 1996 Feasibility and validity of three computer-assisted neurobehavioral tests in 7-year-old children Neurotoxicol Teratol 18 413 420 8866532
Darvill T Lonky E Reihman J Stewart P Pagano J 2000 Prenatal exposure to PCBs and infant performance on the Fagan Test of Infant Intelligence Neurotoxicology 21 1029 1038 11233749
Dietrich KN 2000. Environmental neurotoxicants and psychological development. In: Pediatric Neuropsychology: Research, Theory and Practice (Yeates KO, Ris MD, Taylor HG, eds). New York:Guilford Press, 206–234.
Dietrich KN Bellinger D 1994. The assessment of neurobehavioral development in studies of the effects of pre-natal exposure to toxicants. In: Prenatal Exposure to Toxicants: Developmental Consequences (Needleman HL, Bellinger D, eds). Baltimore, MD:Johns Hopkins University Press, 57–85.
Dietrich KN Krafft KM Bornschein RL Hammond PB Berger O Succop PA 1987 Low-level fetal lead exposure effect on neurobehavioral development in early infancy Pediatrics 80 721 730 2444921
Dietrich KN Succop PA Berger O Hammond PB Bornschein RL 1991 Lead exposure and the cognitive development of urban preschool children: the Cincinnati Lead Study cohort at age 4 years Neurotoxicol Teratol 13 203 211 1710765
Dietrich KN Ware JH Salganik M Radcliffe J Rogan WJ Rhoads GG 2004 Effect of chelation therapy on the neuropsychological and behavioral development of lead-exposed children after school entry Pediatrics 114 19 26 15231903
Dunn L Dunn L 1981. Peabody Picture Vocabulary Test. Rev ed. Circle Pines, MN:American Guidance Service.
Dunn L Padilla E Lugo D Dunn L 1986. Test de Vocabulario en Imagenese Peabody. Circle Pines, MN:American Guidance Service.
Eskenazi B Bradman A Castorina R 1999 Exposures of children to organophosphate pesticides and their potential adverse health effects Environ Health Perspect 107 suppl 3 409 419 10346990
Eskenazi B Harley K Bradman A Weltzien E Jewell NP Barr DB 2004 Association of in utero organopophosphate pesticide exposure and fetal growth and length of gestation in an agricultural population Environ Health Perspect 112 1116 1124 15238287
Fagan JF Singer LT Montie JE Shepherd PA 1986 Selective screening device for the early detection of normal or delayed cognitive development in infants at risk for later mental retardation Pediatrics 78 1021 1026 2431380
Fein GG Jacobson JL Jacobson SW Schwartz PM Dowler JK 1984 Prenatal exposure to polychlorinated biphenyls: effects on birth size and gestational age J Pediatr 105 315 320 6431068
Fifer WP Greene M Hurtado A Myers MM 1999 Cardio-respiratory responses to bidirectional tilts in infants Early Hum Dev 55 265 279 10463790
Frankenburg WK Dodds J Archer P Shapiro P Bresnick B 1992 The Denver II: a major revision and restandardization of the Denver Developmental Screening Test Pediatrics 89 91 97 1370185
Gauger KJ Kato Y Haraguchi K Lehmler HJ Robertson LW Bansal R 2004 Polychlorinated biphenyls (PCBs) exert thyroid hormone-like effects in the fetal rat brain but do not bind to thyroid hormone receptors Environ Health Perspect 112 516 523 15064154
Goldsmith HH 1996 Studying temperament via construction of the Toddler Behavior Assessment Questionnaire Child Dev 67 218 235 8605830
Gregory R Lehman R Mohan P 1976. Intelligence test results for children with and without undue lead absorption. In: Shoshone Lead Health Project (Wegner G, ed). Boise, ID:Idaho Department of Health and Welfare, 120–149.
Guillette EA Meza MM Aquilar MG Soto AD Garcia ID 1998 An anthropological approach to the evaluation of preschool children exposed to pesticides in Mexico Environ Health Perspect 106 347 353 9618351
Hamer D 2002 Rethinking behavior genetics Science 298 71 72 12364769
Haynes EN Kalkwarf HJ Hornung R Wenstrup R Dietrich K Lanphear BP 2003 Vitamin D receptor Fok1 polymorphism and blood lead concentration in children Environ Health Perspect 111 1665 1669 14527848
Hoon AH 1996. Visual impairments in children. In: Developmental Disabilities in Infancy and Childhood (Caupute AJ, Accardo PJ, eds). Baltimore, MD:Brookes Publishing, 461–478.
Hornig M Chain D Lipkin WI 2004 Neurotoxic effects of postnatal thimerosal are mouse strain dependent Mol Psychiatry 9 833 845 15184908
IPCS (International Programme for Chemical Safety) 1995. Environmental Health Criteria 165. Inorganic Lead. Geneva:World Health Organization.
Jacobson J Jacobson S Schwartz O Fein G Dowler J 1984 Prenatal exposure to an environmental toxin: a test of the multiple effects model Dev Psychol 20 523 532
Jacobson JL 2001 Contending with contradictory data in a risk assessment context: the case of MeHg Neurotoxicology 22 667 675 11770888
Jacobson JL Jacobson SW 1996 Intellectual impairment in children exposed to polychlorinated biphenyls in utero N Engl J Med 335 783 789 8703183
Jacobson JL Jacobson SW 2003 Prenatal exposure to poly-chlorinated biphenyls and attention at school age J Pediatr 143 6 780 788 14657828
Jacobson JL Jacobson SW Padgett RJ Brumitt GA Billings RL 1992 Effects of prenatal PCB exposure on cognitive processing efficiency and sustained attention Dev Psychol 28 297 306
Jacobson SW 1998 Specificity of neurobehavioral outcomes associated with prenatal alcohol exposure Alcohol Clin Exp Res 22 313 320 9581634
Kagan J Reznick J Snidman N 1988 Biological basis of childhood shyness Science 240 167 171 3353713
Kahn RS Khoury J Nichols WC Lanphear BP 2003 Role of dopamine transporter genotype and maternal prenatal smoking in childhood hyperactive-impulsive, inattentive, and oppositional behaviors J Pediatr 143 104 110 12915833
Kallen B 1988. Epidemiology of Human Reproduction. Boca Raton, FL:CRC Press.
Kerper LE Ballatori N Clarkson TW 1992 MeHg transport across the blood-brain by an amino acid carrier Am J Physiol 262 761 765
Kerper LE Hinkle PM 1997 Lead uptake in brain capillary endothelial cells: activation by calcium store depletion Toxicol Appl Pharmacol 146 127 133 9299604
Korkman M Kirk U Kemp S 1998. NEPSY—A Developmental Neuropsychological Assessment. San Antonio, TX:Psychological Corporation.
Krasnegor NA Otto DA Bernstein JH Burke R Chappell W Eckerman DA 1994 Neurobehavioral test strategies for environmental exposures in pediatric populations Neurotoxicol Teratol 16 499 509 7845333
Lester BM Tronick EZ Brazelton TB 2004 The Neonatal Intensive Care Unit Network Neurobehavioral Scale Pediatrics 113 641 667 14993524
Lipkin PH 1991. Epidemiology of developmental disabilities. In: Developmental Disabilities in Infancy and Childhood (Capute AJ, Accardo PH, eds). Vol 1. Baltimore, MD:Brookes Publishing, 43–67.
Matthews K Woodall K Stoney C 1990 Changes in and stability of cardiovascular responses to behavioral stress: results from a four-year study of children Child Dev 61 1134 1144 2209183
McCarthy D 1974. McCarthy Scales of Children’s Abilities. San Antonio, TX:Psychological Corporation.
McMichael AJ Vimpani GV Robertson EF Baghurst PA Clark PD 1986 The Port Pirie cohort study: maternal blood lead and pregnancy outcome J Epidemiol Community Health 41 18 25 3711766
Morrell J 1999 The Infant Sleep Questionnaire: a new tool to assess infant sleep problems for clinical and research purposes Child Psychol Psychiatry 4 20 26
Myers GJ Davidson PW 2000 Does MeHg have a role in causing developmental disabilities in children? Environ Health Perspect 108 suppl 3 413 420 10852838
Narahishi T Ginsburg KS Nagata K Song JH Tatebayashi H 1998 Ion channels as targets for insecticides Neurotoxicology 19 581 590 9745915
National Research Council 2000. Toxicological Effects of MeHg. Washington, DC:National Academic Press.
Needleman HL Bellinger DC 1986. Type II fallacies in the study of childhood exposure to lead at low dose: a critical and quantitative review. In: Lead Exposure and Child Development: An International Assessment (Smith MA, Grant LD, Sors AI, eds). Boston:Kluwer Academic Publishers, 293–304.
Needleman HL Schell A Bellinger D Leviton A Allred EN 1990 The long-term effects of exposure to low doses of lead in childhood. An 11-year follow-up report N Engl J Med 322 83 88 2294437
Newborg J Stock JR Wnek L 1984. Batelle Developmental Inventory. Itasca, IL:Riverside Publishing.
NIDA 2000. Early Childhood Neurobehavioral Assessment for the Differential Diagnosis of Fetal Alcohol Syndrome and Alcohol-Related Neurobehavioral Disorder. Bethesda, MD:National Institute on Drug Abuse, the Interagency Coordinating Committee on Fetal Alcohol Syndrome and National Institutes of Health.
Porges S Doussard-Roosevelt JA Portales A Suess P 1994 Cardiac vagal tone: stability and relation to difficultness in infants and 3-year olds Dev Psychobiol 27 289 300 7926281
Raven JC 1998. Raven’s Progressive Matrices. San Antonio, TX:Psychological Corporation.
Rees DC Francis EZ Kimmel CA 1990 Scientific and regulatory issues relevant to assessing risk for developmental neurotoxicity: an overview Neurotoxicol Teratol 12 175 181 2196418
Reif A Lesch KP 2003 Toward a molecular architecture of personality Behav Brain Res 139 1 20 12642172
Rhodes MC Seidler FJ Qiao D Tate CA Cousins MM Slotkin TA 2004 Does pharmacotherapy for preterm labor sensitize the developing brain to environmental neurotoxicants? Cellular and synaptic effects of sequential exposure to terbutaline and chlorpyrifos in neonatal rats Toxicol Appl Pharmacol 195 203 217 14998686
Rodier PM 1994 Vulnerable periods and processes during central nervous system development Environ Health Perspect 102 suppl 2 121 124 7925182
Rodier PM 2004 Environmental causes of central nervous system maldevelopment Pediatrics 113 1076 1083 15060202
Rogan WJ Dietrich KN Ware JH Dockery DW Salganik M Radcliffe J 2001 The effect of chelation therapy with succimer on neuropsychological development in children exposed to lead N Engl J Med 344 1421 1426 11346806
Rohlman DS Bailey SR Anger WK McCauley L 2001 Assessment of neurobehavioral function with computerized tests in a population of Hispanic adolescents working in agriculture Environ Res 85 14 24 11161647
Rohlman DS Gimenes L Eckerman DA Kang SK Farahat FM Anger WK 2003 Development of the Behavioral Assessment and Research System (BARS) to detect and characterize neurotoxicity in humans Neurotoxicology 24 523 531 12900065
Roizen NJ 1996. Hearing loss. In: Developmental Disabilities in Infancy and Childhood (Capute AJ, Accardo PJ, eds). Baltimore, MD:Brookes Publishing, 479–496.
Rothbart MK 1981 Measurement of temperament in infancy Child Dev 52 569 578
Rozman KK Klaassen CD 1996. Absorption, distribution, and excretion of toxicants. In: Toxicology: The Basic Science of Poisons (Klaassen CD, Amdur MO, Doull J, eds). 5th ed. New York:McGraw-Hill, 91–109.
Schantz SL 1996 Developmental neurotoxicity of PCBs in humans: what do we know and where do we go from here? Neurotoxicol Teratol 18 217 228 8725628
Seegal RF 2000 The neurotoxicological consequences of developmental exposure to PCBs Toxicol Sci 57 1 3 10966504
Slikker W Miller RK 1994. Placental metabolism and transfer: role in developmental toxicology. In: Developmental Toxicology (Kimmel CA, Buelke-Sam J, eds). 2nd ed. New York:Raven Press, 245–283.
Slotkin TA 1999 Developmental cholinotoxicants: nicotine and chlorpyrifos Environ Health Perspect 107 71 80 10229709
Stewart P Reihman J Lonky E Darvill T Pagano J 2000 Prenatal PCB exposure and Neonatal Behavioral Assessment Scale (NBAS) performance Neurotoxiol Teratol 22 21 29
TERA 1999. Comparative Dietary Risks: Balancing Risks and Benefits of Fish Consumption. Cincinnati, OH:Toxicology Excellence for Risk Assessment.
Tong S Baghurst P McMichael A Sawyer M Mudge J 1996 Lifetime exposure to environmental lead and children’s intelligence at 11–13 years: the Port Pirie cohort study BMJ 312 1569 1575 8664666
Veenstra-VanderWeele J Cook EH Lombroso PH 2003 Genetics of childhood disorders: 46. Autism, part 5: genetics of autism J Am Acad Child Psychiatry 42 116 118
Wechsler D 1997. Wechsler Adult Intelligence Scale. 3rd ed. San Antonio, TX:Psychological Corporation.
Wechsler D 1999. Wechsler Abbreviated Scale of Intelligence. San Antonio, TX:Psychological Corporation.
Wechsler D 2002. Wechsler Preschool and Primary Scales of Intelligence. 3rd ed. San Antonio, TX:Psychological Corporation.
Weiss B 1997 Pesticides as a source of developmental disabilities Ment Retard Dev Disabil 3 244 256
Weiss B Amler S Amler RW 2004 Pesticides Pediatrics 113 1030 1036 15060196
White RF Gerr F Cohen RF Green R Lezak MD Lybarger J Mack J 1994 Criteria for progressive modification of neurobehavioral batteries Neurotoxicol Teratol 16 511 524 7845334
Whyatt RM Rauh V Barr DB Camann DE Andrews HF Garfinkel R 2004 : Prenatal insecticide exposure and birth weight and length among an urban minority cohort Environ Health Perspect 112 1125 1132 15238288
Wilson JG 1977. Current status of teratology—general principles and mechanisms derived from animal studies. In: Handbook of Teratology (Wilson JG, Fraser RC, eds). Vol 1. New York:Plenum, 47–74.
Zimmerman I Steiner V Pond R 1992. Preschool Language Scale. 3rd ed. San Antonio, TX:Psychological Corporation.
|
16203260
|
PMC1281293
|
CC0
|
2021-01-04 23:41:28
|
no
|
Environ Health Perspect. 2005 Oct 24; 113(10):1437-1446
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.7672
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences 10.1289/ehp.7673ehp0113-00144716203261ResearchMini-MonographAir Pollution Exposure Assessment for Epidemiologic Studies of Pregnant Women and Children: Lessons Learned from the Centers for Children’s Environmental Health and Disease Prevention Research Gilliland Frank 1Avol Ed 1Kinney Patrick 2Jerrett Michael 1Dvonch Timothy 3Lurmann Frederick 4Buckley Timothy 5Breysse Patrick 5Keeler Gerald 3de Villiers Tracy 1McConnell Rob 11 Department of Preventive Medicine, Keck School of Medicine, University of Southern California, Los Angeles, California, USA2 Mailman School of Public Health, Columbia University, New York, New York, USA3 School of Public Health, University of Michigan, Ann Arbor, Michigan, USA4 Sonoma Technology, Inc., Petaluma, California, USA5 Bloomberg School of Public Health, Johns Hopkins University, Baltimore, Maryland, USAAddress correspondence to F. Gilliland, Department of Preventive Medicine, USC Keck School of Medicine, 1540 Alcazar St., CHP 236, Los Angeles, CA 90033 USA. Telephone: (323) 442-1309. Fax: (323) 442-3272. E-mail:
[email protected] authors declare they have no competing financial interests.
10 2005 24 6 2005 113 10 1447 1454 12 10 2004 24 3 2005 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright. The National Children’s Study is considering a wide spectrum of airborne pollutants that are hypothesized to potentially influence pregnancy outcomes, neurodevelopment, asthma, atopy, immune development, obesity, and pubertal development. In this article we summarize six applicable exposure assessment lessons learned from the Centers for Children’s Environmental Health and Disease Prevention Research that may enhance the National Children’s Study: a) Selecting individual study subjects with a wide range of pollution exposure profiles maximizes spatial-scale exposure contrasts for key pollutants of study interest. b) In studies with large sample sizes, long duration, and diverse outcomes and exposures, exposure assessment efforts should rely on modeling to provide estimates for the entire cohort, supported by subject-derived questionnaire data. c) Assessment of some exposures of interest requires individual measurements of exposures using snapshots of personal and microenvironmental exposures over short periods and/or in selected microenvironments. d) Understanding issues of spatial–temporal correlations of air pollutants, the surrogacy of specific pollutants for components of the complex mixture, and the exposure misclassification inherent in exposure estimates is critical in analysis and interpretation. e) “Usual” temporal, spatial, and physical patterns of activity can be used as modifiers of the exposure/outcome relationships. f) Biomarkers of exposure are useful for evaluation of specific exposures that have multiple routes of exposure. If these lessons are applied, the National Children’s Study offers a unique opportunity to assess the adverse effects of air pollution on interrelated health outcomes during the critical early life period.
air pollutionairborneambientCenters for Children’s Environmental Health and Disease Prevention ResearchChildren’s Centerscohort studydirect measurementexposure assessmentmodelingNational Children’s Studypersonal measurement
==== Body
A major study design challenge for the National Children’s Study will be to maximize and characterize exposure contrasts in its cohort of 100,000 pregnant women residing in multiple locations across the United States, thereby enhancing the power to estimate exposure–response relationships from childhood into adulthood. Multiple outcomes are of interest, including pregnancy outcomes, neurodevelopment, asthma, obesity, and pubertal development. Exposures to a wide spectrum of environmental pollutants are being considered for investigation in the study, including air pollutants of indoor and outdoor origin (National Children’s Study 2004).
Given the pollutants and health endpoints currently under consideration, exposure assessment for the variable periods during pregnancy, infancy, and childhood will be needed. For asthma-related outcomes, daily, monthly, yearly, and multiyear exposure metrics with varying time integration periods may be required. For pregnancy outcomes, monthly estimates as well as estimates for critical periods may be needed. For neurodevelopment, monthly, yearly, and multiyear metrics may be most relevant. For these and other outcomes, time-integrated average levels may capture the effects of chronic exposure during specific periods, but more discrete and intense sampling frequency or duration may be needed to better assess specific exposure–response relationships.
The purpose of this article is to summarize exposure assessment lessons learned in the Centers for Children’s Environmental Health and Disease Prevention Research (hereafter Children’s Centers) for air pollutants and health outcomes of National Children’s Study interest. Exposures to allergens and bioaerosols are considered elsewhere in this mini-monograph. Many of the Children’s Centers have active research programs involving the assessment of air pollution in epidemiologic studies (Table 1). On the basis of experience of investigators from these centers, we provide recommendations for air pollution exposure assessment consideration in the study design, population selection, exposure data collection, analysis, and interpretation of findings of the National Children’s Study.
Lessons Learned in Air Pollution Exposure Assessment
An essential design element of environmental epidemiologic studies is the a priori consideration of exposure assessment to ensure that the study exposure range will maximize the ability to evaluate key exposure–response relationships (Navidi et al. 1994, 1999). Study population selection and exposure assessment design are linked. Successful selections require consideration of the developmental time frames of interest and the biologic outcome mechanisms, in addition to understanding the spatial characteristics of airborne indoor and ambient exposures. One potentially successful design strategy is to maximize the number of contrasting pollution profiles among study subjects by using a quasi-factorial approach to select populations distributed over geographic regions with different pollution profiles (and/or including homes with different indoor sources and proximity to specific sources) (Gauderman et al. 2000).
The National Children’s Study proposes to investigate the relationships between patterns and histories of exposure during critical periods and the development of disease in later life. This creates an inherent tension because exposure assessment in large cohort studies requires a compromise between the optimal information obtained from individual measurements and feasibility constraints related to sampling methods, respondent burden, and cost. Feasibility considerations likely dictate that direct measurements will be limited to subsets of subjects monitored for short time periods (“snapshots”) in selected microenvironments, whereas exposure metrics used in chronic effects analyses for the entire cohort will be time-integrated over extended periods (days to months). The proposed size and duration of the National Children’s Study will require the use of modeling to estimate time-integrated exposures for the entire cohort even when direct measurements using snapshots of exposure are available for subsets of the cohort.
Several modeling frameworks are applicable to the National Children’s Study. Basic approaches rely on using questionnaire responses as a surrogate for exposure and on assigning exposures based on air pollutants measured at a central monitor. The latter approach has been successfully employed to detect significant health effects (Dockery et al. 1993; Gauderman et al. 2002; Pope et al. 2002; Ritz et al. 2000; Samet et al. 2000). More refined approaches allow for estimation within communities using dispersion models and information on transport, land use, and meteorology (Brauer et al. 2002; English et al. 1999; Finkelstein et al. 2003; Hoek et al. 2002; Nafstad et al. 2004). Considerations for modeled exposures include the availability of high-quality input data on the appropriate geographic scale and the need for validation and calibration studies to enable exposure uncertainty assignments. There are important limitations of modeling air pollution exposures (Sarnat et al. 2001). Studies indicate that for some pollutants, such as particulate matter (PM) and volatile organic compounds, indoor sources can predominate (Sax et al. 2004; Tonne et al. 2004; Wallace et al. 2004). Any strategy that relies on ambient modeling should also attempt to assess indoor exposures in subsamples of homes and thorough questionnaire or inspection data that examine important potential sources such as smoking habits or the presence of an attached garage. This is especially needed for air pollutants for which indoor sources are often the most significant contributors (Payne-Sturges et al. 2004). Understanding and assessing the role of exposure measurement error in health effects assessment are central issues for the design and implementation of health effect cohort studies (Jerrett and Finkelstein 2005).
Finally, interpretation of National Children’s Study findings will require information about specific pollutant surrogates because of the complex mixture of covarying pollutants in respirable air (Manchester-Neesvig et al. 2003). Pollutants covary because they are emitted from common sources or are produced by common atmospheric chemistry and meteorologic processes. Identification of source contributions within specific geographic regions may enhance interpretability of single pollutant associations with health outcomes (Laden et al. 2000; Samet et al. 2000).
In the following sections, we provide recommendations and issues that may need to be considered in implementing them. These are supported by some specific examples from the Children’s Centers listed in Table 1.
Specific Recommendations
National Children’s Study subject selection.
Study populations should be selected to maximize spatial exposure contrasts for the pollutants of interest. Because multiple pollutants are of interest for the National Children’s Study, priorities must be established to allow identification of individuals with a wide range of exposure profiles for those key pollutants of study interest.
Issues to consider include spatial scale variations of pollutants, in order to select a study population that maximizes exposure contrasts (Table 2). Table 2 identifies the spatial scales of variability for ambient pollutants to consider in the study design for the National Children’s Study. The scales are categorized as regional (100–1,000 km), urban (4–50 km), neighborhood (50 m to 4 km), and household (≤50 m, including outdoor and indoor microenvironments). For some exposures, contrast in exposure can be achieved by considering indoor sources and behavior (e.g., smoking vs. nonsmoking homes), if indoor-source pollutant health effects are of interest. For PM, the spatial scale variability of importance depends on the constituents of interest. For example, elemental carbon (EC) from ambient primary combustion processes varies on urban and neighborhood scales. Indoor sources from combustion also contribute to personal EC exposure (LaRosa et al. 2002). In contrast, particulate sulfates typically vary on a regional scale. To maximize exposure gradients to EC, subjects would need to be selected on a neighborhood scale, such as based on distance to busy roadways. Sulfates’ regional nature would be better reflected in a subject selection scheme involving different regions of the United States.
To select subjects based on exposure contrasts for ambient pollutants (e.g., ozone, sulfate), exposure data on geographic variation in levels and spatial gradients over time are needed. For criteria pollutants, existing data are available from a national network of monitoring stations. Data for many other pollutants of biologic interest may be sparse or nonexistent (e.g., EC and air toxics). In addition, for other pollutants with both indoor and outdoor sources (e.g., PM mass, nitrogen oxides, volatile organic compounds), much of the variability in exposure is driven by indoor source activity and/or very proximate local sources (e.g., traffic). For these pollutants, levels may need to be measured or modeled with the appropriate spatial and temporal resolution in pilot studies to ascertain the appropriate spatial, temporal, and behavioral determinants. In addition to variable pollutant source strengths, subject-specific temporal–spatial–physical patterns of activity may meaningfully affect both within and between-group exposure assignments. Capturing this variability in applicably useful ways for large study population studies is challenging and often a multi-faceted approach using self-administered questionnaires, walk-through surveys, instrument deployments, and sentinel monitoring.
Because several pollutants of interest for the National Children’s Study are regional in nature, subject selection from areas with contrasting pollution profiles is likely to be most informative. The national scope of the National Children’s Study provides the opportunity to maximize the number of study profiles. For example, the constituents of PM <2.5 μm in diameter (PM2.5) within a region are highly correlated, but between regions the correlations may be lower. PM2.5 sulfate is higher in the eastern United States and lower in the western United States, whereas PM2.5 nitrate is lower in the eastern United States and higher in the western United States. Therefore, the comparable effect of these PM2.5 constituents may be separable by study design. Replication of pollution profiles in different regions is also important to allow for effects of geographic variables such as weather and other confounding variables to be controlled in the analyses (Jerrett et al. 2003a, 2003b; Krewski et al. 2000; Peters 1997; Peters et al. 1999a). Exposures within homes with common sources are also highly correlated and may be separated by design.
An example of the integration of these approaches is the Southern California Children’s Health Study (CHS), a study performed by investigators in the University of Southern California (USC)/University of California at Los Angeles Children’s Environmental Health Center. The USC CHS is a multiyear cohort study of several thousand southern California school children (Berhane et al. 2004; Kunzli et al. 2003; Peters 1997). The primary USC CHS research question is whether ambient air pollution causes chronic adverse respiratory health effects during childhood and adolescent growth and development. Almost 12,000 children from schools in 13 southern California communities have been recruited into five cohorts since the study began in 1993.
Communities were selected to maximize differences in outdoor air pollutant concentrations. To distinguish the effects of different pollutants, communities were selected to minimize the spatial correlations between three priority study pollutants [O3, nitrogen dioxide, and PM < 10 μm in diameter (PM10)]. However, the full quasi-factorial design could not be fulfilled because all the potential pollution profiles do not occur in nature. Specific community selections were based on historical air pollution levels for several years before study inception, exposure patterns, and census demographic data. Because of differences in the number of locations at which pollutants were measured and the frequency and type of measurements made, data available for selecting communities were more reliable for O3 than for PM10, and more reliable for PM10 than for NO2. Demographically heterogeneous communities were selected because they would be more likely to exhibit overlapping distributions of confounding risk factors and would allow adjustments for confounding in the analysis. Replication of exposure profiles was employed to improve the chance of including demographically comparable communities and to allow estimation of residual variance within pollution profiles. Additional details have been described previously (Berhane et al. 2004; Peters et al. 1999a, 1999b). This design resulted in contrasting exposure profiles for O3 and a package of correlated pollutants (PM10, PM2.5, and NO2) primarily of mobile source origin. This approach can be extended to other pollutants, such as ultrafine particles whose concentrations may also vary on a localized scale of ≤50 m. Selecting subjects within communities based on the distance between the home and the nearest busy roadway or other traffic density metric may maximize the exposure contrasts of ultrafines within the profiles of other pollutants such as O3.
Other potential valuable exposure sampling designs might consider “matrix” sampling approaches, which would draw on subsets of subjects for specific substudies or specialty projects. In the larger perspective however, maximizing differences in community exposure profiles can provide a rich population base from which to develop and inform multiple studies seeking to optimize the National Children’s Study effort.
Exposure metrics.
Because of the large size, long duration, and diversity of outcomes and exposures of interest in the proposed National Children’s Study, the exposure assessment effort should rely on modeling to provide estimates for the entire cohort, supported by subject-derived questionnaire data. Necessary survey information on temporal–spatial–physical patterns of activity and household characteristics can be collected for the entire cohort, and targeted exposure substudies can be performed in selected subsamples of study subjects.
Issues to consider include modeling for large-scale investigations over long periods (e.g., the National Children’s Study), which is currently the only feasible approach for assigning exposure estimates for the entire cohort. This is especially true for ambient air pollutants that display significant spatial variation on urban, neighborhood, or household spatial scales.
A variety of exposure assessment modeling approaches are available, including proximity-based, geostatistical, land-use regression (LUR), dispersion, integrated meteorologic emission, and hybrid approaches involving personal sampling in combination with one or more of the above methods (Jerrett et al. 2004). Each model varies by data input requirements, software/hardware, technical expertise, and resulting accuracy and extrapolation potential.
Modeled estimates can be refined using targeted substudies designed to measure levels at geographic locations over time on the scale of spatial and temporal variation of the pollutants under study. The time resolution of the exposure estimate needs to be appropriately matched to outcomes to capture effects of frequency, magnitude, and duration of peak or episodic exposure events that may have effects during windows of vulnerability. Long-term average exposures, including average peak levels or hours above threshold levels, are likely more important for relationships with chronic disease, but this assumption needs to be evaluated for specific agents and outcomes of focus in the National Children’s Study.
Data availability and quality for model input are critically important. Central-site monitoring data can be used to assign exposure for outdoor environments, but the utility of this assignment will depend on the relative variability of the pollutant across the sampling area of interest (intra- vs. intercommunity variability issues). Estimates of indoor concentrations require individual information on home operating conditions, home source profiles and activity, factors influencing the penetration of outdoor pollutants and/or the dilution of pollutants of indoor origin (LaRosa et al. 2002; Navidi et al. 1999). Information about temporal, spatial, and physical activity patterns are also important but are likely to have insufficient time resolution over the period of study interest. Broader categories of “usual” patterns of activity, household operation, and susceptibility factors can be considered as modifying factors for the exposure–response relationship using available central-site monitoring data (Gauderman et al. 2000; Janssen et al. 2002).
An existing national system of central site monitors collects continuous data on criteria air pollutants and more limited data on hazardous air pollutants [U.S. Environmental Protection Agency (EPA) 2004]. It is possible to add additional instruments to monitoring sites to measure additional pollutants or speciate PM at reasonable cost. However, the use of central-site monitoring data for epidemiology studies requires a quality assurance activity beyond that which is used for regulatory activities as well as methods to address missing data issues. The Health Effects Institute recently funded a study to compile existing estimates of air toxics into a coherent national database. When available, these data may contribute to the National Children’s Study, and selection of the sampling sites for the National Children’s Study should take into account the location of existing and upcoming monitoring data. No similar monitoring network exists to assess exposure from indoor sources, which may need to rely on questionnaire information and substudies across regions.
Modeling of pollutants with large intra-community variation requires additional community measurements. Substudies can be designed to exploit obtainable information for modeling study subject exposures (Jerrett et al. 2005). These additional microenvironmental measurements can be used for fitting models to better estimate exposure, for model validation, and for assessment of errors in exposure assignments. Calibration studies using repeated personal monitoring may be designed and conducted to validate the exposure estimates and correct for exposure error in the analysis (Berhane et al. 2004; Fraser and Stram 2001; Mallick et al. 2002; Stram et al. 1995).
An illustration of these approaches may be seen in the USC CHS. The USC CHS framework employed a hierarchical approach for estimating exposure, ranging from the coarsest spatial estimates based on community pollutant levels measured at a single central monitor per community, to the finest spatial-scale estimates based on integrated models for individual exposure assessment. The framework involved the following pollutant measurement and modeling levels: a) continuous monitoring of O3, NO2, and PM10, and of PM2.5 mass and composition on a time-integrated 14-day basis, at a central monitoring station in each community; b) measurement of selected pollutants at multiple locations within each community; and c) adjustment of the central site monitor to the levels around children’s homes and schools based on a limited number of field measurements. This framework is augmented by a) modeling of vehicle emissions using geostatistical methods and spatial dispersion models, b) estimating outdoor pollutant concentrations at schools and homes for the entire study population using spatial statistical models in a hybrid microenvironmental approach, and c) modeling individual exposure estimates for the entire study population using unified modeling methods that integrated information with different spatial and temporal resolutions. These unified methods include community monitored pollutant levels, studies of indoor and outdoor levels in homes and schools; step counters; questionnaire-based data on time–activity patterns including commuting patterns, traffic patterns, and housing characteristics; and appropriate accounting of uncertainty in the exposure estimates.
The USC CHS developed a microenvironmental exposure model that, in principle, can provide estimates of exposures to pollutants of ambient origin in five microenvironments. These include residential outdoors, residential indoors, school outdoors, school indoors, and inside vehicles. The exposure model uses individual-level time–activity and housing survey data, residence and school-level traffic model estimates, and community-level air quality measurement data and regional transport factors to estimate short-term and long-term individual exposures. The model estimates show the largest amount of within-community variations in individual exposures of any of the models; however, validating these types of models is difficult and resource intensive (Peters 1997).
Newer modeling strategies such as LUR models are promising. LUR employs the pollutant of interest as the dependent variable and proximate land use, traffic, and physical environmental variables as independent predictors. The methodology seeks to predict pollution concentrations at a given site based on surrounding land use and traffic characteristics. The incorporation of land use variables into the interpolation algorithm detects small-area variations in air pollution more effectively than do standard methods of interpolation (i.e., kriging) (Briggs et al. 1997, 2000; Lebret et al. 2000). These methods are promising for the National Children’s Study because they can be extrapolated, based on land use coverage, without need for extensive monitoring in each location. Most major urban centers maintain land use information, and the U.S. Census has much of the information needed on population density and employment structures. The National Children’s Study could support the monitoring needed to calibrate LUR models that are regionally representative of broad land use and emission patterns. Derived coefficients could then be applied to other places within the region without need for extensive monitoring.
Use of limited substudies for exposure refinement.
Assessment of some exposures of interest will require individual measurements of exposures using snapshots of personal and microenvironmental exposures over short periods and/or in selected microenvironments.
Issues to consider include the large number of interrelated factors that are important in designing exposure substudies. These include the substudy’s purpose, the population sample to include, whether personal or microenvironmental samples should be collected, the respondent burden, study feasibility, sample collection and analytic costs, temporal variation of exposure, subject activity patterns, household operation by residents, and uses in model validation and calibration.
These elements are nicely illustrated in the Columbia Pregnancy Cohort Study (PCS), a study performed by the Columbia University Center for Children’s Environmental Health, which has focused on the effects of pre- and postnatal exposures to air pollution on birth outcomes and neurodevelopmental and respiratory health outcomes in childhood via through recruitment and follow-up of pregnant women and their offspring (Miller et al. 2001; Perera et al. 2003, 2004a; Tonne et al. 2004; Whyatt et al. 2003). In the Columbia PCS, direct air pollution exposure assessment begins in the third trimester of pregnancy with collection of a 48-hr personal sample of PM2.5 and vapors for each pregnant woman. These samples are analyzed for polycyclic aromatic hydrocarbon (PAH) and pesticide concentrations (i.e., a “snapshot” measurement representing “usual” exposure). In a validation substudy, the investigators also collected sequential 2-week integrated indoor samples, analyzed for the same variables as above, for the entire third trimester (preferred over the personal snapshot as an exposure surrogate of third-trimester exposures, but obviously more intensive laborwise, cost-wise, and subjectwise). A home dust sample was also collected during the third trimester from subjects and analyzed for standard allergens relevant to maternal exposures and possible prenatal sensitization, based on evidence emerging from the Columbia PCS (Miller et al. 2001).
Another time interval of study exposure interest was the first 2 years of life, when infants/toddlers spend substantial amounts of time in the home; this may be a critical exposure window for development of allergy and asthma. Columbia PCS homes were visited when the child reached 1 year of age, and a dust sample was collected for allergen analysis. Additional sampling was performed in a subset of 25% of the homes, where 2-week samples of indoor and outdoor air PM2.5, black carbon, and NO2 were collected. These samples are being used to develop and test a spatial LUR model that will then be used to estimate exposures in the full cohort that are representative of those occurring in early childhood.
As a part of its investigations of childhood asthma in Baltimore, Maryland, the Johns Hopkins Center for Asthma in the Urban Environment (JHU Center) has conducted an intervention trial and a cohort study of asthma morbidity (Breysse et al. 2005; Swartz et al. 2004). The exposure assessment efforts for these studies include indoor and outdoor air pollution as well as indoor allergens in approximately 400 homes. The major focus of these studies was indoor air where investigators assessed 3-day average indoor PM10, PM2.5, NO2, O3, and nicotine at 3-month intervals (Breysse et al. 2005). In addition, 3-day time resolved PM was assessed using a data-logging nephalometer. Ambient PM air pollution was assessed using a monitoring site centrally located to the study area.
Results from these studies demonstrate the importance of assessing indoor air. Children, particularly young children, spend the great majority of their time in the home. Others have noted (Wallace et al. 2004) that indoor PM concentrations are generally higher than outdoor levels, and cigarette smoking as well as other household activities are responsible for this increase. In some cases, the PM contribution from cigarette smoking to indoor PM is greater than that penetrating from outdoor air. The JHU Center results indicate, for example, that a single cigarette contributes between 1 and 2 μg/m3 to indoor PM. In addition, a strategy that uses repeat measures allows larger time frame variability to be assessed (e.g., seasonal).
Results from the Michigan Center for the Environment and Children’s Health demonstrate the importance of focusing on the home as an important microenvironment for children’s exposure (Keeler et al. 2002; Yip et al. 2004). An important lesson from these studies is that home-based exposure assessments are feasible for studies involving hundreds of children and need to be considered in the National Children’s Study. This conclusion is particularly true for newborn children who spend essentially all of their time in the home. The microenvironments of importance include the indoor environment in a range of housing types, because there is a growing recognition that housing quality is an important predictor of indoor air pollution and can affect outdoor pollution penetration rates as well as being a general risk factor for poor health (Kingsley 2003).
As described above, the USC CHS experience suggests that exposure assignment accuracy can be improved by conducting substudies with a limited number of measurements extended temporally and spatially. In evaluating the minimal sampling needed to successfully predict long-term exposures in study communities, USC CHS investigators found that the intraclass correlation between estimated annual average of pollutants, based on 2-week subset measurements, and the true annual average was greater than 0.9 for O3, NO2, and nitric oxide in southern California, if two winter, two summer, and one spring sample were obtained. Greater numbers of samples did not appreciably improve the correlation. These results indicate that accurate estimates of the pollutant annual average levels can be obtained at homes, schools, and other central site locations with a limited number of samples. Local measurements can then be combined with concurrent central site measurements to estimate neighborhood and household scale concentrations for the entire cohort. Although the optimum number of samples may differ by region of the country or in different neighborhoods within communities, depending on the pollutants of interest and geographic and temporal variation in the processes driving air pollution, this general strategy may be of use in planning efficient National Children’s Study substudies.
Analytic and interpretation issues.
Understanding issues of spatial/temporal correlations of air pollutants, the surrogacy of specific pollutants for components of the complex mixture, and the exposure misclassification inherent in exposure estimates will be critical in analyzing and interpreting National Children’s Study findings.
Issues to consider include the fact that air pollutants occur as complex mixtures of gases and particles, but coexisting constituents may covary, based on their common sources or photochemical pathways. The ambient level of one pollutant may therefore be a surrogate for other pollutants arising from the same source, so interpretation of findings for individual pollutants must account for this surrogacy (Manchester-Neesvig et al. 2003; Sarnat et al. 2001). Identification of pollutant sources therefore provides a potentially important mechanism to evaluate source-specific health effects and can ultimately lead to effective strategies for reducing population exposure.
Substudies among subjects in differing geographic locations may be useful for defining pollutant relationships. For example, in assessing PM, chemical tracers have been identified that can serve as “fingerprints” for individual sources, or source types, of air pollution (Laden et al. 2000; Manchester-Neesvig et al. 2003; Sarnat et al. 2002). This type of information can be used to apportion contributions to the measured PM mass on a per sample basis, along with providing data critical to the assessment and interpretation of health effects associated with individual chemical components of PM. Quantitative assessments of source contributions for large data sets are often determined using a statistical receptor modeling approach. This type of data analysis is best suited to longitudinal study designs and can be limiting because it may require collection of a large number of samples to obtain robust results.
The recent successful development and deployment of several types of continuous portable PM mass and number monitors offer the potential for producing real-time (< 5-min interval) data. The continuous data collection format of these samplers allows a better understanding of source emission patterns and exposures, especially in urban environments, and can be used to enhance investigations of short-term peak exposures. These highly time-resolved exposure data can be coupled with personal time–activity pattern data to quantitatively identify exposures from specific emission sources. To date, real-time PM samplers do not yet offer the ability to determine PM chemical speciation. A combination of methodologic approaches (employing chemical tracers and continuous PM number and mass count information) may improve the ability to identify specific sources and source types contributing to the measured exposure to PM and other pollutants.
Exposure misclassification is a critical issue for exposure assessment efforts, especially modeled exposures. In most large cohort studies, it is not possible to accurately measure the true personal exposure of individuals over the time interval that is most relevant for the outcomes of interest. Thus, virtually all exposure assessments provide at best estimates of true exposures, with some error. Errors may arise because of temporal factors (e.g., the exposure metric captures only a snapshot of the relevant time interval) or spatial factors (e.g., the exposure metric is collected at a location different from where the subject lives and breathes). Additionally, inherent imprecision in the specific method selected for study application may also result in some measurement error. For the results of the study to ultimately be interpretable, it is important in designing the study for investigators to analyze the nature of the exposure misclassification errors that are likely to be present. Quantitative estimates of exposure errors can be obtained by carrying out calibration substudies where results from more complete exposure metrics are compared with results from the modeled metrics (Berhane et al. 2004; Fraser and Stram 2001; Mallick et al. 2002; Sarnat et al. 2001; Stram et al. 1995). Bayesian statistical frameworks may assist with assessing the impact of measurement error on the exposure–response relationships (Berhane et al. 2004).
Modifiers of exposure–outcome relationships.
“Usual” temporal, spatial, and physical patterns of activity can be used as modifiers of the exposure–outcome relationships. Highly time-resolved activity information over the study period of interest may not be necessary, and is not likely to be available, for all National Children’s Study participants throughout the study. Personal exposure estimates, based on time in microenvironments, are likely to be associated with large uncertainties. “Usual” patterns of activity, such as time usually spent outdoors, can be collected by questionnaire and used as modifiers of exposure–outcome relationships (Gauderman et al. 2002). Activity-level assignments may be important in moving from exposure to delivered dose of an airborne pollutant to the lung. For example, for asthma prevalence and incidence, USC CHS investigators saw little association with community levels of exposure. However, when physical activity was considered, O3 was strongly associated with asthma incidence (where variation entered from increased ventilation rates associated with exercise and likely increased dose to the lung). An important challenge for the National Children’s Study is assessing activity patterns among mothers, infants, and young children.
For extremely large study populations for which individual questionnaires may be impractical to administer and/or collect, randomized sampling schemes or oversampling in certain nested subsamples of possible increased interest may be worth careful consideration.
Use of biomarkers.
Biomarkers of exposure offer utility for evaluation of specific exposures that have multiple routes of exposure. For specific airborne pollutants, exposure assessments may need to consider multiple routes of human exposure. In addition to inhalation, dermal absorption and oral ingestion may be important pathways of exposure for pollutants of interest with regard to young children, infants, and pregnant or lactating mothers. The use of exposure biomarkers is one potentially valuable approach in this area (Weaver et al. 1998). Interpreting the relationship between these markers and exposures, however, is a complex function of the timing and routes of exposure, and of the pollutant toxicokinetics. As discussed above, temporal–spatial–physical patterns of activity will almost surely affect this dynamic in important ways, from modification of ventilation rates to facilitated dermal absorption during periods of elevated, increased, or extended activities. As exposure assessment tools, biomarkers offer the potential advantage of integrating the net effect of all of these factors in producing a given internal dose for a given individual. Such measurements may better represent true health-relevant exposures for an individual than any external measure of exposure can.
Biomarker measurements are substantially integrated into the exposure and health assessment designs of the Columbia PCS. From an exposure perspective, biomarkers focus on DNA-bound PAHs (Perera et al. 2004a, 2004b), pesticides in blood plasma and meconium (Perera et al. 2003; Whyatt et al. 2001, 2003, 2004), and the environmental tobacco smoke (ETS) metabolite cotinine in urine (Perera et al. 2004b), beginning with maternal and infant cord blood samples at birth, and continuing with follow-up assessments in the child at 2 and 5 years of age. PAH-DNA adducts also can be viewed as early measures of procarcinogenic health effects (Perera et al. 2004b). Other effect-related biomarkers focus on the time course of sensitization to environmental allergens, including measurements of maternal, cord-blood, and child IgE, and production of proinflammatory cytokines or proliferation of mononuclear cells in response to specific allergens (Miller et al. 2001).
The integration of newly developed pesticide biomarkers within the epidemiologic design of the Columbia PCS has made possible significant new advances in our understanding of the health effects and patterns of exposures to pesticides among urban women and children (Perera et al. 2003; Whyatt et al. 2001, 2003, 2004). A wide range of pesticides have been shown to be quantifiable in the plasma of women and their newborns, with significant correlations between maternal and cord blood levels in many cases (Whyatt et al. 2003). For some but not all pesticides, correlations also were demonstrated between plasma levels at birth (either cord blood or maternal) and air measurements collected during the third trimester of pregnancy. Cord plasma, but not air, levels of the insecticide chlorpyrifos and diazinon were significantly associated with decreased birth weight and length (Whyatt et al. 2004). Of particular significance, levels of several pesticides in both air and plasma showed significant declines across women enrolled before and after the U.S. EPA insecticide phase-out (Whyatt et al. 2003). Furthermore, associations with adverse birth outcomes were significant only for infants born before the phase-out (Whyatt et al. 2004). These findings illustrate the utility of well-targeted biomarker measurements, in conjunction with health and external exposure measures, for birth cohort studies.
Cotinine and nicotine as markers for ETS, an important source of PM exposure, has a long history of use in biomonitoring. Hair nicotine has the potential to provide estimates of ETS exposure over a 2–3 month period or longer (Jaakkola and Jaakkola 1997), and other nicotine metabolites (e.g. cotinine) may be useful indicators of both exposure and bioavailability.
Summary
The National Children’s Study offers a unique opportunity to understand the adverse effects of air pollution on a broad range of interrelated outcomes during the critical period of early life development and growth. Six recommendations for air pollution exposure assessment are proposed from lessons learned in the Children’s Centers.
National Children’s Study subject selection. Study populations should be selected to maximize spatial-scale exposure contrasts for the pollutants of interest. Because multiple pollutants are of interest for the National Children’s Study, priorities must be established to allow identification of individuals with a wide range of exposure profiles for those key pollutants of study interest.
Exposure metrics. Because of the large size, long duration, and diversity of outcomes and exposures of interest in the proposed National Children’s Study, the exposure assessment effort should rely on modeling to provide estimates for the entire cohort, supported by subject-derived questionnaire data. Necessary survey information on temporal–spatial–physical patterns of activity and household characteristics can be collected for the entire cohort, and targeted exposure substudies can be performed in a selected subsample of study subjects.
Use of limited substudies for exposure refinement. Assessment of some exposures of interest will require individual measurements of exposures using snapshots of personal and microenvironmental exposures over short periods and/or in selected micro-environments.
Analytic and interpretation issues. Understanding issues of spatial–temporal correlations of air pollutants, the surrogacy of specific pollutants for components of the complex mixture, and the exposure misclassification inherent in exposure estimates will be critical in analyzing and interpreting findings from the National Children’s Study.
Modifiers of exposure–outcome relationships. “Usual” temporal, spatial, and physical patterns of activity can be used as modifiers of the exposure/outcome relationships.
Use of biomarkers. Biomarkers of exposure may be required for evaluation of specific exposures that have multiple routes of exposure.
We have learned that there are many challenges to assessing air pollution exposures to children. To overcome these challenges, the National Children’s Study will need to commit extensive resources to exposure assessment activities. With optimal subject selection, exposure estimates can be modeled for the entire cohort, supported by direct measurement of selected pollutants in a subset of the study population. Biomonitoring is likely to be a valuable adjunct to the exposure assessment design, helping to trace the mechanistic linkages between exposures and effects. Prioritization of pollutants of study interest and developmental periods of study focus would allow optimization of the study design for the National Children’s Study to maximize contrasting pollution profiles and enhance the ability to assess exposure–response relationships.
This article is part of the mini-monograph “Lessons Learned from the National Institute of Environmental Health Sciences/U.S. Environmental Protection Agency Centers for Children’s Environmental Health and Disease Prevention Research for the National Children’s Study.”
This work was supported by the National Institute of Environmental Health Sciences (ES009581, ES007048, ES009589, ES009600, ES009142, ES009089, ES003819, ES009606, ES10688), the U.S. Environmental Protection Agency (R826708, R827027, R826724, and R826710), the National Heart, Lung and Blood Institute (HL61768), the Hastings Foundation, the Canadian Institutes of Health Research, and the National Children’s Study.
Table 1 Centers for Children’s Environmental Health and Disease Prevention Research air pollution exposure assessment experience relevant to the National Children’s Study.
Columbia University Johns Hopkins University University of Michigan USC Children’s Health Study University of Southern California
Sample population 500 pregnant women enrolled in the third trimester, and children followed from birth through age 5 ~ 250 children with asthma in urban Baltimore (ages 2–12) 300 children, moderate to severe asthma, 7–11 years of age at baseline ~ 6,000 public school children, 9–18 years of age in four specific age cohorts, from 12 southern California communities 202 Los Angeles public school children, 6–16 years of age with asthma and allergy to house dust mite or cockroach
Outcome(s) Asthma and neurodevelopment; follow-up at multiple time points starting at birth; outcome metrics include questionnaires, biomarkers, clinical assessments, neurobehavioral assessments Asthma severity Daily symptom diaries and pulmonary function (PEF, FEV1) Pulmonary function (PFTs), symptoms (from annual medical and residential histories for 10 years), school-reported absences, food-frequency dietary information, physical activity, smoking and ETS, GxE interactions Asthma severity
Study design Prospective birth cohort study with exposures and outcomes measured at multiple time points starting during the third trimester of pregnancy Longitudinal intervention trial (n = 100); longitudinal cohort study (n = 150); cross-sectional case–control study Longitudinal intervention trial Cross-sectional survey (n ~ 3,600); longitudinal cohort study (n ~ 5,600) Randomized trial of allergen-reduction strategies
Agents assessed Personal PAH and pesticide exposures of mother in third trimester; dust allergens prenatal, 12 months, 36 months, and 60 months; indoor/outdoor PM2.5, black carbon, and NO2 at 12 months in subset; biomarkers for ETS, PAH–DNA adducts, pesticides Indoor/outdoor air pollutants (PM10, PM2.5, O3, nicotine); airborne endotoxin and mouse allergen; allergens in reservoir dust (cockroach, mouse, dust mite, cat, dog) Personal/indoor/outdoor air pollutants (PM10, PM2.5, O3, nicotine); PM components (trace elements, EC, OC, endotoxin) Outdoor air pollutants [O3, NO2, PM10, PM2.5, acid vapor (HNO3, formic, acetic) EC, OC, PM speciation (SO4, NO3, NH4, CI)], PAHs, endotoxin, air toxics, ETS, cigarette smoke Settled allergens (dust mite and cockroach) and endotoxin; cockroach counts
Other exposure determinants GIS assessment of traffic proximity; social condition and stress; home characteristics Home inspection, time– activity data, GIS location, meteorology Home inspection, time– activity data, GIS location, meteorology Annual residential history by written survey; time–activity data, GIS location, traffic density, and proximity Housing characteristics and condition, reported and observed behavior, humidity and moisture
Assessment strategy Prenatal exposures to PAH based on personal sampling and cord blood PAH–DNA adducts at birth; allergen exposures based on dust measures; postnatal air pollution exposures based on prediction model developed in subset Primary exposure assignment based on indoor air pollutants, and allergens; secondary exposure assignment using microenvironmental model with indoor/outdoor air pollution combined with time–activity information Primary exposure assignment using personal/indoor/outdoor air pollutants; secondary exposure assignment using microenvironmental model Primary exposure assignment based on community ambient monitors; secondary exposure assignment using microenvironmental model with outdoor air pollution combined with home characteristics and time– activity information Assessment of only indoor settled dust; no outdoor assessment
Abbreviations: CI, chlorine; EC, elemental carbon; FEV1, forced expiratory volume in 1 sec; GIS, geographic information system; GxE, gene–environment interaction; OC, organic carbon; PEF, peak expiratory flow; PFT, pulmonary function test.
Table 2 Spatial scales of variability for ambient air pollutants.
Compound Regional scale (100–1,000 km) Urban scale (4–50 km) Neighborhood scale (50 m to 4 km) Household scale (≤50 m) outdoors and indoor
Primary PM2.5 constituents
EC from combustion x x x
Organics, including PAHs x x
Metals, including chromium VI, cadmium, lead, beryllium, nickel, arsenic, iron, manganese x x x
Other constituents from road dust, wood smoke, construction dust, and industrial sources x x
Secondary PM2.5 constituents
Sulfate x
Nitrate x x
Ammonium x x
Secondary organics x x
Primary PM2.5–10 constituents
Organics, including PAHs x x x
Metals, including chromium VI, cadmium, lead, beryllium, nickel, arsenic, iron, manganese x x
Other constituents from road dust, wood smoke, construction dust, and industrial sources x x
Primary PM > 10 constituents
Pollen grains x x
O3 x x
Nitric oxide x x
NO2 x x
Sulfur dioxide x x
Carbon monoxide x x
Volatile organic compounds
Benzene x x
1,3-Butadiene x x
Formaldehyde x x
Acetaldehyde x x
Acrolein x x
Vinyl chloride x x
Carbon tetrachloride x x
Chloroform x x
Propylene dichloride x x
Methyl chloride x x
Trichloroethylene x x
Tetrachloroethylene x x
Naphthalene x x
Mercury x x
Bioaerosols, including endotoxin, house dust allergens, fungal spores, and pollen grains, also vary considerably on the household and neighborhood scales; however, they were not included in this analysis.
==== Refs
References
Berhane K Gauderman WJ Stram DO Thomas DC 2004 Statistical issues in studies of the long term effects of air pollution: the Southern California Children’s Health Study Stat Sci 19 3 414 449
Brauer M Hoek G Van Vliet P Meliefste K Fischer PH Wijga A 2002 Air pollution from traffic and the development of respiratory infections and asthmatic and allergic symptoms in children Am J Respir Crit Care Med 166 8 1092 1098 12379553
Breysse P Buckley T Williams D Beck C Kanchanaraksa S Swartz L 2005 Indoor exposures to air pollutants and allergens in the homes of asthmatic children in inner-city Baltimore Environ Res 98 2 167 176 15820722
Briggs D Collins S Elliott P Kingham S Lebret E Pryl K 1997 Urban air pollution GIS: a regression-based approach Int J Geogr Inf Sci 11 699 718
Briggs D de Hoogh C Gulliver J Wills J Elliott P Kingham S 2000 A regression-based method for mapping traffic-related air pollution: application and testing in four contrasting urban environments Sci Total Environ 253 151 167 10843339
Dockery D Pope CA Xu X Spengler J Ware J Fay M 1993 An association between air pollution and mortality in six U.S. cities N Engl J Med 329 1753 1759 8179653
English P Neutra R Scalf R Sullivan M Waller L Zhu L 1999 Examining associations between childhood asthma and traffic flow using a geographic information system Environ Health Perspect 107 761 767 10464078
Finkelstein M Jerrett M DeLuca P Finkelstein N Verma DK Chapman K 2003 A cohort study of income, air pollution and mortality Can Med Assoc J 169 397 402 12952800
Fraser GE Stram DO 2001 Regression calibration in studies with correlated variables measured with error Am J Epidemiol 154 9 836 844 11682366
Gauderman WJ Gilliland GF Vora H Avol E Stram D McConnell R 2002 Association between air pollution and lung function growth in Southern California children: results from a second cohort Am J Respir Crit Care Med 166 1 76 84 12091175
Gauderman WJ McConnell R Gilliland F London S Thomas D Avol E 2000 Association between air pollution and lung function growth in southern California children Am J Respir Crit Care Med 162 4 Pt 1 1383 1390 11029349
Hoek G Brunekreef B Goldbohm S Fischer P Brandt PA 2002 Association between mortality and indicators of traffic-related air pollution in the Netherlands: a cohort study Lancet 360 9341 1203 1209 12401246
Jaakkola M Jaakkola J 1997 Assessment of exposure to environmental tobacco smoke Eur Respir J 10 10 2384 2397 9387970
Janssen NA Schwartz J Zanobetti A Suh HH 2002 Air conditioning and source-specific particles as modifiers of the effect of PM(10) on hospital admissions for heart and lung disease Environ Health Perspect 110 43 49 11781164
Jerrett M Arain A Kanaroglou P Beckerman B Potoglou D Sahsuvaroglu T 2005 A review and evaluation of intra-urban air pollution exposure models J Expo Anal Environ Epidemiol 15 2 185 204 15292906
Jerrett M Burnett RT Goldberg MS Sears M Krewski D Catalan R 2003a Spatial analysis for environmental health research: concepts, methods, and examples J Toxicol Environ Health A 66 16–19 1783 1810 12959844
Jerrett M Burnett RT Willis A Krewski D Goldberg MS DeLuca P 2003b Spatial analysis of the air pollution-mortality relationship in the context of ecologic confounders J Toxicol Environ Health A 66 16–19 1735 1777 12959842
Jerrett M Finkelstein M 2005 Geographies of risk in studies linking chronic air pollution exposure to health outcomes J Toxicol Environ Health 68 13–14 1207 1242
Keeler G Dvonch T Yip F Parker E Israel B Marsik F 2002 Assessment of personal and community-level exposures to particulate matter among children with asthma in Detroit, Michigan, as part of Community Action Against Asthma (CAAA) Environ Health Perspect 110 suppl 2 173 181 11929726
Kingsley G 2003 Housing, health, and the neighborhood context Am J Prev Med 24 3 suppl 6 7 12668191
Krewski D Burnett R Goldberg M Hoover K Siemiatycki J Jerrett M 2000. Reanalysis of the Harvard Six Cities and the American Cancer Society Study of Particulate Air Pollution and Mortality: Phase II. Sensitivity Analysis. Cambridge, MA:Health Effects Institute.
Kunzli N McConnell R Bates D Bastain T Hricko A Lurmann F 2003 Breathless in Los Angeles: the exhausting search for clean air Am J Public Health 93 9 1494 1499 12948969
Laden F Neas LM Dockery DW Schwartz J 2000 Association of fine particulate matter from different sources with daily mortality in six U.S. cities Environ Health Perspect 108 941 947 11049813
LaRosa LE Buckley TJ Wallace LA 2002 Real-time indoor and outdoor measurements of black carbon in an occupied house: an examination of sources J Air Waste Manag Assoc 52 41 49 15152663
Lebret E Briggs D van Reeuwijk H Fischer P Smallbone K Harssema H 2000 Small area variations in ambient NO2 concentrations in four European areas Atmos Environ 34 2 177 185
Mallick R Fung K Krewski D 2002 Adjusting for measurement error in the Cox proportional hazards regression model J Cancer Epidemiol Prev 7 4 155 164 12846486
Manchester-Neesvig JB Schauer JJ Cass GR 2003 The distribution of particle-phase organic compounds in the atmosphere and their use for source apportionment during the Southern California Children’s Health Study J Air Waste Manag Assoc 53 9 1065 1079 13678364
Miller RL Chew GL Bell CA Biedermann SA Aggarwal M Kinney PL 2001 Prenatal exposure, maternal sensitization, and sensitization in utero to indoor allergens in an inner-city cohort Am J Respir Crit Care Med 164 6 995 1001 11587985
Nafstad P Håheim L Wisløff T Gram F Oftedal B Holme I 2004 Urban air pollution and mortality in a cohort of Norwegian men Environ Health Perspect 112 610 615 15064169
National Children’s Study 2004. Study Plan. Rockville, MD:National Children’s Study. Available: http://nationalchildrensstudy.gov [accessed 3 June 2004].
Navidi W Thomas D Langholz B Stram D 1999 Statistical methods for epidemiologic studies of the health effects of air pollution Res Rep Health Eff Inst 86 1 56 10465799
Navidi W Thomas D Stram D Peters J 1994 Design and analysis of multilevel analytic studies with applications to a study of air pollution Environ Health Perspect 102 suppl 8 25 32 7851327
Payne-Sturges DC Burke TA Breysse PN Diener-West M Buckley T 2004 Personal exposure meets risk assessment: a comparison of measured and modeled exposures and risks in an urban community Environ Health Perspect 112 589 598 15064166
Perera FP Rauh V Tsai WY Kinney P Camann D Barr D 2003 Effects of transplacental exposure to environmental pollutants on birth outcomes in a multiethnic population Environ Health Perspect 111 201 205 12573906
Perera FP Rauh V Whyatt RM Tsai WY Bernert JT Tu YH 2004a Molecular evidence of an interaction between pre-natal environmental exposures and birth outcomes in a multiethnic population Environ Health Perspect 112 626 630 15064172
Perera FP Tang D Tu YH Cruz LA Borjas M Bernert T 2004b Biomarkers in maternal and newborn blood indicate heightened fetal susceptibility to procarcinogenic DNA damage Environ Health Perspect 112 1133 1136 15238289
Peters JM 1997. Epidemiologic Investigation to Identify Chronic Health Effects of Ambient Air Pollutants in Southern California: Phase II Final Report. Contract No. A033-186. Los Angeles, CA:University of Southern California School of Medicine, Department of Preventive Medicine.
Peters JM Avol E Gauderman WJ Linn WS Navidi W London SJ 1999a A study of twelve southern California communities with differing levels and types of air pollution. II. Effects on pulmonary function Am J Respir Crit Care Med 159 3 768 775 10051249
Peters JM Avol E Navidi W London SJ Gauderman WJ Lurmann F 1999b A study of twelve southern California communities with differing levels and types of air pollution. I. Prevalence of respiratory morbidity Am J Respir Crit Care Med 159 3 760 767 10051248
Pope CA Burnett RT Thun MJ Calle EC Krewski D Ito K 2002 Lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate air pollution JAMA 287 9 1132 1141 11879110
Ritz B Yu F Chapa G Fruin S 2000 Effect of air pollution on preterm birth among children born in southern California between 1989 and 1993 Epidemiology 11 5 502 511 10955401
Samet J Dominici F Curriero F Coursac I Zeger S 2000 Fine particulate air pollution and mortality in 20 U.S. cities, 1987–1994 N Engl J Med 343 24 1742 1749 11114312
Sarnat JA Long CM Koutrakis P Coull BA Schwartz J Suh HH 2002 Using sulfur as a tracer of outdoor fine particulate matter Environ Sci Technol 36 24 5305 5314 12521154
Sarnat JA Schwartz J Catalano PJ Suh HH 2001 Gaseous pollutants in particulate matter epidemiology: confounders or surrogates? Environ Health Perspect 109 1053 1061 11675271
Sax SN Bennett DH Chillrud SN Kinney PL Spengler JD 2004 Differences in source emission rates of volatile organic compounds in inner-city residences of New York City and Los Angeles J Expo Anal Environ Epidemiol 14 suppl 1 S95 S109 15118751
Stram DO Longnecker MP Shames L Kolonel LN Wilkens LR Pike MC 1995 Cost-efficient design of a diet validation study Am J Epidemiol 142 3 353 362 7631639
Swartz L Callahan K Butz A Rand C Kanchanaraksa S Diette G 2004 Partnering with an inner city community to conduct an environmental randomized clinical trial in asthma Environ Res 95 2 156 165 15147921
Tonne CC Whyatt RM Camann DE Perera FP Kinney PL 2004 Predictors of personal polycyclic aromatic hydrocarbon exposures among pregnant minority women in New York City Environ Health Perspect 112 754 759 15121521
U.S. EPA 2004. About AirData. Research Triangle Park, NC:U.S. Environmental Protection Agency. Available: http://www.epa.gov/air/data/info.html [accessed 18 May 2004].
Wallace L Mitchell H O’Connor G Liu L Neas L Lippmann M 2004 Particle concentrations in inner-city homes of children with asthma: the effect of smoking, cooking, and outdoor pollution Environ Health Perspect 111 1265 1272 12842784
Weaver VM Buckley TJ Groopman JD 1998 Approaches to environmental exposure assessment in children Environ Health Perspect 106 suppl 3 827 832 9646045
Whyatt RM Barr DB Camann DE Kinney PL Barr JR Andrews HF 2003 Contemporary-use pesticides in personal air samples during pregnancy and blood samples at delivery among urban minority mothers and newborns Environ Health Perspect 111 749 756 12727605
Whyatt RM Jedrychowski W Hemminki K Santella RM Tsai WY Yang K 2001 Biomarkers of polycyclic aromatic hydrocarbon-DNA damage and cigarette smoke exposures in paired maternal and newborn blood samples as a measure of differential susceptibility Cancer Epidemiol Biomarkers Prev 10 6 581 588 11401906
Whyatt RM Rauh V Barr DB Camann DE Andrews HF Garfinkel R 2004 Prenatal insecticide exposures and birth weight and length among an urban minority cohort Environ Health Perspect 112 1125 1132 15238288
Yip FY Keeler GJ Dvonch JT Robins TG Parker EA Israel BA 2004 Personal exposures to particulate matter among children with asthma in Detroit, Michigan Atmos Environ 38 31 5227 5236
|
16203261
|
PMC1281294
|
CC0
|
2021-01-04 23:41:28
|
no
|
Environ Health Perspect. 2005 Oct 24; 113(10):1447-1454
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.7673
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences 10.1289/ehp.7674ehp0113-00145516203262ResearchMini-MonographLessons Learned for the Assessment of Children’s Pesticide Exposure: Critical Sampling and Analytical Issues for Future Studies Fenske Richard A. 1Bradman Asa 2Whyatt Robin M. 3Wolff Mary S. 4Barr Dana B. 51 Department of Environmental and Occupational Health Sciences, School of Public Health and Community Medicine, University of Washington, Seattle, Washington, USA2 Center for Children’s Environmental Health Research, School of Public Health, University of California, Berkeley, California, USA3 Columbia Center for Children’s Environmental Health, Mailman School of Public Health, Columbia University, New York, USA4 Department of Community and Preventive Medicine, Mount Sinai School of Medicine, New York, New York, USA5 National Center for Environmental Health, Centers for Disease Control and Prevention, Atlanta, Georgia, USAAddress correspondence to R. Fenske, Department of Environmental and Occupational Health Sciences, Box 357234, University of Washington, Seattle, WA 98195 USA. Telephone: (206) 543-0916. Fax: (206) 616-2687. E-mail:
[email protected] authors declare they have no competing financial interests.
10 2005 24 6 2005 113 10 1455 1462 12 10 2004 24 5 2005 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright. In this article we examine sampling strategies and analytical methods used in a series of recent studies of children’s exposure to pesticides that may prove useful in the design and implementation of the National Children’s Study. We focus primarily on the experiences of four of the National Institute of Environmental Health Sciences/U.S. Environmental Protection Agency/ Children’s Centers and include University of Washington studies that predated these centers. These studies have measured maternal exposures, perinatal exposures, infant and toddler exposures, and exposure among young children through biologic monitoring, personal sampling, and environmental monitoring. Biologic monitoring appears to be the best available method for assessment of children’s exposure to pesticides, with some limitations. It is likely that a combination of biomarkers, environmental measurements, and questionnaires will be needed after careful consideration of the specific hypotheses posed by investigators and the limitations of each exposure metric. The value of environmental measurements, such as surface and toy wipes and indoor air or house dust samples, deserves further investigation. Emphasis on personal rather than environmental sampling in conjunction with urine or blood sampling is likely to be most effective at classifying exposure. For infants and young children, ease of urine collection (possible for extended periods of time) may make these samples the best available approach to capturing exposure variability of nonpersistent pesticides; additional validation studies are needed. Saliva measurements of pesticides, if feasible, would overcome the limitations of urinary metabolite-based exposure analysis. Global positioning system technology appears promising in the delineation of children’s time–location patterns.
childrenexposureGPSorganophosphatespesticides
==== Body
Accurate characterization of children’s exposure to pesticides has proven to be a particularly challenging aspect of the field of exposure assessment. First, the term “pesticides” encompasses a diverse array of chemicals that can potentially produce a wide variety of health effects. Second, exposure of children to pesticides can occur through multiple pathways and routes. For example, the U.S. Environmental Protection Agency (EPA) considers food, drinking water, and residential pesticide use all to represent important sources of exposure, and these exposures can occur simultaneously or sequentially through the routes of ingestion, inhalation, and dermal contact (Cohen Hubal et al. 2000). Certain subpopulations, such as children living in agricultural communities or children whose parents work with pesticides, may be exposed through additional pathways. Third, many pesticides have short residence times in the body, making it difficult to characterize exposures from biologic samples. Finally, chemical exposures may have substantially different health consequences for children depending on the developmental stage during which the exposure occurs, requiring exposure characterization at multiple time points.
Our purpose in this article is to examine sampling strategies and analytical methods associated with a series of recent population studies that have sought to characterize children’s pesticide exposure, and to distill from these experiences a number of lessons learned. In this article, we focus primarily on the experiences of the National Institute of Environmental Health Sciences/U.S. EPA Children’s Centers located at Columbia University, the University of California at Berkeley, Mount Sinai Medical Center, and the University of Washington. We have also included a review of several University of Washington studies that predated establishment of the children’s centers and that were conducted under the auspices of the Pacific Northwest Agricultural Safety and Health (PNASH) Center, sponsored by the National Institute for Occupational Safety and Health, and the U.S. EPA Science To Achieve Results (STAR) Grant Program. This article is not meant to be an exhaustive review of exposure assessment methods, but rather a first-hand commentary on the use of particular methods in our studies. We therefore have not been able to include an analysis of a number of important studies conducted at other institutions, such as the Minnesota Children’s Pesticide Exposure Study (Adgate et al. 2001; Quackenboss et al. 2000) and studies of children’s exposure along the U.S.–Mexican border (U.S. EPA 2004).
In this article we first examine the rationale and methods of exposure data collection in the population studies and then review the substantial challenges associated with the analysis of pesticides in novel and complex matrices, and the interpretation of these analytical findings. It is our hope that experience gained from this work will prove useful to researchers embarking on longitudinal cohort studies, such as the proposed National Children’s Study.
Sampling Strategies in Population Studies
Data used to construct exposure estimates or classifications can be drawn from a variety of sources, ranging from general information regarding pesticide use to personal measurements. Table 1 presents the approaches taken in the studies under review. The first two columns provide source information and environmental measurement methods; the remaining columns categorize various types of exposure samples collected according to age, because different sampling strategies are more or less practical and valuable within these time frames. Table 2 indicates the analytes measured in five biologic sample matrices collected in these studies.
Pesticide source information.
Virtually all children’s exposure studies collect historical and contemporaneous information regarding pesticide use. In most cases, these data are collected through parental questionnaires or interviews and pertain to pesticides in and around the residence. In general, we have found that parents are best able to provide general information regarding the use of products (e.g., control of particular insects, control of weeds) but may not be able to provide detailed information on specific chemicals (Lu et al. 2004; Whyatt et al. 2002). In preliminary analyses of questionnaires administered by the Columbia center, women provided a pesticide product name for only 39% of the pest control methods reported to be used in the home during pregnancy and, in particular, were rarely able to identify the pesticide products used by an exterminator. Further, pesticide products can have the same brand name but contain different active ingredients, further complicating use of questionnaire data in pesticide exposure assessment.
Investigators for most of the reviewed studies have thus gone a step further to visually inspect the pesticide products in the home, sometimes referred to as a pesticide inventory. For example, study staff from the Berkeley center recorded the U.S. EPA registration number and the active ingredients on the label of each home pesticide. The registration number was later entered into a pesticide product database maintained by the California Department of Pesticide Regulation to confirm all active ingredients. Records of commercial pesticide applications can also be accessed during home visits (Berkowitz et al. 2003; Whyatt et al. 2003).
Identification of specific products can be very helpful in determining whether or not a particular class of chemicals has been used in the residence and may inform subsequent sampling plans, but the presence or absence of specific products does not generally enter into the development of an exposure metric for the residents. Frequency of residential pesticide use could be used potentially to sort children into exposure categories, but such an approach has not been fully validated. One study has shown that personal air levels of organophosphate (OP) pesticides were significantly higher among women who reported using exterminator sprays, can sprays, and/or pest bombs during pregnancy compared with those reporting no OP pesticide use (Whyatt et al. 2002, 2003). Another study demonstrated that children whose parents reported garden use of insecticides had higher levels of OP pesticide metabolites than did children whose parents did not use garden insecticides (Lu et al. 2001).
Food can be an important source of pesticide exposure for children, but most of the studies reviewed here have not devoted substantial resources to an evaluation of the dietary pathway. The Mount Sinai center obtained maternal prenatal dietary food frequency data during pregnancy only, with specific information about fish consumption. The Berkeley center also obtained a detailed prenatal food frequency questionnaire. Additional information was also obtained on fruit and vegetable consumption for the pregnant women and, later on, for their children. The Berkeley center and the PNASH center have collected duplicate diets from a relatively small number of children (Fenske et al. 2002a). Such an approach provides very useful quantitative information on exposure but is extremely time-consuming and expensive. A diet diary has also been used to distinguish children whose intake of fresh produce and juices was primarily organic and proved effective in classifying children’s OP pesticide exposure (Curl et al. 2003a).
Studies of children of agricultural workers have focused on potential paraoccupational exposure, collecting data on the transmission of pesticides from the workplace to the home by parents or other adult household members, as well as data on residential proximity to pesticide applications (Bradman et al. 1997; Curl et al. 2002; Eskenazi et al. 2003; Koch et al. 2002; Lu et al. 2000; Simcox et al. 1995). Results to date indicate that both of these pathways can contribute to children’s exposures in agricultural communities and would need to be considered in the design of a study that included rural populations. Studies at the Berkeley center have taken advantage of California’s unique Pesticide Use Reporting system, and researchers there are investigating the use of these data as predictors of pesticide exposure in their cohort (Castorina et al. 2003). The Washington center completed a 2-year intervention to reduce take-home exposure in 2002; the Berkeley center is currently conducting a similar intervention.
Environmental monitoring.
House dust samples have been collected in most of the reviewed studies and have served as a reliable indicator of residential pesticide contamination (studies conducted at the PNASH Center), although not necessarily as a surrogate for children’s exposures (Curl et al. 2002; Fenske et al. 2002b; Lu et al. 2000; Simcox et al. 1995). A practical problem can arise when insufficient dust is available for analysis, as was the case for the Mount Sinai studies. In the Berkeley center studies, the average mass of 509 dust samples was 9 g/m2. The average of the fraction < 150 μm in diameter used for chemical analyses was 7 g/m2. About 20% of the samples had a fine fraction of < 0.5 g total. Most laboratory methods for pesticides require 0.5–2 g dust. It is likely that only a single chemical analysis will be possible for a significant fraction of homes, thus limiting future tests for other chemicals. The Berkeley, Mount Sinai, and PNASH centers have investigated alternate methods of measuring pesticide concentrations in child environments, such as indoor air and surface wipe sampling (Lu et al. 2004). A protocol that is currently being validated involves mailing study participants an alcohol wipe with instruction for wiping dust on the top of a specified doorframe. The sample is then placed in a resealable plastic bag and mailed back to the study team. Advantages include low cost of sample collection and low participant burden. However, research is currently ongoing to determine detection limits and detection frequencies using this method.
The Columbia center has conducted extensive indoor air sampling. For chlorpyrifos and diazinon, the correlation between 48-hr personal air samples collected from the mother during the third trimester and average 2-month indoor air levels over the final 2 months of pregnancy were strong (r > 0.7, p < 0.001) (Whyatt et al. 2003). Air and dust levels were not significantly correlated in a pilot study conducted by the Mount Sinai group; this may have been due to the very small amount of dust collectable in these homes (Markowitz S, personal communication). In addition to the OP pesticides several carbamates and pyrethroids have been measured in personal air samples collected from the mother over 48-hr during pregnancy (Whyatt et al. 2002, 2003).
Evidence of chemicals in a child’s environment does not necessarily provide the basis for a sound exposure metric. Dust, wipe, and indoor air measurements (including personal air samples) have not shown strong associations with biologic measurements (Curl et al. 2002; Whyatt et al. 2003). It is not clear whether the lack of strong associations is due to confounding factors (e.g., dietary exposure), to variability in the biologic measurements (including toxicokinetic considerations (discussed below), or to a relatively weak link between residential contamination and child exposures.
Environmental monitoring in these studies has focused almost exclusively on the home or residential setting and has not yet been extended to child care centers and schools. The Washington studies have included wipe sampling and dust sampling of commuter vehicles of workers to document the movement of agricultural pesticides from the workplace to the home (Curl et al. 2002; Lu et al. 2000).
Hand wipe sampling.
Initial attempts to look at direct child exposures have included the use of hand wipes to collect pesticides from children’s hands. These methods include wiping the child’s hand with sterile gauze dressing pads that have been moistened with isopropanol, or asking the child to place a hand in a bag containing isopropanol (Bradman et al. 1997). Gordon et al. (1999) found excellent correlations between chlorpyrifos in indoor air and corresponding dermal wipes but poor correlations between chlorpyrifos in dust and dermal wipes. Another study reported weak associations between OP pesticide concentrations in hand wipes, house dust, and urinary levels of OP metabolites (Shalat et al. 2003). The Columbia center conducted hand wipes but found all samples to be less than the limit of detection.
Clothing dosimeters.
Other techniques for assessing children’s dermal exposures include use of clothing dosimeters such as cotton gloves, union suits, and socks (Fenske 1993; Lewis 2005). The Berkeley center has experimented with clothing dosimeters in recent studies. Infants (children 6 and 12 months of age) wore precleaned cotton socks and union suits for several hours in their residential environments.
Maternal exposure.
Personal air sampling has been used effectively to monitor maternal exposures during pregnancy by Columbia researchers (Whyatt et al. 2002, 2003). Investigators used motion detectors to determine whether or not the women complied with the request to carry the personal air monitors; motion detectors were installed in the backpacks of randomly selected women. Results were obtained from monitors worn by 113 women for approximately 48 hr each. For the average woman, nearly 95% of the total number of motion detections occurred during waking hours. In addition, 98% of the women self-reported that the air monitor was near them for least 40 of the 48 hr of the personal air monitoring.
This study (Whyatt et al. 2003) also found that levels of several OP and carbamate pesticides measured in the 48-hr personal air samples were significantly correlated with levels in 2-week indoor air samples, indicating that, at least for these pesticides, the 48-hr air samples provided a reasonable estimate of exposure over a longer period during pregnancy. In addition, there was little variability in indoor air levels of the insecticides, and the correlations between each of the insecticides in each of the 2-week air samples were highly significant. In cases where sampling bracketed an application event, it is likely that high levels would be observed initially, increasing temporal variability.
Blood samples have been collected throughout pregnancy to assess body burden of pesticides in the Berkeley, Columbia, and Mount Sinai center studies. No association was seen between insecticide levels in maternal blood collected at delivery and maternal self-reported pesticide use during pregnancy in one study (Whyatt et al. 2003). Weak correlations were seen between pesticide levels in the maternal personal air samples collected during pregnancy and in blood samples collected at delivery (r = 0.10–0.19). However, the correlations were generally stronger when analyses were restricted to women for whom the personal air sample was collected within a month of collection of the blood samples at delivery (r = 0.13–0.45). Maternal and umbilical blood insecticide levels (chlorpyrifos, diazinon, the propoxur metabolite 2-isopropoxyphenol, and bendiocarb) at delivery were highly correlated, indicating that the pesticides are readily transferred to the fetus during pregnancy. Significant inverse associations were seen between chlorpyrifos in umbilical cord blood and both birth weight and length, whereas no association was seen between chlorpyrifos in maternal personal air samples and the same measures of fetal growth (Whyatt et al. 2004). These results suggest that the biomarkers may better reflect exposure from all routes, not only the amount of insecticides absorbed by the mother but also the amount of the absorbed dose that has been transferred to the developing fetus (Whyatt et al. 2004).
Urine samples have also been collected from women during pregnancy in several studies. Investigators at the Berkeley center found that pesticide metabolites in samples collected in the first and third trimester were not correlated. Within-person variability was approximately two times higher than between-person variability, suggesting that more urine samples collected during pregnancy would improve exposure classification (Eskenazi et al. 2004). A moving estimate of the coefficient relating dimethyl OP metabolite levels to shorter gestation was used to show that exposures in later pregnancy may be associated with shorter pregnancies. Blood cholinesterase levels were inversely correlated with gestational duration, consistent with findings for dimethyl OP pesticide metabolites, although no significant correlation between blood cholinesterase and urinary metabolite levels was observed.
The Mount Sinai center collected urine samples in the third trimester of pregnancy and found that approximately 70% of the women in the cohort had been exposed to pesticides, but no associations were found between these biologic levels and pesticide questionnaire data (Berkowitz et al. 2003). In a preliminary analysis of data from the Columbia center, weak but significant correlations were seen between average chlorpyrifos and diazinon levels in indoor air samples collected over the final 2 months of pregnancy and their respective metabolites in urine samples collected biweekly from the mothers over the same time frame.
In summary, it is unlikely that questionnaire data alone can prove adequate for exposure classification of women during pregnancy. However, it appears that systematic monitoring through personal air sampling and biologic monitoring in combination with questionnaire data would yield useful exposure data for epidemiologic investigations.
Perinatal exposure.
Several novel sampling methods are under development to determine perinatal exposure levels, including sampling of amniotic fluid, meconium, and cord blood. A pilot study from the Berkeley center of 100 amniotic fluid samples, slated for disposal after amniocentesis, were analyzed for a number of pesticides and their metabolites, including the OP pesticides (Bradman et al. 2003). Target analytes were detected with frequencies ranging from 5 to 70%. Levels were low compared with levels reported in urine, blood, and meconium. Because of risks to the fetus, amniotic fluid typically can be collected only when medically indicated amniocenteses are conducted, usually around 18–20 weeks of gestation, or during scheduled cesarean sections. Therefore, the population sampled will not necessarily be representative of a larger population of pregnant women. For women already undergoing this procedure, the collection of amniotic fluid for research purposes is noninvasive and causes no additional risk.
At the Columbia center, meconium samples were collected from 20 newborns and analyzed for OP pesticide metabolites (Whyatt and Barr 2001). Detection frequencies were very high for some of these analytes, but others were not detected. Metabolite levels were similar to those seen in adult urine in population-based research. Metabolites were stable at room temperature over 12 hr. These initial results indicate that the measurement of pesticide levels in meconium has promise as a bio-marker of prenatal exposure.
Cord blood has been sampled in three studies. Mount Sinai center investigators collected cord blood for enzyme, lead, and gene analyses. The Mount Sinai group relied on hospital staff for cord blood retrieval, with prenotification of impending delivery and a note on the chart, with the result that 59% of the participants’ cord blood was obtained. Columbia center investigators reported that successful collection of these samples required that a member of the research staff team follow the progress of the labor, go to the labor room before delivery to remind the delivery room staff that the woman is in the study, and assist with the sample collection. Umbilical cord blood was obtained by syringing the blood into heparinized syringes at the point the cord enters the placenta. To date, a cord blood sample has been obtained from 81% of the infants in the study. An average of 29 mL (range, 2–58 mL) was collected per delivery, with > 22 mL collected in 75% of deliveries and ≥30 mL collected in 50% of the deliveries (Whyatt et al. 2003). The Berkeley center investigators reported a similar proportion of cord blood samples collected and found that successful collection of cord blood required close cooperation with hospital staff to develop procedures that eliminated risks of inadvertent sticks (Eskenazi et al. 2003).
In summary, the perinatal sampling procedures described here are in the early stages of development and will need additional study and validation. However, they hold promise for collecting quantitative exposure data at a critical stage of child development.
Infant and toddler exposure.
Traditional urine bags have been used in clinical settings and have proven useful for pesticide-related studies in children (Royster et al. 2002). The Berkeley center has been successful collecting urine from children 6–24 months of age who were not toilet trained. Urine was collected by applying pediatric urine bags to the children during office or home visits (Eskenazi et al. 2003). When children were not able to produce a void during scheduled contacts, study staff trained parents to apply the urine bag at home and to then place the urine in a clean cup provided to them. The parent was instructed to call the field office as soon as the void was produced, and study staff then retrieved the sample.
Cotton inserts have also been used to recover urine from diapers (Hu et al. 2000). However, the most promising development for sampling infants and toddlers who are not yet toilet trained appears to be extracting the metabolites from the diaper gel matrix, although this method still needs to be evaluated for multiple groups of pesticides (Hu et al. 2004).
Preschool children’s exposure.
Urine samples have been collected in nearly all studies of pesticide exposure among preschool children. Urine samples have been analyzed for common metabolites, such as the dialkylphosphate (DAP) compounds or for compound-specific metabolites [e.g., 3,5,6-trichloro-2-pyridinol (TCPy) for chlorpyrifos]. Major exposure assessment issues of concern are duration of collection (spot samples vs. 24-hr samples) and frequency of sampling.
Collection of single urine voids, often referred to as spot urine samples, has been selected as a primary sampling strategy for several practical reasons. The burden it places on study participants is relatively low, and sample processing and analysis are manageable and affordable. However, several studies have now determined that pesticide metabolite concentrations in children’s spot urine samples can exhibit high intraindividual (within-child) variability (Adgate et al. 2001; Koch et al. 2002). In studies in which it is possible to collect only a single urine sample per day, the first morning void is preferred, because the urine is more concentrated, the collection period is longer (usually > 8 hr), and it appears this sample is most representative of the daily average (Kissel et al. 2005). Collection of repeated spot urine samples during a single day or over several days is one means of addressing the issue of intraindividual variability. These repeated measures can be averaged to produce a more stable estimate of exposure and would allow evaluation of exposures during specific windows of vulnerability.
Collection of complete 24-hr urine samples has become a standard part of many occupational exposure studies but has generally been viewed as impractical for small children. Several studies reviewed here have attempted to collect 24-hr samples but have been only partially successful. A recent study (Kissel et al. 2005) of 25 children in a low-income, low-literacy population by the Berkeley center provided intensive training of participants, detailed record keeping by participants, use of small refrigerators, and daily contact by research staff to improve compliance; it was estimated that 28% of participants provided complete samples, an additional 12% were likely complete, 52% missed one or two voids, and 8% likely missed more than two voids.
Several of the centers have also collected blood samples from children postnatally. The Columbia center has employed a pediatric phlebotomist to draw blood when children came to the center for the developmental assessment. Samples were collected from 98% of the children that were seen. However, volumes were generally low (an average of 6.8 mL collected at 24 months and 6.2 at 36 months). The Berkeley center hired a pediatric phlebotomist to collect blood for both state-required lead screening and the CHAMACOS (Center for Health Analysis of Mothers and Children of Salinas) study, increasing the rate of blood collection. Repeat blood samples can be collected from young children but are more difficult to obtain than are urine samples.
Children’s activities are an important variable in assessing pesticide exposure. The Berkeley center has used a visually based, low-literacy child activity time line for parents to record child activity and location. The University of Washington center and the PNASH center have employed miniaturized global positioning system (GPS) units to produce detailed documentation of children’s time–location patterns (Elgethun et al. 2003). Recent studies have found that time–location diaries kept by parents produce relatively poor agreement with the GPS measurements, suggesting that such diary data would result in substantial misclassification. The GPS analysis has also shown that transient peak exposures can occur both temporally and spatially and that such exposures are not adequately captured within the resolution of most microenvironmental analysis studies.
School-age children exposure.
Sampling procedures for school-age children are similar to those described above for preschool children. However, as children reach school age, they are more likely to be able to participate more actively in studies. They may be able to assent to study procedures, wear personal sampling devices, collect more complete urine samples, and provide helpful information regarding pesticide sources and their own activities. Here we would stress greater emphasis on personal sampling devices to improve the quality of exposure data for this age group.
Saliva monitoring.
The PNASH center has explored the feasibility of saliva sampling for pesticides in both workers and children (Denovan et al. 2000; Lu et al. 2003). Current saliva sample collection methods require that children chew on a cotton or synthetic plug for approximately 2 min. The plug, containing up to 2 mL of saliva, is then placed in a vial for storage. The plug is similar in size to a dental sponge and could pose a choking hazard to children < 3 years of age. The Berkeley center has experimented with pipettes to directly transfer saliva from a child’s mouth into a collection container. Sample volumes, however, have been < 1 mL. In rare cases, children have spit directly into a beaker. It is not clear that these techniques provide an adequate or appropriate saliva sample for pesticide analysis.
Participation of cohort members in environmental and biologic sampling.
Collection of an array of biologic and environmental samples from women during pregnancy and soon after birth places a burden on study participants and may lead to attrition regarding participation in the exposure assessment component of these studies. Tables 3 and 4 provide data from the birth cohort studies reviewed here to indicate what might be anticipated in the National Children’s Study. Sample sizes are presented for each study and for each relevant time category; the percentage of enrolled study members is then provided for each of the biologic or environmental samples. It is important to recognize that not all of the rates in Tables 3 and 4 are directly comparable. For example, the Berkeley study accepted all eligible enrollees with no condition that they participate in every exposure assessment event; in contrast, enrollment criteria for the Columbia study included collection of a cord blood sample from each participant at delivery. Participation in environmental and biologic sampling tends to drop over time and can be relatively low for certain types of samples. Factors contributing to low participation include reliance on delivery staff, emergency deliveries, inability to schedule appointments that include both parents, mobile populations that are hard to track, and the absence of children from the home at the time of visits by study staff. Participation can also be enhanced; for example, the Berkeley center saw an increase from 64 to 81% between 12 and 24 months for child blood samples because of the hiring of a child phlebotomist who went to each home.
Challenges in the Analysis of Pesticide Exposure Samples
Increased interest in children’s exposure to pesticides has resulted in the generation of large numbers of samples for analysis. In this section we discuss several key issues and lessons learned regarding analysis.
Laboratory capacity.
As studies of the type described here grow larger and a series of longitudinal samples are collected from each participant, the sample size may become too large for the capacity of one or two laboratories. Multiple laboratories should be enlisted for large studies to avoid sample backlogs. As laboratory capacity is improved, it is imperative to produce comparable data across studies, as the U.S. EPA did in its interlaboratory comparison study among the North American laboratories performing DAP analyses (James et al. 2003).
Intra- and interpersonal variability in urine samples.
Several methods have been evaluated to “correct” for the variability in urine dilution across spot samples, the most popular being creatinine (Boeniger et al. 1993). Creatinine excretion varies because of many factors, including the size of the participant, so interindividual variation, especially among diverse populations, is large. Thus, creatinine-adjusted pesticide concentrations should never be compared among individuals of vastly different age groups (i.e., children vs. adults). Changes in creatinine excretion during pregnancy should be thoroughly evaluated before comparing with other women in similar age groups. The validity of creatinine adjustment may also be analyte dependent. Further studies to assess the variability of commonly measured analytes in urine should be conducted to identify the most effective sampling strategies for cohort studies. In all likelihood, sampling for nonpersistent chemicals will require multiple samples taken over the course of the study at regular intervals (e.g., weekly, monthly, semiannually).
Selectivity of analysis.
Selectivity can refer to either the ability of a measurement technique to differentiate a single analyte that is measured from other components of the matrix (i.e., reducing false positives) or the ability of the analyte measured to accurately, and unequivocally, identify exposure to the target chemical of interest. However, high selectivity techniques are costly and require specialized training for operation (Barr et al. 1999). Methods such as immunoassays and less specialized technologies may be employed, but harmonization should be performed to ensure that data generated using different methods are comparable.
The selectivity of the analyte measured to accurately reflect the exposure of interest may depend on the biomarker being measured rather than the measurement technique. Many OP pesticides, for example, can be metabolized to common DAP compounds, so it is not possible to derive chemical-specific exposure estimates from such data. Further complicating the issue, the DAPs, as well as compound-specific metabolites, may be present in environmental media as the environmental degradates of the pesticides (Curl et al. 2003b; Wilson et al. 2004). No studies to date have shown that these environmental degradates can be absorbed and excreted unchanged; but if this does occur, then DAPs and other pesticide metabolites detected in urine would represent exposure to both the pesticide and its degradate. Some metabolites are very selective for the chemical measured. For example, 2-isopropoxy-4-methyl-6-hydroxypyrimidine, a metabolite of diazinon, is selective for diazinon exposure, although potentially the environmental degradates could contribute to the urinary levels as well. In some cases, the parent pesticide can be excreted in urine, such as for the herbicide 2,4-D (2,4-dichlorophenoxy-acetic acid).
One way to unequivocally identify exposure to a particular pesticide is by measuring the intact pesticide, presumably in blood or similar samples, because the intact pesticide is not appreciable in urine. However, blood measurement levels are typically about 1,000 times lower than urinary metabolite measurements; this requires highly sensitive analytical techniques, driving up the cost of analysis. In addition, target chemicals in blood may exhibit some degree of instability. Finally, there are no laboratory methods available for many common use agricultural or home pesticides in blood. Saliva sampling is an attractive alternative to blood sampling, as discussed above.
Sensitivity of analysis.
The sensitivity of an analytical method—the ability of the method to measure the chemical at the desired level—should be considered before a study begins (Barr et al. 1999). The biologic half-lives of nonpersistent chemicals are relatively short, usually on the order of hours or days (Needham and Sexton 2000). Samples collected several days after an exposure event may require ultra-sensitive methods for analyte detection. These measurements must provide adequate sensitivity to allow detection of the chemicals of interest in a sufficient proportion of the population to provide a realistic representation of the populations’ exposure. The current method for analysis of OP pesticide metabolites developed by the Centers for Disease Control and Prevention was used for many but not all of the studies described in this article and has proven to be quite sensitive (Bravo et al. 2002).
Alternative matrices and/or biomarkers.
Pesticides have been measured successfully in saliva (Lu et al. 2003), meconium (Whyatt and Barr 2001), and amniotic fluid (Bradman et al. 2003). Matrices such as meconium may provide longer term dosimeters for exposure to nonpersistent chemicals; saliva may provide a measure of internal dose without the invasiveness of blood sampling. Preliminary studies evaluating the partitioning of chemicals in the various matrices should be conducted that will allow for comparison of data among matrices and validate the usefulness of alternative matrices for biologic monitoring. An alternative matrix that may prove useful is the gel matrix in disposable diapers. Extraction techniques for solid materials may prove practical for the gel matrix and might improve sample collection procedures for infants and children who are not toilet trained.
Quality assurance and control.
A vital component of all biomonitoring methodology is a sound quality assurance/quality control (QA/QC) program. QA/QC procedures supporting these studies have included proficiency testing, repeat measurements of known biologic materials, and round-robin studies to confirm reproducible measurement values among laboratories, as well as field spikes and field blanks to confirm sample integrity.
Sample storage issues.
The time frame for sample testing and long-term storage becomes an issue for large studies. The long-term stability of analytes has been demonstrated for some matrices but not for others, for example, blood. One final logistical complexity is physical freezer space for storage, and the substantial cost of maintaining that storage. Archiving samples in the smallest containers possible would enhance the ability to keep the samples long term under proper storage conditions.
Conclusions
Epidemiologic investigations have often relied on questionnaire information for exposure classification, but this approach alone is unlikely to capture the complexity of children’s pesticide exposure. In contrast to the Agricultural Health Study, for example, which draws on the records of pesticide applicators and has derived a complex exposure algorithm from 40 years of occupational exposure studies (Dosemeci et al. 2002), the everyday use of pesticides in homes, schools, and other child environments is not easily codified, and dietary pesticide exposures can only be inferred from questionnaire data. It seems, therefore, that some level of environmental and/or biologic monitoring will be required for all study participants. The type of sampling needed will depend primarily on the purpose of the study, be it exposure characterization, long-term health outcomes, or short-term toxic response in children. Lessons learned regarding pesticide exposure can be summarized as follows:
Biologic monitoring appears to be the best available method for assessment of children’s exposure to pesticides. However, all pesticide biomarkers have limitations. It is likely that a combination of biomarkers, environmental measurements, and questionnaires will be needed after careful consideration of the specific hypotheses posed by investigators and the limitations of each exposure metric.
Environmental measurements, such as surface wipes and indoor air or house dust samples, can characterize residential pesticide contamination, but their validity for exposure classification has not been established. Their value in epidemiologic studies deserves further investigation.
Emphasis on personal rather than environmental sampling in conjunction with urine or blood sampling is likely to be most effective at classifying exposure.
A focus on maternal exposures during pregnancy is particularly important for making associations with infant health, given the critical developmental stages during this period.
Questionnaires will need to be validated with carefully designed studies that involve personal sampling or biologic monitoring.
Interpretation of urinary metabolites is not straightforward, but because of ease of collection, these samples may provide the best available approach to capturing exposure variability of nonpersistent pesticides in young children; additional validation studies are needed.
Repeated exposure measures will be needed to overcome high intraindividual variability of biologic samples for most pesticides in use today.
Postnatal exposure can also contribute to health effects in early childhood. For infants and young children, it appears possible to collect urine samples for extended periods of time.
Expansion of laboratory capacity will require careful attention to QA/QC and will need to include formal procedures for ensuring inter-laboratory comparability in sample analysis.
Saliva measurements of pesticides, if feasible, would overcome the limitations of urinary metabolite-based exposure analysis.
GPS technology appears promising in the delineation of children’s time–location patterns.
It is clear from this review that the critical tools needed for accurate characterization of children’s pesticide exposure are not yet in place. Most of the work discussed here has been conducted in the past 6–8 years, and many of the exposure methods have been exploratory in nature. Substantial resources will be needed for validation of existing methods, support of novel methods, and enhancement of analytical capabilities. It may be possible to initiate epidemiologic investigations and validation studies simultaneously, if biomarker samples can be properly archived. Whatever sampling strategies are employed for epidemiologic investigations, they will need to be selected to support specific hypotheses and focus on specific pesticides. Studies with substantial exposure assessment activities will be costly but should ultimately pay benefits in terms of the quality of scientific information produced.
This article is part of the mini-monograph “Lessons Learned from the National Institute of Environmental Health Sciences/U.S. Environmental Protection Agency Centers for Children’s Environmental Health and Disease Prevention Research for the National Children’s Study.”
The National Children’s Study provided support for the preparation of the manuscript. The Children’s Center studies were supported by grants ES009605, ES009601, ES009584, and ES009600 from the National Institute of Environmental Health Science and R826709, R826886, R827039, and R827027 from the U.S. Environmental Protection Agency (EPA). University of Washington Pacific Northwest Agricultural Safety and Health Center studies were supported by grants R819186, 916001537, and R82517101 from the U.S. EPA and S147-14/16 and U07/CCU012926 from the National Institute for Occupational Safety and Health (Association of Schools of Public Health and Agricultural Centers Program).
Table 1 Exposure data collected in reviewed studies of children’s exposure to pesticides through 2003.
Studya Pesticide source information Environmental monitoring Maternal exposure Perinatal/infant exposure Preschool children School-age children
Columbia University Birth cohort study Residential pesticide use Indoor air Blood, urine, personal air Cord blood, urine, meconium Blood, urine Blood, urine
Columbia University Prenatal intervention study Residential pesticide use Indoor air Blood, urine Cord blood
University of California at Berkeley Birth cohort study California Pesticide Use reports, home inventory,b proximity to agricultural spray,c parental workd House dust, vehicle dust Blood, urine, breast milk Cord blood, blood, urine Blood, urine, saliva Blood, urine, saliva
University of California at Berkeley Specialized studies California Pesticide use reports, home inventory, proximity of agricultural spray, parental work Indoor/outdoor air, house dust, surface wipef — Diaper and spot urine, amniotic fluid, surface wipef duplicate diet 1st morning void, 24 hr urine, saliva,e CATg —
Mount Sinai Medical Center Birth cohort study Residential pesticide use — Blood, urine Cord blood, urine Urine —
Mount Sinai Medical Center Community cohort Cockroach enumerationh Indoor air, house dust, surface wipes Urine Urine, hand wipesi Urine, hand wipes —
University of Washington Community intervention Residential pesticide use, parental work House dust — — Urine —
University of Washington Community intervention Residential pesticide use, parental work, proximity to agricultural spray House dust, vehicle dust — — Urine —
University of Washington Spray drift exposure Residential pesticide use, aerial applicationj Indoor/outdoor air, residential surfaces, outdoor deposition — — Hand wipes, personal GPSk Hand wipes, personal GPS
PNASH center Agricultural families Residential pesticide use, parental work, proximity to spray — — — Urine —
PNASH center Aggregate exposure Residential pesticide use, home inventory, diet diariesl Indoor air, house dust, surface wipes, duplicate dietm — — Urine, hand wipes —
PNASH center Longitudinal exposure Residential pesticide use, parental work, proximity to agricultural spray — — — Urine —
— , no data.
a References for studies: Columbia University (Carlton et al. 2004; Perera et al. 2003; Whyatt and Barr 2001; Whyatt et al. 2002, 2003, 2004; Berkeley center (Bradman et al. 2003; Castorina et al. 2003; Eskenazi et al. 2003, 2004; Goldman et al. 2004); Mount Sinai center (Berkowitz et al. 2003, 2004; Brenner et al. 2003); University of Washington center (Curl et al. 2002; Elgethun et al. 2003); PNASH center (Fenske et al. 2002a, 2002b; Kissel et al. 2005; Koch et al. 2002; Lu et al. 2000, 2001, 2004; Simcox et al. 1995).
b Home inventory: visual inspection of pesticide products currently in the residence, along with detailed history of pesticide use.
c Proximity to agricultural spray: normally defined as distance between residence and nearest pesticide-treated farmland; more refined analyses include meteorologic data and pesticide application history; determined by GPS technology.
d Parental work: parent or other household member works in agriculture in a job with potential pesticide exposure.
e See Denovan et al. (2000).
f Surface wipe samples in this study included press samples using the modified Edwards-Lioy sampler.
g CAT: child activity time line, developed as a visual, low-literacy diary for child location and activity.
h Cockroach enumeration: conducted before and after integrated pest management (IPM) activities to determine effectiveness of intervention.
i Hand wipes: children’s hands wiped or rinsed with isopropanol solution; requires skin removal efficiency information for interpretation.
j Aerial application: data on application rates, frequency, and duration of commercial pesticide applications near study community.
k Personal GPS: portable GPS units with data-logging capability suitable for studies of small children (Elgethun et al. 2003).
l Diet diaries: 3-day parental diary of all fresh produce (fruits and vegetables) and juices consumed by child, classified as either organic or conventional foods.
n Duplicate diet: representative portions of all foods consumed by child in a 24-hr period.
Table 2 Measured analytes in five biologic sample matrices.a
Study Maternal blood Cord blood Child blood Maternal urine Child Urine
Columbia University center OP insecticides, carbamate insecticides, pyrethroid insecticides, herbicides, fungicides, diethyltoluamide, organochlorine insecticides, PCBs, PAH-DNA, antioxidants, cotinine OP insecticides, carbamate insecticides, pyrethroid insecticides, herbicides, fungicides, diethyltoluamide, organochlorine insecticides, PCBs, PAH-DNA, antioxidants, cotinine, lead, mercury OP insecticides, carbamate insecticides, pyrethroid insecticides, herbicides, fungicides, diethyltoluamide, organochlorine insecticides, PCBs, PAH-DNA, antioxidants, cotinine OP DAP metabolites, specific OP metabolites, carbamate metabolites, pyrethroid metabolites, herbicides, other Collected at 36 and 60 months; stored for future analysis
University of California at Berkeley center Organochlorine insecticides, cholinesterase, PCBs, PON1 status, PBDEs (subset) Organochlorine insecticides, lead, cholinesterase, PCBs, PON1 status Lead OP DAP metabolites, OP-specific metabolites, carbamate metabolites, pyrethroid metabolites, herbicides, other OP DA metabolites, OP-specific metabolites
Mount Sinai Medical Center Organochlorine insecticides, cholinesterase, paraoxonase, PCBs, lead Cholinesterase, lead, paraoxonase Not collected OP DAP metabolites, OP-specific metabolites, pentachlorophenol, pyrethroid metabolites Collected; not yet analyzed
University of Washington center Not collected Not collected Not collected OP DAP metabolites OP DAP metabolites, OP-specific metabolites
PNASH center (University of Washington) Not collected Not collected Not collected Not collected OP DAP metabolites, OP-specific metabolites
Abbreviations: DAP, dialkylphosphate; PAH, polyaromatic hydrocarbon; PBDE, polybrominated diphenyl ether; PCB, polychlorinated biphenyl; PON1, paraoxonase .
a Specific analytes for chemical classes are as follows: OP insecticides (blood): chlorpyrifos, diazinon, dichlorvos, fonophos, malathion, methyl parathion, parathion, phorate, terbufos; carbamate insecticides and metabolites (blood): bendiocarb, carbofuran, propoxur, 2-isopropoxyphenol (propoxur metabolite), carbofuranphenol (carbofuran metabolite), 1-naphthol (naphthalene and carbaryl metabolite); pyrethroid insecticides (blood): trans-permethrin, cis-permethrin; herbicides (blood): acetochlor, alachlor, atrazine, chlorthal-dimethyl, metolachlor, trifluralin; fungicides (blood): chlorthalonil, dicloran, metalaxyl, captan metabolite, folpet metabolite; organochlorine insecticides (blood): p,p-dichlorodiphenyldichloroethylene, p,p-dichlorodiphenyltrichloroethane, o,p-dichlorodiphenyltrichloroethane, dieldrin, heptachlor epoxide, hexachlorobenzene, β/γ-hexachlorobenzene, mirex, oxychlordane, trans-nonachlor; OP DAP metabolites (urine): dimethylphosphate, dimethylthiophosphate, dimethyldithiophosphate, diethylphosphate, diethylthiophosphate, diethyldithiophosphate; OP-specific metabolites (urine): 3,5,6-trichloropyridinol (methyl/ethyl chlorpyrifos), 4-nitrophenol (methyl/ethyl parathion, ethyl p-nitrophenylbenzenethiophosphonate), malathion dicarboxylic acid, acephate, methamidaphos (acephate, methamidaphos), 2-isopropyl-4-methyl-6-hydroxypyrimidine (diazinon), hydroxycoumarin (coumaphos), pirimiphos methyl metabolite, isazaphos methyl metabolite, o-methoate, dimethoate; carbamate metabolites (urine): 2-isopropoxyphenol (propoxur metabolite), carbofuranphenol (carbofuran metabolite), 1-naphthol (naphthalene and carbaryl metabolite); pyrethroid metabolites (urine): 3-phenoxybenzoic acid, cis/trans-dichlorodimethylvinyl cyclopropane carboxylic acid, cis-dibromodimethylvinyl cyclopropane carboxylic acid, 4-phenoxybenzoic acid; herbicides or metabolites (urine): alachlor mercapturate, atrazine mercapturate, acetochlor mercapturate, 2,4-D, metolachlor mercapturate; others: o-phenylphenol, pentachlorophenol, 2,4-dichlorophenol, 2,5-dichlorophenol (paradichlorobenzene metabolite), 2,4,5-trichlorophenol, 2,4,6-trichlorophenol, ethylene thiourea, propylene thiourea.
Table 3 Percent participation of cohort members in biologic sampling procedures in four birth cohort studies.a
Study Baseline 26 weeks gestation Delivery 6 months 12 months 24 months
University of California at Berkeley birth cohortb n = 528 n = 528 n = 528 n = 477c n = 445c n = 425c
Maternal urine 99 94 94 93d — —
Paternal urine — — 51 — — —
Child urine — — — 88 91 90
Maternal blood — 81 — — — —
Cord blood — — 74 — — —
Child blood — — — — 64 81
Breast milk — — 63 93d — —
Mount Sinai Medical Center birth cohortb n = 479 — n = 404e — n = 215 n = 305
Maternal urine 91 — — — — —
Child urine — — — — 100 94
Maternal blood 90 — — — — —
Cord blood — — 59 — — —
Mount Sinai Medical Center IPM cohortb n = 184 — — — n = 112 n = 56f
Maternal urine 98 — — — 96 100
Child urine — — — — 84 82
Columbia University birth cohortb n = 588g n = 588g n = 449h
Maternal urine 82i — — — — —
Maternal blood — — 99 — — —
Cord blood — — 81 — — —
Meconium — — 51j — — —
Child blood — — — — — 71
— , no samples collected at those time periods.
a Percentages are calculated based on sample size provided for each study and time category; percent participation values are for participation in the biologic sampling procedures only and do not reflect retention rates for the cohorts.
b Berkeley cohort (CHAMACOS) recruited in Salinas Valley, California: n = 528 based on live births; total enrolled = 601 (Eskenazi et al. 2003, 2004). Mount Sinai birth cohort of primiparous pregnant women enrolled 1998–2003 (Berkowitz et al. 2003); Mount Sinai IPM cohort (Growing Up Healthy Integrated Pest Management Cohort; Brenner et al. 2003). Columbia birth cohort (Whyatt et al. 2003).
c Based on number of mothers participating rather than children due to several cases of twins.
d Based on number of women breast-feeding 6 months postpartum.
e Severnty-five women were excluded from follow-up for medical complications, very premature births (< 32 weeks gestation or < 1,500 g), delivery of an infant with birth defects, inability to obtain biologic specimens before delivery, change of residence, or refusal to continue to participate.
f Number of participants reached through the end of October 2003.
g Fully enrolled; subjects are considered fully enrolled once the prenatal monitorings and questionnaires had been completed and blood samples (from the mother and/or newborn) had been collected at delivery.
h Number of subjects currently enrolled at the time of the scheduled assessment whether or not the assessment was completed; subjects are dropped from the cohort if no contact is made for 1 year from the last scheduled assessment.
i A single urine sample is being collected from the mothers during pregnancy and is being stored for future analyses. Biweekly urine samples are being collected on a subset of 100 women beginning during the 32nd week of pregnancy through delivery and are being analyzed as indicated in Table 4.
j Collected from a subset of newborns under supplemental funding from the U.S. EPA STAR grant program.
Table 4 Percent participation of cohort members in biologic sampling procedures in three birth cohort studies.a
Study Prenatal 6 months 12 months 24 months
University of California at Berkeley birth cohortc n = 528 n = 473b n = 442b n = 422b
Home inspection/house dust 91 81 86 88d
Mount Sinai Medical Center IPM cohortc n = 184 — n = 112 n = 56
Air sample 100 — 100 100
Hand wipe 50 — 92 100
Toy wipe 75 — 96 100
Dust 96 — 100 100
Columbia University birth cohortc n = 588
48-hr personal air 100 — — —
2-week integrated indoor air 17e — — —
Kitchen dust samples 17e — — —
— , no samples collected at those time periods.
a Percentages are calculated based on sample size provided for each study and time category; percent participation values are for participation in the environmental sampling procedures only and do not reflect retention rates for the cohorts.
b Percent participation at 6, 12, and 24 months based on number of mothers participating rather than children due to several cases of twins.
c Berkeley cohort (CHAMACOS) recruited in Salinas Valley, California (Eskenazi et al. 2003, 2004); Mount Sinai IPM cohort (Growing Up Health Integrated Pest Management Cohort, 1999–2002; Brenner et al. 2003); Columbia birth cohort (Whyatt et al. 2003).
d Percentage permitting home visits at 24 months; no house dust collected.
e Collected from a subset of 100 homes beginning during the 32nd week of pregnancy and continuing through delivery; kitchen dust samples are also collected from a subset of homes.
==== Refs
References
Adgate JL Barr DB Clayton CA Eberly LE Freeman NC Lioy PJ 2001 Measurement of children’s exposure to pesticides: analysis of urinary metabolite levels in a probability-based sample Environ Health Perspect 109 583 590 11445512
Barr DB Barr JR Driskell WJ Hill RH Jr Ashley DL Needham LL 1999 Strategies for biological monitoring of exposure for contemporary-use pesticides Toxicol Ind Health 15 168 179 10188199
Berkowitz GS Obel J Deych E Lapinski R Godbold J Liu Z 2003 Exposure to indoor pesticides during pregnancy in a multiethnic, urban cohort Environ Health Perspect 111 79 84 12515682
Berkowitz GS Wetmur JG Birman-Deych E Obel J Lapinski RH Godbold JH 2004 In utero pesticide exposure, maternal paraoxonase activity, and head circumference. Environ Health Perspect 112 388 391
Boeniger MF Lowry LK Rosenberg J 1993 Interpretation of urine results used to assess chemical exposure with emphasis on creatinine adjustments: a review Am Ind Hyg Assoc J 54 615 627 8237794
Bradman A Barr DB Claus Henn BG Drumheller T Curry C Eskenazi B 2003 Measurement of pesticides and other toxicants in amniotic fluid as a potential biomarker of pre-natal exposure: a validation study Environ Health Perspect 111 1779 1782 14594631
Bradman AM Harnly ME Draper W Seidel S Teran S Wakeham D 1997 Pesticide exposures to children from California’s Central Valley: results of a pilot study J Expo Anal Environ Epidemiol 7 217 234 9185013
Bravo R Driskell WJ Whithead RD Jr Needham LL Barr DB 2002 Quantitation of dialkyl phosphate metabolites of organophosphorus pesticides in humans using GC-MS-MS with isotopic internal standards J Anal Toxicol 26 245 252 12166810
Brenner BL Markowitz S Rivera M Romero H Weeks M Sanchez E 2003 Integrated pest management in an urban community: a successful partnership for prevention Environ Health Perspect 111 1649 1653 14527845
Carlton EJ Moats HL Feinberg M Shepard P Garfinkel R Whyatt R 2004 Pesticide sales in low-income, minority neighborhoods J Community Health 29 231 244 15141898
Castorina R Bradman A McKone TE Barr DB Harnly ME Eskenazi B 2003 Cumulative organophosphate pesticide exposure and risk assessment among pregnant women living in an agricultural community: a case study from the CHAMACOS cohort Environ Health Perspect 111 1640 1648 14527844
Cohen Hubal EA Sheldon LS Burke JM McCurdy TR Berry MR Rigas ML 2000 Children’s exposure assessment: a review of factors influencing children’s exposure, and data available to characterize and assess that exposure Environ Health Perspect 108 475 486 10856019
Curl CL Fenske RA Elgethun K 2003a Organophosphorus pesticide exposure of urban and suburban pre-school children with organic and conventional diets Environ Health Perspect 111 377 382 12611667
Curl CL Fenske RA Elgethun K 2003b Organophosphate exposure: response to Krieger et al. and Charnley [Letter] Environ Health Perspect 111 A269 A270
Curl CL Fenske RA Kissel JC Shirai JH Moate TF Griffith W 2002 Evaluation of take-home organophosphorus pesticide exposure among agricultural workers and their children Environ Health Perspect 110 787 792
Denovan LA Lu C Hines CJ Fenske RA 2000 Saliva biomonitoring of atrazine exposure among herbicide applicators Int Arch Occup Environ Health 73 457 462 11057414
Dosemeci M Alavanja MC Rowland AS Mage D Zahm SH Rothman N 2002 A quantitative approach for estimating exposure to pesticides in the Agricultural Health Study Ann Occup Hyg 46 245 260 12074034
Elgethun K Fenske RA Yost MG Palcisko GJ 2003 Time-location analysis for exposure assessment studies of children using a novel global positioning system instrument Environ Health Perspect 111 115 122 12515689
Eskenazi B Bradman A Gladstone EA Jaramillo S Birch K Holland N 2003 CHAMACOS, a longitudinal birth cohort study: lessons from the fields J Children’s Health 1 3 27
Eskenazi B Harnley K Bradman A Weltzien E Jewell NP Barr DB 2004 Association of in utero organophosphate pesticide exposure and fetal growth and length of gestation in an agricultural population Environ Health Perspect 112 1116 1124 15238287
Fenske RA 1993 Dermal exposure assessment techniques Ann Occup Hyg 37 686 706
Fenske RA Kedan G Lu C Fisker-Andersen JA Curl CL 2002a Assessment of organophosphorus pesticide exposures in the diets of preschool children in Washington State J Expo Anal Environ Epidemiol 12 21 28 11859430
Fenske RA Lu C Barr DB Needham LL 2002b Children’s exposure to chlorpyrifos and parathion in an agricultural community in central Washington State Environ Health Perspect 110 549 553 12003762
Goldman L Eskenazi B Bradman A Jewell NP 2004 Risk behaviors for pesticide exposure among pregnant women living in farmworker households in Salinas, California Am J Ind Med 45 491 499 15164393
Gordon SM Nishioka MG Brinkman MC O’Rourke MK Lebowitz MD Moschandreas DJ 1999 Residential environmental measurements in the national human exposure assessment survey (NHEXAS) pilot study in Arizona: preliminary results for pesticides and VOCs J Exp Anal Environ Epidemiol 9 456 470
Hu Y Beach J Raymer J Gardner M 2004 Disposable diaper to collect urine samples from young children for pyrethroid pesticide studies J Expo Anal Environ Epidemiol 14 5 378 384 15361896
Hu YA Barr DB Akland G Melnyk LJ Needham LL Pellizzari ED 2000 Collecting urine samples from young children using cotton gauze for pesticide studies J Expo Anal Environ Epidemiol 10 6 pt 2 703 709 11138662
James RR Hern SC Robertson GL Schumacher BA 2003. Comparison of Methods for the Determination of Alkyl Phosphates in Urine. U.S. Environmental Protection Agency Report 600/R-03/075. Washington DC:Environmental Protection Agency, Office of Research and Development. Available: http://www.epa.gov/heasdweb/children/pdf/alkyl-phosphates-hern.pdf [accessed 15 July 2004].
Kissel JC Curl CL Kedan G Lu CA Griffith W Barr D 2005 Comparison of organophosphorus pesticide metabolite levels in single and multiple urine samples collected from pre-school children in Washington State J Expo Anal Environ Epidemiol 15 164 171 15187987
Koch D Lu C Jolley L Fisker-Andersen JA Fenske RA 2002 Temporal association of children’s pesticide exposure and agricultural spraying: report of a longitudinal biological monitoring study Environ Health Perspect 110 829 833 12153767
Lewis RG 2005. Residential post-application exposure monitoring. In: Occupational and Residential Exposure Assessment for Pesticides (Franklin CA, Worgan JP, eds). Sussex, UK:John Wiley and Sons, Ltd., 71–128.
Lu C Fenske RA Simcox NJ Kalman D 2000 Pesticide exposure of children in an agricultural community: evidence of household proximity to farmland and take home exposure pathways Environ Res 84 290 302 11097803
Lu C Kedan G Fisker-Andersen J Kissel JC Fenske RA 2004 Multi-pathway organophosphorus pesticide exposures of pre-school children living in agricultural and non-agricultural communities Environ Res 96 283 289 15364595
Lu C Knutson DE Fisker-Anderson J Fenske RA 2001 Biological monitoring survey of organophosphorus pesticide exposure among pre-school children in the Seattle metropolitan area Environ Health Perspect 109 299 303 11333193
Lu C Showlund-Irish R Fenske RA 2003 Biological monitoring of diazinon exposure using saliva in an animal model J Toxicol Environ Health 66 2315 2325
Needham LL Sexton K 2000 Assessing children’s exposure to hazardous environmental chemicals: an overview of selected research challenges and complexities J Expo Anal Environ Epidemiol 10 611 629 11138654
Perera FP Rauh V Tsai WY Kinney P Camann D Barr D 2003 Effects of transplacental exposure to environmental pollutants on birth outcomes in a multiethnic population Environ Health Perspect 111 201 205 12573906
Quackenboss JJ Pellizzari ED Shubat P Whitmore RW Adgate JL Thomas KW 2000 Design strategy for assessing multi-pathway exposure for children: the Minnesota Children’s Pesticide Exposure Study (MNCPES) J Expo Anal Environ Epidemiol 10 145 158 10791596
Royster MO Hilborn ED Barr D Carty CL Rhoney S Walsh D 2002 A pilot study of global positioning system/geographical information system measurement of residential proximity to agricultural fields and urinary organophosphate metabolite concentrations in toddlers J Expo Anal Environ Epidemiol 12 6 433 440 12415492
Shalat SL Donnelly KC Freeman NC Calvin JA Ramesh S Jimenez M 2003 Nondietary ingestion of pesticides by children in an agricultural community on the US/Mexico border: preliminary results J Expo Anal Environ Epidemiol 13 42 50 12595883
Simcox NJ Fenske RA Wolz S Lee I-C Kalman D 1995 Pesticides in housedust and soil: exposure pathways for children of agricultural families Environ Health Perspect 103 1126 1134 8747019
U.S. EPA 2004. US-Mexico Border 2012 Environmental Health Workgroup. Washington, DC:U.S. Environmental Protection Agency. Available: http://www.epa.gov/orsearth/ [accessed 15 July 2004].
Whyatt RM Barr DB 2001 Measurement of organophosphate metabolites in postpartum meconium as a potential bio-marker of prenatal exposure: a validation study Environ Health Perspect 109 417 420 11335191
Whyatt RM Barr DB Camann DE Kinney PL Barr JR Andrews HF 2003 Contemporary-use pesticides in personal air samples during pregnancy and blood samples at delivery among urban minority mothers and newborns Environ Health Perspect 111 749 756 12727605
Whyatt RM Camann DE Kinney PL Reyes A Ramirez J Dietrich J 2002 Residential pesticide use during pregnancy among a cohort of urban minority women Environ Health Perspect 110 507 514 12003754
Whyatt RM Rauh V Barr DB Camann DE Andrews HF Garfinkel R 2004 Prenatal insecticide exposure and birth weight and length among an urban minority cohort Environ Health Perspect 112 1125 1132 15238288
Wilson NK Chuang JC Iachan R Lyu C Gordon SM Morgan MK 2004 Design and sampling methodology for a large study of preschool children’s aggregate exposures to persistent organic pollutants in their everyday environments J Expo Anal Environ Epidemiol 14 260 274 15141155
|
16203262
|
PMC1281295
|
CC0
|
2021-01-04 23:41:28
|
no
|
Environ Health Perspect. 2005 Oct 24; 113(10):1455-1462
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.7674
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences 10.1289/ehp.7675ehp0113-00146316203263ResearchMini-MonographCommunity-Based Participatory Research: Lessons Learned from the Centers for Children’s Environmental Health and Disease Prevention Research Israel Barbara A. 1Parker Edith A. 1Rowe Zachary 2Salvatore Alicia 3Minkler Meredith 3López Jesús 4Butz Arlene 5Mosley Adrian 6Coates Lucretia 7Lambert George 8Potito Paul A. 9Brenner Barbara 10Rivera Maribel 1011Romero Harry 11Thompson Beti 12Coronado Gloria 12Halstead Sandy 131 University of Michigan School of Public Health, Ann Arbor, Michigan, USA2 Friends of Parkside, Detroit, Michigan, USA3 University of California at Berkeley, School of Public Health, Berkeley, California, USA4 California Rural Legal Assistance, Salinas, California, USA5 Department of General Pediatrics, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA6 Community Advisory Board member, Office of Community Health, Baltimore, Maryland, USA7 Principal, Dr. Bernard Harris Sr. Elementary School, President of Community Advisory Board, Baltimore, Maryland, USA8 Center for Childhood Neurotoxicology and Exposure Assessment, Robert Wood Johnson Medical School, University of Medicine and Dentistry of New Jersey, Piscataway, New Jersey, USA9 Executive Director, New Jersey Center for Outreach and Services for the Autism Community (COSAC), Ewing, New Jersey, USA10 Mount Sinai Center for Children’s Environmental Health and Disease Prevention Research, Department of Community and Preventive Medicine, Mount Sinai School of Medicine, New York, New York, USA11 Boriken Neighborhood Health Center, New York, New York, USA12 Fred Hutchinson Cancer Research Center, University of Washington, Seattle, Washington, USA13 U.S. Environmental Protection Agency Region 10, Prosser, Washington, USAAddress correspondence to B.A. Israel, University of Michigan School of Public Health, 1420 Washington Heights, Ann Arbor, MI 48109-2029 USA. Telephone: (734) 764-9494. Fax: (734) 763-7379. E-mail:
[email protected] authors declare they have no competing financial interests.
10 2005 24 6 2005 113 10 1463 1471 12 10 2004 13 6 2005 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright. Over the past several decades there has been growing evidence of the increase in incidence rates, morbidity, and mortality for a number of health problems experienced by children. The causation and aggravation of these problems are complex and multifactorial. The burden of these health problems and environmental exposures is borne disproportionately by children from low-income communities and communities of color. Researchers and funding institutions have called for increased attention to the complex issues that affect the health of children living in marginalized communities—and communities more broadly—and have suggested greater community involvement in processes that shape research and intervention approaches, for example, through community-based participatory research (CBPR) partnerships among academic, health services, public health, and community-based organizations. Centers for Children’s Environmental Health and Disease Prevention Research (Children’s Centers) funded by the National Institute of Environmental Health Sciences and U.S. Environmental Protection Agency were required to include a CBPR project. The purpose of this article is to provide a definition and set of CBPR principles, to describe the rationale for and major benefits of using this approach, to draw on the experiences of six of the Children’s Centers in using CBPR, and to provide lessons learned and recommendations for how to successfully establish and maintain CBPR partnerships aimed at enhancing our understanding and addressing the multiple determinants of children’s health.
children’s healthcollaborative researchcommunity-based participatory researchpartnership
==== Body
Over the past several decades there has been growing evidence of the increase in incidence rates, morbidity, and mortality for a number of health problems experienced by children—for example, asthma and other respiratory diseases (Landrigan et al. 2002; Mannino et al. 2002), developmental disabilities (Barone et al. 2000; Canfield et al. 2003), neuropsychologic disorders (Baldi et al. 2001; Schantz et al. 2003), and childhood cancers (Daniels et al. 1997). The causation and aggravation of these problems are complex and multifactorial, including genetic predisposition, demographic factors, psychosocial stressors, and environmental exposures. Numerous environmental exposures have been identified as contributing factors, including ambient levels of respirable particulate matter (Delfino et al. 2002; Eggleston 2000; Samet et al. 2000), ozone (Buchdahl et al. 2000; Mortimer et al. 2000), pesticides (Eskenazi et al. 2004; Landrigan et al. 2002; Perera et al. 2003), house dust mite and cockroach allergens (Litonjua et al. 2001; Sporik et al. 1999), and environmental tobacco smoke (Gergen et al. 1998; Gold 2000). The burden of these health problems and environmental exposures is borne disproportionately by children from low-income communities and communities of color (Evans and Kantrowitz 2002; Williams and Collins 1995). Recently, researchers and funding institutions have called for increased attention to the complex issues that affect the health of children living in marginalized communities (Schulz et al. 2002; Williams and Collins 1995), and communities more broadly, and have suggested greater community involvement in processes that shape research and intervention approaches, for example, through community-based participatory research (CBPR) partnerships among academic, health services, public health, and community-based organizations (CBOs) (Israel et al. 2003; Minkler and Wallerstein 2003; O’Fallon et al. 2000a). Each of the initial eight Centers for Children’s Environmental Health and Disease Prevention Research (Children’s Centers) funded by the National Institute of Environmental Health Sciences (NIEHS) and U.S. Environmental Protection Agency (EPA) was required to include a CBPR intervention project, and four additional Children’s Centers were subsequently funded (O’Fallon et al. 2000a). In all instances, the partners involved gained tremendous insights into how to conduct CBPR, and the challenges and benefits of using this approach to children’s environmental health research. The purpose of this article is to provide a definition and set of CBPR principles, to describe the rationale for and major benefits of using this approach particularly with environmental health research, to draw on the experiences of six of the Children’s Centers in using CBPR, and to provide lessons learned and recommendations for how to successfully establish and maintain partnerships aimed at enhancing our understanding and addressing the multiple determinants of children’s health.
Definition and Principles of CBPR
Definition of CBPR and community.
Within the field of public health, a number of partnership approaches to research have been called variously community-centered or community-based participatory/involved/collaborative research [for a review, see Israel et al. (1998)]. In addition, there is a large social science literature that has examined research approaches in which participants are actively involved in the process (e.g., Heron and Reason 2001; Jason et al. 2004; Kemmis and McTaggart 2000).
CBPR in public health is a partnership approach to research that equitably involves, for example, community members, organizational representatives, and researchers in all aspects of the research process, in which all partners contribute expertise and share decision making and responsibilities (Israel et al. 1998, 2003). The aim of CBPR is to increase knowledge and understanding of a given phenomenon and integrate the knowledge gained with interventions and policy change to improve the health and quality of life of community members (Israel et al. 1998, 2003). Within the context of CBPR, community is defined as a unit of identity. Units of identity refer to membership in, for example, a family, social network, or geographic neighborhood, and are socially created dimensions of identity (Steuart 1993). Community, as a unit of identity, is defined by a sense of identification and emotional connection to other members, common symbol systems, values and norms, shared interests, and commitment to meeting mutual needs (Steuart 1993). Communities of identity may be geographically bounded, for example, a neighborhood, or may be geographically dispersed, sharing a common identity (e.g., ethnic group, gays and lesbians). A city, town, or geographic area may include multiple overlapping communities of identity or may be an aggregate of individuals who do not have a common identity.
Principles of CBPR.
Based on an extensive review of the literature, Israel et al. (2003) have identified a list of nine principles or characteristics of CBPR: CBPR recognizes community as a unit of identify; builds on strengths and resources within the community; facilitates a collaborative, equitable partnership in all phases of the research, involving an empowering and power-sharing process that attends to social inequalities; fosters co-learning and capacity building among all partners; integrates and achieves a balance between knowledge generation and intervention for mutual benefit of all partners; focuses on the local relevance of public health problems and ecologic perspectives that recognize and attend to the multiple determinants of health; involves systems development using a cyclical and iterative process; disseminates results to all partners and involves them in the dissemination process; and involves a long-term process and commitment to sustainability. There is no one set of principles that will be applicable for all partnerships; rather, all partnerships need to jointly decide what their core values and guiding principles will be, drawing on those presented here, as appropriate. These principles can be considered to be on a continuum, with those listed here being an ideal goal to strive for (Green et al. 2003; Israel et al. 2003).
Benefits/Rationale for Using a CBPR Approach
As discussed in the literature, there are numerous benefits gained from using a CBPR approach (Israel et al. 1998; O’Fallon et al. 2000b). As reviewed elsewhere (Israel et al. 1998), among the key benefits are that it a) ensures that the research topic comes from, or reflects, a major concern of the local community; b) enhances the relevance and application of the research data by all partners involved; c) brings together partners with different skills, knowledge, and expertise to address complex problems; d) enhances the quality, validity, sensitivity, and practicality of research by involving the local knowledge of the participants; e) extends the likelihood of overcoming the distrust of research by communities that traditionally have been the “subjects” of such research; and f ) aims to improve health and well-being of the involved communities.
Overview of the Children’s Centers’ CBPR Partnerships
To better understand the key issues in establishing and maintaining CBPR partnerships based on the experiences of six of the Children’s Centers, in this section we provide a brief description of the community context and structure of community involvement in these centers. Each of the 12 Children’s Centers was invited to participate in the development of this article. Because of time constraints, 6 of the 12 centers were not able to participate. Therefore, the experiences and lessons learned discussed here represent the efforts of the six Children’s Centers described below. The methodology used in writing this article included identification of academic and community partners from each of the six centers to be co-authors; conduct of several conference calls involving co-authors from each of the Children’s Centers to determine major topic areas to be covered; preparation by each center of a written mini-case study covering the topics outlined by the co-authors (based on ongoing conversations and documentation within the respective partnerships and, in some instances, a more formal evaluation of the partnership); synthesis and integration written by the lead author of the strategies, lessons learned, and recommendations discussed in the case studies; and review of the manuscript and revisions made based on the input and perspectives of the co-authors across the six centers.
California Center for Children’s Environmental Health Research at the University of California, Berkeley (California/Salinas center).
Involving the predominantly Latino farmworker community in Salinas Valley, California, the California/Salinas center is a research partnership among the University of California at Berkeley, several state and federal agencies (the California Department of Health Services, the California Environmental Protection Agency, the Centers for Disease Control and Prevention), educational and research institutions (e.g., Stanford University, Battelle Laboratories), and numerous community agencies. The community partners, all within the state’s Salinas Valley in Monterey County, include Clínica de Salud del Valle de Salinas, Natividad Medical Center, South County Outreach Effort, California Rural Legal Assistance, the Grower-Shipper Association of Central California, and the Monterey County Health Department. The overall role of the partners is to advise center researchers in the development, implementation, analysis, and dissemination of culturally appropriate children’s environmental health research in the Salinas Valley.
The center has two advisory boards in the community, a community advisory board (CAB), which advises on all center studies, and an Intervention Farmworker Council (IFC), which was formed specifically to participate in the development and analysis of the intervention study. All partner organizations attend CAB meetings; however, the formal eight-member board itself is composed of representatives from three partner organizations and representatives of four additional organizations: the Monterey County Farm Bureau, the Monterey County Agricultural Commission, the Monterey County Board of Supervisors, and the California Assembly District 28. A representative from the seven-member IFC also sits on the CAB.
Maryland Center for Childhood Asthma in the Urban Environment, Johns Hopkins University (Maryland center).
Involving the primarily low-income African-American community in East Baltimore, Maryland, the Center for Childhood Asthma in the Urban Environment recruited community members to join a CAB. The CAB met monthly with the study team based at the Johns Hopkins University School of Medicine and Bloomberg School of Public Health. Separate meetings by the CAB were also held. Members of the CAB included two school principals, a pastor, a nun assigned to work in the community, two community association presidents, a parent of a child with asthma, health personnel who had worked in the community, and a clinical social worker. The role of the CAB was to provide community input to the research investigators regarding the construction of the control group, recruitment strategy, and data collection to ensure participants received benefit from their involvement in the study.
Michigan Center for the Environment and Children’s Health (Michigan center).
Involving a low-income, predominantly African-American community on the east side, and a low-income largely Latino community in Southwest Detroit, Michigan, the Michigan center is a CBPR partnership governed by a set of CBPR principles (Israel et al. 2001; Parker et al. 2003; Schulz et al. 1998). Community partners have been involved in all aspects and projects of the Michigan center, but they have been most involved with the Community Action Against Asthma (CAAA) research projects. The work of CAAA is guided by a steering committee (SC) composed of representatives from all of the partner organizations: the Detroit Department of Health and Wellness Promotion, the University of Michigan Schools of Public Health and Medicine, the Henry Ford Health System, and seven CBOs: Community Health and Social Services Center, Friends of Parkside, Warren-Conner Development Coalition, Latino Family Services, United Housing Coalition, Detroiters Working for Environmental Justice, and Detroit Hispanic Development Corporation. The SC has been actively involved in all major phases of the research and intervention, including the initial definition of the research questions, the design of all survey instruments, the hiring of key staff, the decision making on how to enroll and retain families in the intervention and study, and the interpretation, dissemination, and translation of research findings.
New Jersey Center for Childhood Neurotoxicology and Exposure Assessment, University of Medicine and Dentistry, New Jersey (New Jersey center).
The community involved in the New Jersey center is the autism community of New Jersey, New York, Pennsylvania, and Connecticut. Drawing from a well-developed and extensive network of autism-based advocacy, support, and research-oriented groups, community groups have been involved with the center from the start. The community-based group Community Outreach and Support of the Autism Community, which is in its 39th year of operation with 4,000 members, is one of the center’s main partners. The partnership involves the Autism Schools, Edens Family of Services, and Douglass Developmental Center of Rutgers University. The partners work with the center on developing the hypothesis, the protocol design, recruitment, outreach, and communications with the autism community of the states involved. The partners participate in, coordinate, and moderate the town meetings the center has with the autism community of New Jersey and other states. Partners are also on the external advisory board of the center.
New York Mount Sinai Center for Children’s Environmental Health and Disease Prevention Research (New York/East Harlem center).
Involving the predominantly African-American and Latino communities in East Harlem, located in northern Manhattan, New York, the New York/East Harlem center formed partnerships among the center’s principal investigators (PIs) and the leadership of two federally qualified community health centers (Boriken Neighborhood Health Center and Settlement Health). Both health centers are governed by boards whose members represent health care consumers and community residents. Medical school investigators and the community partners agreed from the onset that joint decision making and collaboration was needed to design the intervention and research protocols, select and hire field staff, provide oversight to field staff in study recruitment and conduct of the intervention, organize and sustain a CAB, and disseminate information and lessons learned to the local community and to policy makers. An SC composed of the executive director and/or associate director of the health center, a health center physician, the PI, and the project research coordinator was set up at each health center; representatives from the community partner sites attended monthly center meetings at the Mount Sinai School of Medicine.
A CAB composed of 20 active community stakeholders was formed and met semiannually to advise the researchers on dissemination of information and to help design broader community interventions intended to change both individual and institutional behaviors related to pesticide use and pest control. Members included tenant association leaders and members, housing managers, school teachers, parent association leaders, social service agencies, community health providers, and local elected officials.
Washington Center for Child Environmental Health Risks at the University of Washington (Washington center).
Involving the predominantly Hispanic farmworker community in 16 small towns and eight labor camps in lower Yakima Valley of eastern Washington State, the Washington center’s community project is a partnership composed of a variety of CBOs and individuals. Examples of such groups include the local farmworkers’ union, local farmworkers’ clinics, local department of agriculture, State Department of Health, Department of Labor and Industries, U.S. EPA district 10, Washington Growers’ League, farmworker advocates, farmworkers, health care providers, legal representatives, local newspapers, a Spanish-speaking radio station, and university extension offices. The partnership has been formalized into an 18-member CAB that is facilitated by a project coordinator hired from the community and by the CAB. Rules of the partnership emphasize interaction, respect, and the principle that all ideas are freely expressed and discussed. The CAB has been involved in the community project from the beginning, in a number of areas, including providing information regarding the concerns among local residents about pesticide exposure; participating in the design of the data-collection content and procedures, intervention design, recruitment and implementation, publication, and dissemination; and hiring of local staff. A member of the CAB also serves on the center’s external advisory committee.
Key Issues in Establishing and Maintaining CBPR Partnerships: Strategies, Lessons Learned, and Recommendations
Key Components of the Children’s Centers CBPR Partnerships
In keeping with the principles of CBPR listed above, a number of components or dimensions can be incorporated into CBPR partnerships. Table 1 provides a brief picture of how each of the Children’s Centers has addressed these components. An elaboration and analysis of some of these major components, lessons learned, and recommendations for conducting CBPR, based on the experiences of the centers, is provided below.
Definition of Community and Identification and Selection of Community Partners
A critical consideration in establishing a CBPR partnership is deciding how the “community” is defined, who represents the “community,” and how partners are selected (Israel et al. 2003; Koné et al. 2000).
Diverse approaches to definitions of community.
All but one of the Children’s Centers defined the community(ies) involved using geographic boundaries and common characteristics. In urban areas these were more neighborhood based (e.g., East Baltimore, East Harlem), whereas in rural areas the geographic boundaries were more spread out and included multiple small towns. Within each of these geographic communities, there were similar demographic and other characteristics (e.g., predominantly low income, African American, Latino, farmworkers). In addition, all of the communities experienced high incidence and prevalence of the particular environmental issue and/or health problem(s) that were the focus of the overall center. Furthermore, all of the communities have considerable strengths and assets (e.g., social networks, community organizations). Some of the centers involved smaller “communities of identity” (Steuart 1993), as defined above, such as a predominantly African-American neighborhood. Some of the Children’s Centers defined the community as a larger geographic area in which all of the stakeholders needed to be involved. For example, in the California/Salinas and Washington centers, both farmworkers and representatives from agricultural industry organizations were invited to participate. The New Jersey center defined the community as one that has children with autism and involves partners and participants from ethnically and economically diverse groups across several states.
One of the key principles of CBPR is that it recognizes community as a unit of identity and seeks to identify and work with existing communities of identity (Israel et al. 1998, 2003). This approach acknowledges that communities of identity have numerous individual and organizational skills and resources, but that they may also benefit from external skills and resources. Thus, CBPR partnerships may involve individuals and groups that are not members of the community of identity (Israel et al. 2003). For example, although a group of farmworkers might be most appropriately conceptualized as a community of identity for a CBPR effort, there may be some advantages of also including representatives from the agricultural industry, such as their potential role in policy change. When establishing a partnership, it is important to examine the advantages and disadvantages of extending membership beyond the community of identity at the outset. In the farm-worker community example, given the power differentials that exist between farmworkers and growers, the economic dependence of farmworkers, and the history of adversarial relations, it is critical to determine whether farmworkers will be comfortable expressing their opinions and whether their voice will be heard if growers are also at the table. One possible strategy is to start with the most immediate community of identity, that is, farmworkers, and after trust is established, and with their concurrence, bring additional parties into the process. Another strategy, used by the California/Salinas center, was to establish a separate group, the IFC, composed primarily of farmworkers, that was actively involved in the design and implementation of the intervention component of their center.
Different strategies for selection and identification of partners.
Several different strategies were used in the selection and identification of potential partners. A key aspect of several of the Children’s Centers’ approaches was building on prior positive working relationships that existed between academia and the communities involved. For example, the identification of community partners for the New York/East Harlem center was an outgrowth of > 25 years of collaboration between the academic and primary health center partners involved.
Similarly, the Michigan center evolved from an already existing community–academic partnership, the Detroit Community-Academic Urban Research Center (URC) (Israel et al. 2001; Lantz et al. 2001). In 1997, the URC board identified childhood problems related to the environment (e.g., asthma) as a priority area for future research and interventions, and subsequently when the request for proposals for the Children’s Centers was released, the URC board decided to apply. The Michigan center involves many of the original URC academic and community partners as well as new researchers and community organizations with expertise in asthma and/or the environment. Using a somewhat different approach, the New Jersey center selected autism advocacy groups or schools for children with autism that are regionally and nationally recognized by the autism community.
Another viable strategy for identifying and selecting partners is to conduct a community analysis to assess the values, needs, resources, barriers, and facilitators required for community action around an issue (Eng and Blanchard 1990–1991; Thompson et al. 2001). The Washington center conducted a community analysis to gain an increased understanding of the positions of the major participants or groups and to find common ground among the various parties involved. The results indicated a number of common themes as well as a wide disparity among groups in their views on pesticides. These were discussed with a community planning group, which recommended that because of the contention around pesticides, every constituent should be invited to participate in decision making (for more details, see Thompson et al. 2001).
Another consideration in selecting organizations as partners in a CBPR project is identifying who will represent the organization. To the extent possible, individuals who participate on CBPR boards need to be in leadership positions or have the authority to make decisions without always having to ask the leadership. At minimum, they need to have easy access to and the active and visible support of their organization’s leadership (Israel et al. 2001). Although in many instances it is ideal to have top leadership directly involved, such leaders are often constrained by other demands on their time and may be less able to actively participate. Another viable strategy is to have a designated representative and an alternate, with the alternate receiving all mailings and communications and attending meetings when the primary member can not.
Overall Role of Community Partners in CBPR Projects
One of the key concepts in conducting CBPR is the role of participation of the community members and researchers (Wallerstein and Duran 2003). Some of the core questions that need to be addressed include the following: What aspects of the CBPR process are community partners participating in? What level of influence or control do they have over the decisions made? What level of commitment do university partners have to creating an equitable partnership that attends to power differentials? There are a number of different ways in which community participation has been conceptualized, with the major similarity across these different perspectives being the concept of a continuum of control or power, ranging from the low end of the spectrum, where community members serve on advisory boards and have some limited involvement but little influence and control over the project, to the other end, where community members have full control over all aspects of the research process (Arnstein 1969; Balcazar et al. 2004). Not all CBPR partnerships will achieve the same level of community participation.
As shown in Table 1, four of the Children’s Centers have CABs (California/Salinas, Maryland, New York/East Harlem, and Washington) composed of representatives from highly diverse organizations. In most instances, the researchers and staff are not considered members of the CAB, although they frequently attend CAB meetings. Although the same “CAB” name is used across these four centers, the frequency of meetings, purpose of the CAB, and degree of community participation and control differ considerably and, in some instances, have changed over time. (See Table 1 for information on the frequency of meetings and facilitation of meetings.)
The experience of the Maryland center’s CAB shows how the role of community partners evolved over time. The partnership initially functioned to review study protocols and patient education material and assist in defining the target community. The CAB, composed of 10–14 members, was strictly advisory in nature and functioned within a limited sphere. The CAB expressed concerns about its role as being either too limited or too ambiguous because their opinions and input did not appear to influence the work of the research team. With the guidance of the CAB president, several strategies were developed (e.g., educational session, community tour, retreat) to assess the partnership and enhance the working relationships to mutually satisfying levels so that all could benefit. Through this process, the foundation was laid for increased collaboration and establishment of a shared culture. The CAB moved from “advisory” toward sharing “governance” of the project.
The Michigan center provides an example of another approach to organizing a CBPR partnership. The center is guided by an SC composed of representatives from academia, CBOs, and public health and health care institutions and one community member-at-large. The SC members were identified when the grant proposal was being written, with the size ranging from 14 to 17 members over the 5-year project period, and it has met monthly since the center was established. The meetings are co-facilitated by university faculty members at the initial request of the SC.
Role of Community Partners in Specific Stages of the Research Process
Community participation in and influence over each of the areas listed in Table 1 are considered to be a critical component of CBPR partnerships (Israel et al. 1998, 2003; Minkler and Wallerstein 2003). Ideally, any CBPR project involves community partners from the beginning stages, including defining the initial research priorities and questions. In responding to a call for proposals, this requires that either a partnership already exists or that time and resources be available to bring potential partners together to decide on these key issues. Unfortunately, this is often not the case, and researchers may have to approach potential community partners after decisions have already been made regarding the research priorities. It is important to identify partners who share an interest in the priorities selected, and considerable opportunity needs to be provided for input and decision making in subsequent stages of the research.
All of the Children’s Centers actively involved their community partners and greatly benefited from their participation in the design and implementation of the intervention research studies. Community partners can be instrumental in the overall study design. For example, the Maryland center CAB members voiced their concern that each participant be treated the same and receive immediate benefit from their participation, and under their advisement, the investigators changed the control group to a “treat later” group to ensure that all participants received the intervention. Community partners also provide valuable suggestions for specific intervention strategies—for example, a calendar contest in the schools.
Each of the Children’s Centers has greatly benefited from their community partners’ role in the development and implementation of data collection instruments. For example, the involvement of community partners and local staff in meetings and focus group interviews has provided information that resulted in more complete data collection and investigation of areas initially not included by the researchers, including both content and cultural appropriateness of language and methods (Edgren et al. 2005).
The community partners and local staff across all the Children’s Centers have played an active role in the design and implementation of recruitment and retention activities. Their input has been a significant factor in ensuring cultural and linguistic appropriateness and effectiveness in all written materials as well as in understanding the social, economic, political, and housing conditions in the communities involved that have an impact on participant involvement.
In some of the Children’s Centers the community partners were actively involved in guiding data collection activities. In particular, the hiring and training of local community members as data collectors provide the trust needed between the data collectors and respondents to enhance the quality and validity of the data.
The analysis and interpretation of data are areas in which community partners frequently have limited involvement. None of the Children’s Centers involved their community partners directly in data analysis. Given the time demands and technical aspects of data analysis, the lack of community involvement may be most appropriate. However, this may be an area in which community partners are interested in enhancing their skills, and thus, this needs to be discussed among the partners (Israel et al. 2003). What is crucial for all CBPR efforts is that the results of data analyses be fed back to the partners in ways that are understandable, and that the partners engage in a process of interpreting the data (Israel et al. 2003). Community partners are able to provide meaning to results that outside researchers may not have considered, for example, insights into the role of cultural dynamics and other contextual factors. The involvement of community partners in the interpretation of findings also has helped to increase community partners’ knowledge and comfort with research data and results. This has enabled all partners to share more equally in presenting results to study participants and in other settings.
As depicted in Table 1, there are a number of different ways in which community partners are involved in the dissemination of study findings—for example, presentations at meetings, publications, information booklets, newsletters, and radio announcements. Community partners should have the opportunity to be involved as co-authors and co-presenters on publications and presentations, to the extent that they are interested. Researchers need to recognize, however, that obtaining community partner involvement in this regard may require strategies such as face-to-face meetings and discussions of drafts rather than merely sharing written documents and expecting a written response.
To develop and maintain an effective CBPR partnership, and to increase understanding of the factors that contribute to successful partnerships, it is necessary to evaluate the CBPR partnership process, for example, to assess the extent to which and ways in which CBPR principles are followed (Israel et al. 2001, 2003; Lantz et al. 2001; Parker et al. 2003). Such an evaluation can include quantitative and qualitative methods and needs to involve all partners in the process and include regular feedback of results to make changes in how the partnership functions, as needed (Israel et al. 2003; Lantz et al. 2001; Parker et al. 2003; Schulz et al. 2003) (see, e.g., the evaluation conducted at the Michigan center by Parker et al. 2003).
Group Processes Involved
In keeping with the key principles of CBPR, it is critical that every partnership consider how it will strive to achieve shared equity, influence, and control over the decision-making process (Israel et al. 1998, 2003). This requires devoting considerable time and attention to the group’s process (Becker et al. 2005), which may be frustrating for some partners if it is perceived as taking time and resources away from the accomplishments of specific objectives (Israel et al. 2001, 2003; Lantz et al. 2001). A number of characteristics of effective groups are presented in the literature, such as two-way communication, appropriate decision-making procedures, shared power, the ability to resolve conflicts constructively, and the ability to engage the expertise of all members (Johnson and Johnson 2003). The extent to which CBPR partnerships pay attention to group dynamics and achieve these characteristics (i.e., process objectives) has implications for the group’s ability to achieve its short- and long-term goals (i.e., impact and outcome objectives) (Schulz et al. 2003).
The establishment by a partnership of operating norms and procedures that are in accordance with and reinforce the key principles of CBPR (Israel et al. 1998) is a key factor that attends to group dynamics issues through facilitating the trust and relationship building necessary to successfully conduct CBPR. These need to be consistent with the characteristics of effective groups mentioned above (Johnson and Johnson 2003) and to promote understanding and demonstrate competence in working with diverse cultures, for example, regarding class, gender, ethnicity, age, and sexual orientation (Israel et al. 1998). These norms and procedures need to be identified and agreed on by all the partners involved, documented in writing (they do not need to be as formal as by-laws, although they can be), and reviewed periodically to assess the extent to which they are being followed (Israel et al. 1998, 2001).
There is also considerable emphasis in the literature on the value of partnerships jointly developing overarching CBPR principles or core values (Israel et al. 2001), which also helps attend to group dynamics issues. The Maryland center CAB spent several CAB meetings to identify, define, and adopt its core values, which include cultural competence and inclusiveness, meaning that partners recognize, accept, and celebrate their differences and community perspectives are included and valued; and effective and open communication among partners including recognition of participants’ right to know study findings. The New York/East Harlem center’s guiding principles for shared decision making and power sharing between the research institution and the health centers include joint selection of field staff with an emphasis on hiring from the community, and full review and agreement on research protocols, data collection instruments, recruitment and retention strategies, and educational materials. The California/Salinas center’s guiding principles include giving back more to the community than is taken, being culturally sensitive and appropriate, sharing decision making, and providing long-term and sustainable resources to the community.
Compensation for Community Partners
As indicated in Table 1, a range of approaches were used by the Children’s Centers regarding compensating community partners for their involvement. The emphasis on equity as a key principle of CBPR underscores the importance of addressing this issue. The extent and amount of compensation need to be considered by each partnership in the context of the level of involvement (e.g., annual meetings compared with monthly meetings) and by type of organization (e.g., members from agricultural industry and health care systems, compared with farmworkers and CBOs). Although it may not be possible to fully compensate community partners monetarily for the time they contribute to the partnership, adequate recognition of and compensation for their contributions should be provided. In addition to providing direct financial resources and coverage of travel expenses, this could take the form of technical assistance and training. For example, in the New Jersey center, the community-based partners did not want to have any financial ties to the study to ensure their independence, although compensation did occur through the center’s provision of information and assistance with fund raising. The issue of equity can also be considered in terms of resources provided to the community at large—for example, hiring local community members and providing services such as health information at local health and work fairs in the community. The process for deciding how to handle compensation needs to be joint, open, and transparent.
Staff Hired from the Local Community
Another key factor has been the establishment of field offices in the community, and the hiring of local community members as staff who are similar to the project participants (e.g., culture, language). Although setting up a field office is particularly important when the research institution is not located in the community in which the project is involved, it is also worth considering when the academic institution that is within the community is perceived as having limited access or being inhospitable. Across the Children’s Centers, the staff positions for which community members have been hired have included field coordinators, interviewers, other data collectors (e.g., air quality monitoring), and intervention staff (e.g., outreach workers). In some instances, local staff were hired as employees of a community partner organization, whereas in other cases local staff were hired as employees of the academic institution involved. Local staff have played a crucial role in all phases of the projects (e.g., providing feedback on study protocols and data collection instruments, and problem solving implementation issues that arise). Local staff in the California/Salinas, Maryland, Michigan, and Washington centers have been the day-to-day “face” of the project in the community and have provided a bridge among the researchers, community partners, intervention participants, and community members-at-large. This regular interaction has been crucial for building and maintaining the trust necessary to obtain the input needed to conduct culturally appropriate and high-quality CBPR projects. Although some local staff had prior experience working in research and interventions, in other instances relevant training was provided.
Challenges of Using a CBPR Approach for Children’s Environmental Health Research
Some of the major challenges associated with using CBPR that were faced by the Children’s Centers are presented briefly below. Some strategies for overcoming these challenges are presented in the preceding section, and others are discussed further below in the context of overarching lessons learned.
Costs incurred and lack of resources.
There are numerous costs for both community and academic partners involved in CBPR efforts and insufficient resources for overcoming them (Israel et al. 1998; Koné et al. 2000; Minkler 2004). An effective partnership requires time and infrastructure support, for example, to establish and maintain trust, attend meetings, jointly participate in all phases of the research, and foster capacity building. Community partner organizations face financial costs from involvement, such as lack of adequate reimbursement for their time spent participating, as well as opportunity costs for time taken away from other job responsibilities (Koné et al. 2000; Parker et al. 2003). Research investigators are also constrained by the time and costs required (Parker et al. 2003).
Institutional constraints.
Many institutional constraints are faced in conducting CBPR (Israel et al. 1998). Among the challenges faced by the Children’s Centers are university institutional review board (IRB) processes that do not take into account the needs of CBPR projects (e.g., the need to be flexible and revise protocols based on community input), overhead issues, long delays associated with data analysis and returning results to the community, and hiring policies that require traditional job descriptions and educational degrees. Community partners, many of whom are not paid by the project and have numerous other professional responsibilities, may not be supported by their supervisors if their involvement is perceived to be taking time away from other organizational responsibilities.
Lack of trust and respect: institutional history.
Building and maintaining trust both between the university and community as well as at times within community partners are a substantial challenge (Israel et al. 1998; Minkler 2004). For example, when diverse groups of stakeholders are brought together who have a long and adversarial history, such as those representing farmworker and agricultural industry interests, as was the case in California/Salinas, this can present serious difficulties for the partnership. Some key questions that need to be asked here include the following: Is the trust of the board being compromised by trying to bring too many interests to the table? Are CBPR partnerships the appropriate entity to try to bridge longstanding and political tensions that may exist? Does the participation of “all” stakeholders really promote the support of study results and the future translation of findings into policy?
Ensuring community participation and influence.
Related to time constraints and costs, another challenge faced by CBPR partnerships is ensuring community participation and influence (Green and Mercer 2001; Israel et al. 1998; Minkler 2004). Community building is a very important and often overlooked step in building a “collaborative, equitable” partnership, which requires skill and takes time and commitment on the part of all partners to foster participation and shared decision making.
Lack of training and experience in conducting CBPR.
Another challenge is that many researchers and community partners have limited training and experience in conducting CBPR. Although there is a large and growing literature on how to carry out CBPR efforts (Minkler and Wallerstein 2003), many researchers and community partners have not received direct training and have limited opportunity to engage in learning opportunities to strengthen their skills in this area. This is particularly challenging in situations such as the Children’s Centers, where community involvement was a requirement from the funding institutions, and not all researchers fully understood what the implications of that meant.
Different emphasis on goals, values, priorities, and perspectives.
There are a number of areas where community and academic partners may differ in their emphasis on goals, values, priorities, and perspectives (Israel et al. 1998). For example, in several Children’s Centers, community partners were eager to implement the interventions and disseminate preliminary results, whereas researchers were concerned that the premature dissemination of results would contaminate study findings and lead to scientific criticism and consequences for publications and future funding. Challenges also occur given that members of partnerships have, for example, different values, beliefs, and cultures (Israel et al. 1998; Minkler 2004). Importantly, these various differences do not suggest a “right” or “wrong” way that partnerships should operate; rather, they suggest the need to consider and accommodate diverse perspectives.
Different languages and styles of communication.
Another challenge is that members of CBPR partnerships speak different languages and use different styles of communication. One difference that several of the Children’s Centers faced was that most members speak English whereas some speak Spanish. This creates challenges in terms of conducting bilingual meetings, having all materials in Spanish as well as English, and ensuring participation from predominantly Spanish-speaking members. In addition, researchers often use scientific words and language that are not easily understandable, and community partners may use words and colloquialisms that scientists do not understand. Furthermore, researchers at the Children’s Centers often use electronic mail for communicating, frequently needing/expecting quick responses, and some community partners do not have jobs that enable them to be in such frequent email contact, and others do not use email at all.
Overarching Lessons Learned and Recommendations
Throughout this article, we have shared the experiences of the Children’s Centers in using a CBPR approach and provided lessons learned and explicit as well as implicit recommendations for how to conduct CBPR. Building on these, in this section we present several overarching lessons learned and recommendations.
Sufficient time, resources, and benefits are needed for all partners to ensure active and meaningful participation.
Considerable commitment and time are needed to establish and maintain trust.
Jointly developing and following operating norms and CBPR principles/core values are essential.
Acknowledging and addressing power and equity issues are critical.
Funding and academic institutions need to extend their criteria for research excellence and productivity (e.g., the randomized control trial in which one group receives no intervention may not always be feasible or desirable within a CBPR context) and be flexible to incorporate the input of community partners (e.g., IRB review and approval processes).
Commitment to translating research findings into interventions and policies is of utmost importance.
Hiring and training staff from the local community are essential.
Recognizing, respecting, and embracing different cultures of the partners and partner organizations are imperative for successful CBPR efforts.
Concluding Remarks
CBPR is an especially useful approach for working with marginalized communities that experience a disproportionate burden of environmental, health, and other problems and that typically have not been included in deciding what types of research and interventions are most appropriate for and likely to be most effective in their communities. Although it is neither possible nor appropriate to use CBPR in all research studies, other research approaches may benefit from incorporating some of the principles and strategies recommended throughout this article.
With the NIEHS and the U.S. EPA providing the notable exceptions, most organizations supporting health research, especially basic research (e.g., epidemiologic, genetic), do not require researchers to work with communities in the identification, design, implementation, analysis, and dissemination of research. The NIEHS/U.S. EPA’s emphasis on community–academic partnerships has encouraged researchers conducting health effects and exposure research, in addition to those conducting intervention research, to develop such partnerships and to orient their research in ways they previously had not. We hope that the experiences and benefits gained from these Children’s Centers’ partnerships will provide guidance and encouragement to the National Children’s Study and others to incorporate similar CBPR approaches to address environmental and children’s health issues.
This article is part of the mini-monograph “Lessons Learned from the National Institute of Environmental Health Sciences/U.S. Environmental Protection Agency Centers for Children’s Environmental Health and Disease Prevention Research for the National Children’s Study.”
We thank our many colleagues and partners involved in the efforts described here who made these community-based participatory research partnerships possible. We thank S. Andersen for her assistance in the preparation of the manuscript.
We acknowledge with appreciation the support provided by the National Children’s Study for the development of this article. Funding also comes partially from the National Institute of Environmental Health Sciences (grants ES09589, ES011256, ES009601, ES009606, ES009584, ES009605) and the U.S. Environmental Protection Agency (grants R826710, R829391, R826886, R826724, R827039, R826709).
Table 1 Key components of Children’s Centers CBPR partnerships.
Center location
California/Salinasa New York/East Harlemb
Component CAB IC Maryland Michigan New Jersey BNHC SH Washington
Intervention study design
Group randomized controlled trial X Xc X X X
Randomized staggered controlled trial X
Intervention participants Xd
Predominantly low income X X X X X
African American X X X
Latino/Hispanic X X X X
White non-Hispanic X
Partnership title
CAB X X X X
SC X
Intervention council X
IPO X
Members/organizational representatives involved in CAB, SC, intervention council, and IPO
Individual community members X X X X X X
CBOs X X X X X X
Faith-based organizations X X
Local health department X X X X
Community health center/health personnel X X X X X X
Hospitals/integrated care systems X X X
University X X X X
Other governmental agencies (e.g., schools, social service) X X X X X X
Business/industry X X X
Others attend meetings (e.g., staff, faculty) X X X NA X X X
Other organizationse X X X
No. of board/committee members 8 7 10–14 14–17 5 20 18
Frequency of meetings
Monthly X X Xf Xg Xh
Bimonthly X X Xh
Quarterly Xh
Semiannually Xi Xh
Annually X
Location of meetings
Clinic/medical center in community X X X X X
Rotate among community partner organizations X X
Neighborhood school X
Facilitator of meetings
Project staff X X X
Researchers/faculty members X X X
Community members X
Staff and community member co-facilitate X X
Role of community partners in different stages of research/activities
Define initial research questions/priorities X X X
Design/implementation of research/intervention X X X X X X X X
Development of data collection instruments/protocols X X X X X X X
Hire staff X X X X X
Recruitment of participants X X X X X X
Retention X X X X X X NA
Review/comment educational and feedback materials X X X X X X X
Data collection X X X
Data analysis X
Data interpretation X X X X
Dissemination
Review/provide feedback X X X
Scientific papers X X X
Co-present professional meetings X X X X
Co-present community forums/meetings X X X X X
Co-author journal articles/book chapters X X X X
Review/comment newsletters/flyers X X X X X X X X
Input on website development X X
Evaluation of partnership X X X X
Development of additional research proposals/projects X X X X X
Provide entrée/linkages with other community organizations X X X X X
Group processes
Operating norms/ground rules X X X X X X
CBPR/guiding principles/core values X X X X X X X
Dissemination principles X X X X X
Publication review protocol X NA NA
Community partner compensation for participation
Honorarium to organizations X
Honorarium/reimbursement to individuals Xj X X X
Subcontract for services X X X X
Percent of administrative overhead X X
No compensation Xj Xk X
Communication outside of meetings
Minutes X X X X X
Mailings X X X X
E-mail X X X X X
Fax X X X X X
Telephone X X X X X X
In-person meetings X X X X X X
Staff hired from local community
Field coordinator X X X X
Interviewers X X X X
Other data collectors (e.g., home inspection) X X X X
Intervention staff X X X X
Abbreviations: BNHC, Boriken Neighborhood Health Center; IC, intervention council; IPO, individual partner associations; SH, Settlement Health.
a Eight-member CAB developed after funding received to be involved in overall center activities. After 3 years, additional IFCs established to advise center on intervention-related activities.
b Two partnerships were established, one with BNHC at the beginning of the project, and one with SH at the end of the second year, both federally qualified community health centers. The information in this table applies primarily to these two partner organizations. In addition, a CAB composed of 20 active community stakeholders was established by the researchers and two partner organizations and meets semiannually to advise researchers on the translation of results and to provide feedback during the process of the study. Members of the CAB are indicated on the table, but additional information in the table does not apply to the role of the CAB.
c Over time, under advisement of CAB, control group changed to “treat later” group.
d The participants are approximately representative of the demographics of the states involved (i.e., New Jersey, New York, Pennsylvania, Connecticut).
e Examples of other organization members include legal assistance, farm bureau, and agricultural commission.
f Started with monthly meetings for the first 3 years. As recruitment and intervention phase ended, meetings became less frequent.
g Monthly meetings were recommended but did not occur. Most decisions were made by leaders of the partner organizations on an as-needed basis, via the telephone and face-to-face contact.
h Started with monthly meetings, after first year moved to bimonthly and subsequently quarterly, then semiannually.
i Meetings have been on an annual basis with additional feedback provided through subcommittee meetings and one-on-one communications. Meetings currently being conducted semiannually.
j Honorarium provided for one member who missed work time to attend annual meeting; other members were not compensated for their attendance.
k Members of the center actively participate in many activities of the community partners, including fund raising activities and multiple presentations to the community partners on topics such as autism, children’s development, and the effects of environmental exposure.
==== Refs
References
Arnstein SR 1969 A ladder of citizen participation J Am Inst Planners 35 216 224
Balcazar FE Taylor RR Keilhofner GW Tamley K Benziger T Carlin N 2004. Participatory action research: general principles and a study with a chronic health condition. In: Participatory Community Research: Theories and Methods in Action (Jason LA, Keys CB, Suarez-Balcazar Y, Taylor RR, Davis MI, eds). Washington, DC:American Psychological Association, 17–36.
Baldi I Filleul L Mohammed-Brahim B Fabrigoule C Dartigues J-F Schwall S 2001 Neuropsychologic effects of long-term exposure to pesticides: results from the French Phytoner study Environ Health Perspect 109 839 844 11564621
Barone S Jr Das KP Lassiter TL White LD 2000 Vulnerable processes of nervous system development: a review of markers and methods Neurotoxicology 21 1–2 15 36 10794382
Becker AB Israel BA Allen AJ 2005. Strategies and techniques for effective group process in community-based participatory research partnerships. In: Methods in Community-Based Participatory Research for Health (Israel BA, Eng E, Schulz AJ, Parker E, eds). San Francisco, CA:Jossey-Bass, 52–72.
Buchdahl R Willems CD Vander M Babiker A 2000 Associations between ambient ozone, hydrocarbons, and childhood wheezy episodes: a prospective observational study in south east London Occup Environ Med 57 86 93 10711275
Canfield RL Henderson CR Cory-Slecht DA Cox C Jusko TA Lanphear BP 2003 Intellectual impairment in children with blood lead concentrations below 10 ug/dL New Engl J Med 348 1517 1526 12700371
Daniels JL Olshan AF Savitz DA 1997 Pesticides and childhood cancers Environ Health Perspect 105 1068 1077 9349828
Delfino RJ Zeiger RS Seltzer JM Street DH McLaren CE 2002 Association of asthma symptoms with peak particulate air pollution and effect modification by anti-inflammatory medication use Environ Health Perspect 110 A607 A617 12361942
Edgren KK Parker EA Israel BA Lewis TC Salinas M Robins TG 2005 Conducting a health education intervention and an epidemiological research project involving community members and community partner organizations: the Community Action Against Asthma Project Health Promot Pract 6 3 263 269 16020621
Eggleston PA 2000 Environmental causes of asthma in inner city children: the National Cooperative Inner City Asthma Study Clin Rev Allergy Immunol 18 311 324 10981263
Eng E Blanchard L 1990–1991 Action-oriented community diagnosis: a health education tool Int Q Commun Health Educ 11 93 110
Eskenazi B Harley K Bradman A Weltzien E Jewell N Barr D 2004 Association of in utero organophosphate pesticide exposure and fetal growth and length of gestation in an agricultural population Environ Health Perspect 112 1116 1124 15238287
Evans GW Kantrowitz E 2002 Socioeconomic status and health: the potential role of environmental risk exposure Annu Rev Public Health 23 303 331 11910065
Gergen PJ Fowler JA Maurer KR Davis WW Overpeck MD 1998 The burden of environmental tobacco smoke exposure on the respiratory health of children 2 months through 5 years of age in the United States: Third National Health and Nutrition Examination Survey, 1988 to 1994 Pediatrics 101 e8 Available: http://pediatrics.aappublications.org/cgi/content/full/101/2/e8 [accessed 17 August 2005].9445518
Gold DR 2000 Environmental tobacco smoke, indoor allergens, and childhood asthma Environ Health Perspect 108 643 651 10931782
Green LW George MA Daniel M Frankish CJ Herbert CP Bowie WR 2003. Guidelines for participatory research in health promotion. In: Community-Based Participatory Research for Health (Minkler M, Wallerstein N, eds). San Francisco, CA:Jossey-Bass, 27–52.
Green LW Mercer SL 2001 Can public health researchers and agencies reconcile the push from funding bodies and the pull from communities? Am J Public Health 91 1926 1929 11726367
Heron J Reason P 2001. The practice of cooperative inquiry: research “with” rather than “on” people. In: Handbook of Action Research: Participative Inquiry and Practice (Reason P, Bradbury H, eds). Thousand Oaks, CA:Sage, 179–188.
Jason LA Keys CB Suarez-Balcazar Y Taylor RR Davis MI eds. 2004. Participatory Community Research: Theories and Methods in Action. Washington, DC:American Psychological Association.
Israel BA Lichtenstein R Lantz P McGranaghan R Allen A Guzman JR 2001 The Detroit Community-Academic Urban Research Center: development, implementation, and evaluation J Public Health Manag Pract 7 1 19 11680026
Israel BA Schulz AJ Parker EA Becker AB 1998 Review of community-based research: assessing partnership approaches to improve public health Annu Rev Public Health 19 173 202 9611617
Israel BA Schulz AJ Parker EA Becker AB Allen A Guzman JR 2003. Critical issues in developing and following community-based participatory research principles. In: Community-Based Participatory Research for Health (Minkler M, Wallerstein N, eds). San Francisco, CA:Jossey-Bass, 56–73.
Johnson DW Johnson FP 2003. Joining Together: Group Theory and Group Skills. Boston, MA:Allyn and Bacon.
Kemmis S McTaggart R 2000. Participatory action research. In: Handbook of Qualitative Research (Denzin NK, Lincoln YS, eds). Thousand Oaks, CA:Sage, 567–605.
Koné A Sullivan M Senturia KD Chrisman NJ Ciske SJ Krieger JW 2000 Improving collaboration between researchers and communities Public Health Rep 115 243 248 10968762
Landrigan PJ Schechter CB Lipton JM Fahs MC Schwartz J 2002 Environmental pollutants and disease in American children: estimates of morbidity, mortality, and costs for lead poisoning, asthma, cancer, and developmental disabilities Environ Health Perspect 110 721 728 12117650
Lantz P Viruell-Fuentes E Israel BA Softley D Guzman JR 2001 Can communities and academia work together on public health research? Evaluation results from a community-based participatory research partnership in Detroit J Urban Health 78 495 507 11564852
Litonjua AA Carey VJ Burge HA Weiss ST Gold DR 2001 Exposure to cockroach allergen in the home is associated with incident doctor-diagnosed asthma and recurrent wheezing J Allergy Clin Immunol 107 41 47 11149989
Mannino DM Homa DM Akinbami LJ Moorman JE Gwynn C Redd SC 2002 Surveillance for asthma—United States, 1980–1999 MMWR Surveill Summ 51 1 13
Minkler M 2004 Ethical challenges for the “outside” researcher in community-based participatory research Health Educ Behav 31 6 684 697 15539542
Minkler M Wallerstein N eds. 2003. Community-Based Participatory Research for Health. San Francisco, CA:Jossey-Bass.
Mortimer KM Tager IB Dockery DW Neas LM Redline S 2000 The effect of ozone on inner-city children with asthma Am J Respir Crit Care Med 162 1838 1845 11069823
O’Fallon LR Tyson F Dearry A 2000a Improving public health through community-based participatory research and education Environ Epidemiol Toxicol 2 201 209
O’Fallon LR Tyson F Dearry A 2000b. Executive summary. In: Successful Models of Community-Based Participatory Research: Final Report (O’Fallon LR, Tyson FL, Dearry A, eds). Research Triangle Park, NC:National Institute of Environmental Health Sciences, 1–3.
Parker EA Israel BA Brakefield-Caldwell W Keeler GJ Lewis TC Ramirez E 2003 Community Action against Asthma: Examining the Partnership Process of a Community-Based Participatory Research Project J Gen Intern Med 18 558 567 12848839
Perera FP Rauh V Tsai WY Kinney P Camann D Barr D 2003 Effects of transplacental exposure to environmental pollutants on birth outcomes in a multiethnic population Environ Health Perspect 111 201 206 12573906
Samet JM Dominici F Curriero FC Ciyrsac I Zeger SL 2000 Fine particulate air pollution and mortality in 20 US cities, 1987–1994 N Engl J Med 343 1742 1749 11114312
Schantz SL Widholm JJ Rice DC 2003 Effects on PCBs exposure on neuropsychological function in children Environ Health Perspect 111 357 376 12611666
Schulz AJ Israel BA Lantz P Instrument for evaluating dimensions of group dynamics within community-based participatory research partnerships Eval Program Plann 26 3 249 262
Schulz AJ Israel BA Selig SM Bayer IS Griffin CB 1998. Development and implementation of principles for community-based research in public health. In: Research Strategies for Community Practice (MacNair RH, ed). New York:Haworth Press, 83–110.
Schulz AJ Williams DR Israel BA Lempert LB 2002 Racial and spatial relations as fundamental determinants of health in Detroit Milbank Q 80 677 707 12532644
Sporik R Squillace SP Ingram JM Rakes G Honsinger RW Platts-Mills TAE 1999 Mite, cat, and cockroach exposure, allergen sensitisation, and asthma in children: a case-control study of three schools Thorax 54 675 680 10413718
Steuart GW 1993 Social and cultural perspectives: community intervention and mental health Health Educ Q 20 suppl 1 99 111
Thompson B Coronado G Puschel K Allen E 2001 Identifying constituents to participate in a project to control pesticide exposure in children of farmworkers Environ Health Perspect 109 443 448 11427394
Wallerstein N Duran BM 2003. The conceptual, historical, and practice roots of community-based participatory research and related participatory traditions. In: Community-Based Participatory Research for Health (Minkler M, Wallerstein N, eds). San Francisco, CA:Jossey-Bass, 27–52.
Williams DR Collins C 1995 US socioeconomic and racial differences in health: patterns and explanations Annu Rev Sociol 21 349 386
|
16203263
|
PMC1281296
|
CC0
|
2021-01-04 23:41:27
|
no
|
Environ Health Perspect. 2005 Oct 24; 113(10):1463-1471
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.7675
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0064616203219PerspectivesEditorialEditorial: Children’s Health and the Environment: A Transatlantic Dialogue Landrigan Philip J. Department of Community & Preventive Medicine, Mount Sinai School of Medicine, New York, New York, E-mail:
[email protected] Giorgio Institute of Child Health, IRCCS Burlo Garofolo, Trieste, ItalyPhilip J. Landrigan is the Ethel H. Wise Professor and chair of the Department of Community and Preventive Medicine, and director of the Center for Children’s Health and the Environment at the Mount Sinai School of Medicine.
Giorgio Tamburlini is a pediatrician and a child public health expert. He has been involved in research and development in the field of child public health at international level since 1990, as a consultant for WHO and other international agencies. He is currently Scientific Director of the Institute of Child Health “Burlo Garofolo” in Trieste, Italy.
The authors declare they have no competing financial interests.
10 2005 113 10 A646 A647 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
Important steps to protect children against environmental threats to health have been taken over the past decade in both the United States and Europe This progress is based on the shared recognition that infants and children are very different from adults in their exposures and their susceptibility to toxic chemicals and that they therefore require special protections in risk assessment, regulation, and law. But despite this common scientific foundation, developments in children’s environmental health (CEH) on the two sides of the Atlantic have been quite different. These contrasting and sometimes complementary advances reflect the differing social, legal, and regulatory cultures of the two continents.
Recognition among policy makers of the unique vulnerability of children had its origins in the United States and dates from the publication in 1993 of the National Research Council (NRC) report Pesticides in the Diets of Infants and Children (National Research Council 1993). This report found striking differences between children and adults in exposure as well as in susceptibility to toxic chemicals. The report identified large gaps in regulatory practice and called for expansion of toxicologic testing to assess threats to development. It also urged reform of risk assessment and regulation to enhance protection of children. The central contribution of the NRC report was to elevate consideration of the vulnerability of children from the specialized area of pediatrics to the broad realm of national policy formulation.
The recommendations of the NRC report were incorporated into federal policy in the United States in 1996 through the Food Quality Protection Act (FQPA), the principal U.S. statute governing use of pesticides (FQPA 1996). The FQPA affirms the unique vulnerability of children. It requires explicit consideration of children in risk assessment and mandates child-protective safety factors in regulation. These principles were reaffirmed in April 1997 in an Executive Order on Children’s Environmental Health and Safety requiring all agencies of the U.S. government to consider children’s health and safety in all policy decisions (Clinton 1997). This U.S. experience was shared internationally and led to promulgation in May 1997 of the Miami Declaration on Children’s Environment Health, a declaration approved unanimously by the environmental ministers of the G-8 nations (Environment Leaders' Summit of the Eight 1997).
In recent years, CEH policy development has slowed in the United States, but the nation continues to make important contributions in research and medical practice. These advances date from 1998, when the National Institute of Environmental Health Sciences and the U.S. Environmental Protection Agency established a national network of Centers for Children's Environmental Health and Disease Prevention Research, now 11 in number. These interdisciplinary centers have matured into strong generators of scientific knowledge, they have produced unprecedented gains in environmental pediatrics, and they have been effective incubators of the careers of young scientists and physicians who will become the next generation of leaders in CEH.
Research in the centers has made important contributions to understanding of the environmental causes of asthma, neurobehavioral disorders, endocrine dysfunction, autism and low-level lead toxicity. Findings from this work are already guiding disease prevention. Reports highlighting the accomplishments of the centers are presented in this issue of Environmental Health Perspectives.
The National Children’s Study (NCS) is a second emerging research development in the United States (National Children’s Study 2005). The NCS is a prospective epidemiologic study that will follow 100,000 children—a statistically representative sample of all children born in the United States—from (or before) conception to 21 years of age. The goal is to identify the factors in the environment—chemical, biologic, physical, and psychosocial—that alone or in combination influence children’s health, growth, development, and risk of disease (Trasande and Landrigan 2004).
In pediatric practice, a major U.S. innovation has been development of a network of Pediatric Environmental Health Specialty Units (PEHSUs) [Agency for Toxic Substances and Disease Registry (ATSDR) 2005]. This network now includes 11 sites across the United States, and PEHSUs have also been established in Canada, Mexico, Spain, and (soon) Argentina. PEHSUs are clinical units designed to diagnose and treat children with diseases of toxic environmental origin, to improve access to expertise in pediatric environmental medicine, and to educate health care practitioners about environmental threats to children’s health.
In Europe, progress in protecting children against environmental threats initially lagged behind that in the United States. In recent years, however, advances in Europe have leapfrogged ahead.
CEH first emerged as a policy issue in Europe in 1999 at the third Ministerial Conference on Environment and Health, held in London. This Conference cited the Miami Declaration (Environment Leaders' Summit of the Eight 1997) and emphasized the importance of protecting children from environmental exposures. It identified priority areas for action and started a process that lead to the Fourth Ministerial Conference.
The Fourth Ministerial Conference, held in Budapest in 2004, was a watershed event for CEH. Preparation was coordinated by the World Health Organization (WHO) European Office and the European Environment and Health Committee. It involved eight large meetings with active involvement of all 52 member states of the WHO European Region, as well as nongovernmental organizations and representatives of industry and trade unions. The goal was to define the rationale, structure, and objectives of a novel Children’s Environment and Health Action Plan for Europe (CEHAPE; Ministers of Health and Environment in the European Region of the WHO 2004). Two major products of this process were a) a thorough review of the scientific evidence on children’s environmental health, published by the WHO European Office and the European Environment Agency (Tamburlini et al. 2002); and b) a study that quantified for the first time the burden of disease in children and adolescents in Europe related to environmental exposures (Valent et al. 2004).
At the Budapest Conference, CEHAPE was approved at the highest political level (Valent et al. 2004), thus affirming the commitments of all 52 European member states to mitigation of environmental threats to children’s health. An important feature of CEHAPE is its recognition that children in particularly adverse conditions, such as war or extreme poverty, are at highest risk of injuries, psychological trauma, acute and chronic infections, chronic diseases, disability, and death. CEHAPE urges that special emphasis be placed on preventing these conditions and their risk factors.
The challenge now confronting the European member states is to implement CEHAPE. This work is being coordinated by the WHO European Office–CEH Unit in Rome. Policy tools guiding this process include a book summarizing evidence on children’s environmental health; tools showing countries how to transform the CEHAPE framework into national action plans; a compilation of successful experiences in prevention; and a set of CEH indicators (Licari et al. 2005; Nemer and von Hoff, in press).
Following the commitments made in London and Budapest, the European Commission (EC) has strengthened its focus on CEH. It has provided funding to Member States for implementation of CEHAPE, and has developed far-reaching policies and action plans:
REACH (Registration, Evaluation and Authorization of CHemicals)
SCALE (Science, Children, Awareness raising, Legislation, Evaluation)
The European Environment and Health Strategy
The 2004–2010 Environment and Health Action Plan.
REACH, proposed by the EC in October 2003, is the most ambitious of these proposals (European Commission 2003). It presents a new European regulatory framework for chemicals, and its goal is to close the current gap in knowledge of the toxicity of chemicals. REACH requires that safety and toxicity information be made publicly available on all chemicals produced or imported in Europe in volumes > 1 ton/year per manufacturer/importer. Under REACH, the burden of proof to establish the safety of a chemical will be on industry. Innovation of safer substances will be encouraged under REACH by providing exemptions for research and development. If fully adopted, REACH will hasten the end of the vast ongoing toxicologic experiment in which chemicals are being tested on children worldwide instead of in the laboratory.
REACH is currently under examination by the European Parliament. In a recent public hearing, the current EC—which seems more attentive than its predecessor to the concerns of industry—discussed the importance of balancing children’s health against the competitiveness of European industry.
Among researchers and pediatric practitioners in Europe, CEH seems to be gaining momentum, though at a slower pace than in the United States. CEH is increasingly a focus of epidemiologic investigation, with recent studies examining the effects on children of air pollution and neurotoxicants. Large prospective cohort studies of children are under way in several nations (for example, in the United Kingdom and the Nordic countries) or are being started (Italy). A new research consortium focusing on CEH has been established in Trieste, Italy, under the leadership of the Children’s Hospital.
The evolution of research, practice, and policy in CEH on the two sides of the Atlantic has been a fascinating and interconnected process. Great progress has been made, but this progress raises questions. What, for example, will be the impact of European policy initiatives on policy in the United States? Will adoption of REACH puncture U.S. complacency on chemical testing? Will CEHAPE prod the United States into setting benchmarks and national goals for CEH? And what will be the impacts on policy in Europe of the new science now emerging from the Children’s Environmental Health Centers in the United States and soon to come from the National Children’s Study? Will the information emerging from this research result in bans or restrictions on classes of chemicals?
And what of the Developing World? Will the new science and policies emerging from the industrially developed nations influence the industrializing countries? In this time of globalization, the nations of the Developing World host ever more hazardous industries and ever more toxic chemicals, as those industries and chemicals become more and more unwelcome on the two sides of the North Atlantic. What will be the consequences?
These are critically important questions. If we consider protection of the health of our children an important value, we must confront them.
Support for this work was provided by the Mount Sinai Center for Children’s Environmental Health and Disease Prevention Research (P01 ES009584), the Mount Sinai Superfund Basic Research Program (P42 ES007384), and the U.S. Environmental Protection Agency (RD 83171101).
==== Refs
References
ATSDR 2005. Pediatric Environmental Health Specialty Units (PEHSU). Atlanta, GA:Agency for Toxic Substances and Disease Registry. Available: http://www.atsdr.cdc.gov/HEC/natorg/pehsu.html [accessed 12 September 2005].
Clinton WJ 1997 Executive Order 13045: Protection of children from environmental health and safety risks Fed Reg 62 19883 19888
European Commission. REACH (Registration, Evaluation and Authorization of Chemicals) 2003. Available: http://europa.eu.int/comm/environment/chemicals/reach.htm [accessed 12 September 2005].
FQPA 1996. Food Quality Protection Act of 1996. Public Law 104–170.
Environment Leaders' Summit of the Eight 1997. Declaration of the Environment Leaders of the Eight on Children's Environmental Health. Miami: May 1997, Available: http://www.g8.utoronto.ca/environment/1997miami/children.html [accessed 12 September 2005].
Licari L Nemer L Tamburlini G 2005. Children’s health and environment : developing action plans. Copenhagen:WHO European Office. Available: http://www.euro.who.int/eprise/main/WHO/InformationSources/Publications/Catalogue/20050812_1 [accessed 30 August 2005).
Ministers of Health and Environment in the European Region of the WHO CEHAPE (Children’s Environmental Health Action Plan for Europe). 2004. Declaration. Available: http://www.euro.who.int/document/e83335.pdf [accessed 12 September 2005].
National Children’s Study 2005. Homepage. Available: http://nationalchildrensstudy.gov/index.cfm [accessed 8 September 2005].
National Research Council 1993. Pesticides in the Diets of Infants and Children. Washington, DC:National Academy Press.
Nemer L von Hoff K eds. In press. Children’s Health and Environment Case Studies Summary Book. Copenhagen:WHO European Office Available: http://www.euro.who.int/childhealthenv/Policy/20040921_1 [accessed 12 September 2005].
Tamburlini G von Ehrenstein O Bertollini R eds. 2002. Children’s health and environment: a review of evidence. A joint report from the European Environment Agency and the WHO Regional Office for Europe. Copenhagen:European Environment Agency. Available: http://www.euro.who.int/document/e75518.pdf [accessed 1 January 2005].
Trasande L Landrigan PJ 2004 The National Children’s Study: a critical national investment Environ Health Perspect 112 A789 790 15471708
Valent F Little D Bertollini R Nemer L Barbone F Tamburlini G 2004 Burden of disease attributable to selected environmental factors and injury among children and adolescents in Europe Lancet 363 2032 2039 15207953
|
16203219
|
PMC1281297
|
CC0
|
2021-01-04 23:41:28
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A646-A647
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.113-1281297
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0065016203220PerspectivesDirector's PerspectiveThe Need for Exposure Health Sciences Schwartz David A. MDDirector, NIEHS and NTP, E-mail:
[email protected] Brenda PhDSpecial Assistant to the Director, E-mail:
[email protected] Samuel H. MDDeputy Director, NIEHS, E-mail:
[email protected] 2005 113 10 A650 A650 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
Although potentially hazardous environmental exposures are ubiquitous, the development of diseases from direct environmental causes is, fortunately, limited. For most diseases, environmental exposures represent one of several factors contributing to the development and progression of disease. Other factors include variation in genetic susceptibility, the presence of other conditions or diseases, diet, activity level, and medications taken. In addition, whether an individual develops disease as a function of environmental exposure also depends on the agent, the extent of exposure, and the timing of exposure. Nevertheless, the complex interaction between these factors accounts for the development of the majority of common diseases in the United States, including asthma, diabetes mellitus, atherosclerosis, and many types of cancer. The question we’d like for you to consider is what tools are needed to understand the role of environmental exposures in the development of common complex diseases.
Traditional methods for human exposure assessment involve measuring exposure concentrations in the ambient environment and extrapolating to potential points of human contact, as well as measuring the concentration of parent compounds or their metabolites in biological samples (biomonitoring). However, both these methods are problematic: measures of the ambient environment represent only rough estimates of an individual’s exposures, and biomonitoring provides only transient estimates of exposure.
We believe the time is right to advance a new science focused on the interface between exposures and human health. Technologies currently exist for the global analysis of genes (genomics), gene transcripts (transcriptomics), proteins (proteomics), and metabolites (metabolomics). Emerging fields, such as medical imaging, nanotechnology, and sensor technology are beginning to yield products that affect the way we conduct research and are proving useful in homeland security and biodefense. Our view is that these technologies also represent important opportunities for providing tools that could advance our understanding of disease etiology by providing quantitative methods to assess the temporal and biological response to multiple environmental exposures. Ideally, these new technologies will generate insight on exposures across the exposure–disease continuum, from the point of human contact to the internal dose to the early biological response (see figure above).
Our vision at the NIEHS is to use environmental health sciences to understand human disease and improve human health. Fundamental to this vision is the ability to quantify an individual’s exposure, as well as the unique characteristics that account for individualized responses to common exposures. To achieve this ability, we must develop an exposure biology initiative that will provide the same degree of individual-level precision that is being achieved through the sequencing of the human genome.
The question to consider is what tools are needed to understand the role of environmental exposures in the development of common complex diseases.
While this will not be easy, the impact of this scientific infrastructure on advancing our understanding of disease etiology and pathogenesis will be profound; in the absence of this infrastructure, it will remain difficult to evaluate the relationship between environmental exposures and human health. To the extent feasible, new technological developments should complement efforts that are ongoing in the public and private sectors, such as biodefense and national health surveillance. Realistically, investments will be needed in long-term research and training programs to enhance the precision and applicability of environmental exposure measurements to human health research.
As part of its strategic planning process, the NIEHS will establish priorities and develop a plan to move exposure health sciences forward. The emphasis will be on advancing our knowledge of gene-environment interactions in model disease processes. Deciphering the environmental and genetic risk factors for disease development and progression, specifically the interaction between environmental exposures and gene differences, will enable researchers and clinicians to develop better strategies for modifying risk and treating disease. Hence, the ability to develop, validate, and correlate exposure-response indicators with allele variation will be critical to our success in reducing the burden of human disease. As always, we look forward to your comments.
|
16203220
|
PMC1281298
|
CC0
|
2021-01-04 23:41:27
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A650
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.113-1281298
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0065316203222PerspectivesCorrespondenceVinyl Chloride and U.S. EPA Research Price Courtney M. CHEMSTAR, American Chemistry Council, Arlington, Virginia, E-mail:
[email protected] author is employed by the American Chemistry Council, a trade association representing the chemical industry, including manufacturers of vinyl chloride. The author has had, may have, or may in the future have investment interests in chemical companies, but does not consider these to constitute competing financial interests.
10 2005 113 10 A653 A654 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
A commentary by Sass et al. (2005), “Vinyl Chloride: A Case Study of Data Suppression and Misrepresentation,” is itself a case study in misrepresentation. The inclusion of such an article in this peer-reviewed publication stands in contrast to its stated mission to publish “balanced” and “objective” information. Sass et al. (2005) did not include or address recent studies in characterizing the weight of the scientific evidence related to vinyl chloride and made inaccurate and unsupported allegations about the integrity of U.S. Environmental Protection Agency (EPA) scientists and the rigorous peer review process utilized by the U.S. EPA.
Sass et al. (2005) asserted that there is a “scientific consensus that [vinyl chloride] is a multisite carcinogen in humans and experimental animals,” referencing 21 articles, only 3 of which were published during the past 15 years. They failed to mention or seriously discuss 7 articles noted below that were published in scientific journals since 1997 and update many of the studies Sass et al. cited and reach the opposite conclusion. These and other recent peer-reviewed studies and reviews fully support the U.S. EPA’s conclusion that “the association [between vinyl chloride and cancers other than the liver] is weak and any estimated increase in mortality from cancer at these sites is likely to be less than for liver cancer” (U.S. EPA 2000).
Authors of these articles include Aaron Blair, the chief of the Occupational Studies Section of the National Cancer Institute (NCI), who stated that epidemiologic evidence shows a strong exposure–response relationship for angiosarcoma of the liver, but not for other types of cancer (Blair and Kazerouni 1997). In a more recent review, McLaughlin and Lipworth (1999) reached the same conclusion:
Occupational vinyl chloride exposure has not been conclusively causally linked to any adverse health outcome, with the exception of angiosarcoma of the liver.
Even more recently, Bosetti et al. (2003) stated that
The aggregate data are reassuring in excluding any excess risk of death from lung, laryngeal, soft tissue sarcoma, brain and lymphoid neoplasms, as well as cirrhosis.
Recently published updates of cancer incidence in European and American industry-wide cohorts of workers exposed to vinyl chloride provide a firm basis for the conclusion that vinyl chloride exposure is not causally associated with brain cancer and the other tumors mentioned by Sass et al. (2005). The European study (Ward et al. 2001) was conducted by scientists affiliated with the National Institute of Occupational Safety and Health (NIOSH) and the International Agency for Research on Cancer (IARC). The authors found no evidence of an increase in cancers other than the liver. Similar, though less definitive, results were published by Mundt et al. (2000) in an update of the American cohort. A recent meta-analysis of these cohorts by IARC scientists further supports the conclusion reached by the U.S. EPA (Boffetta et al. 2003).
Given the strength and uniformity of the evidence supporting the U.S. EPA’s position, it is striking that Sass et al. (2005) did not address it. Instead, they claimed that the U.S. EPA yielded to advocacy by chemical manufacturers, implying that the U.S. EPA relied in part upon unpublished data. As noted above, however, the articles upon which the U.S. EPA placed primary reliance are published, in a few cases, by academic scientists sponsored by industry (e.g., Mundt et al. 2000), but for the most part by scientists affiliated with some of the most prestigious government-supported organizations engaged in cancer research (e.g., NCI, IARC, NIOSH).
Finally, it is not accurate that industry unduly influenced the review process for vinyl chloride nor that the potency factors published in the IRIS (Integrated Risk Information System) database (U.S. EPA 2000) are insufficiently protective (Norman 2002). The former comment disparages the U.S. EPA scientists who spent 5 years and went through two external peer reviews to make sure that relevant current science was reflected. The latter fails to recognize that a pharmacokinetic (PK) approach to risk assessment was supported over 20 years ago by the National Academy of Sciences and that the PK model for vinyl chloride used by the U.S. EPA—which predicted the actual incidence of angiosarcoma of the liver in the early cohorts of exposed workers—has been peer-reviewed, published, and validated (Clewell et al. 2001; Reitz et al. 1996).
==== Refs
References
Blair A Kazerouni N 1997 Reactive chemicals and cancer Cancer Causes and Control 8 473 490 9498905
Boffetta P Matisane MD Mundt KA Dell LD 2003 Meta-analysis of studies of occupational exposure to vinyl chloride in relation to cancer mortality Scan J Work Environ Health 29 3 220 229
Bosetti C LaVecchia C Lipworth L McLaughlin JK 2003 Occupational exposure to vinyl chloride and cancer risk: a review of the epidemiologic literature Euro J Cancer Prevention 12 427 430
Clewell HJ Gentry PR Gearhart JM Allen BC Andersen ME 2001 Comparison of cancer risk estimates for vinyl chloride using animal and human data with a PBPK model Sci Total Environ 274 37 66 11453305
McLaughlin JK Lipworth L 1999 A critical view of the epidemiologic literature on health effects of occupational exposure to vinyl chloride J Epidemiol Biostat 4 253 275 10764240
Mundt KA Dell LD Austin RP Luippold RS Noess R Bigelow C 2000 Historical cohort study of 10,109 men in the North American vinyl chloride industry, 1942–72; update of cancer mortality to 31 December 1995 Occup Environ Med 57 774 781 11024202
Norman WC 2002 More replies to “Bad Science.” Environ Forum 19 2 May-Jun 4 5
Reitz R Gargas M Andersen ME Provan WM Green TL 1996 Predicting cancer risks from vinyl chloride exposure with a physiologically based pharmacokinetic model Toxicol Appl Pharmacol 137 253 267 8661351
Sass JB Castleman B Wallinga D Vinyl chloride: a case study of data suppression and misrepresentation Environ Health Perspect 113 809 812 16002366
U.S. EPA 2000. Toxicological Review of Vinyl Chloride, Available: http://www.epa.gov/iris/toxreviews/ [accessed 2 September 2005].
Ward E Boffetta P Andersen A Colin D Comba P Deddens JA 2001 Update of the follow-up of mortality and cancer incidence among European workers employed in the vinyl chloride industry Epidemiology 12 710 718 11679801
|
16203222
|
PMC1281299
|
CC0
|
2021-01-04 23:41:29
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A653-A654
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.113-a653
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a00654PerspectivesCorrespondenceVinyl Chloride: Sass et al. Respond Sass Jennifer Beth Health and Environment Program Natural Resources Defense Council, Washington, DC, E-mail:
[email protected] Barry Environmental Consultant, Garrett Park, MarylandWallinga David Institute for Agriculture and Trade Policy, Minneapolis, MinnesotaJ.B.S and D.W. are employed by environmental nonprofit organizations with an interest in ensuring that regulations of toxic chemicals are as health protective as feasible. B.C. is an independent consultant in toxic substances control and has no competing financial interests regarding the subject matter of this letter.
10 2005 113 10 A654 A655 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
We provided documentation and extensive references to support two claims: industry urged the U.S. Environmental Protection Agency (EPA) to downplay data suggestive of cancer risks in tissues other than the liver, and the U.S. EPA reduced the cancer potency estimate of vinyl chloride in accordance with industry input. The American Chemistry Council (ACC) is a trade association representing over 150 companies that produce and use chemicals, including the Dow Chemical Company (Midland, MI), Georgia Gulf Corporation (Atlanta, GA), and Occidental Chemical Corporation (Los Angeles, CA) (ACC 2005). These three companies are also full members of the Vinyl Institute (Arlington, VA), whose stated goal is to “promote and protect the vinyl industry and the markets it serves” (Vinyl Institute 2005). Price, a lawyer with the ACC, maintains that our commentary is inaccurate because a) studies published since 1997 “reach the opposite conclusion”; b) by demonstrating industry influence on the U.S. EPA assessment of vinyl chloride we are disparaging of U.S. EPA scientists; and c) the pharmacokinetic (PK) model used by the U.S. EPA has been “peer-reviewed, published, and validated.”
In response to Price’s first point, recent studies confirm earlier findings instead of the opposite. At the time of the U.S. EPA assessment (U.S. EPA 2000) there were over 20 scientific articles and two independent reviews by the International Agency for Research on Cancer (IARC 1979, 1987) suggesting that vinyl chloride is a multisite carcinogen in humans and experimental animals. Recent reviews and data support the IARC conclusions. Of the references listed by Price, three are reviews without new data (Blair and Kazerouni 1997; Boffetta et al. 2003; Bosetti et al. 2003), two describe a PK model (Clewell et al. 2001; Reitz et al. 1996), and two contribute new data that neither refute previous studies nor support ACC claims (Mundt et al. 2000; Ward et al. 2001). One of these is a North American multicenter investigation, discussed in our commentary, which reported modest “excesses of brain cancer” and “cancer of connective and soft tissue” (Mundt et al. 2000). The second new study is a European multicenter investigation that is inconclusive regarding risks of nonliver cancers (Ward et al. 2001). Price also references a meta-analysis that actually reported an excess in brain cancer [standardized mortality ratio (SMR) = 1.26; 95% confidence interval (CI), 0.98–1.62) and soft-tissue sarcomas (SMR = 2.52; 95% CI, 1.56–4.07) (Boffetta et al. 2003); the authors of this meta-analysis concluded that “increased mortality from lung and brain cancers and from lymphatic and hematopoietic neoplasms cannot be excluded.” This is consistent with an Italian study that reported increased lung cancer deaths among polyvinyl chloride (PVC) baggers (RR = 3.04; 95% CI, 1.15–7.99) (Gennaro et al. 2003).
Price’s letter and the U.S. EPA assessment (U.S. EPA 2000) both reference a review article by National Cancer Institute authors (Blair and Kazerouni 1997). Blair stressed that his findings do not support disregarding possible risks of cancer outside the liver and that potent carcinogens such as vinyl chloride are unlikely to affect only one organ site (Blair A, personal communication). In overall mortality, a slightly increased rate of a common cancer such as lung cancer may lead to more deaths than a more markedly increased rate of a rare cancer such as liver angiosarcoma.
We believe that the U.S. EPA’s close relationship with industry compromises credibility. The ACC met with U.S. EPA regulators to discuss a vinyl chloride assessment at least 2 years before public notification of an assessment process. At the urging of the ACC (Price 1999), the U.S. EPA eliminated a statement that there is “suggestive epidemiological evidence that cancer of the brain, lung, and lymphopoietic system” associated with vinyl chloride exposure. The U.S. EPA also removed a 3-fold uncertainty factor that had been included to account for possible tumor induction at such sites, after an ACC letter called the factor “ill advised” (Price 1999). The result is that the U.S. EPA assessment (U.S. EPA 2000) does not adequately warn the public of the potential carcinogenicity of vinyl chloride suggested in the scientific literature, and the risk estimate is weakened 3-fold.
The PK model has not been validated, as stated by Price. The PK model developed by industry consultants and used by the U.S. EPA in its assessment has not been validated because assumptions used in the model have not been tested. Importantly, although the model is limited to liver effects only, the implicit assumption that all metabolism occurs in the liver is incorrect (IARC 1987; McFayden et al. 1998; U.S. EPA 2000).
By using a model limited to liver cancer, the U.S. EPA made a radical departure from its cancer guidelines, recommending that the cumulative risks of all tumor types be included in a cancer assessment (U.S. EPA 1999, 2005). The 1999 carcinogen guidelines under which vinyl chloride was assessed (U.S. EPA 1999) state that
In the analysis of animal bioassay data on the occurrence of multiple tumor types, the cancer potencies should be estimated for each relevant tumor type that is related to exposure, and the individual potencies should be summed for those tumors.
This inclusive approach is reconfirmed in the 2005 guidelines (U.S. EPA 2005). This protective approach was not taken by the U.S. EPA in its assessment of vinyl chloride cancer risks.
==== Refs
References
ACC 2005. American Chemistry Council Homepage. Available: http://www.americanchemistry.com/ [accessed 1 September 2005].
Blair A Kazerouni N 1997 Reactive chemicals and cancer Cancer Causes Control 8 473 490 9498905
Boffetta P Matisane MD Mundt KA Dell LD 2003 Meta-analysis of studies of occupational exposure to vinyl chloride in relation to cancer mortality Scan J Work Environ Health 29 3 220 229
Bosetti C LaVecchia C Lipworth L McLaughlin JK 2003 Occupational exposure to vinyl chloride and cancer risk: a review of the epidemiologic literature Euro J Cancer Prevention 12 427 430
Clewell HJ Gentry PR Gearhart JM Allen BC Andersen ME 2001 Comparison of cancer risk estimates for vinyl chloride using animal and human data with a PBPK model Sci Total Environ 274 37 66 11453305
Gennaro V Ceppi M Montanaro F 2003 Reanalysis of mortality in a petrochemical plant producing vinyl chloride and polyvinyl chloride [in Italian] Epidemiol Prev 27 4 221 225 14651027
IARC (International Agency for Research on Cancer) 1979 Vinyl chloride, polyvinyl chloride and vinyl chloride-vinyl acetate copolymers IARC Monogr Eval Carcinog Risk Chem Hum 19 377 438 374232
IARC (International Agency for Research on Cancer) 1987 Vinyl chloride IARC Monogr Eval Carcinog Risks Hum Suppl 7 373
McFayden MC Melvin WT Murray GI 1998 Regional distribution of individual forms of cytochrome P450 mRNA in normal adult human brain Biochem Pharmacol 55 6 825 30 9586955
Mundt KA Dell LD Austin RP Luippold RS Noess R Bigelow C 2000 Historical cohort study of 10,109 men in the North American vinyl chloride industry, 1942–72; update of cancer mortality to 31 December 1995 Occup Environ Med 57 774 781 11024202
Price CM 1999. Letter from the Chemical Manufacturers Association to the U.S. EPA regarding the Toxicological Review of Vinyl Chloride. Arlington, VA:Chemical Manufacturers Association.
Reitz R Gargas M Andersen ME Provan WM Green TL 1996 Predicting cancer risks from vinyl chloride exposure with a physiologically based pharmacokinetic model Toxicol Appl Pharmacol 137 253 267 8661351
Sass JB Castleman B Wallinga D Vinyl chloride: a case study of data suppression and misrepresentation Environ Health Perspect 113 809 812 16002366
U.S. EPA 1999. Draft Revised Guidelines for Carcinogen Risk Assessment (External Draft, July 1999). Washington, DC:U.S. Environmental Protection Agency, Risk Assessment Forum. Available: http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=54932 [accessed 13 June 2005].
U.S. EPA 2000. Toxicological Review of Vinyl Chloride. Available: http://www.epa.gov/iris/toxreviews/ [accessed 15 June 2005].
U.S. EPA 2005. Guidelines for Carcinogen Risk Assessment and Supplemental Guidance for Assessing Susceptibility from Early-Life Exposure to Carcinogens. Available: http://cfpub2.epa.gov/ncea/raf/recordisplay.cfm?deid=116283 [accessed 13 June 2005].
Vinyl Institute 2005. Biography of the Vinyl Institute's President, Tim Burns. Available: http://www.vinylinfo.org/aboutvi/burns.html [accessed 12 June 2005].
Ward E Boffetta P Andersen A Colin D Comba P Deddens JA 2001 Update of the follow-up of mortality and cancer incidence among European workers employed in the vinyl chloride industry Epidemiology 12 710 718 11679801
|
0
|
PMC1281300
|
CC0
|
2021-01-04 23:41:29
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A654-A655
|
utf-8
|
Environ Health Perspect
| 2,005 |
nan
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0065516203223PerspectivesCorrespondenceRisk-Based Consumption of Dioxin-Contaminated Farmed Salmon Middaugh John P. Arnold Scott M. Verbrugge Lori A. Alaska Division of Public Health, Anchorage, Alaska, E-mail:
[email protected] authors declare they have no competing financial interests.
10 2005 113 10 A655 A656 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
In their article, “Risk-Based Consumption Advice for Farmed Atlantic and Wild Pacific Salmon Contaminated with Dioxins and Dioxin-like Compounds,” Foran et al. (2005) present recommendations for consumption of salmon containing dioxin-like compounds (DLCs) based on three risk assessment approaches.
Relying strictly on risk assessment to develop fish consumption advice has many shortcomings and may actually do more harm than good (Arnold et al. 2005; Egeland and Middaugh 1997). Risk assessment is only part of the risk management process when developing fish consumption advice [U.S. Environmental Protection Agency (EPA) 1996]. Other factors need to be considered when developing fish consumption advice, such as the nutritional and health benefits of consuming fish and the cultural, societal, and economic impacts of reduced fish consumption (U.S. EPA 1996). Ignoring these factors may place an undue burden on a local population by removing a relatively inexpensive protein source that would likely be replaced by a less healthy substitute (Arnold et al. 2005; Egeland and Middaugh 1997). Decisions to limit fish consumption should only be made at the local level because local public health officials are most aware of the local aforementioned impacts and the actual concentration of contaminants in locally caught fish [Arctic Monitoring and Assessment Programme (AMAP) 2002; Arnold et al. 2005; Hites et al. 2004].
In addition to measuring contaminant concentrations in fish, human biomonitoring is a very useful tool to measure actual levels of contaminant exposure in targeted “at-risk” populations rather than relying solely on calculated estimates of exposure. Biomonitoring should ideally be performed in any identified at-risk population to verify that a problem actually exists before advising people to reduce fish consumption. This is especially true in populations that rely on locally caught fish as their primary protein source and that have few inexpensive healthy alternatives. Fortunately, recent evidence suggests that the average concentration of DLCs in the general U.S. population is declining (U.S. EPA 2000).
Two aspects of this study’s (Foran et al. 2005) methodology are problematic because they lead to inappropriately conservative estimates of health risk. First, the majority of people in the United States do not eat salmon skin, and as the authors noted, cooking has been shown to reduce DLCs in fish tissue. Because DLCs partition to fatty tissues including skin, measuring DLCs in raw fish with the skin on will overestimate the amount of exposure to DLCs and overestimate the consumption risks. Second, when assessing health risks posed by salmon consumption, Foran et al. (2005) estimated the number of meals of salmon per month that would limit dioxin intake to 20% above the U.S. average “background” adult intake level of 65 pg toxic equivalents (TEQ)/day. This approach ignores the fact that the background rate incorporates both freshwater and saltwater fish consumption (U.S. EPA 2000). In effect, Foran et al. (2005) double counted dioxin exposure through fish consumption. They also ignored the fact that a person who chooses not to consume a salmon meal would likely substitute another protein source that also contains trace quantities of DLCs.
==== Refs
References
Arctic Monitoring and Assessment Programme (AMAP) 2002. Arctic Pollution 2002. Oslo, Norway:Arctic Monitoring and Assessment Programme. Available: http://www.amap.no./ [accessed 18 May 2005].
Arnold SM Lynn TV Verbrugge LA Middaugh JP 2005 Human biomonitoring to optimize fish consumption advice: reducing uncertainty when evaluating benefits and risks Am J Public Health 95 3 393 397 15727965
Egeland GM Middaugh JP 1997 Balancing fish consumption benefits with mercury exposure Science 278 1904 1905 9417640
Foran JA Carpenter DO Hamilton MC Knuth BA Schwager SJ 2005 Risk-based consumption advice for farmed Atlantic and wild Pacific salmon contaminated with dioxins and dioxin-like compounds Environ Health Perspect 113 552 556 15866762
Hites RA Foran JA Carpenter DO Hamilton MC Knuth BA Schwager SJ 2004 Global assessment of organic contaminants in farmed salmon Science 303 226 229 14716013
U.S. EPA 1996. Guidance for assessing chemical contaminant data for use in fish advisories. Volume 3. Overview of risk management. Washington, DC:U.S. Environmental Protection Agency. Available: http://www.epa.gov/ost/fishadvice/vol3/doc3ndx.html [accessed 18 May 2005].
U.S. EPA 2000. Draft Dioxin Reassessment. Washington, DC:U.S. Environmental Protection Agency. Available: http://cfpub.epa.gov/ncea/cfm/part3.cfm?ActType=default [accessed 18 May 2005].
|
16203223
|
PMC1281301
|
CC0
|
2021-01-04 23:41:29
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A655-A656
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.113-a655
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a00656PerspectivesCorrespondenceDioxin-Contaminated Farmed Salmon: Foran et al. Respond Foran Jeffery A. Midwest Center for Environmental Science and Public Policy, Milwaukee, Wisconsin, E-mail:
[email protected] David O. Institute for Health and the Environment University at Albany, Rensselaer, New YorkHamilton M. Coreen AXYS Analytical Services Ltd., Sidney, British Columbia, CanadaKnuth Barbara A. Schwager Steven J. Cornell University, Ithaca, New YorkThe authors declare they have no competing financial interests.
10 2005 113 10 A656 A657 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
Middaugh et al. suggest that relying strictly on risk assessment to develop fish consumption advice has many shortcomings. We agree. They also argue that risk assessment is only part of the risk management process. Although we separate risk assessment and risk management, we agree conceptually that risk management decisions often must be based on more than just the results of a quantitative risk assessment.
What Middaugh et al. fail to recognize is that our report on dioxins in salmon (Foran et al. 2005) was not intended to serve as a fish consumption advisory. Such advisories should be left to appropriate state, federal, and international organizations charged with protection of public health. Rather, we reported risk-based consumption advice that would be triggered by dioxin-like compounds (DLCs) in farmed Atlantic and wild Pacific salmon using two different approaches; the World Health Organization (WHO) tolerable daily intake (TDI) for DLCs and a margin-of-exposure approach advocated by the U.S. Environmental Protection Agency (EPA 2002). We also reported cancer risks, based on the proposed U.S. EPA cancer slope factor for DLCs (U.S. EPA 2002) that would be generated at particular salmon consumption levels. Our results demonstrate clearly that consumption of some farmed Atlantic salmon, even at relatively modest levels, raises human exposure to DLCs above the lower end of the WHO TDI and considerably above background DLC intake for adults in the United States. Further, consumption at these levels poses elevated cancer and noncancer health risks.
Middaugh et al. suggest that human biomonitoring should be used rather than relying on calculated estimates of exposure, presumably to generate fish consumption advice. We strongly disagree, particularly in the case where the exposure source (farmed Atlantic salmon) is not localized. This is a global problem that would require human biomonitoring on immense temporal and spatial scales. In this case, quantitative risk assessment, which includes an assessment of chemical fate, transport, exposure, and effects, is an appropriate surrogate for human biomonitoring. Further, given our vast knowledge of the toxicokinetic behavior and toxicologic effects of dioxin and other bioaccumulative compounds in farmed Atlantic salmon, requiring human biomonitoring before issuing consumption advice is akin to continuing a clinical trial of a drug where unacceptable adverse effects have already been demonstrated. Clearly, responsible public health professionals should strenuously object to such an approach.
Middaugh et al. suggest that two aspects of our study are problematic. First, they argue that measuring contaminants in skin-on fillets may overestimate contaminant concentrations in edible fish tissue and, ultimately, human exposure. We addressed this issue in our article (Foran et al. 2005). We encourage Middaugh et al. to reexamine our conclusion that most studies of the effects of preparation (including removal of skin) and cooking on contaminant concentrations in fish tissue
suffer from small sample sizes, questionable data analyses, inconsistent analytical techniques, inconsistent data presentation, and variability in initial and postintervention contaminant concentrations within and among species, preparation techniques, and cooking techniques. Deficiencies in study design and variability in contaminant reductions preclude development of a useful quantitative correction factor for the effects of preparation and cooking on contaminant burden. As a result, reductions in exposure and risk associated with reduction in contaminant concentrations from preparation and cooking cannot be evaluated quantitatively; thus, we have not incorporated the effects of cooking and preparation in our risk assessments.
Second, Middaugh et al. are correct in stating that we did not adjust for the existing background concentration that incorporates DLC exposure via fish consumption. However, we did assess such exposures and concluded that they were so low, compared with exposure to DLC through consumption of farmed Atlantic salmon, as to be inconsequential in our risk assessment calculations.
Finally, we regret that Middaugh et al. ignored two critically important conclusions of our work. First, in all of our articles (Hites et al. 2004a, 2004b; Foran et al. 2004, 2005) that address contamination of salmon sold commercially, we provided information that will allow and encourage consumers to choose other fish, including wild Pacific salmon, as well as other sources of beneficial n-3 fatty acids. Second, our work has exposed serious deficiencies and inconsistencies in national and international approaches to the management of contaminants in commercially sold fish. These deficiencies and inconsistencies must be resolved so that consumers can confidently choose and consume fish with lower contaminant concentrations while continuing to accrue the health benefits of fish consumption.
==== Refs
References
Foran JA Carpenter DO Hamilton MC Knuth BA Schwager SJ 2005 Risk-based consumption advice for farmed Atlantic and wild Pacific salmon contaminated with dioxins and dioxin-like compounds Environ Health Perspect 113 552 556 15866762
Foran JA Hites RA Carpenter DO Hamilton MC Mathews-Amos A Schwager SJ 2004 A survey of metals in farmed Atlantic and wild Pacific salmon Environ Toxicol Chem 23 9 2108 2110 15378985
Hites RA Foran JA Carpenter DO Hamilton MC Knuth BA Schwager SJ 2004a Global assessment of organic contaminants in farmed salmon Science 303 226 229 14716013
Hites RA Foran JA Schwager SJ Knuth BA Hamilton MC Carpenter DO 2004b Global assessment of polybrominated diphenyl ethers in farmed and wild salmon Environ Sci Technol 38 19 4945 4949 15506184
U.S. EPA 2002. Draft Dioxin Reassessment. Washington, DC:U.S. Environmental Protection Agency. Available: http://cfpub.epa.gov/ncea/cfm/part3.cfm?ActType=default [accessed 18 January 2005].
|
0
|
PMC1281302
|
CC0
|
2021-01-04 23:41:29
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A656-A657
|
utf-8
|
Environ Health Perspect
| 2,005 |
nan
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a00658PerspectivesCorrespondence“Inert” and Active Ingredients: Séralini Responds Séralini Gilles-Eric Laboratoire de Biochimie et Biologie Moleculaire Université de Caen, Caen, France, E-mail:
[email protected] author declares he has no competing financial interests.
10 2005 113 10 A658 A658 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
Surgan raises interesting points in his analysis. This interest has been confirmed by reactions of agriculture authorities all over the world after publication of the article by Richard et al. (2005).
Indeed, scientific problems do exist in the registration of pesticides today, when chronic toxicity tests are conducted with the active ingredient alone—which is generally the case. First of all, chemists from companies may work hard for several years to find the right formulation that best amplifies the effects of the active ingredient. his formulation will allow penetration and stability and/or bioaccumulation of the active ingredient within plant, fungi, or insect cells, for instance, to reach the best toxicity. If there are any side effects in other animal or human cells, these will be also amplified by adjuvants, and thus not measured in chronic toxicity tests with the active ingredient alone. The active compound absorption by skin is generally calculated in the presence of formulated adjuvants, but this is clearly a short-term study and not sufficient to detect, for example, endocrine disruption or carcinogenesis, possibly promoted in vivo by the described synergy. This should even necessitate further care in case of the use of formulated products such as glyphosate-based herbicides on tolerant, edible plants.
As a matter of fact, most genetically modified crops have been modified and selected only to tolerate high-formulated herbicide absorption, but the plants are not submitted for registration requiring chronic toxicity studies involving long-term feeding of animals. Moreover, in the case of environmental pollution, active pesticide ingredients may encounter detergents or other lipohilic xenobiotics with comparable effects other than those of their own adjuvants, for instance, forming microvesicles to penetrate the cells. These combined effects should also be taken into account in authorized thresholds of pollution in order to avoid effects on wildlife or humans.
==== Refs
Reference
Richard S Moslemi S Sipahutar H Benachour N Seralini G-E 2005 Differential effects of glyphosate and Roundup on human placental cells Environ Health Perspect 113 716 720 15929894
|
0
|
PMC1281303
|
CC0
|
2021-01-04 23:41:27
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A658
|
utf-8
|
Environ Health Perspect
| 2,005 |
nan
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0066016203226EnvironewsForumCancer: Save Our (Young) Skins! McGovern Victoria 10 2005 113 10 A660 A660 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
Traditionally an adult disease, melanoma—the deadliest form of skin cancer—is on the rise in both children and adults around the world. In the United States, the overall rate of increase across the population was 2.8% per year between 1981 and 2001, according to data from the National Cancer Institute’s Survey of Epidemiology and End Results. People under age 20, a group in which melanoma is rare, have faced an overall 1.1% annual increase in disease incidence over the same period. But the rate among 10- to 24-year-olds has increased by 3.0%, according to research in the 20 July 2005 issue of the Journal of Clinical Oncology.
Julie Lange, an assistant professor of surgery and oncology at Johns Hopkins University School of Medicine in Baltimore, says, “Part of the apparent rise may be that cases ten or twenty years ago were not as likely to be reported to a tumor registry.” Reporting is more complete today, she says, and in some areas outpatient cases are now routinely reported along with inpatient cases. Improved reporting methods are not the whole story, though. “The incidence probably truly is increasing—it’s a fairly consistent finding,” Lange says.
Melanoma in children occurs so rarely that annual rate increases are measured in fractions of cases per million, so relatively small numbers of new cases can produce substantial percent changes in incidence. “From a public health burden point of view, saying it has increased from five cases to six cases per million children over a decade is more appropriate,” explains Ahmedin Jemal, the American Cancer Society’s program director for cancer occurrence.
The picture across the full human population is complex. “What we are seeing in adults, at least in Australia, is that amongst the older generation, their rates of melanoma are still climbing. We’re seeing the effects of their sun exposure fifty, sixty, seventy years ago,” says David Whiteman, a senior research fellow at the Queensland Institute of Medical Research in Brisbane, Australia. Increased attention to sun exposure seems to be working. “Amongst the younger [adult] cohort—the under-fifties and particularly the under-forties and younger—we’re seeing that their rates of melanoma are not as high as previous birth cohorts at the same age.”
Sun exposure and experience of blistering sunburns have been identified as important risk factors for adult melanoma. “Because we believe that UV exposure increases melanoma risk in adults, we are assuming that the same is true for children—whether there are other important factors for kids today, no one knows,” Lange says.
Whiteman’s group did a case–control study of childhood melanoma in Queensland in the 1990s to look for other such factors. “We were very interested in . . . exposure to pesticides, exposure to other chemicals, other environmental factors,” he says, “but we really found no differences in [those] exposures.” The group did find, however, that children with melanoma had more large noncancerous moles, heavier facial freckling, and less ability to tan compared to children without melanoma; they were also more likely to have a family history of the disease. These findings appeared in the January 1997 issue of the International Journal of Cancer.
Factors not yet investigated may also play a role. The Harvard Nurses’ Health Study, a long-term prospective study of risk factors for chronic diseases in women, has shown an association between orange juice consumption and melanoma in adult females. The investigators hypothesize that a photosensitizing compound in oranges may contribute to risk, says Diane Feskanich, an assistant professor of medicine at Harvard and an investigator on the study. However, a parallel study in men, not yet published, did not find the same strong association. “Whether there are photosensitizing foods is an open question,” she says. “Certainly there are drugs that warn you ‘don’t go out in the sun if you’re taking this.’”
Awareness of the risks of sun exposure has improved, according to Lange. “The population in general is more aware today of the potential danger from the sun than twenty or thirty or forty years ago,” she says. The same is true in Australia, which has among the world’s highest incidence of the disease. “The current generations of children are probably getting less sun exposure and fewer episodes of sunburn,” says Whiteman.
But better awareness of the major risks has not necessarily translated into complete protection of children. Even grasping the extent of older children’s exposure to the best-known risk factor, UV light, can be difficult. Despite prevention messages, many teenagers and young adults still want suntans. “The use of indoor tanning facilities is common among teenagers,” Lange says (in a 2003 survey, 47% of white girls aged 18 or 19 had used tanning beds three or more times). “Teenagers practice a lot of risky behaviors, and exposure to UV light is one of those behaviors.”
Burning youth.
Despite better awareness of the risks of sun exposure, melanoma is climbing among young people, a heretofore largely unaffected group.
|
0
|
PMC1281304
|
CC0
|
2021-01-04 23:41:28
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A660
|
utf-8
|
Environ Health Perspect
| 2,005 |
nan
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a00663EnvironewsForumEHPnet: The Endocrine Society and the Society for Endocrinology Dooley Erin E. 10 2005 113 10 A663 A663 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
The endocrine system encompasses the thyroid gland, the hypothalamus, the pancreas, the adrenal cortex, the thyroid, the parathyroid, and the male and female reproductive glands. Two large and well-established societies, The Endocrine Society and the Society for Endocrinology, serve the practitioners of this field. Both groups have established websites to keep their members and the general public aware of newsworthy events and developments in the field, and to educate those laypeople wishing to learn more about the subject.
The site of The Endocrine Society, http://www.endo-society.org/, features quick links so that information and materials can be pulled up by either subject or visitor’s role (e.g., clinician, student, volunteer worker). Among the 16 subjects featured are cardiovascular function, diabetes/insulin, genetics/genomics, male reproduction, and female reproduction. Also gathered in one area of the homepage are quick links to the society’s five publications, membership information, a member directory, and a subscription page.
The Endocrine Society’s homepage features two news sections, one of news in the general media, the other of updates within the society itself. Also available from the homepage is information on the society’s annual meeting, other society events, and related external events. Visitors can also select a quick link to The Hormone Foundation, the public education affiliate of the society, which offers basic information about the endocrine system, its function, and its associated diseases and disorders.
In the Press Room portion of the site is information on the America Weighs In campaign. This program focuses on educating the media, policy makers, and the general public about the role that endocrinologists play in researching and treating obesity. The Press Room also features a link to The Endocrine Edge, a free monthly online newsletter geared toward the public with the latest news from the society and the field of endocrinology.
The Society for Endocrinology site, http://www.endocrinology.org/, provides information about the Bristol, England–based organization and its programs. The society is affiliated with five journals, all accessible from a page on the website. The society also publishes a free quarterly newsletter, The Endocrinologist, which contains society news, general news, and feature articles.
The website also offers listings of grants and fellowships, society conferences, training courses sponsored by the society, and a calendar of events. The society sponsors a number of travel grants and has five research grant programs. There is also a page on books of interest to those in the field, which includes ordering information and short descriptions.
In 1997, the Society for Endocrinology established a committee for endocrinology nurses. This committee organizes conferences and an annual training course, and publishes the quarterly Endocrine Nursing News newsletter with reports from members and meeting notes. Past issues are available on the society website through the Endocrine Nurses link. Another subgroup are the Young Endocrinologists. Formed to support endocrinologists for up to six years after they receive their Ph.D., this group runs educational courses, provides career advice, and organizes special sessions at the Society of Endocrinology and British Endocrine Society meetings.
|
0
|
PMC1281305
|
CC0
|
2021-01-04 23:41:27
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A663
|
utf-8
|
Environ Health Perspect
| 2,005 |
nan
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0066416203227EnvironewsNIEHS NewsChildren’s Centers Study Kids and Chemicals Phillips Melissa Lee 10 2005 113 10 A664 A668 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
Many studies in recent years have documented that whether they’re used to spray in the kitchen or spray in the field, pesticides have a way of getting into almost all human environments. Pesticide exposure isn’t a great idea for adults, but it poses a particular concern in regards to children. These smallest humans, who spend a lot of time close to the floor and with their hands in their mouths, can encounter much higher doses relative to their body weights. And because they are still growing and developing, children are often more vulnerable to adverse effects of these and other environmental exposures. Likewise, the developing fetus may be especially vulnerable to the effects of pesticide exposure in utero.
In 1998, the NIEHS joined with the U.S. Environmental Protection Agency (EPA) to create eight centers across the country where scientists study environmental influences on children’s health. Today there are 11 centers. Several of these centers, including those at Columbia University and Mount Sinai School of Medicine in New York City, the University of California (UC), Berkeley, and the University of Washington (UW) in Seattle, have focused their efforts on pesticide exposures—how they occur, and the effects they cause in utero and during early childhood. These centers have also studied exposures to other environmental toxicants such as polycyclic aromatic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs), and environmental tobacco smoke.
These studies are showing that children in certain communities have elevated exposures to toxicants early in their development and that some of these exposures can lead to slightly stunted fetal growth, shorter gestation, and suboptimal neurodevelopment, as well as to predisposition to diseases such as asthma. Additional studies are showing that the potential for damage from these chemical exposures may be affected by genetic susceptibility of both the child and the mother. Thus, interactions between genes, the environment, and the timing of exposure can all contribute to a later susceptibility to develop diseases and disorders.
Columbia University
“Early-life exposures, even occurring in the womb, appear to be important determinants of that child’s respiratory health and development later on,” says Frederica Perera, director of the center at Columbia University and a professor of public health. “We have enormous opportunities to prevent these diseases and conditions.” At Columbia University, researchers have set up a cohort study to analyze exposure to pesticides and PAHs during pregnancy and very early childhood, a time of susceptibility that Perera says has not been adequately studied in the past.
Since 1998, nearly 700 pregnant Dominican and black women from Washington Heights, Harlem, and the South Bronx have enrolled in the study. Researchers take urine samples from mothers during pregnancy and blood samples from their babies at birth, sample ambient air in the mothers’ environment during pregnancy, and administer questionnaires and biomarker assessments repeatedly over the child’s early years.
Perera and her colleagues found that all the women in their cohort—and, therefore, their developing babies—were exposed to PAHs from vehicle exhaust and to at least one neurotoxic pesticide during pregnancy. In the February 2003 issue of EHP they reported finding that high PAH levels in a mother’s air samples correlated with having a smaller baby at birth. In papers published in the American Journal of Respiratory and Critical Care Medicine between 2000 and 2002, the team further reported that high prenatal exposure to certain PAHs was related to an increased likelihood that children would show asthma precursor symptoms and allergic responses to cockroach, mouse, and dust mite allergens at 2 years of age.
“We also see evidence that [PAH] exposures can influence cancer risk,” Perera says. Prenatal exposure to PAHs was associated with DNA abnormalities in the babies’ blood. This type of permanent genetic alteration has been linked to increased risk of cancer in children and adults. Also, PAH-induced DNA damage in the babies, in conjunction with exposure to secondhand tobacco smoke, was associated with significantly lower weight and smaller head circumference at birth—both signs of potential future developmental and learning problems.
The pesticide exposures of mothers and children in these urban communities occurred mainly due to insect and rodent infestations in poor-quality housing. In utero exposure to two organophosphate pesticides, chlorpyrifos (then the most widely used pesticide in New York City) and diazinon, resulted in an average birth weight reduction of 6.6 ounces, says Robin Whyatt, an associate professor of clinical environmental health sciences and co-deputy director of the Columbia center.
In 2000, however, the EPA announced that chlorpyrifos and diazinon would both be banned as household pesticides, and “the levels of pesticides in air during pregnancy and in the blood of both mothers and newborns were sharply reduced,” says Perera. By the time samples were taken in the spring of 2001, researchers no longer saw an association between organophosphate exposure and low birth weight.
Columbia researchers are also involved in a number of other studies, including an intervention project to reduce toxic pesticide use in public housing, says Perera. Residents are taught integrated pest management techniques, including removing pest food sources, sealing cracks and crevices, and using low-toxicity pesticides such as baits, gels, and boric acid. The families involved are also given lidded trash containers, pest-proof food containers, trash bags, and cleaning supplies.
Mount Sinai School of Medicine
At the Mount Sinai center, scientists are also using a pregnancy cohort of about 400 New York women of different ethnicities who gave birth at Mount Sinai Hospital from 1998 to 2003. During this period, the study focused on pesticide exposure and how genetic variations in the paraoxonase 1 (PON1) enzyme—which detoxifies organophosphate pesticides in the body—modify response to pesticides.
Maternal blood samples were taken during the third trimester, and PON1 activity was assessed. In the March 2004 issue of EHP, the Mount Sinai researchers reported finding that infants exposed to chlorpyrifos in utero were born with smaller head circumferences, but only if their mothers also had low levels of PON1 activity, says center scientist James Wetmur, a professor of microbiology and human genetics.
In 2003, these studies of development and genetic susceptibility moved away from organophosphate pesticides, largely because levels of these chemicals dropped after the EPA ban on residential use, says Mary Wolff, director of the Mount Sinai center and a professor of community and preventive medicine and oncological sciences. Researchers at Mount Sinai are now focusing on in utero exposures to endocrine-disrupting chemicals often found in plastics such as phthalates and phenols such as bisphenol A.
The center preserves biologic samples from all cohort members for future studies, Wolff says, so researchers will be able to analyze phthalate and phenol levels in maternal prenatal urine samples and correlate these levels with birth outcomes and with subsequent growth and neurodevelopment. Wetmur and his team are also switching gears to search for enzymes that metabolize phthalates and phenols, as well as for genetic variation in these enzymes that might affect birth or growth outcomes.
In a separate study in East Harlem, center researchers have found that integrated pest management is effective at controlling cockroaches. In addition to reducing or eliminating exposure to toxic pesticides, the long-term cost of this method—including building repairs—is lower than standard chemical-based pest control, making it available to lower-income residents. According to a report in the October 2003 EHP, “The costs of adopting building-wide integrated pest management in a typical East Harlem apartment building were calculated to be $46–69 per unit in the first year (including repairs) and $24 per unit per year in subsequent years,” compared to $24–46 per unit per year, not including repairs, for traditional chemical-based control. In coming years, this intervention project will next look at how the built environment affects exposures to endocrine-disrupting chemicals.
Another study to find evidence of health effects of PCBs showed that early-life exposure to these chemicals in animals can affect neuroendocrine development. Led by neuroendocrinologist Andrea Gore, then at Mount Sinai and now at the University of Texas at Austin, researchers discovered that these chemicals directly influence brain cells called gonadotropin-releasing hormone (GnRH) neurons. These neurons control reproduction in all vertebrates, and disruption in their growth or activity can lead to fertility problems, Gore says. In the October 2002 issue of the Journal of Neuroendocrinology, Gore and colleagues reported that a more estrogenic PCB mixture, Aroclor 1221, stimulated GnRH expression, while the less estrogenic Aroclor 1254 had both stimulatory and inhibitory effects, depending on the transcript measured.
University of California, Berkeley
As at Columbia and Mount Sinai, researchers at UC Berkeley are conducting a prospective cohort study of pregnant women and their children. Most women and children enrolled in this study—about 600 pairs total—come from low-income Mexican immigrant farmworker families who live in California’s Salinas Valley.
The first goal of the study has been to understand levels and routes of exposure to pesticides and other environmental contaminants among pregnant women and children, says Brenda Eskenazi, the center director and a professor of maternal and child health and epidemiology. Researchers have collected samples of urine, breast milk, blood, and house dust. They are determining the relationship of urinary pesticide metabolites in pregnant women and children with levels of pesticides in house dust, parental occupation, and nearby agricultural pesticide applications. They are also videotaping young children to identify behaviors that may expose them to environmental chemicals.
They’ve found that pregnant women in their cohort show abnormally high urinary levels of organophosphate pesticide metabolites, with about 15% of them likely exceeding the maximum cumulative exposure levels advised by the EPA. Organophosphate metabolites were higher in 6-month-old babies if the children lived with an agricultural worker. These metabolite levels were also significantly correlated with season, with urine collected in the summer showing the highest concentrations of pesticides. Levels of pesticide metabolites in urine rose as the children passed 6 months, likely because their activity levels—especially hand-to-mouth behavior—increased as they grew older.
A second goal of the study is to examine the health effects of these pesticide exposures in the children of exposed mothers. As in the Columbia and Mount Sinai studies, children in this cohort will be followed through at least age 7 to determine whether prenatal and childhood exposures have altered their cognitive development, growth, or respiratory health. UC Berkeley scientists have already found that high maternal organophosphate exposure during pregnancy correlated with shorter gestation duration, but no associations were found between organophosphate exposure and infant birth weight, length, or head circumference. A UC Berkeley center study published in the March 2005 issue of NeuroToxicology showed that newborns whose mothers had high levels of pesticide metabolites during pregnancy were more likely than other babies to have abnormal reflex functioning soon after birth.
The UC Berkeley center’s projects also include a randomized intervention study to see what types of preventive measures best discourage pesticide transmission from farmworkers to their children, Eskenazi says. Other projects include examining pesticide levels in amniotic fluid and breast milk, monitoring ambient pollen and mold levels, and studying mechanisms of pesticide and allergen effects on neural and immune functions.
University of Washington
The UW center also is measuring the extent of pesticide exposure in agricultural communities. Building upon previous UW center research in the Yakima Valley, center researchers have found that children of orchard workers can be exposed to pesticides that are transported on the clothing, boots, and skin of their farmworker parents, says center director and professor of environmental health Elaine Faustman. These studies also linked children’s exposure with specific agricultural crops, which will be detailed in upcoming unpublished papers. Such findings will allow the UW center to intervene more effectively in preventing the occupational take-home pathway for pesticide exposure in children.
UW scientists have also developed a laser-based method that allows them to monitor pesticide spray drift in real time. They’ve shown that pesticides can volatilize unexpectedly in certain conditions, especially in extreme heat—so even though time has elapsed since a crop spraying, it still may not be safe for children to go near the fields. These results should influence EPA recommendations for safety near agricultural fields after pesticide application, Faustman says.
A major part of UW research has focused on genetic susceptibility to the neurotoxic effects of organophosphate pesticides. Using data gathered by the UC Berkeley center, UW researchers have shown that people with certain forms of the PON1 gene break down chlorpyrifos more efficiently than people with different forms, although all forms detoxify diazinon at the same rate.
However, knowing which genetic variant a person has does not tell you what level of PON1 is present in the blood, says UW research professor of medical genetics Clem Furlong. Knowing the activity levels of the enzyme is important in determining how well a person will metabolize organophosphates and the potential for health impacts from organophosphate pesticide exposure.
“I think it’s extremely important to emphasize that, because epidemiologists continue to try and estimate risk only by doing genotype,” Furlong says. “You really need to look at the functional status of individuals.” It takes nearly a year for infants to begin making the amount of PON1 they will have as adults, and this may lead to increased vulnerability to exposure during this time, says Furlong. Maternal PON1 can provide some protection in utero, but “if you have a mother with extremely low PON1 levels, this is a serious concern—there’s no ability of that fetus to protect itself,” he says.
In animal studies, UW researchers have examined the mechanisms through which pesticides cause neurotoxicity. They’ve found that different pesticides can have very different influences on cell proliferation, differentiation, and death during brain development, and all of these effects are dependent upon dose and time of exposure during development. For example, in the March 2004 issue of Toxicological Sciences the team reported that chlorpyrifos induced apoptosis in primary cortical neurons cultured from embryonic and newborn rats. Currently, says Faustman, center researchers are expanding studies in mice to see how the combination of exposure and genetic susceptibility affects behaviors in the animals.
Where to Go From Here
The research coming out of these children’s centers over the past seven years has revealed that there are still far more unknowns than knowns, says Nina Holland, an adjunct professor of environmental health sciences at Berkeley and member of the UC Berkeley center. There are also discrepancies between some of the findings emerging from different centers. For example, the Columbia center’s report in the July 2004 issue of EHP that in utero exposure to chlorpyrifos or diazinon resulted in an average birth weight reduction of 6.6 ounces contrasted with a UC Berkeley study in the same issue, which found no adverse relationship between fetal growth and any measure of in utero organophosphate pesticide exposure (in fact, that team found increases in body length and head circumference associated with some exposure measures). But the overall finding, Holland says, is that “we have to pay much more attention to potential effects of pesticides on very young children.”
Center researchers are translating their experimental results into interventions, educational materials, community forums, press releases, and newsletters that can be used by parents, health care providers, and policy makers to improve the environmental health of local children. For example, the Columbia center has established a community educational campaign called “Healthy Home, Healthy Child.” Through this program, they have surveyed parents and caretakers of children in Harlem, Washington Heights, and the South Bronx to determine what these people are most concerned about. Then they’ve compiled tip sheets on topics such as air pollution, tobacco smoke, nutrition, pesticides, and lead poisoning, and they’ve distributed these on the street and at community health fairs and public forums. Center researchers have also trained staff at community centers to deliver health workshops to many different types of local groups, such as parent–teacher associations, churches, after-school programs, and foster care agencies. They also send summaries of their findings—in English and Spanish—to all the mothers involved in these cohort studies.
One important focus for the future is the National Children’s Study, according to Nsedu Obot Witherspoon, who is executive director of the Children’s Environmental Health Network. Slated to begin enrolling in the fall of 2007, the proposed study will follow 100,000 children from preconception or early pregnancy through adulthood, examining the effects of many different environmental exposures on various health outcomes. Leaders of the study include the NIEHS, the National Institute of Child Health and Human Development, the Centers for Disease Control and Prevention, and the U.S. EPA.
Researchers from all of the children’s centers and from the National Children’s Study should be able “to work hand in hand and will provide a wealth of information we would have otherwise never had,” Obot Witherspoon says. “It’s going to be phenomenal.”
Home is where the exposure is.
A myriad of exposures in low-income urban housing—including vehicle exhaust and pesticides used in homes—contribute to conditions ranging from cancer to low birth weight.
Mothers, babies, and chemicals.
Researchers are studying whether variations in the enzymes that metabolize the phthalates found in some plastic bottles correlate with later birth and growth outcomes.
More than one way in.
Through pathways both expected and surprising, children of farmworkers have higher pesticide exposure than the general population.
Waiting for the future.
A study of Mexican immigrant farmworkers will follow children through at least age 7 years to monitor possible effects of prenatal and childhood exposure to pesticides.
|
16203227
|
PMC1281306
|
CC0
|
2021-01-04 23:41:28
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A664-A668
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.113-a664
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a00667EnvironewsNIEHS NewsHeadliners: Smoking: The Role of the Parent in Deterring Child Smoking, as Seen by Rural Native American and White Parents Phelps Jerry 10 2005 113 10 A667 A667 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
Kegler MC, Malcoe LH. 2005. Anti-smoking socialization beliefs among rural Native American and white parents of young children. Health Educ Res 20(2):175–184.
Studies suggest that there are differences between the races in parental “anti-smoking socialization”—that is, how parents influence their children’s expectations regarding the feasibility, acceptability, and consequences of smoking cigarettes. For instance, black parents are more likely than white parents to set ground rules regarding tobacco use for their children, and are less likely to assume that teens will inevitably experiment with smoking. Now Lorraine Halinka Malcoe and NIEHS grantee Michelle C. Kegler of Emory University have compared antismoking socialization beliefs among rural white and Native American parents. Better information on how beliefs vary racially could help shape more effective ways of teaching parents to deter their children from smoking.
Teen smoking rates vary significantly between racial and ethnic groups. According to data from the Centers for Disease Control and Prevention for the year 2000, 31.8% of white high school students reported smoking in the past 30 days. Hispanic students were next at 22.6%, followed by Asian Americans at 20.6%, and blacks at 16.8%. Data on smoking among Native American teenagers are not as readily available, but some studies have indicated the rate among Native Americans overall is comparable to or higher than that of whites. In 2000, 36% of adult Native Americans smoked, compared with 24.1% of white adults.
The study showed that Native American and white parents were similar in their antismoking socialization beliefs with one exception: Native American parents were less likely to believe that schools are better than parents at teaching children about the dangers of smoking. Less educated parents were more likely to believe that strictly forbidding children to smoke only makes them want to smoke more. Consistent with earlier results, parents of both races had less stringent beliefs and a lesser sense of parental efficacy compared to black parents.
Methods to bolster antismoking socialization beliefs of less-educated parents may be important in preventing children in low-income rural communities with high smoking rates from beginning to smoke. Although limited in size and scope, this study provides evidence that future research should focus on ways to increase parental communication of antismoking beliefs and assessment of whether such interventions result in lower rates of smoking onset.
|
0
|
PMC1281307
|
CC0
|
2021-01-04 23:41:28
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A667
|
utf-8
|
Environ Health Perspect
| 2,005 |
nan
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a00668EnvironewsNIEHS NewsBeyond the Bench: Virtual School Tillett Tanya 10 2005 113 10 A668 A668 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
In today’s world of high-speed interconnection, technology in the classroom helps keep students interested and engaged in the learning process. Taking advantage of this favorable avenue of instructional opportunity, the Community Education and Outreach Program (COEP) of the NIEHS Center in Molecular Toxicology at Vanderbilt University, in conjunction with the university’s Center for Science Outreach (CSO), has developed an innovative interactive videoconference teaching program known as “Virtual Scientist in the Classroom.” The program creates a direct connection between Vanderbilt University faculty and students all over the country, allowing university researchers to lecture on environmental health topics related to the work they are performing in their own laboratories.
“Through the center’s involvement with outreach and education, we are able to provide reliable, up-to-date, and cutting-edge science to classrooms throughout Tennessee and the U.S.,” says Bradley Hawkins, the COEP director. “In addition, the students are able to interact with our researchers in a manner that was not available just a few years ago.”
The program relies on volunteer faculty with diverse research interests—neuroscience, diabetes mellitus, biomedical engineering, physics, molecular toxicology, and chemistry, for example—who create their own presentations and conduct the sessions in real time in the CSO virtual learning studio (all presentations are taped and archived for future multi-classroom sessions). The topics for presentations to date have included how medicines are developed, how chemicals damage DNA, and the importance of micronutrients. The format of each session is left to the discretion of the expert presenters, and may include anything from PowerPoint slides to movie clips, live virtual tours of lab facilities, even real-time electrocardiograms. One physics professor presented the theory of relativity in character as Albert Einstein.
The sessions of 30–45 minutes can be presented to one school at a time or to multiple site audiences. Scientist–student interaction is a main component of the sessions; questions and feedback from students are expected and encouraged. By using a communications bridge capable of connecting to multiple sites within a videoconference session, the researchers open up the world of scientific discovery to students in classrooms all over the state of Tennessee and beyond. When the program was created in 1999, it primarily reached out to middle and high school students in Tennessee, but has grown to include videoconferences to children in 75 schools in 20 states.
Typically four to six sessions on varied topics are offered each semester. Teachers can find complete descriptions of each session online at http://www.vanderbilt.edu/cso/ and can register there for each session. The sessions are free for Tennessee students, although a charge is applied for out-of-state schools. Once teachers have registered, they can download supplemental lesson material and will receive e-mailed confirmation and detailed instructions for participation. In a new feature, researchers answer questions that arise after each session, and the 1- to 2-minute video response also is archived on the site.
The continued commitment and enthusiasm of the contributing faculty members is a cornerstone of the program, and helps keep the videoconference sessions relevant and timely. “I believe that as researchers we need to take an active role in helping to educate and inform the public about issues related to adverse health outcomes upon exposure to poisons, to educate the public about sources of poisons in food and air, and the mechanisms by which they affect our health,” says Michael Aschner, a professor of pediatrics and pharmacology who has presented on the subject of chemical insults to the brain. Hopefully, he says, educational outreach programs can help bridge the gap between public understanding and public perception of toxicology.
M(aven)TV?
Vanderbilt University specialists use the Internet to connect kids with science straight from the lab.
|
0
|
PMC1281308
|
CC0
|
2021-01-04 23:41:28
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A668
|
utf-8
|
Environ Health Perspect
| 2,005 |
nan
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0067016203228EnvironewsFocusAre EDCs Blurring Issues of Gender? Hood Ernie 10 2005 113 10 A670 A677 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
Although scientists have postulated a wide range of adverse human health effects of exposure to endocrine-disrupting chemicals (EDCs), the nexus of the debate is the concern that prenatal and childhood exposure to EDCs may be responsible for a variety of abnormalities in human sexuality, gender development and behaviors, reproductive capabilities, and sex ratios. Scientists today are asking hard questions about potential human effects: Do EDC exposures impair fertility in men or women? Can they cause sexual organ malformations, stunted reproductive development, or testicular or breast cancer? Do fetal exposures to EDCs alter sex phenotypes? Do they change later gender-related neurobiological characteristics and behaviors such as play activity and spatial ability? Could such exposures even be involved in the etiology of children born with ambiguous gender?
EDCs include a spectrum of substances that can be loosely classified according to their known or suspected activity in relation to sex hormone receptors and pathways. The most-studied and best known are the environmental estrogens, which mimic estradiol and bind to estrogen receptors (ERs). ER agonists include the pesticide methoxychlor, certain polychlorinated biphenyls (PCBs), bisphenol A (BPA; a high production volume chemical used to make polycarbonate plastic), pharmaceutical estrogens such as diethylstilbestrol (DES) and ethinyl estradiol, and phytoestrogens, which occur naturally in many plants, most notably in soybeans in the form of genistein and related substances. There are a few known ER antagonists, or antiestrogens. Antiandrogens, or androgen receptor (AR) antagonists, include the fungicide vinclozolin, the DDT metabolite p,p′-DDE, certain phthalates (a group of chemicals used to soften polyvinyl chloride plastics), and certain other PCBs. And there are other types of EDCs that affect particular endocrine targets. The various EDCs differ greatly in their potencies relative to natural hormones, and in their affinity for target receptors. Some have been shown to act via non–receptor-mediated mechanisms, for example by interfering with hormone synthesis.
In many well-documented cases of high-level fetal exposures to known EDCs such as DES, certain PCBs, and DDT, the answer to the question of whether exposure is associated with gender-related effects is clearly yes. But high-level exposures such as these are relatively rare and isolated. The debate today centers on low-dose exposures—generally defined as doses that approximate environmentally relevant levels—and the idea that low-dose intrauterine exposure to some EDCs during certain critical windows of development can have profound, permanent impacts on subsequent fetal development and adult outcomes.
Critics of this idea maintain that thus far there is no credible evidence to suggest that low-dose exposures cause any adverse human health effects. But if low-dose exposures were confirmed to be the threat that proponents of the concept insist they are, public health would clearly be at risk, regulatory agencies’ risk assessment approach would need to be revised, and certain common chemicals—including some that are massively produced and economically important—would likely disappear from the marketplace.
In a June 2000 EHP review article on human health problems associated with EDCs, Stephen Safe, director of the Center for Environmental and Genetic Medicine at Texas A&M University, concluded that “the role of endocrine disruptors in human disease has not been fully resolved; however, at present the evidence is not compelling.” Frederick vom Saal, a developmental biologist at the University of Missouri–Columbia, disagrees, particularly in light of the research that’s been presented in the years since that review. “The jury is not out on human effects,” he says. “In terms of the amount of information we have in animals and the amount of information we have in humans, clearly there is a huge difference, but that’s a lot different than saying the jury is out on whether EDCs influence humans.” One thing both scientists might agree on, though, is that right now there are still more questions than answers.
A Delicate Process
The endocrine system, comprising the hypothalamus, pituitary, testes, ovaries, thyroid, adrenals, and pancreas, is one of the body’s key communications networks. It regulates the function of specific tissues and organs by secreting hormones that act as precise chemical messengers. Development and regulation of the reproductive system is one of the major functions of the endocrine system.
Sex determination and development begin early in gestation, with the differentiation of the embryonic gonad into either testes or ovaries. If the Sry gene is present on the Y chromosome, it will, when activated, trigger a complex cascade of hormonal events that ultimately results in the birth of a baby boy with all of the requisite male equipment in place and functioning properly. In the absence of the Sry gene, the end product of the process will be a baby girl. The female phenotype is considered to be the “default” pathway for mammalian reproductive development.
Differentiation and development of the sexual organs continues throughout gestation under the guidance of the various sex hormones (such as estrogen and testosterone) produced by the endocrine system. For males and females alike, the entire process of reproductive development is exquisitely sensitive to minute changes in levels of the sex hormones, particularly during certain critical windows of development.
In papers published in the Journal of Animal Science throughout 1989, vom Saal demonstrated this sensitivity in a series of mouse experiments. These studies showed that in multiple-birth species it was possible for adjacently positioned male and female fetuses to transmit tiny amounts of hormones to each other, with pronounced phenotypic consequences. “We found that a difference of about a part per billion of testosterone and about twenty parts per trillion of estradiol [endogenous estrogen] actually predict entirely different brain structures, behavioral traits, enzyme levels, and receptor levels in tissues, hormonal levels in the blood—there is nothing you look for that . . . doesn’t differ in these animals,” says vom Saal.
Such a delicately timed and precisely controlled process presents a myriad of opportunities for perturbation from exposure to EDCs. These chemicals mimic hormones, and can disrupt differentiation and development in a wide variety of ways, by duplicating, exaggerating, blocking, or altering hormonal responses. The developing fetus and early neonate may lack the protective metabolic mechanisms present in adults that help detoxify and break down chemicals, maintaining homeostasis in the system. Also, tissues are rapidly dividing and differentiating in the fetus, and such a high level of cell activity is vulnerable to disruption of normal development. With such small body mass in the fetus and child compared to an adult, exposure levels may be amplified in terms of relative dosages reaching target tissues. And sometimes, exogenous EDCs may show very low binding to plasma hormone-binding proteins and thus roam the body in an unbound state, with unknown effects.
Much of what remains to be discovered about the impacts of EDC exposures on the fetus relates to a new concept called the developmental origins of health and disease (until recently known more commonly as the fetal basis of adult disease). “People are just now recognizing that this is indeed a possibility,” says NIEHS scientist Retha Newbold, a pioneer in the study of endocrine disruption who has spent decades researching the effects of exogenous estrogens, particularly DES. “Developmental exposure to low doses of EDCs may not lead to malformation or to anything you can look at and immediately recognize as a problem,” she says. “But it still could have long-term effects, such as alterations in metabolism, alterations causing cancer later on, or alterations causing infertility.”
Evidence of Effects
Reproductive and developmental abnormalities linked to EDC exposures have now been documented in birds, frogs, seals, polar bears, marine mollusks, and dozens of other wildlife species. For example, alligators in Lake Apopka—one of Florida’s most polluted lakes due to extensive farming activities around the lake, the presence of a sewage treatment facility, and a major 1980 spill of pesticides including DDT and DDE—have been shown to have been “feminized.” That is, zoologist Louis J. Guillette, Jr., and colleagues first reported in the August 1994 EHP, the males have shortened penises and low levels of testosterone, while the females have excessive levels of estrogens. Sex reversal (in which an animal of one sex matures with the reproductive organs and capabilities of the other sex) and skewed sex ratios (in which there is an unusually greater proportion of one sex than the other) have been seen in several fish populations, particularly colonies living in close proximity to pulp and paper mills and sewage treatment plants. Other reports have shown reproductive effects among wildlife resulting from exposure to EDCs excreted into the water supply by women taking birth control pills.
Many of the adverse outcomes seen in wildlife populations have been replicated in laboratory experiments, confirming the role of EDCs in their occurrence. Among the papers reporting such confirmation were a May 1997 article in EHP, in which Guillette, D. Andrew Crain, and colleagues replicated alterations in steroidogenesis (the production of sex hormones) in alligators. More recently, in the December 2004 issue of EHP, Jon Nash and colleagues showed that long-term laboratory exposure to environmental concentrations of the pharmaceutical ethinyl estradiol caused reproductive failure in zebrafish.
According to a report on EDCs published in volume 75, issue 11/12 (2003) of Pure and Applied Chemistry by the Scientific Committee on Problems of the Environment/International Union of Pure and Applied Chemistry (SCOPE/IUPAC), more than 200 animal species are either known or suspected to have been affected by these chemicals. “The weight of evidence for endocrine disruption in wildlife is really overwhelming,” says Joanna Burger, a professor of cell biology and neuroscience at Rutgers University who cochaired the SCOPE/IUPAC project.
The SCOPE/IUPAC report was less definitive on the extent of human effects of endocrine disruptors. “It is too early to reach firm conclusions about whether human populations are seriously at risk from potential exposures to [EDCs], and further vigilance is clearly required,” the authors wrote. “However, it is somewhat reassuring that after substantial research in the past decade, there have been no conclusive findings of low-level environmental exposures to [EDCs] causing human disease.”
The report further notes, however, that “[c]hemical interferences with steroid biosynthesis and metabolism can produce adverse health effects, even though the inducing agent would not be detected as an [EDC] using receptor-based test systems. This is an important area of study because some examples of [endocrine disruption] occurring in animals derive from exposure to inhibitors of steroidogenic enzymes such as 5α-reductase and aromatase. Some such agents are known to be active in humans and are used successfully in the treatment of a range of human hormonal conditions.” The authors suggested that evaluation of such effects will require integrated screening that incorporates in vitro and in vivo technologies.
A comprehensive report issued in 2002 by the World Health Organization’s International Programme on Chemical Safety, titled Global Assessment of the State-of-the-Science of Endocrine Disruptors, reached similar conclusions. The report stated that “although it is clear that certain environmental chemicals can interfere with normal hormonal processes, there is weak evidence that human health has been adversely affected by exposure to endocrine-active chemicals. However, there is sufficient evidence to conclude that adverse endocrine-mediated effects have occurred in some wildlife species.” Citing the fact that studies to date examining EDC-induced effects in humans have yielded inconsistent and inconclusive results, the group wrote that, although that explains their characterization of the evidence as weak, “[that] classification is not meant to downplay the potential effects of EDCs; rather, it highlights the need for more rigorous studies.”
The Global Assessment further states that the only evidence showing that humans are susceptible to EDCs is currently provided by studies of high exposure levels. There is, in fact, clear evidence that intrauterine EDC exposures can alter human reproductive tract development and physiology. The most thoroughly characterized example is DES, the synthetic estrogen prescribed to millions of pregnant women in the United States and elsewhere from the 1940s to the 1970s to prevent miscarriage. The drug is known to have caused a rare form of vaginal cancer in thousands of daughters of women who took DES, as well as a variety of adverse reproductive tract effects in both the daughters and sons of those women.
The DES situation could be seen as a worst-case scenario for prenatal EDC exposure—the deliberate delivery of a potent estrogenic chemical in high doses. Viewed another way, it has provided researchers a rare opportunity to study the effects of prenatal EDC exposure in a relatively controlled fashion, with a well-defined population and well-characterized exposure to a single potent agent.
Over the course of her research, Newbold has developed a mouse model of DES exposure that has proven extremely useful in studying the effects of DES and other environmental estrogens, particularly those outcomes that may be manifested only later in life. “With the experimental model, there are a lot of questions we can ask with DES that will tell us about the weaker environmental estrogens,” she says. “We can change the timing of exposure and the amount of exposure, and we can look at different target tissues.”
The animal model has replicated numerous abnormalities reported in DES-exposed humans, and has also predicted some human outcomes. “We have published documentation [see, for example, the October 1985 issue of Cancer Research and volume 5, issue 6 (1985) of Teratogenesis, Carcinogenesis, and Mutagenesis] that a number of the reproductive anomalies seen in DES-exposed mice, such as retained testes and abnormalities in the oviduct in females, were also later reported in DES-exposed humans,” says Newbold.
The Phthalate Connection
But reliable correlations between animal data and human outcomes have proven elusive, particularly when it comes to showing an association between human exposures to environmental EDCs at ambient levels (that is, unrelated to spills or other acute contamination events) and adverse health effects. That may be about to change for one class of chemicals—phthalates.
Phthalates are commonly used in a wide variety of consumer products such as solvents, soft plastics, and cosmetics. The National Health and Nutrition Examination Survey showed that the majority of the U.S. population carries a measurable body burden of several phthalates. There is an extensive body of literature regarding the effects of prenatal phthalate exposure in rodents. Those effects include an association between intrauterine exposure and abnormalities in male animals in a biomarker known as anogenital distance (AGD), or the distance between the rectum and the base of the penis. AGD has been shown to be a sensitive measure of prenatal antiandrogen exposure. This pattern of genital dysmorphology has come to be known as the “phthalate syndrome.”
In the first study to look at the link between AGD and EDC exposure in humans, Shanna Swan, a professor of obstetrics and gynecology at the University of Rochester, and her colleagues collected data from 85 mother–son pairs participating in the Study for Future Families, a multicenter pregnancy cohort study. The mothers’ urine was analyzed for the presence of several phthalate metabolites, and the infant boys, aged 2–36 months, were examined for genital developmental characteristics, including AGD, which was standardized for weight to develop an anogenital index (AGI).
Although the researchers found no sign of frank genital malformations or disease, they did discover an association between elevated concentrations of four phthalate metabolites in the mothers and shorter-than-expected AGI in the infants, as reported in the August 2005 issue of EHP. And, importantly, shortened AGI was found in infants exposed prenatally to phthalate metabolites at concentrations comparable to those found in one-quarter of the U.S. female population. The boys with short AGI were also significantly more likely to have incomplete testicular descent (cryptorchidism). “We know that incomplete testicular descent is a risk factor for poorer semen quality, lower sperm counts, [impaired fertility], and testicular cancer,” says Swan. Although it is obviously impossible to predict adult outcomes, she says these infants may be at risk of testicular dysgenesis syndrome (TDS) in the future.
TDS is a concept put forth by Danish researcher Niels Skakkebæk and colleagues, in which four adverse male reproductive end points—impaired semen quality, cryptorchidism, hypospadias (abnormal location of the urethra), and testicular cancer—are risk factors for each other. Says Swan, “The idea is that the development of the testis is interrupted in fetal life, and that this has consequences in adult life, as well as at birth. That certainly is something we’ve seen in rodents, and this study is the first evidence we’ve seen of TDS in humans.”
Swan’s study is among the first to combine a population-based, measurable, low-level EDC exposure, observed physiologic effects, and solid biological underpinnings. Even skeptic Safe says that this is the kind of study needed to begin to answer the many questions about EDCs and human health. “This looks to be a good approach, and suggests a correlation,” he says. “Whether it’s causal of anything and whether it holds up or not, I don’t know. It needs to be repeated in different locations and with more and more integrated measurements.” Swan plans to do just that, as well as to follow up on her current pregnancy cohort by measuring gender role behaviors in both the male and female children, who are now between 2 and 5 years old.
The Phthalate Esters Panel of the American Chemistry Council, a trade organization based in Arlington, Virginia, maintains that “there is no well-established and credible evidence for adverse effects [due to phthalates] in humans at environmentally relevant doses,” says panel manager Marian Stanley. With regard to Swan’s study, Stanley says, “It correlated some effects in infant males with some lower-molecular-weight phthalates, particularly diethyl phthalate, for which effects in rodents occur only at very high doses, and which is not considered to pose reproductive or developmental concerns by reviewing government agencies.”
Stanley also points to questions about the biomarker used in the study. “The measurement that was used is something that I think is still subject to debate. You see the AG distance in rodents, and while it is a marker of something, it is certainly not a biological effect,” she says. “I think the study has been overinterpreted by lots of other people [besides] the authors of the study.”
EDCs and Sex Ratios
Sex ratio—the proportion of male to female live births—is very constant on a worldwide basis, typically ranging from 102 to 108 male births for every 100 female births. In recent years, however, a number of reports have suggested that environmental and occupational exposures to EDCs may be altering the sex ratio within given human populations.
In one such study, appearing in the July 2005 edition of Human Reproduction, a group of Swedish researchers analyzed blood and semen samples from 149 fishermen to investigate whether exposure to the persistent organochlorine pollutants CB-153 (a PCB) and p,p′-DDE affected the proportion of Y-and X-chromosome-bearing sperm. They discovered that elevated exposure levels of both chemicals were positively associated with a higher proportion of Y-chromosome sperm. The researchers conclude that their findings add to evidence that exposure to persistent organic pollutants may alter the offspring sex ratio, with the higher proportion of Y-chromosome sperm likely tending to lead to a higher proportion of male births.
A study appearing in the October 2005 issue of EHP takes an epidemiologic approach to the issue. Constanze Mackenzie, a member of the Faculty of Medicine at the University of Ottawa, and colleagues report a distinct skewing of the sex ratio within members of the Aamjiwnaang First Nation community near Sarnia, Ontario. They found a severe decline in the proportion of boys born among the Aamjiwnaang over the last five years, and a lesser though still significant decline over the past ten years. Although no causal factors were determined, the authors note that the community is located in immediate proximity to several large petrochemical, polymer, and chemical plants, and that previous studies—such as those following the 1976 industrial accident in Seveso, Italy—have shown that exposure to contaminants such as EDCs can impact sex ratios within small communities near such industrial facilities. The authors suggest that further assessment should be pursued to identify potential exposures among community members. [For more details on this study, see “Shift in Sex Ratio,” p. A686 this issue.]
How Low Do They Go?
When is a hypothesis no longer a hypothesis, but a validated scientific concept ready to drive regulatory and policy decision making? When it comes to the so-called “low-dose hypothesis” regarding the biological activity or adverse effects of low-dose exposures to EDCs, that is the key question. The issue has been debated for years, since vom Saal’s group first published in the January 1997 issue of EHP their findings of enlarged prostate in male mice whose mothers had been fed low doses of BPA. Today, the controversy over whether vom Saal’s findings have been sufficiently replicated, and whether the U.S. Environmental Protection Agency (EPA) should revise its risk assessment process to reflect the potential for adverse effects of low-dose EDCs, is still going strong.
Some proponents of the low-dose hypothesis argue that the traditional toxicologic approach to risk assessment is an inappropriate method to assess EDCs. The current protocol assumes a linear dose-dependent response to chemical exposures, determines the lowest level at which there is an observed adverse effect, and then adds a safety factor to arrive at an official reference dose—the daily human intake assumed to be safe. Experimental work by vom Saal and others has postulated that EDCs exhibit a U-shaped dose–response curve, with biological activity stimulated at very low doses—often several orders of magnitude below current reference doses—as well as very high doses.
Proponents also state that the process of endocrine disruption itself is inherently different from many other toxicologic processes, affecting a variety of highly sensitive pathways (especially in the fetus) via novel mechanisms of action, many of which are as yet poorly understood. Also, they say, endocrine-signaling pathways that mediate responses to EDCs have evolved to act as powerful amplifiers, resulting in large changes in cell function occurring in response to extremely small concentrations.
One chemical that has become a lightning rod in the debate is BPA. By vom Saal’s count, there are now more than 100 published peer-reviewed studies showing significant biological effects of low doses of BPA (almost half published within the last two years) compared to 21 reporting no effect. He is convinced that widespread exposure to BPA poses a threat to human health.
Not so, claims Steve Hentges, executive director of the Polycarbonate Business Unit of the American Plastics Council: “For our purposes, what we have to know is, does BPA cause health effects in humans at any relevant dose, particularly at the levels at which people are actually exposed? When you look at all of the evidence together, and in particular look at the comprehensive studies that are designed to look for health effects, you don’t find them.”
The industry group also believes that the weight of evidence does not support the concept of a low-dose effect for BPA. “And it’s not just us saying that,” says Hentges. “Indeed, every government body worldwide that’s looked at it has reached effectively the same conclusion in terms of how they regulate BPA or consider regulating it.” He acknowledges that there has been quite a bit of new research activity in this area within the past few years, but states that “even though new research has been conducted, we believe that the weight of evidence has not shifted.”
Where does the EPA stand on these issues? The agency’s Office of Research and Development is in the midst of implementing a multiyear plan to set the EPA’s agenda and goals in the area of EDC research. The plan is part of the agency’s Endocrine Disruptors Research Program, a five- to ten-year research agenda it started in 2001 to look comprehensively at the science surrounding EDC exposures and effects. The integrated program was launched at about the same time that a congressional mandate, under the 1996 Food Quality Protection Act, directed the EPA to develop a screening and testing program for EDCs.
The EPA’s stance is that the jury is still out on both the public health impacts of EDCs and the need to incorporate low-dose methodologies into the agency’s risk assessment protocols. Elaine Francis, director of the Endocrine Disruptors Research Program, says the EPA needs to conduct a lot more research before any definitive public health statements can be made about this class of compounds. “When you look at such a diverse group of organisms that have been impacted in wildlife, and certainly laboratory rodent species,” she says, “there is enough concern that we recognize the importance of developing a body of work in humans to try to characterize any impact [EDCs] might be having on humans.”
The agency is currently funding three research grants in the area of low-dose EDC exposures, partly in response to the conclusions reached in a 2000 peer review and subsequent report on the low-dose issue held by the National Toxicology Program at the EPA’s request. In the 2001 Report of the Endocrine Disruptors Low-Dose Peer Review, that expert panel acknowledged that low-dose effects had been sufficiently documented at that point in time for the EPA to consider revisiting its current testing paradigm.
“The general consensus was that more work needed to be done in this area,” says Francis. “Since that time, we would still agree that there has not been enough information to indicate that the existing approaches are ones that would not be valid for endocrine disruptors. But we left the door open that we would need to do more research, and the best we could do at this point is to support and promote research in that area, and we’ve done that.”
Vom Saal is of a different opinion: “In the risk assessment process for chemicals as currently conducted, the maximum tolerated dose is used as a reference, and a span of typically not more than fiftyfold in the dose range is the maximum that anyone ever uses in the studies. Studies [from the 1 January 2005 issue of Cancer Research and the April 2005 EHP show] literally millions of fold below that dose range in adverse effects . . . from BPA, and when you have that type of unbelievable discrepancy, for the EPA to come out as it recently did and state that it has no intention of testing low doses as part of the testing process [implies] that you no longer have a scientifically based process—it is an entirely politically driven process, because they are explicitly ignoring the scientific findings that are out there.”
From her perspective, Newbold feels that although there is no question that EDCs have low-dose effects, more research needs to be done to document adverse effects in humans. “We spend an awful lot of time arguing whether there are low-dose effects or not. That just infuriates me,” she says. “There are low-dose effects. There have always been low-dose effects. The question is, are they adverse? We don’t know, and we’ve got to design studies to get answers to that question.” She adds, “In order to take this argument to a whole other level, we’re going to have to have more epidemiology studies. I know it happens with mice, but I don’t know what happens with humans.”
Connecting the Gender Dots
It’s premature to call it a theory; at this point, it barely qualifies as a hypothesis: some observers are putting forth the proposition that prenatal EDC exposures may affect gender identity—how a person identifies him- or herself, regardless of physical characteristics. This idea presupposes two basic concepts: first, that transgenderism (in which a person experiences “gender dysphoria,” a strong feeling of having been born the wrong sex) is physiological in origin, most likely due to events during prenatal neurological development; second, that intrauterine EDC exposures can and do disrupt prenatal neurological development.
A paper in the 2 November 1995 issue of Nature, among other reports, lends credence to the first concept. Jiang-Ning Zhou and colleagues at the Netherlands Institute for Brain Research studied heterosexual men and women, homosexual men, and male-to-female transsexuals. They reported finding a distinctly female brain structure in genetically male transsexuals (men who had gone through hormonal treatment and irreversible sexual reassignment surgery to become women). The volume of the central subdivision of the bed nucleus of the stria terminalis (BSTc), a sexually dimorphic brain area that is essential for sexual behavior, is larger in men than in women. Anatomical study results showed that BSTc volume did not differ significantly between heterosexual and homosexual men, and that BSTc volume was 44% larger in heterosexual men than heterosexual women. In the male-to-female transsexuals, BSTc volume was only 52% that of the reference males—a volume analogous to that seen in the women. The authors write that these findings “support the hypothesis that gender identity develops as a result of an interaction between the developing brain and sex hormones.”
But a study by Wilson C.J. Chung and colleagues published in the 1 February 2002 Journal of Neuroscience complicates this picture. This group, also from the Netherlands Institute for Brain Research, reported that BSTc size differentiation between men and women became significant only in adulthood, implying that the phenomenon may be more effect than cause. The authors do point out, however, that the lack of marked sexual differentiation of the BSTc volume before birth and in childhood does not rule out early gonadal steroid effects on BSTc functions. They point to earlier animal experiments showing that fetal or neonatal testosterone levels in humans may first affect synaptic density, neuronal activity, or neurochemical content during early BSTc development, and that “[c]hanges in these parameters could affect the development of gender identity but not immediately result in overt changes in the volume or neuronal number of the BSTc.”
On the other side of the ledger, in the June 2002 edition of EHP Supplements, Bernard Weiss, a professor of environmental medicine and pediatrics at the University of Rochester, reviewed the existing literature on sexually dimorphic nonreproductive behaviors as indicators of endocrine disruption. Weiss made a strong evidence-based case that “gender-specific regional differentiation of the brain and, ultimately, its expression in behavior are guided by the gonadal hormones,” and that the process is subject to interference by drugs and environmental contaminants. He points out that sex differences in performance and behavior are not—but should be—a recognized criterion in developmental neurotoxicity testing.
So who out there is connecting these dots?
Scott Kerlin is a Ph.D. social scientist at the University of British Columbia. He devotes considerable time to monitoring the international scientific literature on DES and other EDCs as well as to researching and writing about the long-term health effects of pre-natal DES exposure on males. He is himself the son of a woman given DES in pregnancy.
Kerlin recently conducted a survey study of 500 members of the DES Sons International Network, an online resource for men who know or strongly suspect they were exposed to DES in utero. In a paper presented in August 2005 at the International Behavioral Development Symposium in Minot, North Dakota, he reports that more than 150 respondents identified themselves as having any of a variety of gender-related disorders. Kerlin does not claim that DES causes these gender disorders, but feels that his results indicate that such outcomes should be included in research related to the potential effects of prenatal EDC exposures.
The Road Ahead
It’s going to be very difficult to ever conclusively answer the basic question of whether low-level EDC exposures during development are causing deleterious reproductive or gender-related outcomes in humans. Scientists agree that one of the major challenges is to address the issue of mixtures. Typically, researchers look at the impact of one chemical at a time, but environmental exposures regularly involve an unpredictable mix of chemicals, with exposures varying widely in dose and duration. It is unlikely there will ever be a comprehensive understanding of how the many EDCs in mixtures interact with each other and with human physiology.
Convincing epidemiologic evidence of adverse effects in humans is also difficult to come by, but will be necessary to translate scientific findings into concrete actions to protect public health. Swan’s study, one of the first of its kind to appear thus far, may serve as a methodological model for future investigations of low-level EDC exposures.
Do we know enough now that steps should be taken in the policy and regulatory realm? Some observers, taking a precautionary approach, think that we do. For example, there are bills under consideration in the California and New York legislatures to restrict the use of certain phthalates in toys, child care products, and cosmetics, and a California bill would ban the use of BPA in products meant for use by children aged 3 years or younger. Also, the European Parliament voted in 2005 to ban the use of three phthalate plasticizers (DEHP, di-n-butyl phthalate, and benzyl butyl phthalate) in toys and child care items, and to prohibit the use of three others (diisononyl phthalate, diisodecyl phthalate, and di-n-octyl phthalate) in toys and child care items that children can put in their mouths.
Theo Colborn, a professor of zoology at the University of Florida and author of the 1996 book Our Stolen Future, believes the time for action is now. “In the animals, it was at the population level that we really began to realize what was going on,” she says. “If we’re going to wait to see population effects for all of these concerns that we have in the human population, it’s going to be too late.” She points out that we’re already into the fourth generation of individuals who have been exposed in utero to chemicals that had never been used before the mid-1930s or early 1940s.
Swan agrees that there is sufficient knowledge at this point to call EDC exposures a serious threat to public health. “I don’t think it’s necessarily a threat to individuals,” she says, “but I think that as a population we are threatened. I’m not predicting the end of the species or anything like that, but I think the increasingly alarming trends that we’re seeing, in terms of couples that can’t conceive or couples whose babies have undescended testicles, and so on, can have an impact on the population as a whole.”
Other observers are not so sure. Harry Fisch, director of the Male Reproductive Center at Columbia University Medical Center, specializes in the diagnosis and treatment of male infertility. From his clinical perspective, other factors—including other exposures—are more important than EDCs. “The sky is not falling,” he says. “A lot of times there’s extrapolation from high-dose exposure to low-dose exposure. I think one of the biggest culprits for the abnormalities we see that’s been totally ignored is [increased] parental age. Also, we need to look at things we’re doing to ourselves before we start blaming low-level chemicals. For example, what does cigarette smoking do compared to Saran Wrap? What about the diets we eat, the high-fat intakes? Before we start blaming others, we need to look at ourselves to determine the impact of our lifestyles.”
Although plastic wrap may not be responsible for human infertility, the scientific evidence fueling growing concerns about the effects of ambient environmental exposures to EDCs cannot simply be dismissed. “Vigilance is the key word here, because there are so many chemicals out there,” says Burger. “Understanding the effects of chemicals is a three-pronged approach. It’s being sure that we have wildlife models and people who are watching wildlife populations to see quickly if something detrimental happens. It’s having really good epidemiological studies and vigilance of people in various places. And it’s backing those two up with laboratory science immediately when a problem turns up, to try to ascertain the cause quickly.”
Watching wildlife.
Research has documented reproductive and developmental abnormalities linked to EDC exposures in wildlife species such as alligators and polar bears, although what these results mean for humans is still unknown.
A question of Y.
A Swedish study of fishermen exposed to CB-153 and p,p′-DDE associated elevated levels of these chemicals with a higher proportion of Y-chromosome sperm, suggesting that exposure to EDCs could skew the ratio of boys to girls.
Ubiquitous exposure, unknown consequences.
Humans are exposed to EDCs through many routes including pharmaceuticals, air pollution, pesticides, and drinking water, but the effects of environmental exposure are largely unknown.
Gender basis.
In a study of the brain region known as the BSTc, which varies in size by sex, the volume of the BSTc for male-to-female transsexuals was analogous to that seen in women, leading the authors to speculate that the findings “support the hypothesis that gender identity develops as a result of an interaction between the developing brain and sex hormones.”
|
16203228
|
PMC1281309
|
CC0
|
2021-01-04 23:41:28
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A670-A677
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.113-a670
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0067816203229EnvironewsSpheres of InfluenceDamming the Flow of Drugs into Drinking Water Hemminger Pat 10 2005 113 10 A678 A681 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
Roughly 100 pharmaceuticals have now been identified in rivers, lakes, and coastal waters through out Europe and the United States in concentrations of parts per billion to parts per trillion. The first major European studies on this topic—in journals such as volume 67, issue 1–4 (1997) of the International Journal of Environmental Analytical Chemistry and the November 1998 issue of Water Research—examined German ground and surfaces waters, and found occurrences of drugs including cholesterol regulators, analgesics, and antiseizure medications. Since that time, numerous other studies have documented the presence of pharmaceuticals, including potential endocrine disruptors, in other locales as well.
So far there is no evidence of adverse human health effects due to traces of pharmaceuticals in water. But scientists have linked certain pharmaceuticals with disturbing ecosystem changes. For example, in volume 8 (1994) of Chemistry and Ecology, researchers demonstrated that the feminization of fish—male carp and trout producing vitellogenin, an egg protein usually found only in females—was associated with exposure to sewage effluent now known to contain ethinyl estradiol, the active ingredient in birth control pills.
There is much concern about what is not known: ecotoxicity data are available for less that 1% of human pharmaceuticals, according to estimates published in the April 2004 issue of Regulatory Toxicology and Pharmacology. Today, intensive research is under way to investigate the effect of human medications on the environment.
In 1999, in response to these concerns, the European Medicines Agency (EMEA) began drafting guidance that outlined an environmental risk assessment procedure to accompany pharmaceutical companies’ applications to market new drugs in Europe. The latest draft was published in January 2005, after several revisions, and the public comment period closed in April 2005. Scientists and pharmaceutical companies alike hope the guidance will be finalized later this year.
The proposed European guidance is the first to recommend long-term ecotoxicity testing for environmental risk assessment of pharmaceuticals from the outset of the proposed testing program (in contrast, U.S. Food and Drug Administration [FDA] requirements for chronic ecotoxicity testing come later in that agency’s assessment). The European guidance is also the first to take into account the possibility of environmental effects from extremely low concentrations of bioactive substances, such as endocrine disruptors.
If finalized, the guidance could call for substantially more testing of new drugs than has been demanded thus far. Its implementation would also generate much-needed chronic ecotoxicity data. “The main advance in this draft guideline is that we really address this issue and get more information on the toxicity of these compounds,” says Thomas Heberer, an environmental chemist at the Technical University of Berlin and coauthor of many papers on the topic, including the 1997 International Journal of Environmental Analytical Chemistry report.
What the Draft Guidance Covers
The draft guidance outlines the risk assessment procedure for new active pharmaceutical substances, their metabolites, and possibly excipients (the inert substances in which a drug is delivered) if they are deemed similar to chemicals with known adverse environmental effects. It does not apply to drugs already on the market. If an environmental risk is found, the guidance recommends that the manufacturer take appropriate precautionary and safety measures to limit the product’s environmental impact. The guidance specifically recommends the labeling of pharmaceuticals when there is a possibility of an environmental risk, to educate people about how best to dispose of expired or unused medicines.
The guidance applies only to potential environmental risks that are a consequence of people storing, taking, and excreting medicines. The potential risks posed by the manufacture of drugs are not addressed, nor does the guidance apply to “orphan” drugs used only to treat rare diseases. Separate guidance governs medicinal products containing genetically modified organisms.
Proposed EMEA Protocols
The EMEA risk assessment protocol is a tiered process that begins with a rough calculation of the aquatic predicted environmental concentration (PEC) of the new drug. During this Phase I prescreening, substances whose PEC is deemed too low to be of concern to environmental health are ruled out for further assessment. Vitamins, electrolytes, amino acids, peptides, and proteins are exempted by the guidance because they are not tailored active ingredients (unlike, for example, a drug that interacts with a receptor) and thus are deemed “unlikely to result in significant exposure of the environment.” However, the guidance does note that certain substances that are likely to cause effects at very low concentrations, such as endocrine disruptors, may need to be addressed regardless of the quantity released into the environment.
Phase II begins with Tier A testing, which aims to determine the aquatic fate and effects of the drug. Its degradability, potential to bioaccumulate, adsorption on sewage sludge, and toxicity to sewage microbial populations are evaluated from the results of standard tests also used in the FDA risk assessment. Also included in Tier A of the EMEA protocol is the long-term testing of fish, Daphnia (water fleas), and algae to assess the predicted “no effect” concentration (PNEC) of the new drug for each of these species. The PEC is further refined at this stage in the EMEA assessment by taking into account the pharmaceutical company’s projected sales forecast for the drug.
The risk assessment is terminated if the outcome of Tier A testing results in a PEC lower than the PNEC. However, if the PEC is greater than the PNEC in either water, sediment, the sewage treatment plant, or soil (where sewage sludge has been spread as a fertilizer), this indicates a potential risk, and further Tier B testing is initiated. These tests follow the protocol in the European Technical Guidance Document to further investigate the risk posed by the drug to the environment. For instance, where there is a potential risk to soil, tests would be conducted to determine the drug’s biodegradation in soil, its toxicity to soil invertebrates, and its acute effects on plants and soil microorganisms.
At this stage, data on the drug metabolism and excretion profile may be consulted to allow a more accurate calculation of the PEC and determine whether metabolites need to be tested. The EMEA guidance recommends that metabolites exceeding 10% of the drug residue should be assessed for environmental risk. If this round of testing indicates that the PEC of the drug will be greater than the PNEC, then pharmaceutical companies following the European approach must propose recommendations to limit the drug’s impact on the environment.
There are two major differences between the proposed EMEA approach and the existing FDA approach. First, the FDA protocol turns to chronic testing only if acute testing indicates a risk or if there is an indication that the drug could bioaccumulate. The latest scientific research suggests that acute testing is not a reliable indicator of all chronic effects, however, and the EMEA document reflects this finding.
Second, the trigger concentrations of pharmaceuticals that prompt risk assessment under the FDA and EMEA guidance differ by a factor of 10 when dilution is taken into account. “The way the two guidelines express this trigger may be confusing,” says Virginia Cunningham, director of environmental sustainability sciences for GlaxoSmithKline. She explains that the EMEA’s trigger of 0.01 microgram per liter (μg/L) reflects a surface water concentration, whereas the FDA’s 1.0 μg/L trigger reflects an “expected introduction concentration,” or the concentration of a compound in sewage effluent.
The EMEA trigger of 0.01 μg/L is calculated from the maximum daily dose of the drug per patient and the assumption that 1% of the population is treated daily with the drug; this is divided by the amount of waste-water per person per day and a dilution factor of 10. The FDA trigger corresponds to a PEC in surface water of 0.1 μg/L, assuming a dilution factor of 10, and is calculated from manufacturers’ sales estimates.
The consideration given to metabolites and the provision for the introduction of scientific experts into the risk assessment process—both part of the revisions to the 2003 guidance—are welcomed by scientists. “It allows for experts to be drawn into the discussion and give their opinions rather than be sticking blindfolded to a number,” says Evelyn O’Brien, a scientist in the Ecotoxicology Workgroup at the University of Konstanz in Germany and coauthor of a discussion of the draft guideline published in the July 2004 Trends in Biotechnology.
One caution added by zoologist Theo Colborn, whose seminal 1996 work Our Stolen Future uncovered the dangers of endocrine disruptors in the environment, is that conflict of interests for experts working in academia but funded by drug companies must be revealed. “The important thing is,” she says, “that in [the United States] they’re selecting experts to do things like this on campuses where the particular department that that individual is working in oftentimes receives tremendous amounts of grant money from the pharmaceutical company. Openly admitting conflict of interest is so important.”
The EMEA website notes that members of the agency’s scientific committees “are not permitted to have any direct financial or other interests in the pharmaceutical industry. . . . They are required to make an annual declaration of their financial interests and also any indirect interests which could relate to the pharmaceutical industry.” Colborn also hails the guidance for including excipients as well as active ingredients in the risk assessment process. For instance, phthalates such as diethyl phthalate and dibutyl phthalate, used as plasticizers in the coating of some site-directed drugs, may be a potential source of phthalates for people taking these drugs, as reported in the May 2004 issue of EHP.
Limitations of the Guidance
There are certain serious, though perhaps unavoidable, limitations to the guidance. One is the fact that they are not retroactive. “The only thing that [researchers] are concerned about is that the guidance only concerns those pharmaceuticals that are not yet on the market,” says Heberer. “It’s our main concern about this guideline, but compared to the situation in the past it’s really an advance.” But even if future legislation required the environmental risk assessment of drugs already on the market, the big question would be who should do the testing since the originator of a drug is often no longer the main manufacturer.
Another major problem is that monitoring may be difficult. “There are problems detecting certain substances that have been on the market for years,” says O’Brien. Examples of such hard-to-detect drugs include the antidepressants known as selective serotonin reuptake inhibitors (which include Paxil, Prozac, and Zoloft). “So the analysis can be quite difficult,” she says, “and that’s one of the main stumbling features.”
Further, it is not clear how drugs that pose risks will be handled, apart from the addition of labels to recommend appropriate disposal of expired drugs. Another emerging area of concern in North America and Europe alike is the disposal of used birth control patches and hormone replacement patches. Because pharmaceuticals can save lives, the guidance does not suggest removing them from the market even when a risk is found.
“I think there’s going to be a lot of emphasis on labeling, and also on treatment processes,” says Alistair Boxall, a senior lecturer at York University and Central Science Laboratory in England. “So perhaps if you’ve got a hospital where cancer drugs are being used, it may be that we have to start putting treatment processes on the end of the [sewer] pipes of those hospitals to remove some of the drugs.”
Drug take-back programs for expired pharmaceuticals are in place in parts of Europe, so labeling drugs with instructions to return unused portions to a pharmacy makes sense. By comparison, in the United States, the Controlled Substances Act complicates such schemes because it prohibits patients from transferring controlled medicines to anyone other than a law enforcement official. However, a drug return program has recently been legislated (though not implemented) in Maine.
Another limitation, also difficult to avoid, is that the draft guidance only briefly addresses the possibility of additive or synergistic effects, noting that an assessment factor of 10 is applied to the PNEC to account for extrapolation from lab data to field impacts. “It’s worth pointing out that the guidance is written as if the concern is for a single drug in isolation,” says Christian Daughton, chief of the environmental chemistry branch at the Environmental Protection Agency National Exposure Research Laboratory. “But if a drug shares a common mechanism of action with other drugs, or even other pollutants, there’s the possibility for additive effects.”
Some scientists and drug companies are concerned that assumptions in the guidance could lead to unrealistic PECs. The initial calculation assumes the worst-case scenario: that the drug is not metabolized or degraded at all, so the full dose ends up in the environment (this is one of 30 points raised by the Pharmaceutical Research and Manufacturers of America in their comments on the guidance). But others worry that actual concentrations in the environment could be higher than the calculated PEC due to the guidance’s assumed 1:10 dilution factor for sewage effluent entering rivers. In farming areas, water levels drop precipitously in dry weather when water is drawn for crops and cattle, so the 1:10 dilution factor could be too high. Colborn, a Colorado resident, says, “Most of the river water that’s in this part of the West is coming from returned sewage treatment plants.” O’Brien argues the same point in cities where the influx of people stretches the capacity of sewage treatment plants.
Another problem noted by O’Brien is that peak or seasonal variations are not taken into account—flu epidemics, drought, or heavy snowfall could temporarily increase drug concentrations in specific places to values higher than the calculated PEC. Colborn also comments that local use of pharmaceuticals differs, reflecting, for example, recent visits by pharmaceutical representatives telling doctors about new drugs. “To estimate that pharmaceuticals will be released homogeneously across a particular region is, I think, mistaken,” she says. Daughton addressed these and related issues in greater detail in the May 2003 issue of EHP.
One worry for pharmaceutical companies is that the increased amount of testing required could translate into costly delays for the release of new drugs. About 50 new drugs come onto the market in the United States each year, and approximately a dozen of those are predicted to occur above the trigger concentration requiring them to undergo the first level, or Tier A, of risk assessment testing.
But only one new drug in the last few years has gone on to the next level to be tested for environmental risks through chronic ecotoxicity tests, according to Florian Zielinski, a chemist at the FDA Center for Drug Evaluation and Research. “In fact, in the States, almost all pharmaceuticals in the Tier A assessment will come out at under one microgram per liter,” says Chris Metcalfe, a professor in the Environmental and Resource Studies Program at Trent University in Ontario, “whereas in the EU there will be a fair number of pharmaceuticals which will move from the Tier A to the Tier B as a result of their lower thresholds.” British labs put about 20 new pharmaceutical products on the market each year.
Forging Ahead
Since neither the EMEA guidance nor its U.S. sister document addresses pharmaceuticals already on the market, there is much research into whether wastewater treatment can economically remove pharmaceuticals. Increased retention time within treatment plants, chlorination, ozonation, and the natural reduction of a compound’s mass or concentration over time due to processes such as biodegradation all increase the removal of some drugs from wastewater; more advanced treatments such as adding activated carbon or reverse osmosis can remove even more. “But there’s never a silver bullet,” says Shane Snyder, research and development project manager of the Southern Nevada Water Authority. “There’s always a catch.”
The catch with ozone treatment is that it forms bromate, which is a regulated disinfection by-product; with chlorination, the catch is that chlorine combines with ammonia in the sewage treatment system to form chloramines, which are not strong oxidants and so cannot break down compounds such as estrogens. However, chlorination can destroy almost all the estrogens if ammonia is removed first, says Snyder. But even with the use of reverse osmosis (which removes pharmaceuticals down to parts per trillion) and the addition of activated carbon, there’s the problem of what to do with the retained contaminants.
Although Europe has been at the forefront of recognizing and addressing the potential environmental hazard posed by pharmaceuticals, other countries are perhaps beginning to catch up. In the United States, for example, the Federal Interagency Task Group on Pharmaceuticals and Personal Care Products was formed in September 2004. This group comprises seven federal agencies and is chaired by the FDA. The group had its first face-to-face meeting in July 2005 to identify federal research needs and gaps. One of the questions raised was how much of the estrogen in wastewater comes from synthetic sources.
In Canada, the Environmental Impact Initiative was formed in 2001 in response to growing evidence that pharmaceutical substances are being found in the environment. The initiative, which accepted public comments through September 2005 on proposed options for regulating these substances, may result in new rules for the environmental assessment of substances in products regulated under the Food and Drugs Act, according to Health Canada. Japan is also in the process of formulating a plan for environmental risk assessment of pharmaceuticals with sales exceeding one ton per year.
In the meantime, the EMEA draft guidance is seen as an appropriate response to an emerging issue which includes possible risks not just from pharmaceuticals but also from personal care products. “What has come into the scientific literature is that most pharmaceuticals do not show acute ecotoxicity, so the whole mindset is shifting to chronic toxicity, and I think the EMEA guideline is trying to reflect that,” says Cunningham. “None of the people I talk to have a problem with that.”
|
16203229
|
PMC1281310
|
CC0
|
2021-01-04 23:41:28
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A678-A681
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.113-a678
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0068216203230EnvironewsInnovationsGetting the Lead Out of Electronics Black Harvey 10 2005 113 10 A682 A685 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
The electronics industry is learning to do without: it is having to abandon one of its long-time staples, lead–tin solder. For decades lead–tin solder has been used to attach electronic components to printed wiring boards. However, with the body of evidence pointing to serious adverse health effects of lead, the search for a replacement has spawned intense effort in the electronics industry and in universities. Now scientists think they may have found some promising leads: solders made of alternative alloys and polymer formulations known as electrically conductive adhesives (ECAs).
The Linchpin of Electronics
Solder is the “linchpin of electronics manufacturing,” says Jack Geibig, acting director of the Center for Clean Products and Clean Technologies at the University of Tennessee. “Without it, it’s difficult to achieve a proper electronic connection that is durable and reliable.”
Lead has been ideal for solder. In fact, says Carol Handwerker, chief of the metallurgy division at the National Institute of Standards and Technology, “The whole electronics infrastructure was designed around the melting point and physical properties of [lead].” Lead is malleable and thus easy to work with, and it doesn’t fracture, she says. When lead is combined with tin in the correct proportion (63% tin to 37% lead), the resulting alloy has a low melting point of 183°C, which is another advantage, Geibig says: “If you’re not operating at really high temperatures, you have more control over processes, so that the processes aren’t sensitive to slight temperature variations, which are costly to control.” Low temperatures also mean less strain on the equipment and materials (such as printed circuit board and components) that must be heated as part of the assembly process.
The main impetus for the industry to leave lead behind is a ban on lead in electronics imposed by the European Union. Under the Restriction of Hazardous Substances directive, as of 1 July 2006 lead must be replaced by other substances in electronic equipment. (The directive also bans mercury, cadmium, and hexavalent chromium.) Any electronic components bound for Europe are subject to the ban.
Lead is not a problem when contained in electronic equipment, says Robert Donkers, an environmental counselor for the European Commission who is based in Washington, DC. However, when electronic components are deposited in landfills, he says, people may scavenge for equipment and break it open, or the lead may leach out of landfills and into drinking water. The risk is compounded in countries that receive massive imports of electronic waste. In China, for example, unprotected workers, including many children, strip recyclables out of electronic components in a cottage industry of sorts [see “e-Junk Explosion” in the April 2002 issue of EHP].
Lead exposure, even at low levels, is well known for its harmful effects on children, resulting in lowered IQ. Lead also affects the ability to pay attention. Children exposed to low levels may appear hyperactive and irritable, according to the American Academy of Child and Adolescent Psychiatry. The current maximum allowable level for blood lead in the United States is 10 micrograms per deciliter (μg/dL).
Alternative Alloys
The main approach to replacing lead in solder has been to look for other metals as substitutes. Electronics manufacturers began to look for alternative metals in the 1990s, notes Handwerker, when now-abandoned proposals were being discussed in the United States to ban lead in electronics.
Ronald Gedney, a consultant for the International Electronics Manufacturing Initiative (iNEMI), a technology consortium, has been intimately involved in the search for alternatives. He says that a search by industry experts for possible replacements for lead–tin solder winnowed down 75 metal alloy alternatives to about half a dozen. “We decided the biggest benefit for the industry would be to pick one solder, concentrating our development and research efforts on one alloy and making it work,” he says.
The industry eventually selected a tin–silver–copper combination as offering the most reliability and ease to work with as a replacement. The formulation—95.5% tin, 3.9% silver, 0.6% copper—is also known as SAC solder, for the first letters of the chemical symbols of each of the elements (Sn, Ag, Cu). “Tin–silver–copper appears to have at least as good reliability if not higher reliability than tin–lead,” says Handwerker.
Furthermore, according to a 2005 draft report issued by the U.S. Environmental Protection Agency titled Solders in Electronics: A Life-Cycle Assessment, silver was “rarely encountered above the detection limit” in synthetic landfill leachate created to test the stability of electronics components. Silver—which is regulated as a hazardous material—is toxic to aquatic life.
With a melting point of 217°C, SAC solder also is closest in melting point to the conventional lead–tin solder. This does mean, however, a yet-unquantified increase in energy use. Furthermore, the higher temperature may pose problems for the electronics industry. Higher temperatures mean more stress on components and the entire manufacturing process, notes Geibig. Higher temperatures also mean increases in the time it takes to make products, because more time is required to heat and cool the products during the course of their manufacture.
SAC solder is used widely in the industry today. However, many of the components being made could not withstand the higher temperatures, says C. Michael Garner, director of materials technology operations at Intel: “That required re-engineering and getting new materials, not only for newer products but for older products. All the older products that had been in production for ten or fifteen years had to be converted over to high temperatures.” He says it has taken a massive effort to integrate the new solder into production processes.
There are also short-term consequences of using the new solder. Anytime there is a change in materials, there is a learning curve in using the new materials, says Karl J. Puttlitz, who managed IBM’s efforts to reduce lead in its products before he retired last year. He anticipates the occurrence of more manufacturing defects as a result of the changeover. “We can expect that at least initially the failure rates [of products] will increase,” he says. In fact, he notes the industry has asked for exemptions to the EU lead ban in certain critical electronic components where lives and security might be involved, such as equipment used in hospitals, until a track record is established with consumer goods such as cell phones and digital cameras. (The EU directive does permit exemptions to the lead ban if replacing lead is technically or scientifically impractical or if negative health, environmental, or safety consequences of replacing lead outweigh the benefits of the ban.)
A Stickier Approach
A more experimental alternative to lead–tin solder is the use of ECAs. These are polymers, such as silicone or polyamide, containing tiny flakes of metals such as silver. The polymers adhere to the printed circuit boards, and the metal flakes conduct electricity.
ECAs offer a range of advantages, notes C.P. Wong, a professor in the School of Materials Science and Engineering at the Georgia Institute of Technology who is regarded by many in the field as the leading researcher in this new technology. Silver’s electrical conductivity is very high, and its electrical resistance is very low, he points out. “If the current-carrying capability [can be boosted], ECAs can replace solder,” he says.
And there is another benefit. The temperature required to apply ECAs to circuit boards is far lower than that required for lead-based solder—150°C compared to 183°C. “You save energy, number one,” says Wong. “Number two, you subject all the components to lower temperatures and thus less thermomechanical stress. That enhances their reliability.”
Preliminary studies comparing parts using ECAs instead of solder, such as a Finnish study presented in 2000 at the 4th International Conference on Adhesive Joining and Coating Technology in Electronics Manufacturing, suggest that ECAs boast a much tighter bond than solders—perhaps an order of magnitude better, says James Morris, a professor of electrical and computer engineering at Portland State University. But he adds this research has to be replicated before it is regarded as valid.
ECAs are available for a small number of applications requiring low power—for instance, liquid crystal displays—though they are not ready for the marketplace in general, where greater amounts of current are needed. Wong is working to enhance their ability to carry current. He is adding molecules of dicarboxylic acid to the silver flakes, which provides a link between the flakes, allowing for efficient and quick conduction of electric current. “It looks like it can be as good as or even better than lead–tin solder. We demonstrated that it works [in a presentation at the March 2005 national meeting of the American Chemical Society], but we still need further research and development,” says Wong.
Wong and his collaborators are also using another means to boost the capacity to carry current—self-assembled monolayers. These are single layers of sulfur-containing molecules known as thiols that are attached to gold pads in the electronic device. At less than 10 angstroms (10 ten-billionths of a meter) in length, the molecules chemically bind to the gold pads in the device and the board, providing a direct electrical connection.
Still more work is needed on these structures, however, because they begin to fail structurally if the component heats up above 150°C. And there are other concerns about ECAs. With time, notes Wong, the ability of ECAs to conduct electricity drops, and resistance to electricity increases. Another concern is moisture. “Water is absorbed by polymers, in general,” says Morris. That can encourage corrosion, he says, and may cause other as yet unknown problems, he says.
Wong also points to the need for ECAs to become tougher so they can withstand the force of being dropped. One way to do this, says Wong, is to develop polymers that are rubberized and made more elastic, so they won’t break. Finally, Garner reiterates that these materials have not been reliable for carrying moderate to high amounts of current under normal operating conditions.
Wong and Morris are optimistic that with more research and development, ECAs can be successful alternatives to lead–tin solder. And Puttlitz does see a place for them in consumer electronics such as cell phones and digital cameras, which are not “mission critical” applications where reliability is a matter of life and death as in medical monitoring equipment or aircraft electronics.
Solder Replacement Soldiers On
Even as efforts to replace lead in solder move ahead, there still appear to be concerns about the impact that newly implemented metals will have on human and environmental health. “The alternatives to lead have not been researched as well as lead in terms of potential health and environmental impacts,” says Oladele A. Ogunseitan, a professor of environmental health, science, and policy at the University of California, Irvine. “When the Europeans said industry must get rid of lead, they didn’t say you must replace lead with something that is obviously safer,” he notes wryly. It is important, he adds, to keep looking for lead alternatives that are environmentally benign.
Indeed, the draft Solders in Electronics report indicates that no metallic alternative to lead is free from environmental concerns. For instance, whereas lead may pose a greater public health problem than SAC solder, the latter uses noticeably more energy than lead–tin solder.
But the presence of today’s substitutes is good enough for Donkers. “Since there are alternatives, we have chosen not to have lead in the products anymore,” he says. And while he does acknowledge that there are relatively few data on the impact of the current lead solder alternatives, he asserts that “in terms of active policy, you cannot always wait till you have complete certainty, because in the meantime a lot of people get exposed [to lead].”
Sticking with the problem.
Electrically conductive adhesives are one alternative to lead–tin solder being tested in the search for healthier electronics.
The alloy alternative.
Tin–silver–copper solder offers a safer solder than the lead–tin alloy, and research is continuing to address limitations on its use.
Current advances.
Researcher Grace Yi Li holds samples of electrically conductive adhesives being studied at Georgia Tech’s School of Materials Science and Engineering. Such adhesives may one day replace lead-based solders.
==== Refs
Suggested Reading
Geibig JR Socolof ML 2005. Solders in Electronics: A Life-Cycle Assessment (Draft). Environmental Protection Agency, Office of Pollution Prevention and Toxics. Available: http://www.epa.gov/dfe/pubs/solder/lca/index.htm
Li Y Moon K Wong CP 2005 Electronics without lead Science 308 5727 1419 1420 15933187
Puttlitz KJ Stalter KA eds. 2004. Handbook of Lead-Free Solder Technology for Microelectronic Assemblies. New York, NY: Marcel Dekker.
Schoenung JM Ogunseitan OA Saphores J-DM Shapiro AA 2004 Adopting lead-free electronics: policy differences and knowledge gaps J Ind Ecol 8 4 59 85
|
16203230
|
PMC1281311
|
CC0
|
2021-01-04 23:41:29
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A682-A685
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.113-a682
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a00687EnvironewsScience SelectionsLead in Cocoa Products: Where Does Contamination Come From? Taylor David A. 10 2005 113 10 A687 A688 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
Manufactured cocoa products frequently have higher lead concentrations than other foods, even though cocoa beans, the main ingredient, have some of the lowest reported lead levels for any natural food. In 2001 the Codex Alimentarius Commission, an international body based in Rome, proposed reducing the maximum permissible level of lead in cocoa products by half, to 100 nanograms per gram (ng/g) for cocoa butter and 1,000 ng/g for cocoa powder. At a March 2002 meeting in West Africa, where most of the world’s cocoa supply originates, producers agreed that to reduce lead in their products, they needed research to identify the source of contamination. Now a U.S.–Nigerian research team has uncovered some of the first clues about where the lead is coming from [EHP 113:1344–1348].
Lead contamination of candies has been recognized as a problem since 1820, when a British study found the poison widespread in London confectionary products. In recent years, documented lead content in candy has ranged from a mean concentration of 21 ng/g in milk chocolate bars in an Australian study to an average of 1,920 ng/g in chocolates seen in research in India. In Nigeria, a 1999 study found an average of 310 ng/g lead in cocoa powders. (For comparison, the mean U.S. lead concentration for apples is 20 ng/g, 200 ng/g for dry table wine, and 100 ng/g for canned pineapple.) Lead is known to cause anemia, muscle weakness, and brain damage, with children particularly susceptible to effects.
In the current study, the researchers studied the lead isotopic compositions of cocoa beans and shells from six farms in Nigeria’s top three producing states to determine if soil or farm sources might be the cause of lead contamination. The team took bean and sediment samples and homogenized them to make composites for soil, beans, and cocoa bean shells for each farm. They analyzed lead concentrations using high-resolution inductively coupled plasma mass spectrometry to make preliminary isotopic measurements, followed by thermal ionization mass spectrometry measurements.
The lead concentrations for cocoa beans ranged from less than 0.103 to 1.78 ng/g, averaging 0.512 ng/g—among the lowest lead concentrations reported for any food. The average concentrations found in the cocoa bean shells, however, was about 320-fold higher (160 ng/g). Soils showed a range of isotopic compositions overlapping those of the shells.
The cocoa bean shells all had an extremely similar isotopic composition, indicating a singular source of contamination, perhaps leaded gasoline. The authors conclude that although the soil may have caused a small degree of the contamination, the narrower range of isotopic composition in the shells suggested the more singular source of contamination was the true culprit. According to the paper, cocoa bean shells are known to be very effective at removing lead from solutions. So, although they provide excellent protection of the bean inside, the shells may also serve to contaminate the cocoa beans during fermentation or drying.
The team also compared the cocoa beans with finished cocoa products and found much higher lead concentrations and greater variability in the isotopic composition among the finished products. They therefore deduced that most of the contamination occurred after the cocoa left the farm stage.
The researchers conclude that while cocoa bean shells may be one source of lead, most contamination occurs during shipping or processing of the beans and in manufacturing. Further research on those stages of the process will help to isolate the source.
Searching for the golden ticket.
Cocoa beans are naturally low in lead, but cocoa products frequently are not. Now researchers are following new clues to identify the source of the contamination.
|
0
|
PMC1281312
|
CC0
|
2021-01-04 23:41:27
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A687-A688
|
utf-8
|
Environ Health Perspect
| 2,005 |
nan
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a00688EnvironewsScience SelectionsMoms and Mercury: Fine-Tuning Fish Consumption During Pregnancy Hood Ernie 10 2005 113 10 A688 A688 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
Due to ongoing concerns that high mercury intake via fish can cause adverse neurologic effects in the developing fetus, the U.S. Food and Drug Administration now recommends that expectant mothers should limit their consumption of fish to two or fewer meals per week. But pregnant women shouldn’t throw the baby out with the bathwater. A new study by a group of Harvard researchers suggests that this advice, which could result in many pregnant women eliminating fish from their diets altogether, may be denying some babies substantial neurocognitive benefits gained from important nutrients found in fish, such as n-3 polyunsaturated fatty acids [EHP 113:1376–1380].
The scientists sought to determine whether fish consumption during pregnancy is harmful or beneficial to fetal brain development. To do this, they examined associations of maternal fish consumption during pregnancy, maternal hair mercury levels (a sensitive marker of organic mercury body burden) at delivery, and infant cognition at age 6 months. Study subjects were 135 mother–infant pairs who participated in Project Viva, a prospective pregnancy and child health cohort study in eastern Massachusetts.
The mothers completed questionnaires about fish consumption during their second trimester. That period of time was used to best coordinate temporally with the mercury exposure reflected in maternal hair samples, which were taken at delivery. The questions concerned how much and what categories of fish (canned tuna, dark meat, light meat, shellfish) the women ate.
Mothers consumed an average of 1.2 servings of combined fish categories per week. Their mean hair mercury level was 0.55 part per million (ppm), with 10% of the samples higher than 1.2 ppm, the current U.S. reference dose. Fish consumption was directly correlated with hair mercury levels.
Infant cognition was assessed using a test called visual recognition memory (VRM). In the VRM test, which has been shown to correlate with later IQ, the child is first shown two identical photographs of an infant’s face, side by side, at a standardized distance. Then, one of the photos is replaced with a photo of another infant’s face. By tracking the percentage of time the baby looks at each photo, a novelty preference score is derived, reflecting the infant’s ability to encode a stimulus into memory, to recognize that stimulus, and to look preferentially at a novel stimulus.
Mean VRM score among the children was 59.8, with a range of 10.9–92.5. After accounting for characteristics such as maternal age and education level, higher fish intake was found to be associated with higher infant cognition, especially after adjusting for mercury levels, which had a dose-dependent negative impact on the infants’ cognition. For each additional weekly serving of fish, the infants’ VRM score was 4.0 points higher. Conversely, the researchers found that an increase of 1 ppm in hair mercury was associated with a decrement in VRM score of 7.5 points. The babies with the highest cognition scores were from mothers who had eaten more than two weekly fish servings but had mercury levels of 1.2 ppm or less.
Although the results may seem contradictory, the authors suggest that the most cognitive benefit is derived by mothers eating fish types with the combination of relatively little mercury and high amounts of beneficial nutrients. However, since the study assessed maternal fish consumption of four broad categories, there is no information presented on associations with specific types of fish. The researchers say that future studies could incorporate more detailed dietary information to help pregnant women make informed decisions about which fish meals are better or worse for their children’s cognition.
Ultimately, the message behind these findings is that pregnant women should continue to eat fish, but should try to choose varieties known to be low in mercury and high in nutrients, such as canned light tuna and sardines. Finding the most appropriate balance between risk and benefit may be challenging in this situation, but given the strong associations found in the current study, making the right decisions about which fish to eat during pregnancy, and how often, may be even more important than previously suspected.
Eating for two, thoughtfully.
Despite the threat posed by high mercury levels in certain types of fish, new findings suggest a healthy prenatal diet most likely should include some low-mercury seafood.
|
0
|
PMC1281313
|
CC0
|
2021-01-04 23:41:28
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A688
|
utf-8
|
Environ Health Perspect
| 2,005 |
nan
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a00689AnnouncementsNIEHS Extramural UpdateNew Program to Identify Outstanding New Investigators 10 2005 113 10 A689 A689 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
The National Institute of Environmental Health Sciences as launched an innovative new program of R01 research grants specifically designed for new investigators to support their research and to enhance their career progression: the Outstanding New Environmental Scientist (ONES) Award. Through this program, the NIEHS hopes to identify talented scientists who are in the early, formative stages of their career, who have not yet been awarded their first R01 grant support, and who intend to make a long-term career commitment to research related to the mission of the NIEHS.
These awards will offer several unique features to make them attractive to new investigators:
Funds earmarked for program ($3.6 million)
Review by a Special Emphasis Panel with scientific expertise targeted to the mission of the NIEHS and to the applicant pool
A unique budget structure that will allow applicants to request up to $250,000 base funds in all 5 years, plus an additional $150,000 per year in years 1–2 and $25,000 in years 3–5 for equipment or other resource development and/or career enhancement activities.
The ONES award is designed to be highly competitive. In addition to the unique features outlined above, only a limited number—six in fiscal year 2006—will be awarded. In addition, to further encourage universities to identify their best new investigators as potential applicants, only one application per school or college within an institution will be accepted.
The mission of the NIEHS is distinguished from that of other Institutes by its focus on research programs seeking to link the effects of environmental exposures to the cause, mechanisms, moderation, or prevention of a human disease, disorder, or relevant pathophysiologic process. Proposals will have to address research topics that fall within this mission.
Letters of intent are due 21 November 2005, and applications are due on 21 December 2005.
For the full text of the announcement, please refer to the NIH Guide for Grants and Contracts at http://grants.nih.gov/grants/guide/rfa-files/RFA-ES-05-005.html.
Contact
Dr. Carol Shreffler | 919-541-1445, [email protected]
Dr. Pat Mastin | 919-541-3289, [email protected]
|
0
|
PMC1281314
|
CC0
|
2021-01-04 23:41:27
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A689
|
utf-8
|
Environ Health Perspect
| 2,005 |
nan
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a00690AnnouncementsFellowships, Grants, & AwardsFellowships, Grants, & Awards 10 2005 113 10 A690 A691 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
Outstanding New Environmental Scientist (ONES) Award
An essential element of the mission of the NIEHS is the support and career promotion of the future generation of exceptionally talented and creative new scientists who will further the understanding of the impact of environmental exposures on human health. The NIEHS supports a number of training and fellowship programs for pre- and postdoctoral training, and mentored career development awards for faculty in the early stages of their career development. Primary among these are the Ruth Kirschstein National Research Service Awards for pre- and postdoctoral training, the Career Development Awards for clinically trained scientists (K08 and K23), and the Mentored Quantitative Research Career Development Awards to support the career development of scientists with quantitative and engineering backgrounds who wish to integrate their expertise with biomedicine. In addition, in 1999 the NIEHS instituted the Transition to Independent Positions Program to address the progression of individuals from postdoctoral positions to faculty positions. In this career development award the individual applies for the grant while still in a postdoctoral position, and the grant for start-up funding is awarded at the institution where the candidate accepts the faculty position. However, even with these career development mechanisms in place, to fulfill its mission of assuring a cadre of productive environmental health science investigators for the future, NIEHS needs to initiate further imaginative programs to identify the best new biomedical investigators and facilitate their establishing vibrant, independent research programs in the environmental health sciences.
To identify outstanding scientists at the formative stages of their career and assist them in launching an innovative research program with a defined impact in the environmental health sciences, the NIEHS is establishing a program of R01 research grants intended for researchers who have not received their first R01 research grant. It is designed to be highly competitive, and only a limited number will be awarded per year.
Research programs supported by this announcement seek to promote career advancement of the most highly creative and promising new scientists who intend to make a long-term career commitment to research in the mainstream of the environmental health sciences, and bring innovative, ground-breaking research initiatives and thinking to bear on the problems of how environmental exposures affect human biology, human pathophysiology, and human disease.
The R01 applications in this program are distinguished from other R01 research grants in that the applications 1) incorporate a statement of career goals in the environmental health sciences; 2) include a discussion of previous research experience and achievements in addition to the research proposal; 3) may include active participation of an external advisory committee; 4) require demonstration of the commitment by the institution to actively support the research program development of the principal investigator (PI); and 5) include a separate budget specifically devoted to equipment and career enhancement activities.
Research projects proposed in response to this Request for Applications will be expected to have a defined impact on the environmental health sciences and be responsive to the mission of the NIEHS, which is distinguished from that of other Institutes by its focus on research programs seeking to link the effects of environmental exposures to the cause, mechanisms, moderation, or prevention of a human disease or disorder or relevant pathophysiologic process. For purposes of this announcement, all applications must focus on a specific human disease, dysfunction, pathophysiologic condition, or relevant human biologic process and propose to study a specific environmentally relevant toxicant. Examples of environmentally relevant toxicants include industrial chemicals or manufacturing byproducts, metals, pesticides, herbicides, air pollutants, and other inhaled toxicants, particulates or fibers, and fungal, bacterial, or biologically derived toxins. Agents considered nonresponsive include, but are not limited to, alcohol, chemotherapeutic agents, radiation that is not a result of an ambient environmental exposure, drugs of abuse, pharmaceuticals, and infectious or parasitic agents, except when these are disease co-factors to an environmental toxicant exposure to produce the biological effect.
Applicants involving animal exposures must include a justification of how the exposure paradigm is relevant to human exposure and clearly discuss the link between the exposure and the relevant human disease in the Background and Significance section of the application. The applicant should also discuss the potential for translating the research—applying the ideas, insights, and discoveries generated through the basic inquiry to the treatment or prevention of human disease. Applicants proposing epidemiologic research should address how the significant associations revealed in the studies could be confirmed in the laboratory setting.
The ONES program would be evaluated on a continuing basis by the NIEHS, to assess the impact of the program on the portfolio of the NIEHS as well as on the progression of the awardees' careers. Metrics to be used include, but are not limited to, publications (numbers and impact factors of publications); academic promotion of PIs; awards, invited talks at national/international symposia, students and postdoctorals trained in the PI's laboratory, and honors received by PIs; committee service by PIs; and subsequent grant support awarded. The design of the program evaluation will be determined by the Program Analysis Branch of the Division of Extramural Research and Training. PIs of awarded ONES grants must provide information for the evaluation and any subsequent program evaluations for up to 10 years after the award.
This funding opportunity will use the R01 Grant mechanism. This funding opportunity uses the just-intime budget concepts. It also uses the nonmodular budget format described in the PHS 398 application instructions (see http://grants.nih.gov/grants/funding/phs398/phs398.html). A detailed categorical budget for the "Initial Budget Period" and the "Entire Proposed Period of Support" is to be submitted with the application. For further assistance contact GrantsInfo, 301-435-0714 (telecommunications for the hearing impaired: TTY 301-451-0088) or by e-mail:
[email protected]
Because the nature and scope of the proposed research will vary from application to application, the size and duration of each award will also vary. Although the financial plans of the NIEHS provide support for this program, awards pursuant to this funding opportunity are contingent on the availability of funds and the receipt of a sufficient number of meritorious applications.
Applications must be prepared using the most current PHS 398 research grant application instructions and forms. Applications must have a Dun & Bradstreet (D&B) Data Universal Numbering System number as the universal identifier when applying for federal grants or cooperative agreements. The D&B number can be obtained by calling 866-705-5711 or through the web site at http://www.dnb.com/us/. The D&B number should be entered on line 11 of the face page of the PHS 398 form.
The deadline for receipt of letters of intent is 20 November 2005, with 22 December 2005 the deadline for receipt of applications. The complete version of the RFA is available at http://grants.nih.gov/grants/guide/rfa-files/RFA-ES-05-005.html
Contact: Carol Shreffler, Division of Extramural Research and Training, National Institute of Environmental Health Sciences, P.O. Box 12233, EC-23, 111 T.W. Alexander Drive, Research Triangle Park, NC 27709 USA, 919-541-1445, fax: 919-541-5064, e-mail:
[email protected]. Reference: RFA-ES-05-005
Centers for Children’s Environmental Health and Disease Prevention Research
This initiative continues NIEHS’s and EPA’s intent to foster advances in children’s health by supporting innovative, state-of-the-art Research Centers examining the adverse health effects of environmental exposures among children. Both agencies are interested in reducing environmental risks to children and alleviating the societal burden of environmentally-induced disease/dysfunction.
Collaborative, multidisciplinary research approaches are required to explore the dynamic interaction of children and the environment. Centers are expected to have fully coordinated programs that incorporate exposure and health effects research to support the development and validations of novel health promotion strategies. A Center should identify a central theme or focus of its research effort so that the proposed projects are responsive to the specific research area of children’s environmental health included in this Center program.
The program emphasizes integration of basic and population sciences while utilizing a community-based participatory research (CBPR) approach. By bridging gaps between basic and clinical research and between institutional researchers and community members, this program aims to improve our knowledge regarding detection, treatment, and prevention of environmentally-induced disease/dysfunction in children.
The long-range goals of this program are 1) to stimulate new and expand existing research on the role of environment in the etiology of disease/dysfunction among children, 2) to develop novel effective intervention and prevention strategies, and 3) to promote translation of basic research findings into applied intervention and prevention methods, thereby enhancing awareness among children, their families, and health care practitioners regarding detection, treatment, and prevention of environmentally related diseases and health conditions.
A spectrum of scientific approaches is expected to create a truly multidisciplinary working environment where basic research can inform clinical research. These may include: 1) mechanistic research including pathophysiology of target-organ system; 2) toxicologic research; 3) molecular and cellular sciences; 4) clinical research; 5) public health research including epidemiology; 6) exposure assessment and remediation; 7) behavioral and social sciences; 8) cost/benefit; and 9) social policy research.
Each Center must produce a synergistic research environment that allows each research effort to share the creative strengths of the others. Each Center, by supporting interrelated projects and collaborating investigators, is expected to yield results beyond those achievable were each project pursued separately and without formal interaction among the participating investigators. The demonstration of synergy among the projects and the multidisciplinary nature of the work are critical components of this program. Ultimately, the expected outcomes of this research program include 1) the generation of cutting-edge science, 2) the development of local and national networks of children’s environmental health research professionals, and 3) peer-reviewed data relevant to better understanding vulnerability and risk. Expected outputs from each Center include 1) contributions to the scientific literature and 2) translational outreach and communication tools developed with and applicable for the effected community of concern.
For this RFA, NIEHS and EPA will accept applications focused on environmentally mediated disorders/dysfunctions of the nervous and/or endocrine systems. Applicants must study an environmental agent/chemical/stressor to which there is human exposure and the potential for in utero exposure. This includes any endocrine active chemical(s) or organic solvents, particulate matter (PM), pesticides, nutritional supplements, phytochemicals or metals; nutrition and social and cultural factors cannot be considered alone. Applicants are encouraged, however, to incorporate these factors in assessing the effects of previously described environmental exposures. These areas are of interest: 1) mental retardation, cerebral palsy, autism spectrum disorders, visual and or hearing impairment, attention deficit and hyperactivity disorders, and affective disorders; 2) delays or deficits in domains of neurodevelopment such as cognition, motor, sensory; 3) thyroid/pituitary dysfunction, puberty and sexual maturation, reproductive development, sexual dimorphic phenotypes, growth disorders or impairments including but not limited to childhood obesity; 4) reproductive outcomes such as preterm delivery, birth defects; 5) other outcomes associated with nervous and/or endocrine system disruption.
Each Center will propose an overall research mission and plan that is responsive to the objectives of the NIH Center Program. The application must contain a minimum of three research projects, including the first two listed below: 1) laboratory basic research project; 2) clinical research project; 3) other research project(s). At least one of the three projects must use the CBPR process as defined below. Applications lacking the first two projects and a project using the CBPR process will be considered nonresponsive and returned without review.
Each Center must also include the Community Outreach and Translation Core (COTC) and the Administrative Core. Applications lacking these two cores will be considered nonresponsive and returned without review.
Laboratory-based research projects may include mechanistic studies of environmental agents that contribute to adverse health outcomes in children as well as research that will improve our basic understanding of pathophysiology, molecular genetics, or cell biology of developmental processes. Basic mechanistic research may pertain to the disciplines of toxicology, cell and molecular biology, physiology, psychology, genetics, or other relevant fields, and methods may include animal models, in vitro systems, and/or human clinical specimens.
Clinical research is conducted with human subjects (or on material of human origin such as tissues, specimens, and cognitive phenomena) for which an investigator (or colleague) directly interacts with human subjects. Excluded are in vitro studies that utilize human tissues that cannot be linked to a living individual. For this RFA, two types of clinical research would be accepted: 1) patient-oriented research (mechanisms of human disease, therapeutic interventions, clinical trials); 2) epidemiologic and behavioral studies. Clinical research that explores gene and environmental interactions in risk of disease/dysfunction is highly encouraged
At least one other “research project(s)” must be proposed. The additional project(s) must be thematically related and integrated with the above two projects. For example: 1) Studies characterizing pathways of exposure; the magnitude, frequency, duration, and time-pattern activities that lead to contact in children and quantifying contact rates of children with exposure media, contaminant transfer efficiencies, and uptake rates in children; 2) research on behavioral factors that affect exposure and or the ability to reduce exposure; 3) research that characterizes the economic and social impact of children’s illnesses on society; 4) evaluation of how scientific information about children’s environmental health affects policy, social change, and changes in clinical and public health management of these diseases.
In CBPR, scientific inquiry is such that community members, persons affected by the health condition, disability, or issue under study, or other key stakeholders in the community’s health can be full participants in each phase of the work (conception–design–conduct–analysis–interpretation–conclusions–communication of results). CBPR is characterized by substantial community input in the development of the project.
Community refers to populations that may be defined by geography; race; ethnicity; sex/gender; sexual orientation; disability, illness, or other health condition; or to groups that have a common interest or cause, such as health or service agencies and organizations, health care or public health practitioners or providers, policy makers, or lay public groups with public health concerns. Community-based organizations may be involved in the research process as members or representatives of the community. Organizations as varied as tribal governments and colleges, state or local governments, independent living centers, other educational institutions such as junior colleges, advocacy organizations, health delivery organizations (e.g., hospitals), health professional associations, nongovernmental organizations, and federally qualified health centers are possible community partners. Each Center must include 1) Description of Cores and 2) a COTC and an Administrative Core.
A COTC is required to develop, implement, and evaluate strategies to translate and apply the scientific findings of the Center into information for the public, policy makers, and clinical professionals to use to protect the health of children. This must include personnel from one or more of the following areas: health educators, nurses, members of community or faith-based organizations, members of organizations that advocate for research and services pertaining to children’s health, members of professional societies of health care professionals, and state and local health departments or medical service organizations. Examples of activities considered responsive are creating training materials for health professions, developing new ways to disseminate research findings to the broad audience of stakeholders, and assessing community understanding of research results and plans for action. A Center will devote at least 10% of its budget to the COTC.
Each Center must include an Administrative Core unit to provide oversight, coordination, and integration of Center activities and establish an External Advisory Committee (EAC) to the Center Director. The EAC should consist of three to five scientists with expertise appropriate for the Center’s research focus, plus one representative from a community-based organization involved in community-based research. Representation from a state or local health department is also encouraged. At least 67% of Committee members should be from outside the grantee institution. The membership of the EAC must be approved by the funding agency. The EAC is to help evaluate the merit, value, and contribution of research projects and the relevance and importance of individual organizational elements to the overall goals of the Center. The membership of the EAC must be approved by the Participating Agencies postreview; names should not be submitted in the application. Individuals in senior leadership positions should provide intellectual, administrative, and scientific leadership for the Center and are critical to its overall effectiveness and evolution. These individuals should be in place and committed to a defined percent effort. Please submit only a description of proposed protocols and planned committee by representation and area of expertise. If awarded, you must provide an identifiable list of membership to the EAC for approval by the funding agencies.
Each Center may support other cores that provide a technique, service, or instrumentation that will enhance ongoing research efforts, such as animal resources, cell/tissue culture, pathology, biostatistics, molecular biology, neuropsychology, neuroimaging, analytical chemistry, exposure assessment, genotyping, and resequencing. Budgeted Center projects and external research projects may have access to these cores. The application should provide a total operational budget for each facility core together with the percentage of support requested from the Center grant. The application should explain the organization and proposed mode of operation of each core, including a plan for usage, priority setting, allocation of resources, and any applicable charge-back system.
Within the Center Program, PIs are encouraged to support training of new investigators within the proposed projects. New investigators should have a doctoral degree with < 8 years of postdoctoral experience at the time of application, and have demonstrated outstanding abilities in basic, clinical, or population-based research, such as postdoctoral research in an academic, industry, or government environment. However, years of clinical training will not count against the limitation. Ineligible individuals include current and former PIs on EPA STAR Grants or NIH research projects (R01), subprojects of program projects (P01), or Center Grants with research components (P50) or equivalent research grant awards. Applicants will be expected to devote at least 50% time and effort to the award and have a long-term commitment to research in the environmental health sciences.
This funding opportunity will use the NIH P01 award mechanism and the EPA’s Office of Research and Development, STAR Grant awards. As an applicant, you will be solely responsible for planning, directing, and executing the proposed project.
Applications must be prepared using the most current PHS 398 research grant application instructions and forms. Applications must have a D&B Data Universal Numbering System (DUNS) number as the universal identifier when applying for Federal grants or cooperative agreements. The D&B number can be obtained by calling (866) 705-5711 or through the web site at http://www.dnb.com/us/. The D&B number should be entered on line 11 of the face page of the PHS 398 form.
This funding opportunity uses the just-in-time budget concepts. It also uses the non-modular budget format described in the PHS 398 application instructions ªsee http://grants.nih.gov/grants/funding/phs398/phs398.html). A detailed categorical budget for the “Initial Budget Period” and the “Entire Proposed Period of Support” is to be submitted with the application. For further assistance contact GrantsInfo, 301-435-0714 (telecommunications for the hearing impaired: TTY 301-451-0088) for by e-mail:
[email protected].
The deadline for receipt of letters of intent is October 23, 2005, with November 24, 2005 the deadline for receipt of applications The complete version of the RFA is available at http://grants.nih.gov/grants/guide/rfa-files/RFA-ES-05-004.html.
Contact: Kimberly A. Gray, Division of Extramural Training and Science National Institute of Environmental Health Sciences, PO Box 12233, EC-21, 111 T.W. Alexander Drive, Research Triangle Park, NC, 27709 USA, 919-541-0293, fax: 919-316-4606, e-mail:
[email protected]; Chris Saint, National Center for Environmental Research, U.S. Environmental Protection Agency (8723R), 1200 Pennsylvania Ave. NW, Washington, DC 20460 USA, 202-564-6909, fax: 202-565-2448, e-mail:
[email protected]; Nigel Fields, National Center for Environmental Research, U.S. Environmental Protection Agency (8723F), 1200 Pennsylvania, Ave. NW, Washington, DC 20460 USA, 202-343-9767, fax: 202-233-0677, e-mail:
[email protected]. Reference: RFA-ES-05-004
|
0
|
PMC1281315
|
CC0
|
2021-01-04 23:41:28
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A690-A691
|
utf-8
|
Environ Health Perspect
| 2,005 |
nan
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0652a16203221PerspectivesCorrespondenceMissing Link?: Alachlor and Semen Quality Gustafson David Monsanto Company, St. Louis, Missouri, E-mail:
[email protected] author is employed by Monsanto, a manufacturer of alachlor.
10 2005 113 10 A652 A652 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
The article by Swan et al. (2003) suggested that alachlor exposure was linked to reduced sperm quality in fertile men; after publication of the article, Monsanto (St. Louis, MO) began a detailed examination of the issue because the findings were entirely unexpected and inconsistent with both our information and the extensive published literature on alachlor. Most surprisingly, alachlor mercapturate (AM) was reportedly found in the urine of 92% of study participants in Columbia, Missouri. Because this metabolite arises exclusively from exposure to the parent compound, few, if any, detects would be expected based on declining alachlor use (National Agricultural Statistics Service 2005) and our detailed understanding of its biological and environmental properties (Feng et al. 1994). Also, extensive water monitoring studies submitted to the U.S. Environmental Protection Agency (EPA) have shown that parent alachlor occurs very infrequently in both potable wells (Holden et al. 1992) and drinking water from surface water sources (Hackett et al. 2005), thereby calling into question the plausibility of such widespread exposure.
We met with Swan and colleagues at the University of Missouri in Columbia (UMC), and with the personnel at the Centers for Disease Control and Prevention (CDC) who conducted the analyses, to discuss our surprise at the findings. Our concern about the reported frequent detections of alachlor in urine was heightened when we learned that the liquid chromatography/mass spectrometry–mass spectrometry method employed by the CDC (Olsson et al. 2004) included no confirmatory ions, a standard technique for avoiding false positives in the analysis of urine (Department of Health and Human Services 1998). An 18-month collaboration ensued, which included numerous discussions and a round-robin study conducted between Monsanto and the CDC, with involvement by UMC researchers, to assess the performance and transferability of the methods used by each laboratory.
After successful completion of the round-robin study, several frozen urine samples retained from the original study were sent to Monsanto. In results that we intend to submit for publication, we found no detectable level of AM (limit of detection < 0.10 ppb) in samples that the CDC reported to contain up to 3 ppb (Swan et al. 2003). Sample degradation does not explain this difference, because we have data demonstrating AM stability under such conditions. Our analyses followed Good Laboratory Practice standards [U.S. Environmental Protection Agency (EPA) 1989] and included confirmatory ions. We also analyzed urine samples of 52 volunteers from agricultural areas across North America, none of which actually contained detectable AM, but 11 of which showed false positives when confirmatory ions were not used.
The CDC has now modified its method to include confirmatory ions for alachlor. We are confident that little or no AM would have been detected had they included confirmatory ions in the original analysis. The CDC previously declined Monsanto's request that they analyze the original samples using a modified method with confirmatory ions. However, after receiving an earlier version of this letter, the CDC quickly performed new analyses of 14 retained frozen samples and informed us that analysis with confirmatory ions validated their original findings. Unfortunately, the CDC has not provided us with sufficient data to confirm the validity of the new method and results. Monsanto continues to believe the detections are spurious. From our perspective, the only possible resolution of the matter at this time would be for the retained samples to be sent to an independent third-party laboratory for confirmatory analysis.
We understand Swan et al. are now having urine samples from other similar agricultural areas analyzed using a confirmatory method, and that AM is no longer being routinely detected. This would affirm that alachlor exposure is rare and the alleged link to semen quality is implausible. It would be very informative to identify the apparent interferent using high-resolution mass spectrometry methodology, but our collaboration has clearly shown that it was not AM. The evidence presented by our analyses of the samples provided by Swan, supported by our successful performance in the round-robin analysis of fortified samples, demonstrates that the reported detections of alachlor were most likely attributable to an interferent. The data refute any link between alachlor exposure and reduced sperm quality in fertile men.
Editor’s note: Swan et al. chose not to respond because the criticism was aimed at the CDC; see Barr’s response below.
==== Refs
References
Department of Health and Human Services 1998 Federal workplace drug testing programs; mandatory guidelines Fed Reg 63 63483 63484
Feng PCC Sharp CR Horton SR 1994 Quantitation of alachlor residues in monkey urine by ELISA J Agric Food Chem 42 316 319
Hackett AG Gustafson DI Moran SJ Hendley P van Wesenbeeck I Simmons ND 2005 The Acetochlor Registration Partnership Surface Water Monitoring Program for four corn herbicides J Environ Qual 34 877 889 15843651
Holden LR Graham JA Whitmore RW Alexander WJ Pratt RW Liddle SK 1992 Results of the National Alachlor Well Water Survey Environ Sci Technol 26 935 943
National Agricultural Statistics Service 2005. NASS Database: Agricultural Chemical Statistics Search—By Year. Available: http://www.pestmanagement.info/nass/app_statcs1_year.cfm [accessed 5 April 2005].
Olsson AO Baker SE Nguyen JV Romanoff LC Udunka SO Walker RD 2004 A liquid chromatography-tandem mass spectrometry multiresidue method for quantification of specific metabolites of organophosphorous pesticides, synthetic pyrethroids, selected herbicides, and DEET in human urine Anal Chem 76 2453 2461 15117183
Swan SH Kruse RL Liu F Barr DB Drobnis EZ Redmon JB 2003 Semen quality in relation to biomarkers of pesticide exposure Environ Health Perspect 111 1478 1484 12948887
U.S. EPA (U.S. Environmental Protection Agency) 1989 Good laboratory practices Fed Reg 54 34052 34074
|
16203221
|
PMC1281316
|
CC0
|
2021-01-04 23:41:29
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A652a
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.113-a652a
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0652b16203221PerspectivesCorrespondenceMissing Link: Barr and Needham Respond Barr Dana B. Needham Larry L. National Center for Environmental Health Centers for Disease Control and Prevention, Atlanta, Georgia, E-mail:
[email protected] authors declare they have no competing financial interests.
10 2005 113 10 A652 A653 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
In January 2004, the Monsanto Company contacted our laboratory at the Centers for Disease Control and Prevention (CDC) regarding their concern about the association between alachlor (a Monsanto product) exposure and semen quality reported by Swan et al. (2003). As a result, we provided Monsanto with detailed information about our methodology for alachlor exposure assessment by measuring its urinary metabolite alachlor mercapturate (AM). In addition, we participated in a study in which Monsanto sent 25 urine samples to the CDC for analysis. Monsanto had spiked 15 of these samples with AM (< 1.5 ng/mL), and 10 unspiked samples were collected from a field study; all were blinded to the CDC. We did not detect AM (> 0.1 ng/mL) in any of the field samples, including one with an alleged interferent; thus, the method did not produce false-positive results. For the spiked urine samples, the CDC and Monsanto measurements showed excellent correlation (r = 0.9881; p < 0.0001), although the Monsanto measurements averaged about 30% higher. Similarly, the CDC sent samples representing a broader range of concentrations (~ 1–100 ng/mL) to Monsanto for blinded analysis; again, the results were comparable.
At Monsanto’s request, residual samples from those originally tested by Swan et al. (2003) were sent to them for analysis. Because of sample volume constraints, Monsanto pooled individual samples to produce three samples with concentrations of < 0.1 ng/mL, approximately 0.2 ng/mL, and approximately 3 ng/mL. Monsanto did not detect AM in any of the pooled samples; thus, they concluded that the CDC obtained false-positive results possibly caused by a putative interferent. We suggested that Monsanto use the CDC method in its laboratory to assess whether they observed the interferent. Although Monsanto originally agreed to do this, they reportedly did not do so.
The addition of confirmation ions does increase confidence in measurements, although the method used by Swan et al. (2003) was peer-reviewed, published in Analytical Chemistry, and included many components that produce highly reliable results (Olsson et al. 2004). We have since acquired technology that allowed us to measure AM with a similar limit of detection while including confirmation ions. Using both the older method (Olsson et al. 2004) and a newer one (Norrgran et al., in press), we analyzed 14 properly archived samples that were split from samples originally analyzed and reported by Swan et al. (2003) and compared all data. In these samples, the AM levels were similar to those previously obtained (r = 0.9912; p < 0.0001) (Norrgran et al., in press) and showed good agreement using either method (r = 0.9999; p < 0.00011) (Norrgran et al., in press). We recently shared with Monsanto chromatograms of a urine sample with low levels of AM as determined by all three analyses and provided sufficient information with which to evaluate the methodology. Furthermore, we offered to discuss these new results with Monsanto, but they have not accepted this offer.
Finding AM concentrations in urine samples collected in 2000 from men in Missouri is not unlikely. Several studies have detected alachlor with high frequency in Midwestern groundwaters and surface waters (Battaglin et al. 2000; Lerch and Blanchard 2003) near the time and location our sampling occurred. Thus, although we do not frequently detect AM in general population samples, we were not surprised to find it in urine samples collected from this region. Also, contrary to Gustafson’s claim, we have not yet analyzed any field samples from other agricultural areas using our new method.
We strive to present quality human exposure assessment data. We have been assessing alachlor-related exposures since 1994; in fact, we were the first to report that AM was the primary human metabolite of alachlor (Driskell et al. 1996). Our laboratory uses both the highest caliber instrumentation and isotopically labeled internal standards, which result in high-quality, validated exposure-assessment methods capable of producing reliable and consistent results. Furthermore, our laboratory is certified to analyze human biological samples according to the Clinical Laboratory Improvement Amendment (1988), which requires extensive quality control and assurance, semiannual blinded proficiency testing, continued verification and documentation of operational parameters, and recertification every 2 years.
We do not know why Monsanto did not obtain similar results when analyzing pooled urine samples left over from the original analyses. Possible false-negative analyses could result from multiple confirmation ions that limit the sensitivity of detecting low concentrations, degradation of AM in the samples that had undergone several thaw-refreeze cycles, or inadvertent dilution of AM during the pooling process. However, the results from our analysis of properly archived specimens from 14 of the same persons from the original study provide strong evidence that our first analyses were, indeed, correct. Perhaps, when we have more details on Monsanto’s methodology and sample handling procedures, we can further explore potential reasons for the discrepancy between our results.
==== Refs
References
Clinical Laboratory Improvement Amendment 1988. Public Law 100–578.
Battaglin WA Furlong ET Burkhardt MR Peter CJ 2000 Occurrence of sulfonylurea, sulfonamide, imidazolinone, and other herbicides in rivers, reservoirs and ground water in the Midwestern United States, 1998 Sci Total Environ 248 123 133 10805233
Driskell WJ Hill RH Jr Shealy DB Hull RD Hines CJ 1996 Identification of a major human urinary metabolite of alachlor by LC-MS/MS Bull Environ Contam Toxicol 56 853 859 8661877
Lerch RN Blanchard PE 2003 Watershed vulnerability to herbicide transport in northern Missouri and southern Iowa streams Environ Sci Technol 37 5518 5527 14717159
Norrgran J Bravo R Bishop A Restrepo P Whitehead RD Needham LL In press. Quantification of six herbicide metabolites in human urine. J Chromatogr B Biomed Sci Appl.
Olsson AO Baker SE Nguyen JV Romanoff LC Udunka SO Walker RD 2004 A liquid chromatography—tandem mass spectrometry multiresidue method for quantification of specific metabolites of organophosphorus pesticides, synthetic pyrethroids, selected herbicides, and DEET in human urine Anal Chem 76 2453 2461 15117183
Swan SH Kruse RL Liu F Barr DB Drobnis EZ Redmon JB 2003 Semen quality in relation to biomarkers of pesticide exposure Environ Health Perspect 111 1478 1484 12948887
|
0
|
PMC1281317
|
CC0
|
2021-01-04 23:41:29
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A652b-A653
|
utf-8
|
Environ Health Perspect
| 2,005 |
nan
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0657aPerspectivesCorrespondenceDisease and “Broken Windows” Meléndez Leo Public Health Seattle-King County Local Hazardous Waste Management Program, Seattle, Washington, E-mail:
[email protected] author declares he has no competing financial interests.
10 2005 113 10 A657 A657 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
Frumkin’s editorial in the May 2005 issue of EHP (Frumkin 2005) was very interesting and enlightening. On page A291, Frumkin cited several studies that endorse the “broken windows theory,” noting that
Part of this effect may well be due to the disorder and squalor of the environment. Poor people and people of color are disproportionately exposed to “broken windows.”
It is interesting that the “broken windows” are considered to cause disease and health inequity. What happened first: the “broken windows,” or the lack of social skills and the abandonment of the population who live in such places? As a scientist, I find it very difficult to accept that “broken windows” are associated with the number of cases of gonorrhea and are associated with causality. The cases of venereal diseases (VD) are more related to the social skills and social behaviors of the people living in the community. They also have a lack of respect for property, and destruction of property often occurs.
If we say the reverse is plausible, what would happen if we got a grant and fixed all of the “broken windows” in a particular community, with no other intervention, and observed the trend of VD? With the assumptions and inferences made in Frumkin’s editorial, this would have a positive effect in reducing cases of VD. My instincts tell me that this would not be the case. The “broken windows” are a consequence of the behaviors of that particular community and they are not the cause of the behaviors. The “broken windows” are what I consider “collateral damage” of people lacking the necessary social skills to overcome certain challenges, such as socioeconomic stress and the lack of maintenance provided by building owners. These people show their frustration and anger many times against property, as well as other people.
==== Refs
Reference
Frumkin H 2005 Health, equity, and the built environment [Editorial] Environ Health Perspect 113 A290 A291 15866747
|
16203225
|
PMC1281318
|
CC0
|
2021-01-04 23:41:29
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A657a
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.113-1281318
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0657bPerspectivesCorrespondence“Broken Windows”: Frumkin Responds Frumkin Howard Rollins School of Public Health Emory University, Atlanta, Georgia, E-mail:
[email protected] author declares he has no competing financial interests.
10 2005 113 10 A657 A657 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
I thank Meléndez for his careful reading of my editorial and for raising the very reasonable question of whether “broken windows”—an indicator of neighborhood squalor—are causally related to poor health.
Clearly the relationship between features of the built environment—including signs of degradation and outcomes such as behavior and health—is very complex. Many of the causal arrows are probably bidirectional. True clinical trials, which might help disentangle and clarify specific causal pathways, are difficult to carry out, as Meléndez points out. However, at least two interesting studies approximate a trial and are informative.
First, in the mid-1990s, former New York City police commissioner William Bratton implemented a “fixing broken windows” approach—enforcing nuisance laws, cleaning up graffiti, and so on. This approach was credited with a substantial subsequent decrease in street crime (Bratton 1995; Bratton and Knowlner 1998; Kelling and Coles 1996). Second, the Moving to Opportunity trial in the mid-1990s enrolled over 3,000 families in high-poverty neighborhoods of Baltimore, Maryland; Boston, Massachusetts; Chicago, Illinois; Los Angeles, California; and New York, New York. They were randomly assigned to receive housing vouchers usable in low-poverty neighborhoods or to remain where they were. Although the results were variable, families moving to low-poverty neighborhoods did experience improvements in several aspects of physical and mental health (Orr et al. 2003). So while these effects are not simple, there is some evidence that less-chaotic, disordered environments may predict better health.
Perhaps the fundamental issue is that in poor communities, environmental factors and social factors are inextricably intertwined. Our efforts to understand their effects on health, and to improve people’s lives, need to focus on the root causes of both poverty and environmental hazards.
==== Refs
References
Bratton W Knobler P 1998. Turnaround: How America’s Top Cop Reversed the Crime Epidemic. New York:Random House.
Bratton WJ 1995 The New York City Police Department’s civil enforcement of quality-of-life crimes J Law Policy 3 3 447 464
Frumkin H 2005 Health, equity, and the built environment [Editorial] Environ Health Perspect 113 A290 A291 15866747
Kelling G Coles C 1996. Fixing Broken Windows. New York: Free Press.
Orr L Feins JD Jacob R Beecroft E Sanbonmatsu L Katz LF 2003. Moving to Opportunity for Fair Housing Demonstration: Interim Impacts Evaluation. Washington DC:U.S. Department of Housing and Urban Development, Office of Policy Development and Research. Available: http://www.huduser.org/Publications/pdf/MTOFullReport.pdf (accessed 8 June 2005).
|
0
|
PMC1281319
|
CC0
|
2021-01-04 23:41:29
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A657b
|
utf-8
|
Environ Health Perspect
| 2,005 |
nan
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0657c16203224PerspectivesCorrespondenceToxicity Tests: “Inert” and Active Ingredients Surgan Michael H. Environmental Protection Bureau New York State Attorney General’s Office, New York, New York, E-mail:
[email protected] author is the chief scientist in the New York State Attorney General’s Environmental Protection Bureau and was actively involved in the 1996 action against Monsanto.
10 2005 113 10 A657 A658 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
The findings of Richard et al. (2005) are an important addition to our understanding that the health and environmental effects of formulated pesticide products are not fully reflected in tests conducted on the active ingredient(s) alone. It has been long known that the adjuvants (commonly and misleadingly called “inert” ingredients) may be toxic and may enhance or supplement the toxic effects of the active pesticidal ingredient.
In the case of glyphosate-containing products, this phenomenon was well demonstrated in the data submitted to the (EPA) by the registrant (Monsanto), and summarized by the U.S. EPA in the Reregistration Eligibility Document (RED) for glyphosate (U.S. EPA 1993). For example, based on the registrant’s own tests of acute toxicity to freshwater fish, the U.S. EPA classified technical grade glyphosate as “slightly toxic” to “practically non-toxic” and formulated products ranged from “moderately toxic” to “practically non-toxic.” Tested alone, the surfactant adjuvant (identified as “inert”) was “highly toxic” to “slightly toxic.” Similar differences were reported in tests of acute toxicity to freshwater invertebrates.
Based in part on the data in the glyphosate RED (U.S. EPA 1993), the New York State Attorney General’s office successfully pursued an action against Monsanto in 1996 (Attorney General of the State of New York 1996). At that time, Monsanto was making advertising claims about the toxicity of the Roundup products based on data from tests on the active ingredient alone. Such claims are scientifically unfounded and inherently deceptive. The Attorney General’s action was facilitated by the availability of at least some limited information about the inert ingredients and their toxicity. That same sort of information enabled Richard et al. (2005) to conduct their study.
Unfortunately, that is not always the case, and for many pesticide products, little or no information about the identity of inert ingredients is publicly available. Registrants are generally required to conduct acute toxicity tests on formulated products, but they traditionally conduct chronic toxicity tests on the active ingredient alone. Even when formulated products are tested, the identity of inert ingredients is rarely revealed in the open literature, publicly available regulatory documents, or product labels. Therefore, independent research is stymied, and the public is ill-informed in the marketplace.
==== Refs
References
Attorney General of the State of New York 1996. In the Matter of Monsanto Company, Respondent. Assurance of Discontinuance Pursuant to Executive Law § 63(15). New York:Attorney General of the State of New York, Consumer Frauds and Protection Bureau, Environmental Protection Bureau.
Richard S Moslemi S Sipahutar H Benachour N Seralini G-E 2005 Differential effects of glyphosate and Roundup on human placental cells Environ Health Perspect 113 716 720 15929894
U.S. EPA 1993. Reregistration Eligibility Decision (RED). Glyphosate. EPA-738-R-93-014. Washington, DC:U.S. Environmental Protection Agency. Available: http://cfpub.epa.gov/oppref/rereg/status.cfm?show=rereg [accessed 1 September 2005].
|
16203224
|
PMC1281320
|
CC0
|
2021-01-04 23:41:28
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A657c-A658
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.113-a657c
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0661aEnvironewsForumChildren’s Health: Child Survival Gets TV Boost Burton Adrian 10 2005 113 10 A661 A661 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
A recently launched campaign known as Rx for Child Survival gets into top gear 1–3 November 2005 with the help of a six-part TV series on global health narrated by actor Brad Pitt. The series, Rx for Survival—A Global Health Challenge, which was produced by WGBH/NOVA Science Unit and Vulcan Productions and will be aired by PBS stations nationwide, aims to help Americans better understand global health problems. The series also highlights the plight of children under age 5, of whom more than 6 million die every year in the developing world from diseases that could be prevented or treated for just a few dollars.
The TV series is part of a multimedia project that also includes the efforts of Time magazine, Penguin Press, NPR, and an interactive website (http://www.pbs.org/wgbh/rxforsurvival/campaign/index.html). Says Paula Apsell, senior executive producer for WGBH/NOVA, “With the power of television to extend our message into eighty-six million living rooms each week, one of the most visited dot-org websites in the world, the local reach of three hundred forty-eight member stations across the U.S., and a far-reaching impact campaign forged on the precept of partnership, PBS is in a unique position to help Americans learn more about the world’s most pressing issues and to show them ways to do more to make the world a better place.”
With funds initially provided by the Bill & Melinda Gates Foundation and The Merck Company Foundation, the campaign’s goal is not just to inform but also to encourage Americans to donate and raise money for public health interventions in the world’s poorest countries. Some 88 cents of every dollar raised will be spent in the field to provide vaccines against measles and tetanus, insecticide-treated netting to prevent malaria, vitamin supplementation, oral rehydration packs for diarrhea (which kills 1–4 million children per year), antibiotics, and anti-malarial agents.
The actual field work will be performed by CARE and Save the Children, humanitarian organizations with delivery infrastructures already in place. The initial recipient countries will be Afghanistan, Mali, Mozambique, Nepal, Nicaragua, Sierra Leone, and Vietnam.
“Simple and affordable tools exist to save six million young lives lost each year from preventable causes like diarrhea and pneumonia. But these tools are not reaching the children who need them most,” says Charles MacCormack, CEO of Save the Children. This is because humanitarian groups cannot afford to buy and distribute them. “Through Rx for Child Survival, Americans can help these children survive and thrive,” says MacCormack.
Why another fundraising initiative? The answer is simple, says Jorge Alvar, head of the World Health Organization Department of Communicable Disease Control: “Children are dying, and we can prevent much of this tragedy at just a few dollars per head. This campaign, which aims to inform and mobilize Americans—citizens of the world’s richest nation—could be of great help in that task.”
Hope for better things.
The Rx for Child Survival campaign is a bid to better child health world-wide using simple tools such as insecticide-treated netting to combat malaria-bearing mosquitoes.
|
0
|
PMC1281321
|
CC0
|
2021-01-04 23:41:27
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A661a
|
utf-8
|
Environ Health Perspect
| 2,005 |
nan
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0661bEnvironewsForumThe Beat Dooley Erin E. 10 2005 113 10 A661 A663 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
Mercury’s Afterlife?
A report released by the New England Zero Mercury Campaign says that dental fillings in cremated corpses emit about 2.5 tons of mercury each year, with the amount expected to double by 2025. Dental amalgams are 50% mercury by weight. Although the use of mercury-containing amalgams is declining steadily, 34 tons of mercury are still used for dental purposes each year. The EPA has met with the American Dental Association to discuss ways to further reduce the use of mercury in fillings.
More people today die with some or all of their teeth in place. In addition, more people are choosing cremation. The April 2005 report is online at http://www.cleanwateraction.org/mercury/pdf/NEZMC_ReportCard_DentalMercury.pdf.
Climate Change Hits the Road
Cities are the biggest consumers of electricity and therefore the primary generators of the greenhouse gases that cause global warming. Now the British Council, Great Britain’s international agency for promoting education and cultural relations, has launched its US$7-million two-year ZeroCarbonCity campaign to educate people in 100 cities across 60 countries about climate change. The campaign, begun in March 2005, teaches how decisions at all levels, from urban planning to personal choices people can make every day, can contribute to or help mitigate the effects of climate change. A traveling photographic exhibition is visiting all 100 cities as part of the campaign. Related debate transcripts and publications are available online at http://www.britishcouncil.org/zerocarboncity.htm?mtk=8.
Personal Products Keep Organic Label
People who prefer to buy organic cosmetics, dietary supplements, and pet food can breathe a sigh of relief—in August 2005, the USDA ruled that the use of the “USDA Organic” label is permissible on those products. The ruling reverses an earlier decision that putting the green and white label on such items went beyond the original intent of the labeling program, implemented in 2002. Following the original ruling, the Dr. Bronner’s Magic Soaps organic body care company and the Organic Consumers Association filed a suit against the USDA, a move seen as the leading factor behind the reversal.
Taiwan Touts Trash Sorting
Taiwan, with the acreage of Belgium but twice the population, has 200 landfills. In two years these will be full, leaving the island nation dependent on some 20 trash incinerators that emit pollutants such as dioxin. To help curb the flow of waste into the incinerators, new recycling laws have been enacted that fine residents almost US$200 for not sorting their trash properly.
The laws, now in force only in Taipei, will be implemented across Taiwan by January 2006 in an effort to cut the number of trash incinerators to five within 20 years. Over 90% of Taipei residents are reportedly complying with the new rules.
Pandemic Prevention
Officials from 192 countries agreed upon revised regulations for notifying the WHO of all major disease outbreaks and suspected bioterrorism events at a May 2005 meeting. Until now, only outbreaks of cholera, plague, and yellow fever had to be reported to the organization. The regulations, which come into effect in 2007, also require that the WHO assist member countries in responding to such threats and in fostering greater international cooperation in outbreak response. The health ministers and other officials who signed the regulations hope the new system will help contain outbreaks of infectious diseases such as SARS and influenza before they spread globally.
Less Gummy Gum
There’s no doubt about it: chewed chewing gum is hard to clean up if it ends up anywhere except in a trashcan. In the United Kingdom alone, over US$260 million is spent each year by municipalities on gum cleanup, and the methods used include abrasive chemical cleaners, pressure washing, and scraping. Now the University of Manchester in England and the company Green Biologics are developing a biologically based cleaner, TP-GUM™, that is cost-effective and nontoxic. The new product, which uses enzymes to break down the chemical structure of chewing gum at low temperatures and pressure, is easier to use and less damaging to surfaces than conventional gum cleanup methods.
|
0
|
PMC1281322
|
CC0
|
2021-01-04 23:41:28
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A661b-A663
|
utf-8
|
Environ Health Perspect
| 2,005 |
nan
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0662aEnvironewsForumGenomics: HapMap Complete Schmidt Charles W. 10 2005 113 10 A662 A662 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
The International HapMap Project, a consortium of researchers and funding agencies from the United States, Japan, China, Nigeria, Canada, and the United Kingdom, is set to release a dramatically enhanced version of its haplotype map. The newly revised HapMap will be formally introduced on 26 October 2005 at the annual meeting of the American Society of Human Genetics in Salt Lake City. This information will provide researchers with an effective shortcut to map the genes contributing to particular diseases and drug responses.
The HapMap currently characterizes a total of 4 million common DNA sequence variants known as single-nucleotide polymorphisms (SNPs). With the HapMap, scientists are better able to investigate the genetic components of many complex disorders, such as asthma, cancer, and obesity. Mark Daly, an associate member of the Broad Institute, a research collaboration of universities, research centers, and hospitals in Cambridge, Massachusetts, says the HapMap shows where common SNPs are located on human DNA, and how they are distributed among populations in different parts of the world. “The HapMap allows us to accelerate our understanding of genetic variation and its relationship to disease,” he says.
Most SNPs are inherited in blocks, or haplotypes, on the chromosome. Each haplotype typically carries “tag” SNPs that characterize the haplotype as a whole and thus can be used to predict the identity of the other SNPs in the same block. For example, if researchers found that a certain tag SNP showed up consistently in studies of bipolar disorder, that tag could provide some indication of the other nearby SNPs on the chromosome—SNPs that may act in concert to exert some effect on the individual phenotype. Researchers can then look more closely at those neighboring SNPs to see whether and how they contribute to a given disease. The HapMap project has identified 250,000–400,000 such tag SNPs.
Phase I of the project, which was completed in March 2005, characterized 1 million SNPs in the genomes of 269 individuals from four sampled populations: the Yoruba people of Nigeria, Han Chinese from Beijing, Japanese people from Tokyo, and a group in Utah with ancestry from Western and Northern Europe.
In Phase II, the HapMap increased the SNP density characterization in these populations to 4 million. According to Daly, this expanded number encompasses the vast majority of common SNPs thought to exist in human beings.
Lisa Brooks, program director of the Genetic Variation Program at the National Human Genome Research Institute, says that researchers will seek to validate the current HapMap’s findings in additional populations, including African Americans, Mexican Americans, and others. “Phase II has given us a better genomewide HapMap,” she explains. “This is a wonderful resource for mapping genes affecting complex diseases.”
|
0
|
PMC1281323
|
CC0
|
2021-01-04 23:41:27
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A662a
|
utf-8
|
Environ Health Perspect
| 2,005 |
nan
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0662bEnvironewsForumLegislation: NYC Adopts Pesticide Laws Claudio Luz 10 2005 113 10 A662 A662 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
In response to the growing evidence that chemical pesticide use has potential human health consequences, New York City has adopted two new laws that aim to reduce exposures to toxic pesticides. The pesticide phase-out under these laws, signed in May 2005, will be complete by November 2006.
Under the NYC Pesticide Reduction Law, city agencies and their contractors must phase out the use on city property of pesticides that are known or suspected to cause cancer or developmental effects, and must adopt less toxic alternatives for pest control. Under the Neighbor Notification Law, the city must opt into a state law requiring that commercial lawn pesticide applicators provide 48 hours’ advance notice to adjacent neighbors before spraying pesticides on lawns, trees, and shrubs.
“These bills put New York City at the forefront of the national effort to move pest control in a new direction, away from poisons and towards prevention,” says Laura Haight, senior environmental associate at the New York Public Interest Research Group, one of the organizations that spearheaded community-based campaigns for the laws.
Pesticides are extensively used in densely populated cities. Cockroaches, mice, and rats thrive in multifamily dwellings, where excessive moisture, structural cracks and crevices, abundant food sources, crowded apartments, and overstuffed closets provide nutrition and shelter for pests. In the New York City metropolitan area—which in the late 1990s accounted for more than a quarter of the total pesticide use in the state—these conditions are magnified by the sheer size of the urban center, where more than 8 million people live in 800 square kilometers.
“One of the most important potential effects from both laws may be the reduction of exposures to pesticides in schoolchildren,” says Claire Barnett, executive director of the Healthy Schools Network, an advocacy organization that helped push the laws through. It is expected that these laws could potentially reduce exposure to pesticides for over 1 million children in the city’s 1,500 public schools, as well as hundreds of thousands of other residents.
What made the NYC Pesticide Reduction Law feasible is that there are effective alternatives to pesticide use, says Barbara Brenner, principal investigator of an NIEHS-funded study at the Center for Children’s Environmental Health and Disease Prevention Research at Mount Sinai School of Medicine. Data published by Brenner’s group in the October 2003 EHP showed that reducing the breeding habitats for pests and using agents like boric acid that are nontoxic to humans effectively reduced cockroach infestation in an inner-city environment.
Says Brenner, “Cockroach, mouse, and rat infestation is a very real and serious problem in both indoor and outdoor environments throughout New York City. . . . However, traditional chemical pesticide spraying has not controlled the problem, bringing with it health risks and hazards of its own. Recognition of this dilemma by New York City government represents official recognition of both the problem and the need to now use proven least-toxic methods.”
City council member James Gennaro, who cosponsored both bills, says, “The active participation of community organizations and scientists were both vital to the success of this landmark legislation. . . . Frankly, I don’t believe this legislation would be law today without the involvement of these two essential groups.” He adds, “I firmly believe that this legislation will have tangible health benefits for large numbers of New York City residents.”
|
0
|
PMC1281324
|
CC0
|
2021-01-04 23:41:29
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A662b
|
utf-8
|
Environ Health Perspect
| 2,005 |
nan
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0686aEnvironewsScience SelectionsPrelude to Intersex in Fish: Identifying a Sensitive Period for Feminization Spivey Angela 10 2005 113 10 A686 A686 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
Field studies have shown a high occurrence of intersex (the presence of both male and female characteristics) and ovotestis (the presence of eggs in the testis) in wild populations of a fish known as roach (Rutilus rutilus) in rivers in the United Kingdom that are downstream from wastewater treatment plants. Furthermore, studies have demonstrated that intersex males are less fertile, which may have population-level effects. However, to date, scientists have been unable to induce intersex in male fish with controlled exposures to wastewater effluents. A study conducted at The University of Exeter now shows that the sensitive period for feminization of the reproductive duct—in which the male testis forms an ovary-like cavity—may occur earlier than previously thought, and raises new questions about the conditions that lead to actual germ cell disruption [EHP 113:1299–1307].
Many questions persist about the causes of and the most vulnerable life stages for various types of sexual effects induced by estrogenic chemicals in wastewater effluent. In this study, the researchers collected two different U.K. wastewater effluents and exposed wild roach at two life stages: during early life and development of the gonads (from fertilization up to 300 days post-hatch) and as adults producing germ cells following annual spawning. These adults included one group of fish that had been raised in clean water and another that had hatched and grown to maturity in the wild.
Both effluents induced synthesis of vitellogenin (an estrogen-dependent yolk precursor and biomarker of estrogen exposure) at both life stages, with the extent of this induction correlating with the steroid estrogen content of the effluent. Previous studies have demonstrated that feminization of the sperm duct to form an ovary-like cavity occurs when exposure to effluent comes during the time of sexual differentiation, which in roach occurs from 50 to 150 days post-hatch. This study showed alteration of the sperm duct with an exposure earlier in life, from fertilization to 60 days post-hatch, before any signs of sexual development appear. The alteration, furthermore, was permanent, persisting even after 240 days’ maintenance in clean water after exposure.
However, no ovotestis was observed in any of the juvenile fish. There was also no evidence of ovotestis in post-spawning adult male roach raised in a clean environment and subsequently exposed to effluent. There was evidence that the wild males had previously been exposed to estrogenic stimuli, as some of males had ovotestis when the study began. The severity of this condition increased slightly during the study period, but the increase occurred across both exposed and control fish and thus appeared unrelated to the study effluent exposure.
The authors suggest possible explanations that need further study—one is that ovotestis is induced only by effluents with greater levels of estrogenic chemicals than those used in the study. The researchers evaluated the effluents for content of two chemicals previously implicated in causing intersex—steroidal estrogens and alkylphenols—and found that these levels were similar to concentrations reported in wastewater effluents in the United Kingdom and worldwide. They emphasize that chemical content and interactions ideally should be taken into account when trying to determine the conditions that lead to sexual effects.
The results of these studies raise the possibility that ovotestis may be a result either of longevity of exposure or of programming in early life that manifests itself as fish mature sexually. Previous findings from the authors support this idea by showing that the severity of intersex increases with age. The authors are further exploring these possibilities now with a laboratory study of roach that includes an environmentally relevant estrogen exposure of two years’ duration.
|
0
|
PMC1281325
|
CC0
|
2021-01-04 23:41:29
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A686a
|
utf-8
|
Environ Health Perspect
| 2,005 |
nan
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0686bEnvironewsScience SelectionsShift in Sex Ratio: Male Numbers Sink in Great Lakes Community Hood Ernie 10 2005 113 10 A686 A687 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
Sex ratio—the proportion of male to female live births—can be an important indicator of the reproductive health of a population, whether animal or human. This figure is typically fairly constant. For example, the worldwide human sex ratio ranges from 102 to 108 male births for every 100 female births; in other words, male babies make up about 50.4–51.9% of live births worldwide. Now, however, investigators have documented a significant skewing of the human sex ratio in a population located in a heavily polluted Great Lakes area [EHP 113:1295–1298].
In response to concerns about a shifting sex ratio among members of the Aamjiwnaang First Nation community near Sarnia, Ontario, a team of Canadian researchers examined birth records for the group from the years 1984–2003 as part of a broader community-based investigation. The researchers discovered that, as community members had suspected, there had been a significant and precipitous shift in the sex ratio.
The expected sex ratio in Canada is 51.2% male babies to 48.8% female babies. For the period 1984–1992, that ratio held fairly constant among this community. In the period 1993–2003, however, male babies made up only 41.2% of live births. The five-year period from 1999 to 2003 showed an even sharper decline, with male babies making up 34.8% of live births. According to the researchers, although there is normal variation in sex ratio within populations, the deviation in this case appears to be outside the normal range.
Although there is as yet no direct evidence linking this human sex ratio decline to environmental exposures, the circumstantial evidence suggests there may be a connection. The Chippewas of the Aamjiwnaang reserve reside within the St. Clair River Area of Concern, situated immediately adjacent to several large petrochemical, polymer, and chemical industrial plants. The area is one of Canada’s largest concentrations of industry. Prior soil and sediment assessment has shown that the reserve land is heavily contaminated with pollutants such as polychlorinated biphenyls, polyaromatic hydrocarbons, hexachlorobenzene, mirex, a variety of potentially toxic metals, volatile organic compounds, phthalates, and dioxins; many of these are known or suspected endocrine disruptors.
As the investigators point out, past studies have documented reproductive outcomes in wildlife populations within the same region, including reduced hatching success, altered sexual development, and changes in sex ratios. Scientific suspicion has long been focused on environmental endocrine disruptor exposures as the root cause of these effects.
The authors acknowledge that there are many other potential factors that could influence the declining sex ratio they describe. But the combination of close proximity to industrial facilities emitting known endocrine-disrupting chemicals and the documented adverse reproductive outcomes in wildlife populations in the region leads them to conclude that further investigations are warranted into the types and routes of chemical exposures—via air, water, food, soil, and sediment—for this population. A community health survey designed to explore health concerns among residents of the reserve is in progress, including information on potential covariates that may influence the sex ratio, such as parental age or smoking.
Chemical culprit?
The Sarnia–Lambton area in Ontario is home to Chemical Valley as well as the Aamjiwnaang First Nation community, which has experienced a significant skewing of the ratio of male to female babies born in recent years, leading some to question whether environmental exposures may be to blame.
|
0
|
PMC1281326
|
CC0
|
2021-01-04 23:41:28
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A686b-A687
|
utf-8
|
Environ Health Perspect
| 2,005 |
nan
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0706a16203231AnnouncementsBook ReviewPreventing Childhood Obesity: Health in the Balance Schwimmer Jeffrey B. Jeffrey B. Schwimmer is assistant professor of pediatrics in the Devision of Gastroenterology, Hepatology, and Nutrition; Department of Pediatrics; University of California, San Diego. He is the director of Weight and Wellness at the Children's Hospital and Health Center.10 2005 113 10 A706 A706 Koplan Jeffrey P. , Liverman Catharyn T. and Kraak Vivica I.
Preventing Childhood Obesity: Health in the Balance .
.
2005 . . National Academies Press . : Washington, DC . . ISBN: ISBN: 0-309-09315-5. . $44.952005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
By Jeffrey P. Koplan, Catharyn T. Liverman, and Vivica I. Kraak
Washington, DC:National Academies Press, 2005. 414 pp. ISBN: 0-309-09315-5, $44.95 cloth
On a local daily level, what do 9 million obese children look like? In the typical classroom in America today there are four or five obese students. At an urban high school, set in a low-income neighborhood, serving an ethnic minority population, one might find nearly half the teens overweight or obese. Pediatric clinics are increasingly confronted with children and adolescents suffering from obesity-related health problems. Because of the global complexity of factors influencing the development of obesity and the speed at which childhood obesity has become commonplace, our society is unprepared to adequately address the crisis.
Preventing Childhood Obesity: Health in the Balance is a book-length report based on a 24-month study by a national committee of thought leaders from across the nation in response to the epidemic of childhood obesity. Three years ago, directed by the U.S. Congress and the Centers for Disease Control and Prevention, the Institute of Medicine (IOM) charged a committee to develop a prevention-focused plan to decrease the prevalence of obesity in children and youth in the United States.
The book presents a broad range of data to build the case for launching a large-scale effort befitting a national public health priority. Although the prevalence of obesity in America’s youth and its health and psychosocial consequences are well documented, limited data are available on how to effectively reverse the crisis. Like an exploring cartographer, to prevent childhood obesity we must discover and record the routes at the same time we are traversing them; to wait until they are delineated is not an option. Such a belief led the committee to recommend actions “based on the best available evidence—as opposed to waiting for the possible evidence.” They produced 10 recommendations on national priority, industry, nutrition labeling, advertising and marketing, multimedia and public relations, community programs, built environment, health care, schools, and home.
Many of these areas are fraught with controversy in the interface of government, industry, and schools. Even the term used—obesity—carries a controversial definition; the word “obese” is loaded with negative connotations. However, in the context of health, the word is not interchangeable with “overweight.” This latter term trivializes the importance of the problem. The committee took a major step forward by justifying and adopting a definition for pediatric “obesity” rather than “overweight”: For children 2–17 years of age, “obesity” is a body mass index ≥the 95th percentile for age and sex.
Of particular interest is the growing realization of the importance of the built environment in the societal decrease in physical activity and dietary quality. To address the relationship between the built environment and childhood obesity, the National Institute of Environmental Health Sciences convened national conferences in 2004 and 2005. The IOM committee recommended that local governments, private developers, and community groups expand opportunities for physical activity, including recreational facilities, parks, playgrounds, sidewalks, bike paths, routes for walking or bicycling to school, and safe streets and neighborhoods, especially for populations at high risk of childhood obesity.
The goal set by Healthy People 2010 is to reduce the proportion of obese children and adolescents to 5% by 2010. Given that 16% of children are obese, 5 years is simply not enough time to undo a problem of this magnitude. The committee states that “it will require a long-term commitment spanning many years and possibly decades because the epidemic has taken years to develop and will require persistent efforts and the investment of sustained resources to effectively ameliorate.” The IOM committee’s charge did not include the issue of treatment. Although preventive efforts hold promise for future generations, for millions of children and adolescents prevention comes too late. Therefore, secondary and tertiary prevention (treatment) must also become part of the national plan.
In 2001 the U.S. Surgeon General issued a call to action, sounding the alarm that obesity was a crisis. In 2005, this book provides a working document for taking action. It is incumbent on all of us to read this book and take action. Through our collective actions, this roadmap will become a living document that will grow and change over time. The committee stressed the need for evaluation of those efforts—to learn from both success and failure. The ultimate success will be to make the social norm one that prioritizes a healthy diet, activity level, and body weight for all children.
|
0
|
PMC1281327
|
CC0
|
2021-01-04 23:41:28
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A706a
|
utf-8
|
Environ Health Perspect
| 2,005 |
nan
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0706b16203231AnnouncementsNew BooksNew Books 10 2005 113 10 A706 A706 2005Publication of EHP lies in the public domain and is therefore without copyright. All text from EHP may be reprinted freely. Use of materials published in EHP should be acknowledged (for example, ?Reproduced with permission from Environmental Health Perspectives?); pertinent reference information should be provided for the article from which the material was reproduced. Articles from EHP, especially the News section, may contain photographs or illustrations copyrighted by other commercial organizations or individuals that may not be used without obtaining prior approval from the holder of the copyright.
==== Body
Biodefense: Principles and Pathogens
Michael S. Bronze, Ronald A. Greenfield, eds.
Norwich, UK:Horizon Scientific Press, 2005. 838 pp. ISBN: 1-904933-12-2, $380
Chemoinformatics in Drug Discovery
Tudor I. Oprea, Raimund Mannhold, Hugo Kubinyi, Gerd Folkers, eds.
Hoboken, NJ:John Wiley & Sons, Inc., 2005. 515 pp. ISBN: 3-527-30753-2, $185
Climate Crash: Abrupt Climate Change and What It Means for Our Future
John D. Cox
Washington, DC:National Academies Press, 2005. 224 pp. ISBN: 0-309-09312-0, $27.95
Decision Making for the Environment: Social and Behavioral Science Research Priorities
Garry D. Brewer, Paul C. Stern, eds.
Washington, DC:National Academies Press, 2005. 296 pp. ISBN: 0-309-09540-9, $45
Encyclopedia of Genetics, Genomics, Proteomics and Bioinformatics
Michael J. Dunn, Lynn B. Jorde, Peter F.R. Little, Shankar Subramaniam, eds.
Hoboken, NJ:John Wiley & Sons, Inc., 2005. 3,700 pp. ISBN: 0-471-84974-6, $1,660
Environmental Economics for Tree Huggers and Other Skeptics
William K. Jaeger
Washington, DC:Island Press, 2005. 240 pp. ISBN: 1-55963-668-8, $22.50
EU Environmental Law: Challenges, Change and Decision-Making
Maria Lee
Oxford, UK:Hart Publishing, 2005. 270 pp. ISBN: 1-84113-410-4, $50
Fundamentals of Ecogenetics
Lucio G. Costa, David L. Eaton, eds.
Hoboken, NJ:John Wiley & Sons, Inc., 2005. 516 pp. ISBN: 0-471-46781-2, $110
Global Public Health Communication: Challenges, Perspectives and Strategies
Muhiuddin Haide
Sudbury, MA:Jones and Bartlett Publishers, 2005. 450 pp. ISBN: 0-7637-4776-9, $68.95
International Public Health: Diseases, Programs, Systems, and Policies
Michael Merson, Robert Black, Anne Mills
Sudbury, MA:Jones and Bartlett Publishers, 2005. 775 pp. ISBN: 0-7637-2874-8, $87.95
Nutrigenomics: Concepts and Technologies
James Kaput, Raymond L. Rodriguez, eds.
Hoboken, NJ:John Wiley & Sons, Inc., 2005. 320 pp. ISBN: 0-471-68319-1, $79.95
Proteomics for Biological Discovery
Timothy D. Veenstra, John R. Yates
Hoboken, NJ:John Wiley & Sons, Inc., 2005. 320 pp. ISBN: 0-471-16005-9, $59.95
Scarcity and Growth Revisited : Natural Resources and the Environment in the New Millennium
R. David Simpson, Michael A. Toman, Robert U. Ayres
Washington, DC:RFF Press, 2005. 320 pp. ISBN: 1-933115-10-6, $70
The Politics of the Global Oil Industry
Toyin Falola, Ann Genova
Westport, CT:Praeger, 2005. 280 pp. ISBN: 0-275-98400-1, $44.95
Toxicogenomic Technologies and Risk Assessment of Environmental Carcinogens: A Workshop Summary
Committee on How Toxicogenomics Could Inform Critical Issues in Carcinogenic Risk Assessment of Environmental Chemicals, Committee on Emerging Issues and Data on Environmental Contaminants, National Research Council
Washington, DC:National Academies Press, 2005. 55 pp. ISBN: 0-309-09700-2, $12
|
0
|
PMC1281328
|
CC0
|
2021-01-04 23:41:28
|
no
|
Environ Health Perspect. 2005 Oct; 113(10):A706b
|
utf-8
|
Environ Health Perspect
| 2,005 |
nan
|
oa_comm
|
==== Front
PLoS Comput BiolPLoS Comput. BiolpcbiplcbploscompPLoS Computational Biology1553-734X1553-7358Public Library of Science San Francisco, USA 1629235310.1371/journal.pcbi.001005605-PLCB-RA-0144R3plcb-01-06-01Research ArticleBioinformatics - Computational BiologyEvolutionGenetics/GenomicsGenetics/Genome ProjectsGenetics/Comparative GenomicsGenetics/EvolutionEubacteriaNoneUse of Artificial Genomes in Assessing Methods for Atypical Gene Detection Construction and Use of Artificial GenomesAzad Rajeev K Lawrence Jeffrey G *Department of Biological Sciences, University of Pittsburgh, Pittsburgh, Pennsylvania, United States of AmericaBorodovsky Mark EditorGeorgia Institute of Technology, United States of America* To whom correspondence should be addressed. E-mail: [email protected] 2005 11 11 2005 1 6 e5629 6 2005 29 9 2005 Copyright: © 2005 Azad and Lawrence.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.Parametric methods for identifying laterally transferred genes exploit the directional mutational biases unique to each genome. Yet the development of new, more robust methods—as well as the evaluation and proper implementation of existing methods—relies on an arbitrary assessment of performance using real genomes, where the evolutionary histories of genes are not known. We have used the framework of a generalized hidden Markov model to create artificial genomes modeled after genuine genomes. To model a genome, “core” genes—those displaying patterns of mutational biases shared among large numbers of genes—are identified by a novel gene clustering approach based on the Akaike information criterion. Gene models derived from multiple “core” gene clusters are used to generate an artificial genome that models the properties of a genuine genome. Chimeric artificial genomes—representing those having experienced lateral gene transfer—were created by combining genes from multiple artificial genomes, and the performance of the parametric methods for identifying “atypical” genes was assessed directly. We found that a hidden Markov model that included multiple gene models, each trained on sets of genes representing the range of genotypic variability within a genome, could produce artificial genomes that mimicked the properties of genuine genomes. Moreover, different methods for detecting foreign genes performed differently—i.e., they had different sets of strengths and weaknesses—when identifying atypical genes within chimeric artificial genomes.
Synopsis
Bacterial genomes contain genes that come from two sources; although most genes are inherited directly from parent cells at cell division, others may come into the genome from an unrelated organism. Often, these foreign genes can be detected because their sequences have compositional properties that differ from those of other genes in the genome. Methods for detecting atypical genes are difficult to assess because there are no genuine genomes wherein the histories of all genes are known. Here, the authors describe a method for creating artificial genomes that mimic the properties of genuine genomes, including containing “foreign” genes. The researchers used these constructs (a) to evaluate existing methods for finding foreign genes based on their atypical properties and (b) to test a new method for finding atypical genes. The researchers found that existing methods differ in their abilities to detect genes from different sources and that combining different methods can improve overall performance. The new method for finding atypical genes—which also identified sets of genes that share their unusual properties—worked very well in identifying potentially foreign genes in artificial, chimeric genomes.
Citation:Azad RK, Lawrence JG (2005) Use of artificial genomes in assessing methods for atypical gene detection. PLoS Comput Biol 1(6): e56.
==== Body
Introduction
With the number of genome sequences accumulating at a rapid pace, evidence for rampant lateral gene transfer among prokaryotes has increased dramatically [1−4]. Significant advances have been made in understanding this evolutionary phenomenon, and current research is aimed at understanding the impact of gene transfer rather than at demonstrating its occurrence [5−8]. Although inferences regarding the scope and impact of lateral gene transfer rely on the accurate and consistent identification of putative foreign genes, methods for objective, robust quantification of the lateral gene transfer have been difficult to devise. Unlike gene identification, where experimental validation of predictions is possible, it is difficult to ascertain the evolutionary history of a gene. In addition, there has been no platform available to test the efficacy and performance of methods for the identification of foreign genes. As a result, classification of genes as native or laterally transferred uses various sets of indirect evidence, and the scope and objectivity of each approach are debatable [9−13].
There are two primary strategies used to detect genes introduced by lateral gene transfer: parametric methods and phylogenetic approaches [3,14]. Phylogenetic methods detect putatively transferred genes by virtue of an unduly large degree of similarity among genes found in otherwise unrelated taxa and/or by the absence of orthologs in closely related taxa. The efficiency and reliability of this approach thus have a dependence on the depth and breadth of the sequence database and often rely on interpretation of discrepancies in relationships reflected by phylogenetic trees, themselves imperfect summaries of sets of relationships [15]. In contrast, parametric methods use the genome sequence of an organism to detect the genes that are atypical relative to the majority of genes in the genome; commonly used discriminant criteria include single nucleotide composition (SNC), dinucleotide composition (DNC), and codon usage bias (CUB).
While these two approaches are often used in concert to estimate the amount of genetic material transferred into a genome [3,14], parametric approaches are often invoked to assess whether particular genes may have been recently acquired because these analyses use only the information contained within the target genome and therefore do not require sister taxa for comparison. In addition, the results often appear to be more readily interpreted. Yet the efficacy of parametric methods lies in their ability to discriminate between typical and atypical genes, and to date no objective criteria have been offered to measure the robustness of parametric methods. This is due in part to the lack of genomes wherein the evolutionary histories of all genes are known with certainty.
As a result, critical issues remain relating to the discordant sets of atypical genes found by different methods for any species [9,10]. Both Ragan [9] and Lawrence and Ochman [14] speculated that different methods test different null hypotheses, thus leading to nonconvergent results. Moreover, each parametric method will necessarily balance the two types of classification error (failure to identify some foreign genes due to their similarity to native genes and misclassification of native genes as foreign due to some unusual character). This will lead to incongruent sets of putatively foreign genes being identified due to dissimilar thresholds for detection. Although these drawbacks could be alleviated by employing multiple identification methods and standardizing their classification error rates, the biases and error rates of most methods are not known.
Here we develop an approach to assess the abilities of parametric methods to detect atypical genes, thereby suggesting routes for establishing a unified approach for the identification of laterally transferred genes using multiple, complementary parametric approaches. To this end, we have developed a method for the creation of artificial, chimeric genomes using a generalized hidden Markov model (HMM) [16−19]. These artificial genomes reproduce the critical statistical properties of genuine genome sequences and therefore serve as valid test beds for evaluating both new and existing methods for the detection of laterally transferred genes. First, the genes composing the core of a genome—i.e., those genes likely not to have been introduced by lateral gene transfer and thus representing the spectrum of mutational signatures native to that genome—were obtained by using a novel gene clustering algorithm based on the Akaike information criterion (AIC) [20,21]; core genes were classified as “typical” by virtue of their nucleotide compositions, DNCs, and CUB patterns. Second, native genes were grouped using a k-means clustering algorithm that used relative entropy as a distance measure to decide the convergence of the algorithm [22]. Third, multiple gene models were derived according to these groups, so that artificial genomes could be generated by a generalized HMM using these gene models to represent the variability found among genuine “core” genes.
A set of artificial genomes modeled after genuine bacterial genomes was obtained. Chimeric genomes were generated as the mosaic collection of genes sampled randomly from different artificial genomes. Therefore, in these genomes, the evolutionary histories of genes as “native” or “transferred” were known with certainty. Using these artificial chimeric genomes, we tested the performance of several existing parametric methods for the detection of putative foreign genes, as well as novel methods for atypical gene identification based on the AIC. We discuss a framework for integrating multiple approaches, thereby allowing for more robust identification of foreign genes.
Results
Generating Artificial Genome Sequences
An artificial genome generator was constructed that produced protein-coding sequences and intergenic sequences using Markov models trained on genuine bacterial genome sequences. Protein-coding sequences were created by multiple, fifth-order, inhomogeneous Markov models; noncoding sequences were created by a homogeneous Markov model of noncoding sequence accounting for hexamer statistics. Separate models were derived for genes on leading and lagging strands. Structural RNAs, promoters, transcription terminators, and other features not commonly used in the identification of foreign genes were not included in genome models. The distributions of lengths of both coding and noncoding regions corresponded to those of the genome being modeled.
All gene sequences in a bacterial genome cannot be accurately described by a single model; the probabilistic nature of the HMM would necessarily result in artificial genomes that failed to represent the variability among gene sequences seen in genuine genomes. For example, the genuine Escherichia coli genome contains far more variable genes than are contained in an artificial genome created with a single model accounting only for variability between genes encoded on the two DNA strands (Figure 1A and 1B). The spectrum of genes in genuine genomes results from numerous selective regimes acting upon genes in a single genome; e.g., genes experience a range of selection for CUB [23,24]. To resolve this problem, Markov models for protein-coding sequences were trained on sets of genes that reflected distinct directional mutational biases. To create appropriate training sets, genes within genuine genomes were grouped by their similarity in nucleotide composition, DNC, or CUB; segregation into distinct classes was achieved via the k-means clustering algorithm described by Hayes and Borodovsky [22] using relative entropy as a distance measure. As expected, artificial genomes generated by the HMM begin to recapitulate variability seen within genuine genomes when multiple gene models are used; e.g., if the E. coli genome was described by three or nine models, the resulting artificial genomes contained a more representative assortment of genes (Figure 1C and 1D) than did artificial genomes generated from a single gene model (Figure 1B).
Figure 1 Variability within Genuine and Artificial E. coli Genomes Created with Variable Numbers of Gene Models
The variability of percent GC at third-codon positions of genes is shown within the genuine E. coli genome (A), as well as artificial genomes created using one (B), three (C), and nine (D) gene models. Genes were clustered according to frame-specific DNC; μ and σ represent the mean and standard deviation of the distribution. For comparison between graphs, colors demarcate corresponding ranges of GC content.
Optimizing the HMM for Generating a Genome Sequence
While increasing the numbers of models will allow the variability of genuine genomes to be more accurately represented, this tactic necessarily provides fewer genes in the training sets for each model. To optimize the HMM for number of gene models, we compared the distributions of nucleotide compositions and CUBs of genes within artificial genomes generated by the HMM to those in their genuine counterparts. As artificial genomes became more complex, the variability of such parameters among genes began to approximate that seen in their cognate genuine genomes. To measure the difference between artificial and genuine genomes, we calculated the cumulative χ2 of the differences of the three frame-specific percent GC distributions, using the distributions of these values in genuine genomes as the “expected” values. The cumulative χ2 values were plotted as a function of number of gene models; the minimum value in this curve was used to determine the minimal number of gene models required to encompass the directional mutational bias implicit in a genome.
Analysis of artificial E. coli genomes shows that the cumulative χ2 difference decreases sharply as the number of gene models increases until an optimum number of models is reached (Figure 2), after which increasing the number of gene models in the HMM did not result in any significant change. As very large numbers of gene models are used, the cumulative χ2 difference increases, as apportioning fewer numbers of genes into each model decreases the accuracy of the HMM. For the three discriminant criteria—SNC, DNC, and CUB—tested in the k-means clustering algorithm, variability of nucleotide composition within the E. coli genome can be quite closely approximated by using about 10–12 gene models (Figure 2). Closest approximation used somewhat larger numbers of gene models, but the improvement in fit was only marginal; optimal numbers for the artificial E. coli genome were 12, 14, and nine models for clusters formed using the SNC, DNC, and CUB criteria, respectively.
Figure 2 Goodness-of-Fit between Genuine and Artificial Genomes Created with Variable Numbers of Gene Models
Genes within the genuine E. coli genome were clustered by nucleotide composition, frame-specific DNC, or CUB patterns. Correspondence between genuine and artificial genomes was calculated as the χ2 of the distributions of percent GC for the three-codon positions. Small χ2 values correspond to closer approximations.
In artificial genomes constructed with the optimum number of gene models, the variability in nucleotide composition at each codon position closely approximated that seen in the genuine E. coli genome; the plot for percent GC at third-codon positions is shown in Figure 3, although clustering using CUB criteria performed less well (see Figures 2 and 3D). To examine variability in CUB, we created factor maps from the first and second axes of correspondence analyses using software developed by McInerney [25]. In the plot for genuine E. coli genes (Figure 4A), the shape of the now-famous “rabbit head,” as first described by Médigue et al. [26], is evident. Here, the majority of E. coli genes share a similar CUB, highly expressed genes form one “ear,” and laterally transferred genes—bearing more unusual CUBs—form the other “ear.”
Figure 3 Variability within Genuine and Artificial E. coli Genomes Using Different Sets of Gene Models
The distribution of percent GC of the third-codon positions of genes within the genuine E. coli genome (A), as well as artificial E. coli genome generated from E. coli genes clustered by SNC (B), DNC (C), or CUB (D). Artificial genomes were constructed using the optimal number of gene models (see Figure 2); μ and σ represent the mean and standard deviation of the distribution. For comparison between graphs, colors demarcate corresponding ranges of GC content.
Figure 4 Correspondence Analysis of CUB
The first axes—indicating variability in usage among 59 synonymous codons—are plotted for genuine E. coli genes (A) and genes from artificial genomes (see Figure 2) created from gene models sampling groups of genes clustered by SNC (B), DNC (C), or CUB (D) criteria.
This shape is also apparent in the factor maps obtained for artificial genomes created from genes clustered by either the DNC and CUB criteria (Figure 4). However, this distribution is not evident for genes clustered by similarity in nucleotide composition (Figure 4B), indicating that CUB information is lost. That is, these sets contain genes with disparate CUBs, resulting in less-informative models. The factor map for the genome based on genes clustered by the CUB criterion also appears to be more fragmented (Figure 4D), likely because each model was trained on a set of genes with highly similar CUB profiles. These observations led us to conclude that the HMM with gene models derived after clustering genes using DNC as a discriminant criterion is most effective in modeling the mutational bias patterns specific to a prokaryotic organism. That is, it captures genic complexity in both nucleotide composition and structure.
Extracting the Core Genes of a Genome
In developing artificial genomes for evaluating parametric methods for detecting atypical genes, we wished to create chimeric genomes with genes “donated” from different artificial genomes, each modeled after a different genuine genome. Yet most genuine genomes include both foreign and native genes [3,27], potentially confounding the training sets selected to represent the variability of native genes within genomes. Therefore, we must eliminate from the HMM training sets any atypical genes likely to have been recently introduced through lateral gene transfer. While the number of vertically inherited genes decreases as one compares genes that are more distantly related [27], the majority of genes in bacterial genomes have been resident there for sufficient time to acquire similar sequence characteristics [28−30]. That is, robust models representing the spectrum of native genic variation within a genome can be created if the most atypical genes are first excluded.
We identified genes likely corresponding to the native, vertically inherited “core” genome using a parametric clustering method based on the AIC (see Materials and Methods). As expected, the number of genes in the core genome depended on the criteria used to cluster them. For example, by applying the AIC gene clustering algorithm to a set of 4,255 E. coli protein-coding genes, 3,026, 2,643, and 3,031 genes were identified as native genes when using frame-specific single nucleotide bias, frame-specific dinucleotide bias, and CUB as discriminant criteria, respectively. Here we chose the set of 2,141 genes identified by all three criteria, representing the high-confidence set of core genes; this AIC-generated core was used for subsequent analyses.
Correspondence analysis of the core E. coli genome, similar to that shown in Figure 4, shows that the “ears” of the rabbit head—representing both atypical genes and highly expressed native genes—have disappeared (Figure S1). The removal of very highly expressed genes from the E. coli core genome is neither unexpected nor unwanted. Because highly conserved genes are both transferred less frequently [8] and more readily identifiable as “native” due to their readily identified functions, refining parametric methods to detect them is unnecessary. Therefore, the core genome represents a framework against which all atypical genes can be detected.
Aside from their sequence properties, the identities of genes included and excluded from the E. coli core follow predictable patterns. As expected, genes for “housekeeping” metabolism—those directing amino acid biosynthesis and central metabolism—were included in the core genome. Three classes of genes were noted to be excluded. First, mobile genetic elements (transposons and genes within prophages) were excluded, likely because of their unusual CUB. Second, other genes of known foreign origin, identified through either parametric analysis [31] or phylogenetic analysis [13], were also excluded (e.g., genes of the phn, rhs, hsd, rfb, and lac operons). Third, highly expressed genes—e.g., those encoding ribosomal proteins and elongation factors—were also excluded, as predicted from the correspondence analysis. Overall, the number of genes in the core genome is comparable to the number of protein-coding genes shared between E. coli and its sister taxon, Salmonella, that are greater than 300 nucleotides in length. These data indicate that this approach does provide a reasonable collection of genes that would reflect the major portion of the spectrum of native mutational biases. More important, it is against this variability that atypical genes must be detected; therefore, these genes represent ideal candidates for the construction of artificial genomes.
Generating Artificial Core Genomes and Chimeric Genomes
The core genes of a bacterial genome were obtained as described above and were segregated into distinct classes by the k-means gene clustering algorithm using frame-specific DNC as the discriminant criterion. Given the performance of the HMM in representing the variability within complete genomes, we expected even better performance when the most atypical genes were excluded from the training sets. The number of gene models was selected using the optimization technique described above. The gene models derived from these clusters were used in the HMM to generate artificial core genomes reflecting the characteristics of the cognate genuine core genomes; the number of genes created by each gene model was proportional to the number of genes in its training set. As was the case when entire genomes were being modeled, parametric properties such as the frame-specific nucleotide composition (Figure S2) and CUB (see Figure S1) of genes in the artificial core genome reflect those of the genuine core genome being modeled. The variability of genes within the artificial core genome—reflecting the range of that seen in genuine core genomes—again justifies the use of HMM with multiple gene models.
To create artificial genomes that have experienced simulated lateral gene transfer events, the core genomes of several prokaryotic organisms were modeled by the genome generator; for each core genome, the optimum number of gene models was used. Chimeric genomes were then generated as mosaics of genes taken randomly from several synthetic genomes in predefined proportions. In this way, artificial genomes can be created with varying proportions of foreign genes from a large number of sources. More important, the history of genes in these artificial genomes—i.e., whether genes are “native” or “foreign”—is known with absolute certainty. Because each core genome is described by multiple gene models, several hundred gene models may be used to create even the most simplistic chimeric genome, thereby providing the high degree of variability among genes observed in genuine genomes.
Evaluating Parametric Methods for Detecting Atypical Genes
Numerous chimeric genomes were generated and analyzed by the parametric methods to detect atypical genes (see Materials and Methods). We present here the results from analyses of mosaic artificial genomes containing 4,000 genes, with the majority (85%) generated from the E. coli core gene models. The “foreign” genes were modeled after core genomes derived from Archaeoglobus fulgidus (1%), Bacillus subtilis (1%), Deinococcus radiodurans (2%), Haemophilus influenzae Rd (2%), Methanococcus jannaschii (1%), Neisseria gonorrhoeae (1%), Ralstonia solanacearum (2%), Sinorhizobium meliloti (2%), Synechocystis PCC6803 (1%), and Thermotoga maritima (2%). We implemented several methods to identify atypical genes; in this case, the artificial E. coli core—contributing 85% of the genome—was considered to be the recipient genome, and the ten other artificial genomes were considered to be donors for simulated lateral gene transfer events. To evaluate the performance of each method, two error rates were considered. Type I error (false negative) was calculated as 100 – sensitivity, where sensitivity is the percentage of foreign genes correctly identified as foreign. Type II error (false positive) was calculated as 100 – specificity, where specificity is the percentage of predicted foreign genes that were actual foreign, i.e., created by a model trained on non−E. coli genes.
As expected, there was a tradeoff between type I and type II errors, i.e., as methods became more sensitive in detecting foreign genes (lower type I error), they were also less specific and misclassified more native genes as putatively foreign (higher type II error). As an example, Figure 5A shows the results for Karlin's dinucleotide method [32], where the threshold parameter determines which genes are considered sufficiently atypical to be deemed foreign. This tradeoff is seen for all methods examined (Figure 5B). As expected, more conservative thresholds result in lower type II error and higher type I error. The use of artificial genomes enables users of these algorithms to evaluate the stringency of their threshold criteria prior to application of these methods on genuine genome sequences. Alternatively, one could use the differential performance of the method to assign confidence values to atypical gene assignments, i.e., genes declared “foreign” at low threshold values would have higher confidence than those declared foreign at high threshold values, where type II error was greater. To compare the performance of different methods, we established optimal threshold criteria that minimized the average error rate (Figure 5A).
Figure 5 Tradeoffs in Error Rates in Methods for Detecting Atypical Genes
(A) Type I error, type II error, and mean error for predicting foreign genes according to Karlin's DNC method [32]; the dashed line indicates the minimum mean error.
(B) Tradeoffs in error rates for several methods of gene detection.
The performances of several methods for identifying foreign genes—each using threshold criteria that minimized their mean error rate—are compared in Table 1. Several results are notable. First, it is clear that the efficiency of detecting foreign genes depends on the source of the gene. For example, Karlin's codon usage method performed well in identifying genes from A. fulgidus, R. solanacearum, and M. jannaschii but comparatively poorly in identifying genes donated from B. subtilis, N. gonorrhoeae, or Synechocystis PCC6803 (Table 1). Second, sets of foreign genes detected well by some parametric methods were not detected as well by others. For example, Karlin's dinucleotide method did well in identifying foreign genes introduced from Synechocystis PCC6803 but not from D. radiodurans; Karlin's CUB method had the opposite tendency, performing poorly in identifying foreign genes from Synechocystis PCC6803 and doing fairly well with those from D. radiodurans. Third, it is clear that—at least in identifying genes from this test set—some methods are more robust than others; the average error rates showed substantial variation. Some methods minimized both type I and type II errors (visualized on Figure 5B as curves that approach the intersection of the axes) better than others. As a point of comparison, identifying foreign genes solely on the basis of atypical nucleotide composition can show very low type II error (indicating that few suspected foreign genes are actually native) but very high type I error (indicating that many foreign genes were not identified).
Table 1 Error Rates of Parametric Methods for Detecting Atypical Genes in an Artificial E. coli Genome
The k-Means Clustering Algorithm Fails to Identify Genes from Variable Sources
The k-means clustering algorithm has been implemented on genuine genomes to group genes into either two or three clusters, where one cluster is labeled as foreign [22]. When applied to chimeric artificial genomes, this method produced high values of both types of error for k = 2 (two clusters, Table 1). This result is not unexpected, because not all atypical genes are alike and would not be segregated into a single cluster. For k = 3, one of the three clusters contained predominantly (>95%) native genes and one cluster contained predominantly (>95%) foreign genes. The third cluster typically contained approximately 60% native genes, and assignment of this third gene cluster as either native or foreign would produce either a high type I or high type II error.
If the weakness of the k-means method lay in the high variability of foreign genes in artificial genomes, then reducing the complexity of the artificial genome should improve the performance of this method. Therefore, we constructed another set of artificial genomes with 75% E. coli−derived genes and the remaining genes from five other artificial genomes (modeled after A. fulgidus, M. jannaschii, B. subtilis, R. solanacearum, and H. influenzae, at 3%–6% abundance per genome). Using these less-complex genomes, the k-means clustering algorithm performed better, and the mean error of 13.0% compared favorably with the error rates of other methods (Table 2). In addition, while Hayes and Borodovsky [22] initiated their analyses using cluster seeds derived from the GenMark algorithm, we found that random cluster seeds were equally effective (Table 2). When the proportion of E. coli genes was increased to 85%, type II error remained the same and type I error increased slightly to 24.5% (data not shown). We conclude that when foreign genes are less diverse, the k-means method performs better. Similar improvements were not observed for other methods (Table 2), and the AIC-based approaches remained the most robust.
Table 2 Error Rates of Parametric Methods for Detecting Atypical Genes in an Artificial E. coli Genome
Using the AIC to Identify Atypical Genes
We used artificial genomes as a platform to test the implementation of a novel method for detecting foreign genes using the AIC [20]. Here, genes within chimeric, artificial genomes were clustered using either nucleotide composition, DNC, or CUB as the discriminant criterion (see Materials and Methods). Initially, genes were assigned to individual gene clusters (i.e., clusters containing a single gene). The pairwise distances between clusters were assessed using the AIC, and the closest clusters were merged if ΔAIC was negative, i.e., if the N – 1 cluster model better described the data than did the N cluster model. This process was repeated until cluster merger was no longer significant (see Materials and Methods).
The largest cluster was inferred to contain “native” genes, because native genes would be the most numerous genes in a genome; smaller clusters were inferred to contain foreign genes that failed to be merged with the primary cluster because of their atypical sequence features. This approach of assigning a single native gene cluster worked well for the analysis of artificial genomes, where unusual native genes have been excluded from the “core” genomes (see Figure S1). When applied to genuine genomes, additional clusters containing highly expressed genes would also be denoted native; this assignment should not be problematic or contentious, because the ancestry of these genes is rarely in doubt [33].
Two features of the AIC-based approach are salient. First, the number of clusters arrived upon by this method is not predetermined, as it is with the k-means algorithm [22]. Because the numbers and features of foreign genes cannot be predicted, the AIC-based clustering method avoids arbitrary assignment of genes into clusters. Second, clusters may contain a single gene if they were never merged with other gene clusters. In this way, foreign genes that are not similar to other genes are still identified as foreign. That is, the AIC clustering method does not derive a description of foreign genes and cluster them together; rather, typical genes are identified and grouped together, and foreign genes are those that do not fall into the cluster of native genes. Third, foreign genes that have similarity with each other are clustered, which serves as a form of validation. That is, groups of genes with suspected common foreign origin—e.g., the E. coli phn operon [34] or the Salmonella cob operon [35]—should fall into the same cluster.
The error rates produced by the new AIC-based gene clustering methods show that they perform very well, outperforming the other methods described (see Figure 5B; Tables 1 and 2). For example, in examining artificial genomes with laterally transferred genes from ten sources (Table 1), the mean error rates for the AIC-based methods (12%−15%) were far lower than Karlin's dinucleotide (37%–39%) or CUB (26%−28%) methods. Overall, the AIC clustering method using DNC performed the best on these data, minimizing both type I and type II errors (Figure 5B). Similar results are seen when analyzing the five-donor genome case (Table 2). In addition, the overall performance of this method did not rely heavily on the value of the “tuning” parameter (see Materials and Methods), which is analogous to threshold parameters of other methods. As seen in Table 2, all methods show a tradeoff between type I and type II errors; for the AIC-based methods, small adjustments in the tuning parameter did not dramatically alter performance. The performance of the AIC-based methods does not reflect the composition of the core genomes, which were generated via an AIC-based clustering algorithm. When core genomes extracted using Kullback-Leibler (K-L) distance were used to train Markov models used to generate artificial genomes, nearly identical results were obtained (Figure S3).
Performance in Classifying the Short Open Reading Frames
Short open reading frames are commonly misclassified as putative foreign genes when parametric methods are applied [14]. Although short genes may encapsulate useful biological information in their structure, they may appear as noise in the statistical analysis. There must be a minimum length beyond which a gene fails to provide robust data for statistical analysis, but this threshold is not obvious; in many analyses, it has been arbitrarily set to 400 nucleotides [14]. In addition, different methods may have different sensitivities vis-à-vis short genes. We examined the performance of the methods used to detect atypical genes as a function of gene length (Figure 6). For most methods, one can easily conclude that genes in excess of 250 nucleotides can be easily classified; therefore, the threshold of 400 nucleotides is valid, although somewhat more conservative than is necessary. The exception to this trend is Karlin's CUB method, which performed poorly in classifying short genes but improved as gene length increased (Figure 6). This behavior was not solely the result of CUB providing insufficient information for identification of short genes; the AIC-based clustering method that uses CUB as a discriminant criterion performed well in identifying short, foreign genes.
Figure 6 Performance of Parametric Methods in Classifying Short Genes
The error rate in proper classification of genes as native or foreign as a function of gene length was assessed for genes within chimeric, artificial genomes.
Discussion
Artificial Genomes Provide a Useful Evaluation Platform
The performances of several methods were evaluated by a test system using chimeric artificial genomes, which has allowed us to critically analyze the limitations of parametric methods for detecting laterally transferred genes. These results provide to our knowledge the first comparative assessment of the abilities of parametric methods. The tradeoff between type I and type II errors has been evaluated, and differential performance in detecting genes from different source genomes has been demonstrated. In addition, methods using the same discrimination criterion—e.g., CUB implemented by both Karlin et al. [36] and the AIC-based method described here—have shown significantly different results, suggesting that alternative analytical approaches using similar data are worth pursuing.
Genomes are enormously complex sequences, and it would be fair to consider even domains of genes to represent sequences under unique selective constraints. In addition, genes are organized into operons and are regulated in complex networks; each level of complexity imparts characteristic details that could be modeled at the sequence level. Considering phylogenetic paradigms where the interactions are at the genome level, thus obviating the need to look at more obtrusive levels of complexity, genes that have evolved under similar conditions can be described by a distinct model. While simple models for artificial genome construction based on nucleotide or hexamer statistics (e.g., GenRGenS [http://www.lri.fr/~denise/GenRGenS/]) are suitable for examination of regulatory interactions or performance of artificial life simulations [37,38], more sophisticated models are required to accurately assess the performance of algorithms in detecting atypical genes in genuine genomes.
We exploited the directional mutational bias driving genome evolution to optimize the HMM for a minimal number of gene models. The artificial genomes we constructed represent a simplification of the complexity underlying a genome. The factor maps for genes of artificial genomes show some discontinuity, representing the centers of clusters as compared to the continuous distribution observed for real genome (see Figure 4). The finite number of gene clusters used to train gene models does not reproduce the subtle complexities of bacterial genomes; rather, the gene clusters represent the major trends observed among the core genes. Some apparently unusual (atypical) genes left unfiltered by the core extraction method are thus not represented in an artificial genome. An artificial genome is intended to model certain characteristic variations among genes that are exploited in the detection of foreign genes. Other complexities of genome sequences were not modeled but could be included if they were deemed useful or important.
The performance of atypical gene identification methods could be examined with or without additional, more complex information included. For example, strand bias was included in our artificial genome generator, but artificial genomes can be generated that lack strand identity (see Figure S4); therefore, the sensitivity of methods to this aspect of genome complexity could be assayed directly. This optimized HMM lies at the core of the test system developed to assess the performance of parametric methods. The chimeric, artificial genomes provide a level playing ground for parametric methods to perform upon and be evaluated, i.e., we expect methods that perform well in detecting atypical genes in artificial genomes to perform well in classifying the genes in genuine genomes.
Comparative assessment of the parametric methods using the test system that we developed provides several insights. We observed that Karlin's dinucleotide method was outperformed by methods that used codon bias as a discriminant measure (see Figure 5B). However, we also found that a frame-specific dinucleotide measure implemented in an AIC-based clustering algorithm better discriminated the native and foreign genes than did codon bias measures implemented by any other algorithm. Therefore, the performance of a method depends both on the choice of statistic and on the methodology used. Methods like k-means clustering showed a significant variation in performance with the number of donor genomes (see Tables 1 and 2), and setting k = 2 does not seem to be a suitable choice for discriminating the pool of foreign genes from the native one. The donor genes originating from one source genome have distinct variability with respect to other genes, so the two-cluster approach may not always be a viable choice; increasing k can allow the method to create more centers for the genes to cluster around according to the genotypic variability inherent in a genome. Indeed, we have seen that an HMM with multiple gene models derived from gene clusters using the k-means method generates an artificial genome having characteristic variations of its genuine counterpart.
Other Approaches for Deducing Gene Ancestry
In theory, comparing an organism's gene inventory to that of a close relative would provide one measure as to which genes were native (those shared between the two genomes) and which genes were foreign (those unique to the genome of interest). This approach has been applied to analyses of foreign gene detection with some success [13]. This phylogenetic approach has several weaknesses, which can color attempts to tune the performance of methods for atypical gene detection or to validate the analysis of any one genome sequence. First, there are many organisms for which no close relatives have been sequenced; in these cases, there are no suitable genomes to provide a basis for comparison. Second, the presence of a gene only in the taxon of interest may result from gain in that lineage or from loss in the sister lineage; the polarity of this event can be determined only by the analysis of three or more genomes. Third, there is a large degree of variability in gene content even among very closely related taxa—e.g., strains of E. coli share less than half of their species-wide gene inventories [29,39]—which will confound the identification of lineage-specific genes.
Last, and most important, genes shared among two genomes are “native” only from the perspective that they were present in the common ancestor of those two strains. That is, one would arrive upon very different inventories of “foreign” genes if the Salmonella typhimurium genome were compared to the Salmonella typhi genome, the E. coli genome, or the Yersinia pestis genome. To validate and calibrate parametric methods for detecting laterally transferred genes, assignment of genes as being “foreign” or “native” should not rely upon the designation of a particular outgroup taxon.
Combining Approaches for Detecting Foreign Genes
Different sets of putative foreign genes are identified by different parametric methods in genuine genomes [9,10], leading to the conjecture that different methods detect different subsets of foreign genes. We believe that this hypothesis is supported by our finding that different methods for detecting foreign genes performed noticeably different in detecting genes from different sources (see Table 1). Because the identities of foreign genes are known with certainty in artificial genomes, we could test the hypothesis that a combination of methods that performed differently could, in tandem, outperform each method when used alone. Two strategies could then be implemented. One option is to relax discriminant criteria for the methods of atypical gene detection, thus identifying more foreign genes, but at the expense of misclassifying more native genes as potentially foreign (see Figure 5). The final set of putative foreign genes would be defined as those genes identified by all methods (the intersection of all gene sets). We do not favor this approach, because each method has difficulty in identifying particular foreign genes, and one would not expect them to appear in all sets.
Alternatively, one could use more stringent threshold criteria for atypical gene detection, thus misclassifying fewer native genes and minimizing type II error. The final set of putative foreign genes would comprise all atypical genes detected (the union of all gene sets). We favor this approach, because one method should identify some foreign genes that are not identified by the other. In addition, analysis of error rates (see Figure 5) allows us to choose threshold criteria that are conservative for each method. To this end, we identified putative foreign genes in chimeric artificial genomes using two of Karlin's methods, those using DNC and CUB as discriminant criteria. These two methods showed complementary strengths and weaknesses in identifying genes from different donor genomes (see Table 1).
To combine results, we selected threshold criteria that were more conservative than the optimal values, i.e., fewer native genes were misclassified as foreign at the expense of fewer foreign genes being correctly identified. However, when the results of the two methods were combined—i.e., we declared as foreign any gene that was so identified by either method—then the results of the combined methodology outperformed either method alone (Table 3). The mean error rate of the combined method (22.9%) was also less than the mean error rate of the component methods at their respective optimal thresholds (37.7% and 26.1% for Karlin's dinucleotide and codon bias methods, respectively). Therefore, we believe the artificial genome platform has justified the concept of a combined foreign gene identification approach whereby the union of sets of genes identified by different methods is denoted as “foreign.” We believe the strong improvement in detecting atypical genes reflects a “complementarity” of the methods, i.e., atypical genes detected well by one method were not detected well by the other, and vice versa. The three AIC-based methods showed less complementarity (see Table 1). When these methods were used together, the most prominent improvement in performance was observed for the combination of the AIC nucleotide and AIC codon bias methods (mean error rate of 13.8 compared to 14.9 and 15.2 at optimal thresholds for the AIC nucleotide and AIC codon bias methods, respectively; Tables 1 and S1). Understandably, addition of the AIC dinucleotide method yielded no additional improvements (Table S1), likely because this method does not augment detection of classes of genes left undetected by the combination of the other two methods. A notable feature of this analysis is that in all cases the type I error decreased by a large margin while the mean error rate remained almost same or less than those of the component methods at optimal thresholds (Tables 3 and S1). The combination of methods is thus suited for increasing the sensitivity substantially while keeping the number of false-positive results to a minimum.
Table 3 Performance of Combined Parametric Methods for Detecting Atypical Genes in Prokaryotic Genomes
A Novel Method for Detecting Foreign Genes
Novel gene clustering algorithms based on the AIC [20] have also been proposed. These methods use the AIC to cluster genes by any parametric measure (e.g., DNC). These methods compared favorably with the existing parametric methods of atypical gene detection, clearly outperforming them in our test sets (see Tables 1 and 2; Figure 5B). Whereas the k-means clustering algorithm selects an arbitrary number of clusters (k) into which genes are apportioned, the AIC-based clustering algorithm segregates the genes into distinct gene classes reflecting the inherent complexity underlying the given genome. Unlike the current parametric methods, which merely detect unusual genes, it has the ability to distinguish between distinct classes of acquired genes, i.e., it identifies sets of genes that are atypical in a particular way. This property may be useful in identifying genes that were acquired from similar sources and thus bear similar sequence signatures. In addition, this feature may serve as a validation technique, where operons of foreign genes would comprise genes that fall into the same AIC-defined clusters.
The performance of the AIC-based methods was not influenced by use of the AIC in the method for identifying “core” genes for use in training Markov models to generate artificial genomes. To ensure the independence of these methods, we extracted the core genome using a method based on K-L distance (see Materials and Methods). The core of the E. coli genome selected by the K-L method contained 2,445 genes, where 1,788 of the genes were shared with the AIC-generated core. Because the core genomes produced by the two methods contain many of the same genes, the method used to select the core does not appear to bias the composition of the core. Rather, the differences reflect the relative stringency of the selection method. When methods for detecting atypical genes were assessed using chimeric genomes created using models of these core genomes, no significant differences were detected (see Figure S3; compare to Figure 5B). These results support the hypothesis that little, if any, bias remains in the composition of the core genome, and any bias would have been eliminated upon the creation of a chimeric artificial genome from genes created by several hundred Markov models. Therefore, we conclude that this approach provides a robust platform for evaluating the performance of parametric methods for the detection of atypical genes in bacterial genomes.
Conclusions
Identifying the atypical character of a gene is a first step in identifying and quantifying lateral gene transfer events. Even though parametric methods have proved to be very effective in classifying foreign genes, reducing the margin of error is still a challenge. Our probabilistic approach is a step forward in assessing the atypical nature of genes through parametric methods using different null hypotheses and provides a platform for developing an integrated system of approaches that can assign a confidence value to a gene to be called typical or atypical, thus opening a new direction in quantifying lateral gene flow. Use of the HMM allows for artificial chimeric genomes to be generated given any set of prokaryotic genomes. This provides an objective test bed for evaluating the performance of newly proposed methods for atypical gene detection.
Materials and Methods
Genomes.
The complete genome sequences of several prokaryotic organisms—A. fulgidus DSM4304, B. subtilis 168, D. radiodurans R1 chromosome I, E. coli K12, H. influenzae Rd KW20, M. jannaschii DSM2661, N. gonorrhoeae FA1090, R. solanacearum GMI1000, S. meliloti 1021, Synechocystis sp. PCC6803, and T. maritime MSB8—were retrieved from GenBank. Open reading frames were extracted using coordinates provided in the annotation; for assigning genes to leading and lagging strands, the origins and termini of replication were localized using cumulative nucleotide skew [40,41].
Generalized HMM as a descriptor of a genome sequence.
Markov models have been successfully applied in deciphering the complex structural and functional units of a genome [42]. Borodovsky et al. [43] provided a rigorous mathematical framework in the form of a codon position−specific inhomogeneous Markov model for describing protein-coding sequences and a homogeneous Markov model for describing noncoding sequences. Markov models have been used extensively in gene-finding algorithms applied to both prokaryotic and eukaryotic genomes [19,44−47]. At the heart of such algorithms is an HMM incorporating models of distinct sequence types. The problem is formulated as deciphering the sequence of “hidden” states (e.g., protein-coding or noncoding) underlying a DNA sequence.
Given the model parameters, a generalized HMM can be used to either predict desired features in a test sequence (e.g., find sequences that resemble genes) or generate a DNA sequence (e.g., create sequences that resemble genes). A simple, generalized HMM (Figure 7) can generate a genome sequence by choosing an oligonucleotide
(O ∈
, = {A,T,C,G}, where i and j indicate the start and end positions of the oligonucleotide in DNA sequence, respectively) according to output probability distribution Q in state Si. The length of the emitted sequence is determined by the length distribution of the sequences of a distinct state type (e.g., the length of genes or the length of noncoding sequences). Transitions from state Si to Sk (e.g., from a protein-coding state to a noncoding state) are made according to the probability distribution of transitions between states; this process is repeated until a genome sequence of a desired length L is generated.
Figure 7 The HMM Architecture
The oval represents a hidden state, and the square represents observation sequence. Each state emits a string of nucleotides and then makes a transition to another state. The transitions that are allowed between hidden states are shown by line arrows, and the emission of observation sequence is shown by block arrows. This HMM generates one strand of a genome; by including models of reverse complement of protein-coding sequence (“Shadow”), the HMM encapsulates the information of sequences on both strands of DNA helix. The “Reverse Start” and “Reverse Stop” states correspond to the reverse complement of start codons and stop codons, respectively.
Training sets were derived using the annotations provided in the GenBank sequences, and model parameters were obtained as the maximum likelihood estimate. For a gene model, the initial probability P i(O
1
m) of observing an oligonucleotide O
1
m is estimated as
where N i(O
1
m) is the number of occurrences of oligonucleotide O
1
m in phase i in the training data (phase i corresponds to the ith-codon position of the first base of the oligonucleotide). N – m + 1 is the count of all possible oligonucleotides of size m in the training data. The maximum likelihood estimate of the transition probability estimate
is given by
Here,
and
are the counts of the oligonucleotides
and
, respectively, in phase i. For noncoding sequences, the phase consideration is omitted. The values of all other probabilistic parameters (including the state initial and transition probabilities) were obtained as maximum likelihood estimates from the training set. The distribution of the number of genes in the same orientation, as well as the length distribution of protein-coding and noncoding sequences, was estimated from the GenBank annotation.
An HMM with a single gene model will fail to represent the variability of genes within a genome. There are distinct classes of genes that have evolved under different selective constraints, including genes encoded on leading and lagging strands, and variation in selection on codon usage, which requires the HMM to have multiple gene models. The HMM used for the creation of artificial genomes includes separate, multiple models of protein-coding sequence (its reverse complement or the protein-coding shadow) on both the leading and lagging strands, a model of noncoding sequence, and a model of gene orientation (see Figure S4).
Gene clustering by a k-means algorithm using K-L divergence.
To build multiple gene models, genes were segregated into distinct gene classes signifying divergent mutational biases. We adapted the k-means gene clustering method suggested by Hayes and Borodovsky [22] to perform this task. The (dis)similarity of two genes can be quantified in terms of their nucleotide composition or codon usage pattern. To quantify the difference in nucleotide composition between two genes or clusters of genes, F and Q, the difference is defined in the symmetric form of K-L divergence [22] as
where fi and qi denote the relative frequencies quantitating the nucleotide patterns of the DNA sequences of genes F and Q, respectively. To quantify the difference in codon usage pattern, D is defined as
where fc and qc are the codon frequencies, c, normalized in the ath group of synonymous codons to which it belongs, for F and Q, respectively; and na is the size of the ath group of synonymous codons. Note that for a cluster of genes, the center of the cluster is represented by the cumulative frequencies normalized in the respective groups.
The k-means gene clustering algorithm was initialized by selecting open reading frame cluster seeds by distributing genes at random among k clusters and calculating the cluster centers. Genes were reassigned to the cluster with the closest cluster center (in terms of D[F||Q] distance), and cluster centers were recalculated, until all genes resided within clusters with the closest centers. This process was repeated for several random realizations of the cluster seeds to eliminate any bias due to initial cluster assignment; the gene cluster configuration that minimizes the distance function Ψ,
where C denotes cluster of genes, was selected.
Methods to extract the core genes of a genome.
We implemented two alternative approaches to extract the genes forming the “core” genome, those genes experiencing the range of naturally imparted mutational biases. First, genes were sorted into approximately 25 clusters by the k-means clustering algorithm using K-L divergence as the distance measure between clusters (see above). The two clusters whose centers were closest to each other were merged, and this process was repeated until the relative change in the K-L distance, R, between closest clusters exceeds the established threshold of
where D
min(i) is the K-L distance between the two closest clusters at the ith iteration. The largest cluster was retrieved for each of the three discriminant criteria—nucleotide composition, DNC, and CUB—and the genes common to these three sets were taken to represent the core.
A drawback of using the K-L distance measure for finding gene clusters is that one begins with a arbitrary number of gene clusters that are then merged; therefore, the members of the final genome core will be biased both by the composition of the initial clusters and by the degree of variation within a genome. We sought to eliminate this bias by introducing a more rigorous method for cluster formation that did not begin by arbitrarily assigning genes to a fixed number of clusters. Rather, we sought to merge genes into clusters if they were similar to other members of that cluster. Criteria that can be used to select between models include the AIC [20], the minimum description length [48], and the Bayesian information criterion [49]. These approaches are based on the principle of finding the most parsimonious model, thus avoiding underfitting or overfitting models. After testing each criterion of model selection, we converged on the AIC for identification of core genes, which performed well and showed no bias with respect to cluster size or composition. The AIC is defined as
where
is the maximum likelihood and K is the number of free parameters in the model; the best-fitting model minimizes the AIC.
We used the AIC to determine if a one-gene (cluster) model significantly improves upon a two-gene (cluster) model. Among all possible pairings of gene clusters, the one minimizing reduction in likelihood of the gene cluster set was selected and this process was repeated to segregate genes into distinct clusters. The AIC provided the stopping criterion for the clustering procedure. In practice, for N genes, N single gene clusters were examined, and the pair of clusters with the least likelihood decrease were merged, resulting in N – 1 clusters. This process was repeated until the AIC for the merged cluster model was no longer less than the AIC for the separate cluster model. The largest cluster was retrieved for each of the three discriminant criteria—SNC, DNC, and CUB—and the genes common to these three sets were taken to represent the core genome.
To account for base compositional bias, the likelihood function
can be expressed in terms of frequencies of nucleotides (or oligonucleotides) located at specific codon positions. Considering single nucleotide statistics, a 12-dimensional frequency vector with frequencies of elements b ∈ {Ai , Ti , Ci , Gi}, i = 1, 2, 3, which takes into account both base identity and codon position, was used to calculate the maximum likelihood ;
the likelihood for the separate cluster model was obtained as
The likelihood for the merged cluster model was obtained as
where {p(b)} and {p
1(b), p
2(b)} are the probabilities of base b in merged cluster and the two component clusters, respectively. N(b) denotes the count of base b in respective clusters. To allow the merger of the two clusters, the AIC of the merged cluster model should be less than the AIC of the two-cluster model. We assessed this difference as
where K
1 and K
2 are the number of free parameters and
1 and
2 are the corresponding likelihoods in the two random cluster model and one random cluster model, respectively. The likelihood function, and thus the stopping criterion, can be obtained similarly for oligonucleotide statistics and also for CUB consideration.
Methods for detection of atypical genes.
Several widely used, parametric methods used in lateral gene transfer detection were implemented as follows. Karlin's dinucleotide bias [32] was assessed through use of the odds ratio:
where f
XY is the frequency of the dinucleotide XY and f
X is the frequency of the nucleotide X. The dinucleotide average relative abundance difference between two DNA sequences f and g is defined as
If the value of δ for a gene compared to average over all genes in a genome is greater than an established threshold, the gene is classified as foreign. Karlin's codon usage difference [36] of the gene family F relative to gene family C was quantified as
where {f(x,y,z)} is the set of codon frequencies for the gene family F, {c(x,y,z)} is the set of codon frequencies for the gene family C, and {pa(F)} is the set of amino acid frequencies of the genes of F.
The codon frequencies were normalized to one in each amino acid codon family, so that
If C is the set of all genes and F is a single gene, B(F|C) = B(F|all) measures the codon bias of F compared to the average for all genes. If B(F|all) is greater than an established threshold, F is classified as a foreign gene.
CUB was used as a discriminant criterion in the k-means gene clustering algorithm of Hayes and Borodovsky [22], where relative entropy was used as a distance measure of codon usage difference between clusters of genes (see equation 4) in a k-means algorithm. We also implemented base compositional bias and CUB as discriminant criteria in an AIC-based gene clustering algorithm. We have already discussed the utility of the AIC in identifying the genes likely to be forming the native core of a genome. We also tested the performance of an AIC-based gene clustering algorithm in identifying atypical genes. Note that we used a generalized version of the AIC, defined as
where n is the sample size and n0 is a positive constant [50]. For n
0
= n, the generalized version takes the form of a standard AIC (see equation 7). The tuning parameter n
0 was used to optimize the algorithm.
Supporting Information
Figure S1 Correspondence Analysis of CUB
The first axes—indicating variability in usage among 59 synonymous codons—are plotted for 4,255 E. coli genes (A), 2,141 E. coli genes representing the “core” genome (B), and 2,141 genes comprising the artificial E. coli core genome (C). The artificial genome was created from genes clustered by frame-specific DNC.
(826 KB TIF)
Click here for additional data file.
Figure S2 Variability within Genuine and Artificial E. coli Genomes
The percent GC of third-codon positions is plotted for 4,255 E. coli genes (A), 2,141 “core” E. coli genes (B), and 2,141 genes within the artificial “core” E. coli genome (C). The artificial core genome was created from genes clustered by frame-specific DNC; μ and σ represent the mean and standard deviation of the distribution.
(388 KB TIF)
Click here for additional data file.
Figure S3 Tradeoffs in Error Rates for Several Methods of Gene Detection
Artificial genomes were generated from Markov models trained on core genomes extracted using a K-L distance method. Compare to Figure 5B.
(142 KB TIF)
Click here for additional data file.
Figure S4 Cumulative GC-Skew Plots for Genuine and Artificial E. coli Genomes
For each gene, GC skew was calculated as (%G – %C)/(%G + %C) of third-codon positions, corrected for direction of transcription. Beginning with the first gene in a genome sequence, cumulative skew was obtained as the sum of skew values for the preceding genes. The genuine E. coli genome comprised 4,255 protein-coding genes, whereas the artificial genomes comprised 4,000 genes. Cumulative GC-skew plots are shown for artificial E. coli genomes with and without gene models accounting for strand bias during training. Strand bias is evident as large domains of either G-rich or C-rich genes in the genuine E. coli genome and in the artificial genome with strand identity incorporated into the model.
(273 KB TIF)
Click here for additional data file.
Table S1 Performance of Combined Parametric Methods for Detecting Atypical Genes in Prokaryotic Genomes
(42 KB DOC)
Click here for additional data file.
This work was supported by grant MCB-0217278 from the National Science Foundation. We thank Ram Ramaswamy, Antoine Danchin, James McInerney, and Eduardo Rocha for helpful discussions.
Competing interests. The authors have declared that no competing interests exist.
Author contributions. RKA and JGL conceived and designed the experiment. RKA performed the experiments. RKA and JGL analyzed the data and wrote the paper.
Abbreviations
AICAkaike information criterion
CUBcodon usage bias
DNCdinucleotide composition
HMMhidden Markov model
K-LKullback-Leibler
SNCsingle nucleotide composition
==== Refs
References
Doolittle WF 1999 Lateral genomics Trends Cell Biol 9 M5 M8 10611671
Doolittle WF Boucher Y Nesbo CL Douady CJ Andersson JO 2003 How big is the iceberg of which organellar genes in nuclear genomes are but the tip? Philos Trans R Soc Lond B Biol Sci 358 39 57 12594917
Ochman H Lawrence JG Groisman E 2000 Lateral gene transfer and the nature of bacterial innovation Nature 405 299 304 10830951
Koonin EV Makarova KS Aravind L 2001 Horizontal gene transfer in prokaryotes: Quantification and classification Annu Rev Microbiol 55 709 742 11544372
Lawrence JG Hendrickson H 2003 Lateral gene transfer: When will adolescence end? Mol Microbiol 50 739 749 14617137
Koonin EV 2003 Horizontal gene transfer: The path to maturity Mol Microbiol 50 725 727 14617135
Gogarten JP Doolittle WF Lawrence JG 2002 Prokaryotic evolution in light of gene transfer Mol Biol Evol 19 2226 2238 12446813
Nakamura Y Itoh T Matsuda H Gojobori T 2004 Biased biological functions of horizontally transferred genes in prokaryotic genomes Nat Genet 36 760 766 15208628
Ragan MA 2001 On surrogate methods for detecting lateral gene transfer FEMS Microbiol Lett 201 187 191 11470360
Ragan MA 2001 Detection of lateral gene transfer among microbial genomes Curr Opin Genet Dev 11 620 626 11682304
Wang B 2001 Limitations of compositional approach to identifying horizontally transferred genes J Mol Evol 53 244 250 11523011
Guindon S Perrière G 2001 Intragenomic base content variation is a potential source of biases when searching for horizontally transferred genes Mol Biol Evol 18 1838 1840 11504864
Koski LB Morton RA Golding GB 2001 Codon bias and base composition are poor indicators of horizontally transferred genes Mol Biol Evol 18 404 412 11230541
Lawrence JG Ochman H 2002 Reconciling the many faces of gene transfer Trends Microbiol 10 1 4 11755071
Lawrence JG Hartl DL 1992 Inference of horizontal genetic transfer: An approach using the bootstrap Genetics 131 753 760 1628816
Rabiner L 1989 A tutorial on hidden Markov models and selected applications in speech recognition Proc IEEE 77 257 286
Borodovsky M McIninch J 1993 GenMark: Parallel gene recognition for both DNA strands Comp Chem 17 123 133
Azad RK Borodovsky M 2004 Probabilistic methods of identifying genes in prokaryotic genomes: Connections to the HMM theory Brief Bioinform 5 118 130 15260893
Lukashin AV Borodovsky M 1998 GeneMark.hmm: New solutions for gene finding Nucleic Acids Res 26 1107 1115 9461475
Akaike H 1974 A new look at the statistical model identification IEEE Trans Automat Contr 19 716 723
Sakamoto Y Ishiguro M Kitagawa G 1999 Akaike information criterion statistics Berlin Springer 320 p.
Hayes WS Borodovsky M 1998 How to interpret an anonymous bacterial genome: Machine learning approach to gene identification Genome Res 8 1154 1171 9847079
Sharp PM Li WH 1987 The codon adaptation index: A measure of directional synonymous codon usage bias, and its potential applications Nucleic Acids Res 15 1281 1295 3547335
Sharp PM 1991 Determinants of DNA sequence divergence between Escherichia coli and Salmonella typhimurium: Codon usage, map position, and concerted evolution J Mol Evol 33 23 33 1909371
McInerney JO 1998 GCUA: General codon usage analysis Bioinformatics 14 372 373 9632833
Médigue C Rouxel T Vigier P Hénaut A Danchin A 1991 Evidence of horizontal gene transfer in Escherichia coli speciation J Mol Biol 222 851 856 1762151
Charlebois RL Doolittle WF 2004 Computing prokaryotic gene ubiquity: Rescuing the core from extinction Genome Res 14 2469 2477 15574825
Lawrence JG Ochman H 1997 Amelioration of bacterial genomes: Rates of change and exchange J Mol Evol 44 383 397 9089078
Konstantinidis KT Tiedje JM 2005 Genomic insights that advance the species definition for prokaryotes Proc Natl Acad Sci U S A 102 2567 2572 15701695
Lerat E Daubin V Ochman H Moran NA 2005 Evolutionary origins of genomic repertoires in bacteria PLoS Biol 3 e130. DOI: 10.1371/journal.pbio.0030130 15799709
Lawrence JG Ochman H 1998 Molecular archaeology of the Escherichia coli genome Proc Natl Acad Sci U S A 95 9413 9417 9689094
Karlin S 1998 Global dinucleotide signatures and analysis of genomic heterogeneity Curr Opin Microbiol 1 598 610 10066522
Jain R Rivera MC Lake JA 1999 Horizontal gene transfer among genomes: The complexity hypothesis Proc Natl Acad Sci U S A 96 3801 3806 10097118
Metcalf WW Wanner BL 1993 Evidence for a fourteen-gene, phnC to phnP locus for phosphonate metabolism in Escherichia coli
Gene 129 27 32 8335257
Roth JR Lawrence JG Rubenfield M Kieffer-Higgins S Church GM 1993 Characterization of the cobalamin (vitamin B12 ) biosynthetic genes of Salmonella typhimurium
J Bacteriol 175 3303 3316 8501034
Karlin S Mrazek J Campbell AM 1998 Codon usages in different gene classes of the Escherichia coli genome Mol Microbiol 29 1341 1355 9781873
Reil T 1999 Dynamics of gene expression in an artificial genome: Implications for biological and artificial ontogeny Floreano D Nicoud J-D Mondada F Advances in artificial life Berlin Springer-Verlag 457 466
Watson J Geard N Wiles J 2004 Towards more biological mutation operators in gene regulation studies BioSystems 76 239 248 15351147
Welch RA Burland V Plunkett G 3rd Redford P Roesch P 2002 Extensive mosaic structure revealed by the complete genome sequence of uropathogenic Escherichia coli
Proc Natl Acad Sci U S A 99 17020 17024 12471157
Lobry JR 1996 Asymmetric substitution patterns in the two DNA strands of bacteria Mol Biol Evol 13 660 665 8676740
Salzberg SL Salzberg AJ Kerlavage AR Tomb JF 1998 Skewed oligomers and origins of replication Gene 217 57 67 9795135
Durbin R Eddy S Krogh A Mitchison G 1998 Biological sequence analysis: Probabilistic models of proteins and nucleic acids Cambridge Cambridge University Press 356 p.
Borodovsky M Sprizhitsky A Golovanov EI Alexandrov AA 1986 Statistical patterns in the primary structures of functional regions of the genome in Escherichia coli: II. Nonuniform Markov models Mol Biol 20 833 840
Krogh A Mian IS Haussler A 1994 A hidden Markov model that finds genes in E. coli DNA Nucleic Acids Res 22 4768 4778 7984429
Krogh A 1997 Two methods for improving performance of an HMM and their application for gene finding Proc Int Conf Intell Syst Mol Biol 5 179 186 9322033
Larsen TS Krogh A 2003 EasyGene: A prokaryotic gene finder that ranks ORFs by statistical significance BMC Bioinform 4 21
Burge C Karlin S 1997 Prediction of complete gene structures in human genomic DNA J Mol Biol 268 78 94 9149143
Rissanen J 1987 Stochastic complexity J R Stat Soc Ser B 49 223 239
Schwarz G 1978 Estimating the dimensions of a model Ann Stat 6 461 464
Kuha J 2004 AIC and BIC: Comparisons of assumptions and performance Sociol Methods Res 33 188 229
|
16292353
|
PMC1282332
|
CC BY
|
2021-01-05 09:20:35
|
no
|
PLoS Comput Biol. 2005 Nov 11; 1(6):e56
|
utf-8
|
PLoS Comput Biol
| 2,005 |
10.1371/journal.pcbi.0010056
|
oa_comm
|
==== Front
PLoS Comput BiolPLoS Comput. BiolpcbiplcbploscompPLoS Computational Biology1553-734X1553-7358Public Library of Science San Francisco, USA 1629235410.1371/journal.pcbi.001006005-PLCB-RA-0142R2plcb-01-06-02Research ArticleBioinformatics - Computational BiologyEvolutionMicrobiologyEubacteriaArchaeaA Guild of 45 CRISPR-Associated (Cas) Protein Families and Multiple CRISPR/Cas Subtypes Exist in Prokaryotic Genomes A Community of CRISPR/Cas SubtypesHaft Daniel H *Selengut Jeremy Mongodin Emmanuel F Nelson Karen E The Institute for Genomic Research, Rockville, Maryland, United States of AmericaEisen Jonathan A EditorThe Institute for Genomic Research, United States of America* To whom correspondence should be addressed. E-mail: [email protected] 2005 11 11 2005 6 10 2005 1 6 e6029 6 2005 5 10 2005 Copyright: © 2005 Haft et al.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.Clustered regularly interspaced short palindromic repeats (CRISPRs) are a family of DNA direct repeats found in many prokaryotic genomes. Repeats of 21–37 bp typically show weak dyad symmetry and are separated by regularly sized, nonrepetitive spacer sequences. Four CRISPR-associated (Cas) protein families, designated Cas1 to Cas4, are strictly associated with CRISPR elements and always occur near a repeat cluster. Some spacers originate from mobile genetic elements and are thought to confer “immunity” against the elements that harbor these sequences. In the present study, we have systematically investigated uncharacterized proteins encoded in the vicinity of these CRISPRs and found many additional protein families that are strictly associated with CRISPR loci across multiple prokaryotic species. Multiple sequence alignments and hidden Markov models have been built for 45 Cas protein families. These models identify family members with high sensitivity and selectivity and classify key regulators of development, DevR and DevS, in Myxococcus xanthus as Cas proteins. These identifications show that CRISPR/cas gene regions can be quite large, with up to 20 different, tandem-arranged cas genes next to a repeat cluster or filling the region between two repeat clusters. Distinctive subsets of the collection of Cas proteins recur in phylogenetically distant species and correlate with characteristic repeat periodicity. The analyses presented here support initial proposals of mobility of these units, along with the likelihood that loci of different subtypes interact with one another as well as with host cell defensive, replicative, and regulatory systems. It is evident from this analysis that CRISPR/cas loci are larger, more complex, and more heterogeneous than previously appreciated.
Synopsis
The family of clustered regularly interspaced short palindromic repeats (CRISPRs) describes a class of DNA repeats found in nearly half of all bacterial and archaeal genomes. These DNA repeat regions have a remarkably regular structure: unique sequences of constant size, called spacers, sit between each pair of repeats. The DNA repeats do not encode proteins, but appear to be transcribed and processed into small RNAs that may have any number of functions, including resistance to any phage (i.e., virus of bacteria) whose sequence matches a spacer; spacers change rapidly as microbial strains evolve. This work describes 41 new CRISPR-associated (cas) gene families, which are always found near these repeats, in addition to the four previously known. It shows that CRISPR systems belong to different classes, with different repeat patterns, sets of genes, and species ranges. Most of these seem to come and go rather rapidly from their host genomes. These possibly beneficial mobile genetic elements may play an important role in driving prokaryotic evolution.
Citation:Haft DH, Selengut J, Mongodin EF, Nelson KE (2005) A guild of 45 CRISPR-associated (Cas) protein families and multiple CRISPR/Cas subtypes exist in prokaryotic genomes. PLoS Comput Biol 1(6): e60.
==== Body
Introduction
Clusters of short DNA repeats with nonhomologous spacers, which are found at regular intervals in the genomes of phylogenetically distinct prokaryotic species, comprise a family with recognizable features [1–3]. These repeats were first observed by Ishino and colleagues [4] upstream of the iap gene in Escherichia coli and later in other bacterial and archaeal species such as Haloferax mediterranei, Streptococcus pyogenes, and Mycobacterium tuberculosis. Each member of this family of repeats was designated differently by the original authors, leading to a confusing nomenclature: SPIDR (spacers interspaced direct repeats) by Jansen and colleagues [5], SRSR (short regularly spaced repeats) by Mojica and colleagues [2], and LCTR (large cluster of 20-nt tandem repeat sequences) by She and colleagues [6], among others. Based on a systematic characterization in 40 different bacterial and archaeal genomes, Jansen and colleagues [3] proposed, in agreement with Mojica's research group, a new nomenclature for this family of DNA repeats, which are now known as clustered regularly interspaced short palindromic repeats (CRISPRs). Sequencing of the genome of the archaeon Methanococcus (now Methanocaldococcus) jannaschii [7] led to the first detailed characterization of these repeats in a complete genome, where 18 loci were found, most featuring a single copy of a long repeat (LR) or leader sequence and a variable number of regularly spaced short repeats (SRs). In M. jannaschii, two of the LRs were truncated, each ending with a single SR, and one cluster of five SRs had no LR. Similar repeats were identified in the genome of the hyperthermophilic bacterium Thermotoga maritima [8]. The association of these repeats with nearby gene clusters that showed closest similarity to archaeal species and their atypical DNA composition (as measured by χ2 analysis) were called consistent with other observations as evidence of lateral gene transfer (LGT) between archaeal and bacterial species [8]. Together, these findings suggested transfer of repeat-associated DNA within and between prokaryotic genomes.
Four genes, designated CRISPR-associated (cas), are found only in species containing CRISPR, always located near to a repeat locus, and usually oriented head-to-tail as if cotranscribed [3]. The most common arrangement of these genes is cas3–cas4–cas1–cas2. The Cas3 protein appears to be a helicase, whereas Cas4 resembles the RecB family of exonucleases and contains a cysteine-rich motif, suggestive of DNA binding. Cas1 is generally highly basic and is the only Cas protein found consistently in all species that contain CRISPR loci. Cas2 remains to be characterized. None of the other genes in the vicinity of these four cas genes were found to represent protein families specifically associated with CRISPR.
There has been only limited biological characterization of these elements. Efforts to increase the copy numbers of these repeats in the archaeon Haloferax volcanii resulted in altered segregation and reduced viability of the cells [1]; a role in replicon partitioning was suggested. Supporting this, small clusters of repeats are found in two self-transmissible plasmids of the genus Sulfolobus; these plasmids appear more stably maintained than plasmids lacking repeats [9]. The main chromosome of Sulfolobus solfataricus, unlike the plasmids themselves, has both cas gene clusters next to repeats [3] and a genus-specific binding protein (SSO0454) for its own and for the plasmids' repeats [10]. Tang and coworkers [11] found 22 small, nonmessenger RNAs transcribed from CRISPR arrays of the archaeon Archaeoglobus fulgidus, nearly all had sizes just below a repeat plus a spacer with the 3′ end in the middle of the repeat; repeat arrays of S. solfataricus also were transcribed into RNA and processed. Two recent analyses of spacer elements found between individual CRISPRs demonstrate that most have no conclusive origin by sequence similarity; those that do, strikingly, match phage or other types of selfish genetic elements [12,13]. Despite these advances, the functions of both CRISPR and Cas proteins remain unknown. In this study, we present a detailed analysis of four previously defined and 41 newly identified Cas proteins in the microbial species for which we have complete genome sequences.
Results
Identification of New CRISPR-Associated (Cas) Protein Families
In addition to the previously described cas genes (cas1–4), we have detected a number of protein families whose members are found in the vicinity of CRISPRs across multiple prokaryotic species. Hidden Markov models (HMMs) for these families have been constructed and deposited in the TIGRFAMs database (http://www.tigr.org/TIGRFAMs; Table 1). In the present work, the CRISPR-associated protein families are described as a “guild,” i.e., a collection of members that perform somewhat similar work. The guild is presumed to be involved in processes that may include the maintenance of repeat clusters [3], capture of new spacer elements [12,13] and expansion or contraction of clusters, propagation of the leader sequence and repeat clusters within a genome [3,7], transfer of CRISPR and cas genes together to new genomes [8,14,15], and interaction of CRISPR/cas loci with the host cell (see Discussion). From our study, a total of 53 HMMs have been constructed that represent at least 45 different protein families (including Cas1–4). The discrepancy between the number of HMMs and protein families results from two pairs of models for the Cas2 protein and the Cas3 protein (Table 1), which have enough sequence divergence that a single model is not sufficient. Also included is a model for the HD domain of Cas3, which in some cases is a separate polypeptide and in others is absent. Finally, in addition to a model that detects the diagnostic domain of Cas5 (see below), we present five narrower models that detect the five subtype-specific full-length variants of this family (Table 1). Many of these families contain members that belong to clusters of orthologous groups (COGs) [14,16], although the relationship between the HMMs described here and these COGs is imperfect (see Discussion). The functions of these protein families are largely unknown, although distant homologies to characterized proteins, motifs, and domains have been noted in the present study and in previous analyses [14] (Table 1). For example, eight families of CRISPR-associated proteins (Csm3–5, Cmr1, Cmr3, Cmr4, Cmr6, and Csx7) all belong to the repair-associated mysterious protein (RAMP) superfamily [14] as detected by Pfam [17] model PF03787. These RAMP families appear to act in concert since sets of them typically are found in gene clusters (Figure 1).
Table 1 Description of the Different cas Core Genes, CRISPR/Cas Subtypes, and the RAMP Module, Based on the New Cas Protein Families
A, AFigure 1 Distribution of the Different CRISPR/Cas Subtypes across Some of the Prokaryotic Species for Which a Whole-Genome Sequence Is Available
The taxonomy of each species/strain is indicated on the left side of the figure. The CRISPR/cas loci of a number of illustrative examples for the different CRISPR subtypes are displayed on the right side of the figure.
a
E. coli K12-MG1655, O157:H7 EDL933, and O157:H7 VT2-Sakai.
b
Salmonella enterica Paratyphi ATCC9150, serovar Typhi CT18, and Ty2; Salmonella typhimurium LT2 SGSC1412.
c
Y. pestis CO92, KIM, and biovar Mediaevalis 91001; Yersinia pseudotuberculosis IP32593.
d“p” indicates a partial cluster lacking some of the genes usually associated with this subtype, the repeats, or both. Such clusters may represent autonomous functional units, degradation from the common subtype, or cases in which the missing components are supplied by distantly located CRISPR clusters within the same genome.
The assignment of genes to these 45 families has allowed for an analysis of the genomic context in which they are typically found. Three basic types of family context have emerged. First, the “core” cas genes (i.e., cas1–4) are found in a wide range of contexts with respect to the other gene families, whose genes are clustered nearby. Second, subtype-specific genes are found in association with the core genes and others of the same subtype, often with conserved gene order. Finally, modular genes, associated with one another in particular combinations, are always found in genomes containing the core genes, but may be found at distant sites from those clusters.
Based on the observation of such contextual patterns, we have defined two additional core cas genes (cas5 and cas6), eight CRISPR-associated subtypes, and one CRISPR-associated module each of which is described in detail below and presented in Table 1. Each of the subtypes has been named for a genome in which it appears as the only CRISPR locus (e.g., CRISPR subtype Apern after Aeropyrum pernix; see Table 1, Figure 1), and the associated subtype-specific genes have been assigned gene symbols based on a three-letter prefix and a numeral suffix (i.e., csa1) following the cas1–4 model (Table 1, Figure 1). The module (CRISPR RAMP module, cmr1–6) has been named after the RAMP superfamily since four of the six genes appear to be members of this superfamily. Models for a number of families have been constructed for which no contextual pattern has yet been defined, most likely due to an insufficient number of genomes harboring the gene. These have been assigned gene symbols with the prefix “csx” (Table 2).
Table 2 Other CRISPR/Cas Protein Families with No Identified Contextual Pattern
Our assignments of genes to CRISPR-associated families has allowed for the identification of CRISPR/cas loci that span the genomic distance between CRISPR arrays not previously appreciated as forming the same locus (e.g., Bacillus halodurans C-125 and Aquifex aeolicus VF5; see Figure 1). Indeed, it appears to be a rule that virtually every gene found between any two cas genes is strictly CRISPR-associated itself, although it may not be as common as the core cas genes, cas1–cas6. The exceptions are putative transposases, restriction enzymes, and addiction module proteins, or hypothetical proteins with few or no homologs; several examples appear in Figure 1. Frequently we have found that the addition of new genomes to our databases shows such hypothetical proteins to belong to a new family of cas genes. Through this process of slowly building up our library of cas gene families, the patterns of conserved subtypes, previously obscure, has come into sharp focus (Figure 1).
New Core cas Gene Families: cas5 and cas6
The cas core genes (cas1–4) were originally delineated by Jansen and colleagues [3] and are characterized by their close proximity to the CRISPR loci and their broad distribution across bacterial and archaeal species. Although not all cas core genes associate with all CRISPR loci, they are all found in multiple subtypes (Table 3). We have observed a 43-amino acid N-terminal domain, which appears in a single protein in each of five separate CRISPR/Cas subtypes and in a number of currently untyped loci. We designate these families collectively as Cas5. Members average 250 amino acids in length; regions outside the N-terminal domain form subtype-specific families with remote to undetectable homology across subtypes. For this reason, we have included both subtype-specific full-length models and the domain model in the TIGRFAMs library (see Table 1) and have assigned gene symbols with a trailing letter to indicate the subtype variant in question (e.g., cas5e, for the Ecoli subtype variant). The Cas5 domain is found in the M. xanthus DK1622 DevS protein, which has been implicated in a species-specific developmental pathway [18,19], although found within an apparent CRISPR/cas locus of a novel subtype. There may be a distant homology relationship between Cas5 proteins and the RAMP superfamily, though not detected by the Pfam [17] RAMP superfamily model PF03787 (data not shown).
Table 3 Characteristics of the Repeat Arrays Associated with the Different CRISPR/Cas Subtypes
The Cas6 family includes proteins averaging 140 amino acids in length that share a strong homology at the C-terminus, including a GhGxxxxxGhG motif (“h” indicates a hydrophobic amino acid). The cas6 gene is found in association with four separate CRISPR/Cas subtypes (see Table 3) and has a preferred location as the cas gene most distal to the CRISPR (Figure 1).
Description of the Different CRISPR/Cas Subtypes
Tables 1, 3, and 4 and Figure 1 delineate the essential features of the eight CRISPR/Cas subtypes that we have defined thus far, including the subtype-specific (see Table 1) and core (see Table 3) genes involved, the length and periodicity of the repeats and the length distributions of the associated spacers (Table 3), the co-occurrences and subtype fusions observed (see Table 4), and the species in which they are found (Figure 1). Distinctive and notable features of these subtypes will be discussed below. Each subtype is named for the species of a genome sequence in which only that subtype is found.
Table 4 Co-Incidence of the Different CRISPR/Cas Subtypes
CRISPR/Cas Subtype Ecoli (Based on Escherichia coli K12-MG1655)
The Ecoli subtype features five subtype-specific genes and cas1–3. The cas2 gene associated with this subtype is sufficiently diverged from the rest of the Cas2 family so that the construction of a separate HMM (TIGR01873) was necessary. The 61-bp average (see Table 3) periodicity is unique among the subtypes we describe, and we never find an Ecoli-type cluster fused to another type. This subtype is sporadically distributed among bacteria and not found in any of the sequenced Archaea present in the Comprehensive Microbial Resource (http://www.tigr.org/tigr-scripts/CMR2/CMRHomePage.spl) [20]. Saunders and coworkers [21] report a cluster of “bacterial-specific” CRISPR-associated genes in the incomplete genomic sequence of the cold-adapted archaeon Methanococcoides burtonii that prove to be cas3, cse1, cse2, and cse4; we detect a second cluster in this organism with the remaining four required genes.
CRISPR/Cas Subtype Ypest (Based on Various Yersinia pestis Strains)
The Ypest subtype is unique in its lack of a Cas2 homolog. It has the shortest average repeat periodicity (only 60 bp; Table 3), the most well-conserved repeat sequence from one species to another, and easily the narrowest phylogenetic range. It is observed only in several Gammaproteobacteria sp. and one Betaproteobacterium. The Cas3 putative helicase associated with this subtype is sufficiently diverged from the rest of the Cas3 family that the construction of a separate HMM (TIGR02562) was necessary. It is the spacers of the Ypest-subtype repeats in Y. pestis, which were analyzed by Pourcel and colleagues [13]. The cas1 gene associated with these clusters is most closely related to that of the Ecoli subtype (Figure 2), which has the next shortest repeat periodicity.
Figure 2 Molecular Phylogeny of the Cas1 Protein across 54 Prokaryotic Genomes
A representative selection of Cas1 protein sequences were aligned using ClustalW, and columns with greater than 20% gaps were removed. A neighbor-joining tree was calculated in Belvu using the Storm and Sonnhammer distance correction. Trees calculated using more computationally intensive methods showed insignificant differences.
aFrom the preliminary annotation of the Haloferax volcanii genome, currently sequenced at The Institute for Genomic Research (http://www.tigr.org/tdb/mdb/mdbinprogress.html).
CRISPR/Cas Subtype Nmeni (Based on Neisseria meningitidis Serogroup A Z2491)
This subtype is the most abbreviated that we have described, being the only one lacking cas3, cas4, and cas5 and having the shortest average spacer lengths observed (30 bp; Table 3). Jansen and colleagues [3] noted similarity between Cas4 and RecB family exonucleases. Members of the Csn1 family, by contrast, contain an McrA nuclease-like domain [14]. Csn1, a large and likely multidomain protein, may perform the functions of the absent Cas4 and potentially Cas3 as well. A second subtype-specific gene, csn2, is present in some but not all Nmeni CRISPR/cas loci. A characteristic feature of this subtype is a single copy of the repeat (sometimes direct, sometimes inverted), which appears upstream of the first gene in the locus, in addition to the repeat cluster downstream of the last gene. Notably, species bearing this CRISPR/Cas subtype are, without exception, vertebrate pathogens and commensals.
“Three-Gene” CRISPR/Cas Subtypes: Dvulg (Based on Desulfovibrio vulgaris Hildenborough), Tneap (Based on Thermotoga neapolitana DSM4359), and Hmari (Based on Haloarcula marismortui ATCC 43049)
A number of subtypes appear to have a similar overall structure consisting of cas1–4 (and, with the exception of Dvulg [see below] of cas6 as well) and three subtype-specific genes, including the subtype-specific cas5 variants (see Figure 1). Typically, one of these shows homology to the DevR protein, which has been characterized as a negative autoregulator and is the presumed partner of DevS/Cas5 from M. xanthus DK1622 [22], while the last is a large (400–700 amino acids) protein, often containing a CXXC–CXXC motif. These three genes are always adjacent to cas3. Each of these types generally is associated with repeat spacers of a distinct average length (Dvulg, 34; Hmari, 36; and Tneap, 37; Table 3), and all have cas1 genes that are more closely related to one another than to the cas1 genes of other subtypes (see Figure 2). Several genomes contain CRISPR/cas loci that also appear to conform to this structural class, but the number of sequenced genomes containing homologs is currently insufficient to create subtype-specific HMMs (e.g., CRISPR/cas loci of Leptospira interrogans serovar Lai strain 56601 and Fusobacterium nucleatum ATCC 25586). In addition to the repeat cluster immediately downstream of the cas gene operon, CRISPR/cas loci of the Tneap and Hmari types, but not the Dvulg type, frequently have another cluster immediately upstream.
CRISPR/Cas Subtype Apern (Based on A. pernix K1)
This subtype is found only in Archaea and comprises the only described subtype in the Crenarchaeota (although the RAMP module is also observed; see below). Although this subtype is only found in thermophilic species, significance of this correlation is tempered by the fact that the large majority of archaeal species are thermophiles. In Sulfolobus sp. and A. fulgidus DSM4304, csa4 is absent, while csa5, a gene not found in A. pernix, is present, although they have no detectable homology. In similar fashion to the three-gene class of subtypes (see above), csa2 is a distant homolog of devR. The cas1 genes of this subtype are most closely related to those of the three-gene subtypes; indeed, the cas1 gene of M. jannaschii DSM2661 would appear by homology to be a Tneap type and may represent an instance of subtype recombination (Figure 2).
CRISPR/Cas Subtype Mtube (Based on M. tuberculosis Strains CDC1551 and H37Rv)
Although observed as the sole subtype in several genomes, this subtype is more commonly found in genomes containing other subtypes at remote sites and in hybrid, fused loci (e.g., T. maritima; see Figure 1). The subtypes with which Mtube have been observed to associate are Tneap, Hmari, and Apern (Figure 1). The repeats proximal to the Mtube genes in unfused loci have long, but variable average periodicities and spacers (Figure 3). When found in fused loci, the csm genes tend to be distal from cas1 and cas2, which are themselves proximal to the subtype-specific genes of the other subtype in the locus. Additionally, in these hybrid loci, the spacer length is typical of the subtype partner, not of the Mtube subtype. Occasionally, as is observed in Methanosarcina acetivorans C2A, an Mtube locus with a robust repeat array is found lacking all cas core genes, although they are found elsewhere in the genome in association with a CRISPR/cas locus of another subtype. This would suggest that the linkage between the core cas genes and the subtype-specific genes is weaker in this subtype and is somewhat akin to the behavior of the RAMP module (see below). The notable variability of the spacer lengths may also indicate a heterogeneity of origin for these repeat arrays. When the cas1 genes that are proximate to csm genes are incorporated into a tree such as the one displayed in Figure 2, they do not form a single clade, but are found in clades that include cas1 genes associated with those subtypes with which Mtube typically associates (data not shown). Additionally, three of the five subtype-specific genes (csm3–5) are members of the RAMP superfamily, and csm1 is a homolog of cmr2.
CRISPR/Cas RAMP Module
CRISPR/Cas systems include a six-gene module that appears to occur only in genomes that contain other CRISPR systems, whether or not those systems are nearby. The cluster of genes cmr1–6 is observed in a wide range of archaeal and bacterial genomes, but always in the same genome with other subtypes, and most often fused into hybrid loci (see Figure 1; Table 1). Four of the six genes in this module are members of the RAMP superfamily [14]. This RAMP module associates with the three-gene class (Dvulg, Tneap, and Hmari), as well as Apern and Mtube subtypes. We have observed one instance where a RAMP module is found in the same genome with an Ecoli type (in Thermus thermophilus HB8), but in this case the RAMP is actually part of a hybrid locus with an Mtube subtype.
Degraded and Atypical CRISPR/cas Loci
Expansion of the number of CRISPR families from four to 45, definition of CRISPR/Cas subtypes, and examination of over 200 genomes allow reexamination of cas pseudogenes and degenerate CRISPR/cas loci. The genome of Thermoplasma volcanium, e.g., should be viewed not as having lost much of the system present in many other Archaea [14], but rather as having a CRISPR/cas locus of the Mtube type, while many other Archaea have loci of the Apern type (Figure 1). We believe we have observed evidence of ongoing processes of CRISPR locus degradation in multiple independent cases in five separate subtypes (Ecoli, Ypest, Nmeni, Dvulg, and Mtube) in bacteria. Examples include both cas1 and cas2 pseudogenes adjacent to apparently intact subtype-specific genes (allowing subtype identification) and loci in which novel subtype-specific genes are degenerate. A dramatic case of degradation is found in the genome of Coxiella burnetii RSA 493, in which the Ypest locus contains frameshifts and truncations in four of the six genes, and the repeat array consists of only a single exact copy of the Ypest repeat. The Ypest repeat is so well conserved that it can be used to search for genomes in which it is the only remaining trace of a Ypest CRISPR locus. BLAST searches of a consensus sequence created from alignments of the repeat can detect instances with up to four mismatches. We have detected tiny arrays of one or two full-length copies and one additional partial repeat (with the expected spacing of exactly 60 nt) only in several strains of E. coli, Shigella flexneri, and Shewanella oneidensis (all of which are within the phylogenetic range of the Ypest subtype generally).
Atypical CRISPR systems do occur, such as the previously overlooked repeat array in the genome of Thermoplasma acidophilum, where no cas genes are found, and the Ypest system of Zymomonas mobilis ZM4, where the cas gene cluster is far from the characteristic 28-bp repeat cluster. Large distances between cas gene clusters and their closest repeat clusters occur less often than cas pseudogenes adjacent to degenerate repeats. Excluding these rare exceptions, the average distance from the cas gene cluster to the nearest repeat cluster is well below 1,000 bp and varies according to CRISPR/Cas subtype, e.g., 180 bp for Dvulg, 232 bp for Ypest, and 414 bp for Apern. This spacing often accommodates the CRISPR leader sequence.
Molecular Phylogeny of Cas Core Proteins
The definition of the subtypes discussed above was driven by the observation of the conserved contexts of families of distinct genes. As has been mentioned above, the cas core genes are found across various subtypes, most definitively in the case of cas1, which appears to be nearly universal for CRISPR/cas loci. The molecular phylogeny of various Cas core proteins has been explored by the construction of multiple sequence alignments, restriction of those alignments to well-aligned regions, and the calculation of neighbor-joining trees. A representative tree for Cas1 is shown in Figure 2. Trees for other Cas core proteins showed largely the same pattern, although limited to the subtypes in which they are individually found. These trees were robust, showing insignificant differences in branching patterns when a variety of alignment regions and tree-building algorithms were used. The clustering of the Cas core proteins broadly recapitulates the subtype divisions that were defined independently of this information. There would appear to be a limited number of cases where the Cas core proteins do not share the same evolutionary history as their associated subtype-specific proteins.
Discussion
CRISPR is a widely distributed family of repeats in prokaryotes [1–3,5,7,15]. Preliminary insight into their biology came with the discovery that four different protein families occur in prokaryotes only if CRISPRs are present. These proteins are always near a set of these repeats and always include Cas1 [3]. In the current study, we built on these prior findings and established a number of HMM-defined Cas protein families. These protein families have been found to form conserved clusters across multiple genomes, which allowed us to create rules for the identification of specific subtypes of CRISPR/Cas system.
From the study presented here, it is apparent that CRISPR/Cas systems are far more complex than previously appreciated. Forty-five distinct protein families associated with CRISPRs have been identified among the first 200 completed prokaryotic genomes. These are currently represented by 53 HMMs (Tables 1 and 2). These models are sensitive, in that they unambiguously identify genes, and are also selective, in that they do not identify genes in organisms lacking CRISPR/cas loci. The subtype-specific models accurately discriminate between the subtypes but may, infrequently, identify genes in novel CRISPR/cas contexts that, given sufficient additional genomes, would warrant the status of separate subtypes.
Previous work by Makarova and colleagues [14], conducted on a smaller set of available microbial genomes and without the knowledge of the associated CRISPRs, resulted in the identification of some 20 gene families (COGs) proposed to act in DNA repair, many of which contain genes identified by our HMM models. The relationship between these two sets of families is uneven, with some of our HMMs spanning multiple COGs, some COGs spanning multiple HMMs, and some COGs including genes we believe unrelated to CRISPRs. COG0640, e.g., includes eight putative transcriptional regulators in A. fulgidus and five in M. jannaschii, but only MJ0379 and AF1869, one locus in each species, are CRISPR-associated; they encode the Csa3 protein of the Apern-type CRISPR system. These differences are not unexpected, considering the different clustering methods and search algorithms applied to unequal datasets in this case. Their work also introduced the RAMP superfamily[14], to which a number of Cas protein families belong. The proposed helicase, nuclease, and other domains for DNA repair metabolism may instead or in addition act in the processes of CRISPR physiology: mobilization, maintenance, processing, and addition of new spacer elements. To reflect this change in interpretation, we propose renaming the RAMP superfamily from repair-associated to repeat-associated mysterious protein, thus preserving the acronym currently in use.
The groups of gene families that comprise the CRISPR/Cas subtypes appear to have traveled together through evolutionary time as discrete units. Even the core cas genes appear to have the same evolutionary history as their partner subtype-specific genes (Figure 2). The reasonable hypothesis that the Cas proteins interact (i.e., bind to, stabilize, regulate the expression of, cleave, modify, degrade, etc.) with the repeats in their DNA or expressed RNA form is supported by the observation of subtype-specific characteristics of the repeats such as repeat periodicity. Although as demonstrated in this study, CRISPR/cas loci of different subtypes can coexist within the same genome, phylogenetic reconstructions of Cas core proteins do not provide any evidence of switching between subtypes having repeat periodicities of 60, 61, and those with longer periodicities (Figure 2). The RAMP module and the RAMP-like Mtube subtype would appear to deviate from this pattern, showing varying degrees of independence from dedicated cas core genes and their associated repeat periodicities.
It has been previously suggested that cas genes have undergone LGT events based on phylogenetic analyses and conservation of gene order [14], anomalous nucleotide frequencies [8], and the presence of multiple chromosomal CRISPR loci [3]. Our finding that the core cas genes belong to multiple CRISPR subtypes, each with its own sporadic distribution, indicates that this conclusion should be reexamined and reconfirmed. Indeed we have observed several lines of evidence that support the LGT hypothesis: (1) CRISPR/cas loci representing five different subtypes are found on plasmids (subtype Ypest in Legionella pneumophila Lens, subtype Dvulg in D. vulgaris, subtype Hmari in H. marismortui, subtype Ecoli in Photobacterium profundum, and both subtypes Mtube and Ecoli in T. thermophilus HB8). In the case of L. pneumophila Lens, a second, nearly identical copy of the locus is found on the chromosome. (2) In L. pneumophila Paris, by contrast, there is no trace of any gene with homology to the Ypest subtype genes or repeats found in the Lens strain, while an entirely different (untyped) CRISPR locus is found. Differences in CRISPR locus content have been observed between closely related strains of S. pyogenes, Listeria monocytogenes, and T. thermophilus. (3) Comparison of the Ecoli subtype loci from E. coli K12-MG1655 and E. coli O157:H7 EDL933 shows that while Cas1, Cas2, and the surrounding genomic region are nearly identical between K12-MG1655 and O157:H7 EDL933, this similarity does not extend to the rest of the Cas proteins in the cluster. For K12-MG1655, these proteins are most similar to those in Geobacter sulfurreducens, while for O157:H7 EDL933 they are most similar to those of Photorhabdus luminescens. (4) Cas1 proteins found in Porphyromonas gingivalis W83, Vibrio vulnificus YJ016 and Nostoc sp. PCC 7120 are fusion proteins, having a C-terminal Cas1 domain but also a reverse transcriptase domain similar to that found in group II introns. This may represent one mechanism used for mobilization in a subset of CRISPR loci.
Clusters of cas genes and their associated repeats must maintain themselves in prokaryotic populations by reproducing and mobilizing themselves as fast as they are degraded. We see numbers of degenerate CRISPR/cas systems as well as profound differences in cas gene content between closely related species or strains. This is significant, because it implies that the process of replenishing genomes with intact CRISPR loci is frequent. We are inclined to believe that CRISPR/cas loci may, under certain circumstances, confer selective advantages to their host cells and, in these cases, stabilize the loci against degradation. We have yet to observe a single instance of a duplicated cas gene cluster on the chromosome(s) of any species. This is in contrast to selfish genetic elements such as transposons, which persist in a given lineage largely through redundancy.
Plasticity with respect to the number of repeat copies, as well as the extensive differences in the spacers between repeats, is observed in CRISPR loci [2,12,15,23]. The finding that spacer sequences derive from foreign DNA, such as phage and transposons, suggests a defensive capacity for at least some instances of CRISPR system [12,13], but roles in replicon partitioning in the Archaea [1] and regulation of fruiting body development in M. xanthus [19,22] are also suggested. Correlation of repeat expression with CRISPR subtype is in order. Apern subtype repeats are expressed and processed in A. fulgidus and S. solfataricus [11]. Also expressed, in addition to their neighboring cas genes, are the Nmeni repeats of Streptococcus agalactiae (H. Tettelin and J. Dunning Hotopp, personal communication), the Mtube repeats of Staphylococcus epidermidis (S. Gill, personal communication), and the Hmari/Mtube/RAMP module region repeats of T. maritima (data not shown). Five separate markers from the Ecoli-type CRISPR array of G. sulfurreducens were up-regulated 2- to 3-fold when cells were grown with Fe(III) versus fumarate as electron acceptor [24].
We have characterized multiple distinct subtypes of CRISPR/cas loci and demonstrated profound differences in CRISPR system content between closely related strains and species. Beneficial roles may include defense of the host against foreign DNA [12,13] and regulation of the fruiting body development cycle by the DevR and DevS cas genes in the special case of M. xanthus [19,22]. These findings support an emerging model of CRISPR/cas systems. They appear to be portable adaptation modules for their host genomes. They are sufficiently unstable that degenerate forms are often seen and sufficiently mobile that multiple instances of LGT are apparent. Their repeat arrays consistently are among the most rapidly evolving loci seen in strain comparison studies, such that they are the basis of “spoligotyping” [23,25,26]. Both cas gene and repeat expression can be differentially regulated. They can be co-opted by their hosts for new regulatory systems, as seen for a pathway unique to Myxococcus in the interaction between the non-Cas protein FruA and Cas protein DevR. The adaptations they enable may be supplanted later by the evolution of more stable regulatory systems, but in the meantime they may be superbly useful in rapid adaptation, such as in the invasion of a new biological niche.
Materials and Methods
Identification of CRISPRs.
CRISPRs were identified by three methods. Arrays of exact or near-exact repeats were readily detected by REPfind, a part of the REPuter package [27–29]. Sample sequences from known arrays were used to identify additional, smaller repeat clusters by BLASTN (http://blast.wustl.edu). Finally, regions suspected to have few and/or poorly conserved (degenerate) repeats, including regions near otherwise unexplained cas gene clusters, were examined manually with the dot-matrix homology visualization tool dotter [30].
Definition of CRISPR-associated (Cas) protein families.
Cas protein families were identified from the construction of specific HMMs and subsequently deposited in the TIGRFAMs database (http://www.tigr.org/TIGRFAMs) [31]. The construction of HMMs involves refining multiple sequence alignments (also known as seed alignments), building HMMs from these alignments, exhaustively searching protein sequence databases, and selecting cutoff scores above which are found only true positives and below which no false negatives are detected. HMMs for Cas1–Cas4 were recognized among families previously designated “conserved hypothetical protein,” or were constructed for the first time, according to descriptions of these families by Jansen and colleagues [3]. All proteins encoded between or near identifiable cas genes were searched against a series of in-house databases of all available protein sequences and of prokaryotic genome sequences. Those that showed a pattern of matching numbers of similarly positioned proteins were investigated further as candidate new Cas protein families. For many of these families the process is iterative. Significant matches to the current model for the emerging family are found only near CRISPRs (see below) and/or previously identified cas genes. These new sequences are added to the family and realigned, and the revised HMM may then identify additional sequences. More distantly related sequences were distinguished from spurious matches by their contiguity to cas and CRISPR loci in other genomes, by the quality of the revised multiple sequence alignments, and by the improved search sensitivity of the HMM that resulted from adding these sequences to the seed alignment. Completed models were classified as new Cas protein families only if they did not overlap with other Cas HMMs, did not identify a sequence in a species that lacks CRISPRs, and if members of the family were found in at least four different species from at least three different lineages. Furthermore, members of these models had to be encoded adjacent or near to a set of CRISPRs and other cas genes. Iteration was halted rather than allow separate families to coalesce into one if the separate families showed substantially different domain architecture with only local sequence similarity, or if the separate families described separate genes recurringly found near one another in different genomes. The construction of Cas protein HMMs continued in this way until the boundaries of the loci were reached, where additional genes had identifiable non-CRISPR-associated functions, and/or their homologs in other genomes were no longer CRISPR-associated. All designated Cas family HMMs were searched routinely against comprehensive protein databases, which include eukaryotic sequences and CRISPR-negative prokaryotic genomes, to reconfirm specific association with repeats.
CRISPR genome properties.
The presence or absence of CRISPR/Cas systems (with both repeats and sets of Cas proteins that include Cas1) in general and of eight different CRISPR/Cas subtypes (Ypest, Ecoli, Nmeni, Apern, Dvulg, Tneap, Hmari, Mtube, and the RAMP module) are determined by evidence-based rules implemented in the Genome Properties system [32]. Genome Properties is a database system (http://www.tigr.org/Genome_Properties) that can collect both manually curated and automated rule-based assertions of the presence or absence of complex biological systems and their components. States of YES and NO were imported from the work of Jansen and colleagues [3] and corrected in one case (T. acidophilum to YES). The YES state is set for subsequent prokaryotic genomes if Cas1 is detected; repeats are examined subsequent to assignment of the state. The state “none found” is converted to “NO” only if repeats prove absent. Rules for the individual CRISPR/Cas subtypes are based on protein family assignments made by the sets of subtype-specific HMMs listed in Table 1 and on proximity to the specified core Cas proteins for each type listed in Table 3.
This project was funded under awards from the US Department of Energy (DE-FG02-01ER63133, DE-FG02-04ER63935, and DE-FG02-01ER63203). The authors would also like to thank Dr. Robert T. DeBoy for useful discussions.
Competing interests. The authors have declared that no competing interests exist.
Author contributions. DHH and JS conceived and designed the experiments. DHH, JS, and EFM performed the experiments. DHH, JS, EFM, and KEN analyzed the data. DHH, JS, EFM, and KEN contributed reagents/materials/analysis tools. DHH, JS, EFM, and KEN wrote the paper.
A previous version of this article appeared as an Early Online Release on October 6, 2005 (DOI: 10.1371/journal.pcbi.0010060.eor).
Abbreviations
CasCRISPR-associated
COGsclusters of orthologous groups
CRISPRclustered regularly interspaced short palindromic repeat
HMMhidden Markov model
LGTlateral gene transfer
LRlong repeat
RAMPrepair-associated mysterious protein
SRshort repeat
==== Refs
References
Mojica FJ Ferrer C Juez G Rodriguez-Valera F 1995 Long stretches of short tandem repeats are present in the largest replicons of the Archaea Haloferax mediterranei and Haloferax volcanii and could be involved in replicon partitioning Mol Microbiol 17 85 93 7476211
Mojica FJ Diez-Villasenor C Soria E Juez G 2000 Biological significance of a family of regularly spaced repeats in the genomes of Archaea, Bacteria and mitochondria Mol Microbiol 36 244 246 10760181
Jansen R Embden JD Gaastra W Schouls LM 2002 Identification of genes that are associated with DNA repeats in prokaryotes Mol Microbiol 43 1565 1575 11952905
Ishino Y Shinagawa H Makino K Amemura M Nakata A 1987 Nucleotide sequence of the iap gene, responsible for alkaline phosphatase isozyme conversion in Escherichia coli, and identification of the gene product J Bacteriol 169 5429 5433 3316184
Jansen R van Embden JD Gaastra W Schouls LM 2002 Identification of a novel family of sequence repeats among prokaryotes OMICS 6 23 33 11883425
She Q Singh RK Confalonieri F Zivanovic Y Allard G 2001 The complete genome of the crenarchaeon Sulfolobus solfataricus P2 Proc Natl Acad Sci U S A 98 7835 7840 11427726
Bult CJ White O Olsen GJ Zhou L Fleischmann RD 1996 Complete genome sequence of the methanogenic archaeon, Methanococcus jannaschii
Science 273 1058 1073 8688087
Nelson KE Clayton RA Gill SR Gwinn ML Dodson RJ 1999 Evidence for lateral gene transfer between Archaea and bacteria from genome sequence of Thermotoga maritima
Nature 399 323 329 10360571
Greve B Jensen S Brugger K Zillig W Garrett RA 2004 Genomic comparison of archaeal conjugative plasmids from Sulfolobus
Archaea 1 231 239 15810432
Peng X Brugger K Shen B Chen L She Q 2003 Genus-specific protein binding to the large clusters of DNA repeats (short regularly spaced repeats) present in Sulfolobus genomes J Bacteriol 185 2410 2417 12670964
Tang TH Bachellerie JP Rozhdestvensky T Bortolin ML Huber H 2002 Identification of 86 candidates for small non-messenger RNAs from the archaeon Archaeoglobus fulgidus
Proc Natl Acad Sci U S A 99 7536 7541 12032318
Mojica FJ Diez-Villasenor C Garcia-Martinez J Soria E 2005 Intervening sequences of regularly spaced prokaryotic repeats derive from foreign genetic elements J Mol Evol 60 174 182 15791728
Pourcel C Salvignol G Vergnaud G 2005 CRISPR elements in Yersinia pestis acquire new repeats by preferential uptake of bacteriophage DNA, and provide additional tools for evolutionary studies Microbiology 151 653 663
Makarova KS Aravind L Grishin NV Rogozin IB Koonin EV 2002 A DNA repair system specific for thermophilic Archaea and bacteria predicted by genomic context analysis Nucleic Acids Res 30 482 496 11788711
Mongodin EF Hance IR DeBoy RT Gill SR Daugherty S 2005 Gene transfer and genome plasticity in Thermotoga maritima, a model hyperthermophilic species J Bacteriol 187 4935 4944 15995209
Rogozin IB Makarova KS Murvai J Czabarka E Wolf YI 2002 Connected gene neighborhoods in prokaryotic genomes Nucleic Acids Res 30 2212 2223 12000841
Bateman A Birney E Durbin R Eddy SR Howe KL 2000 The Pfam protein families database Nucleic Acids Res 28 263 266 10592242
Ellehauge E Norregaard-Madsen M Sogaard-Andersen L 1998 The FruA signal transduction protein provides a checkpoint for the temporal co-ordination of intercellular signals in Myxococcus xanthus development Mol Microbiol 30 807 817 10094629
Boysen A Ellehauge E Julien B Sogaard-Andersen L 2002 The DevT protein stimulates synthesis of FruA, a signal transduction protein required for fruiting body morphogenesis in Myxococcus xanthus
J Bacteriol 184 1540 1546 11872704
Peterson JD Umayam LA Dickinson T Hickey EK White O 2001 The Comprehensive Microbial Resource Nucleic Acids Res 29 123 125 11125067
Saunders NF Goodchild A Raftery M Guilhaus M Curmi PM 2005 Predicted roles for hypothetical proteins in the low-temperature expressed proteome of the Antarctic archaeon Methanococcoides burtonii
J Proteome Res 4 464 472 15822923
Thony-Meyer L Kaiser D 1993 devRS, an autoregulated and essential genetic locus for fruiting body development in Myxococcus xanthus
J Bacteriol 175 7450 7462 7693658
Schouls LM Reulen S Duim B Wagenaar JA Willems RJ 2003 Comparative genotyping of Campylobacter jejuni by amplified fragment length polymorphism, multilocus sequence typing, and short repeat sequencing: Strain diversity, host range, and recombination J Clin Microbiol 41 15 26 12517820
Methe BA Webster J Nevin K Butler J Lovley DR 2005 DNA microarray analysis of nitrogen fixation and Fe(III) reduction in Geobacter sulfurreducens
Appl Environ Microbiol 71 2530 2538 15870343
Stragier P Ablordey A Meyers WM Portaels F 2005 Genotyping Mycobacterium ulcerans and Mycobacterium marinum by using mycobacterial interspersed repetitive units J Bacteriol 187 1639 1647 15716434
Kamerbeek J Schouls L Kolk A van Agterveld M van Soolingen D 1997 Simultaneous detection and strain differentiation of Mycobacterium tuberculosis for diagnosis and epidemiology J Clin Microbiol 35 907 914 9157152
Kurtz S Choudhuri JV Ohlebusch E Schleiermacher C Stoye J 2001 REPuter: The manifold applications of repeat analysis on a genomic scale Nucleic Acids Res 29 4633 4642 11713313
Kurtz S Ohlebusch E Schleiermacher C Stoye J Giegerich R 2000 Computation and visualization of degenerate repeats in complete genomes Proc Int Conf Intell Syst Mol Biol 8 228 238 10977084
Kurtz S Schleiermacher C 1999 REPuter: Fast computation of maximal repeats in complete genomes Bioinformatics 15 426 427 10366664
Sonnhammer EL Durbin R 1995 A dot-matrix program with dynamic threshold control suited for genomic DNA and protein sequence analysis Gene 167 GC1 GC10 8566757
Haft DH Selengut JD White O 2003 The TIGRFAMs database of protein families Nucleic Acids Res 31 371 373 12520025
Haft DH Selengut JD Brinkac LM Zafar N White O 2005 Genome Properties: A system for the investigation of prokaryotic genetic content for microbiology, genome annotation and comparative genomics Bioinformatics 21 293 306 15347579
|
16292354
|
PMC1282333
|
CC BY
|
2021-01-05 09:18:23
|
no
|
PLoS Comput Biol. 2005 Nov 11; 1(6):e60
|
utf-8
|
PLoS Comput Biol
| 2,005 |
10.1371/journal.pcbi.0010060
|
oa_comm
|
==== Front
BMC GenetBMC Genetics1471-2156BioMed Central London 1471-2156-6-491621266110.1186/1471-2156-6-49Research ArticleTwo previously proposed P1/P2-differentiating and nine novel polymorphisms at the A4GALT (Pk) locus do not correlate with the presence of the P1 blood group antigen Hellberg Åsa [email protected] M Alan [email protected] Martin L [email protected] Division of Hematology and Transfusion Medicine, Department of Laboratory Medicine, Lund University and the Blood Centre, Lund University Hospital, Lund, Sweden2005 7 10 2005 6 49 49 28 3 2005 7 10 2005 Copyright © 2005 Hellberg et al; licensee BioMed Central Ltd.2005Hellberg et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
The molecular genetics of the P blood group system and the absence of P1 antigen in the p phenotype are still enigmatic. One theory proposes that the same gene encodes for both the P1 and Pk glycosyltransferases, but no polymorphisms in the coding region of the Pk gene explain the P1/P2 phenotypes. We investigated the potential regulatory regions up- and downstream of the A4GALT (Pk) gene exons.
Results
P1 (n = 18) and P2 (n = 9) samples from donors of mainly Swedish descent were analysed by direct sequencing of PCR-amplified 5'- and 3'-fragments surrounding the Pk coding region. Seventy-eight P1 and P2 samples were investigated with PCR using allele-specific primers (ASP) for two polymorphisms previously proposed as P2-related genetic markers (-551_-550insC, -160A>G). Haplotype analysis of single nucleotide polymorphisms was also performed with PCR-ASP. In ~1.5 kbp of the 3'-untranslated region one new insertion and four new substitutions compared to a GenBank sequence (AL049757) were found. In addition to the polymorphisms at positions -550 and -160, one insertion, two deletions and one substitution were found in ~1.0 kbp of the 5'-upstream region. All 20 P2 samples investigated with PCR-ASP were homozygous for -550insC. However, so were 18 of the 58 P1 samples investigated. Both the 20 P2 and the 18 P1 samples were also homozygous for -160G.
Conclusion
The proposed P2-specific polymorphisms, -551_-550insC and -160G, found in P2 samples in a Japanese study were found here in homozygous form in both P1 and P2 donors. Since P2 is the null allele in the P blood group system it is difficult to envision how these mutations would cause the P2 phenotype. None of the novel polymorphisms reported in this study correlated with P1/P2 status and the P1/p mystery remains unsolved.
==== Body
Background
The P-related blood groups include four antigens that are predominantly of glycolipid nature and occur in related biosynthetic pathways [1].
The GLOB blood group system [International Society of Blood Transfusion (ISBT) number 028] comprises the P antigen and the GLOB collection (ISBT number 209) includes the Pk antigen and also LKE that is not discussed further here [2]. The P1 antigen is assigned to ISBT system number 003. Five phenotypes depending on the presence or absence of the three antigens, P1, P and Pk, are known (Table 1). The presence of all three antigens results in the P1 phenotype but absence of the P1 antigen causes the P2 phenotype. If both P1 and P are absent the phenotype P2k arises. Absence of P but presence of P1 and Pk results in the P1k phenotype. Absence of all three antigens results in the p phenotype. In each of the phenotypes naturally occurring-antibodies can arise against the missing antigen, invariantly so in the case of P and Pk but less frequently for P1. These phenotypes can be explained biochemically by the presence or absence of some of the enzymes shown to catalyse the pathways shown in Figure 1.
Table 1 The P/GLOB blood groups.
Phenotype Frequency Antigen present on RBC Antibodies in serum
P1 20–90%a P1, Pk, P none
P2 10–80%a Pk, P Anti-P1
p rareb none Anti-PP1Pk
P1k rare P1, Pk Anti-P
P2k rare Pk Anti-PP1
a These frequencies differ significantly between different populations. E.g. the frequency of P1 vs. P2 are virtually the opposite when Caucasians (80 vs. 20%) and Japanese (20 vs 80%) are compared.
b While the frequency of this phenotype has been estimated at 1 per million, two population groups, Swedes and Amish people, have significantly higher numbers (e.g. 141 per million in Västerbotten county in Northern Sweden [21]).
Figure 1 Biosynthetic pathways relating the P1, P and Pk glycolipids.
The P1 antigen is present on hematopoietic cells [3,4] and other cells [5]. The strength of the antigen expression can differ from one person to another and it seems to be dependent on gene dosage [1]. Frequencies of the P1 phenotype vary in different ethnic groups, for example, ~80% of Caucasians are P1, compared to only ~20% of Asians [1]. This may be due to selective pressure since P1 and related antigens can act as cellular receptors for microorganisms and biotoxins [5].
Two genes that code for the Pk- and P-synthesizing enzymes, 4-α-galactosyltransferase (α4GalT, Gb3 synthase) [6] and 3-β-N-acetylgalactosaminyltransferase (β3GalNAcT, globoside/Gb4 synthase) [7], respectively, have been cloned. Subsequently, numerous mutations have been found to explain the rare P1k, P2k [8,9] and p [6,9-12] phenotypes.
The molecular genetic background of the P1 antigen remains unknown. Several theories exist, including one model suggesting that the same α4GalT is able to transfer galactosyl residues to both lactosylceramide and paragloboside but in order to use the latter as the acceptor, a regulatory protein is required [13]. Another model postulates the existence of two different enzymes, and thus two genes, requiring both of them to be inactivated to cause the p phenotype [13]. A third model proposes a single gene with three alleles, one allele coding for an α4GalT that can utilise lactosylceramide and paragloboside as acceptors, one allele using lactosylceramide only and the third allele coding for an inactive form of the transferase [14]. However, none of the known polymorphisms (109A>G, 903G>C, 987G>A) in the coding region of the Pk gene explains the P1/P2 phenotypes [6].
Recently, Iwamura et al. [15] suggested that transcriptional regulation caused by two polymorphisms (-551_-550insC, -160A>G) in the 5'-upstream region of the Pk gene might be the reason for the P1/P2 phenotypes.
The Pk gene was originally thought to comprise two exons, but recent GenBank depositions indicate the presence of three exons with the whole coding region in exon 3, as shown in Figure 2. Various publications have considered different transcription starting points resulting in different numbering of the same nucleotide positions. The numbers used here are described in the legend to Figure 2.
Figure 2 Schematic picture of the exon/intron structure of the A4GALT (Pk) gene and the positions of the 19 polymorphisms investigated. The numbering systems used are as follows: The 5'-region and exon 1 are described using the same numbering as Iwamura et al. [15] for easy comparison. Intron 1 and exon 2 are numbered from the first nucleotide of each element, with the number being preceded by int1- or ex2-, respectively. The adenosine of the translation start codon ATG in exon 3 is defined as nt. 1. Nucleotides upstream from this (untranslated exon 3 and the 3'-region of intron 2) are given as negative numbers. The exons are represented by rectangles and the intervening introns by triangles. Exons 1–3 are 21, 145 and 1059 bp long, respectively and introns 1 and 2 are approximately 25 and 1.5 kbp long, respectively.
In this study we have investigated an extended sequence surrounding the coding region of the Pk gene including untranslated exons and intronic portions as well as potentially regulatory regions 5' or 3'of the transcribed region. Contrary to a previous report [15], we found no clear-cut correlation with the P1/P2 phenotype, neither between previously described polymorphisms in the 5'-regulatory region, nor any of the novel polymorphisms reported in this study.
Results
Screening for the -551_-550insC and -160A>G polymorphisms by PCR-ASP
Seventy-eight samples were screened for the two genetic markers previously [15] suggested to cause the P2 phenotype. The results are summarized in Table 2. Two haplotypes, -550T;-160A and -551_-550insC;-160G, were found whilst the other theoretically possible haplotypes, 550T;-160G and -551_-550insC;-160A, were not detected in this study. Each of the 20 samples, that were phenotyped as P2 were homozygous for both -551_-550insC and -160A>G, which could indicate that these polymorphic positions were indeed P2-specific as proposed. However, 18 of the 58 P1 samples investigated were also homozygous for the same polymorphic markers. Of the remaining P1 samples, 32 were heterozygous at both nucleotide positions and only eight samples were homozygous for the proposed P1-specific combination, -550T;-160A. Twenty-nine of the 32 heterozygous samples were available for analysis by two haplotype-specific PCR reactions (-550T;-160A and -551_-550insC;-160G). All samples were positive in both PCR reactions, indicating that the samples were heterozygous for the combinations -550T;-160A and -551_-550insC;-160G and not for -550T;-160G and -551_-550insC;-160A. Additionally, 26 samples with the rare phenotypes P1k (n = 3), P2k (n = 3) and p (n = 20) were screened. One of the P1k samples was homozygous for the genetic markers reported to be associated with the P2 phenotype, and the other two were heterozygous. Of the P2k samples two had the expected polymorphisms (-551_-550insC;-160G), but the third P2k was heterozygous and thus the first reported sample that lacks the P1 antigen in spite of a genotype not homozygous for the -551_-550insC and -160G markers. Sixteen of the 20 p samples were homozygous for -551_-550insC;-160G, thus consistent with their lack of P1, but one was heterozygous and three were homozygous for -550T;160A. Due to lack of available DNA from these rare individuals the haplotype-specific PCR was only run on one of the four heterozygous samples. As above, this sample was also positive in both PCR reactions.
Table 2 The occurrence of the -550T;-160A and -551_-550insC;-160G haplotypes in individuals with various P phenotypes.
Combination of polymorphisms
Phenotype No. -550T;-160A (homozygous) -550T;-160A -551_550insC;-160G (heterozygous) -551_-550insC;-160G (homozygous)
P1 58 8 32 18
P2 20 0 0 20
P1k 3 0 2 1
P2k 3 0 1 2
p 20 3 1 16
Amplification of the 5'-, 3'- and coding regions of the Pk gene for DNA sequencing
Ten of the 18 P1 samples with P2-associated polymorphisms (-551_-550insC;-160G) were chosen for further investigation by sequencing the 5'-region and the 3'-region, as were all eight P1(-550T;-160A) and nine of the 20 P2 samples. In three samples from each category the whole Pk gene reading frame, located in exon 3, was also sequenced.
In the 5'-region upstream of exon 1, four novel polymorphisms were detected compared to a sequence deposited in GenBank (accession number AL049757). These findings comprised a substitution (-770C>T), an insertion (-107_-106insG) and two deletions (-907_-903del and -17_8del). Interestingly, the latter deletion is located across the border of the 5'-region and exon 1.
In the 3'-UTR, five new polymorphisms were found, four of which were substitutions (1409G>A, 1495C>A, 1523G>A, 1697G>A) and one was an insertion (1592dupG). Figure 2 shows the relative positions of the polymorphisms investigated.
The distribution of the polymorphisms in the three different categories, P1(-550T;-160A), P1(-551_-550insC;-160G) and P2 are shown in Figure 3. As can be seen, none of the polymorphisms is a genetic marker specific for the P2 phenotype. On the other hand, individuals from the P1(-550T;-160A) category are homozygous for 12 of the 16 investigated polymorphisms. However, when individuals from the P1(-551_-550insC;-160G) samples are included such pattern is no longer evident. It can also be noted that the frequency of some of the variants appears to be relatively low, less than ~10 %, for six of the polymorphic sites analyzed.
Figure 3 The distribution of the polymorphic variants in the three different sample categories: P1(-550T;-160A), P1(-551_-550insC;-160G) and P2. The diagram shows data of 16 polymorphisms investigated. The data for the 5'- and the 3'-regions were obtained by sequencing while most of the data for exon 3 originated from PCR-ASP analysis. Eight, P1(-550T;-160A), ten P1(-551_-550insC;-160G) and nine P2 samples were investigated. However, for the 903C>G polymorphism the corresponding numbers were five, five and seven samples. In position -107_-106 all investigated samples had three guanosines instead of two as reported in the reference sequence (AL049757). We therefore do not consider this position as polymorphic in our analysis and it is consequently not included in this figure. cons. = consensus.
While sequencing exon 3, an unexpected mutation, 441G>A, was encountered in a P1 sample of African origin. No other samples, including two of African descent, examined in this study had this particular mutation, which would not alter the amino acid sequence, or any other new polymorphisms.
Haplotype analysis of SNPs in the reading frame and a polymorphism in the 3'-UTR using PCR-ASP
We also performed PCR-ASP utilising polymorphisms in exon 3 and in the 3'-UTR of the Pk gene to determine the cis/trans linkage of SNPs and to establish if combinations of polymorphisms correlated with specific alleles. The SNPs chosen were two of those previously described, i.e. 109A>G and 987G>A. Analysis of the polymorphism 903C>G was considered unnecessary due to its proximity to nt. 987. In addition, of the new polymorphisms the one furthest downstream (1697G>A) was analyzed. The results are summarized in Table 3. None of the haplotypes correlated well with the P1/P2 phenotypes although 83% of the alleles among the P2 samples and 75% of the P1(-551_-550insC;-160G) category samples had the A-1 haplotype, whilst in the P1(-550T;-160A) category samples only 12.5% had this partial haplotype (B-1). The remaining alleles in this latter category consisted of 12.5% of the rare B-3 and 75% of the B-2 haplotype.
Table 3 Correlation of phenotype and haplotype
Phenotype No. of samples Polymorphism Haplotype No. of haplotypes
-550 -160 109 987 1697
P1 10 insC G A G G A-1 15
insC G G A A A-2 5
P1 8 T A A G G B-1 2
T A G A A B-2 12
T A G A G B-3 2
P2 9 insC G A G G A-1 15
insC G G A A A-2 3
Discussion
The molecular background of the P blood group system has been a subject of speculation since its discovery. Antigens formed as a result of closely related biosynthetic pathways are now known to arise from independent genetic loci. However, the current paradox revolves around the 4-α-galactosyltransferase (α4GalT) that synthesises the Pk antigen and possibly also the P1 antigen. If it is indeed exactly the same enzyme synthesizing both, then the P2 and P2k phenotypes should not exist, as the P1-synthesizing α4GalT that is lacking in these individuals should also result in the loss of the Pk and, subsequently, P antigens. The absence of the P1 antigen is not due to a defect in the biosynthesis of paragloboside since this glycolipid is a major precursor of the erythrocyte ABH antigens that are unaffected by the P1/P2 status. The absence of P1 antigen in the P2 phenotype is likely to be caused by inefficient or absent glycosyltransferase activity that can be due to a structural defect in the enzyme itself or more indirectly due to various other factors necessary to ensure the enzyme's optimal expression, localisation and efficiency[1]. Studies are currently in progress to analyse A4GALT mRNA by real-time PCR and other methods.
The P1 and Pk antigens both share the same terminal carbohydrate structure and therefore the question arises whether the same enzyme is, or is not, synthesizing both antigens. Various theories exist but so far none has been proven correct. Recently Iwamura et al [15] proposed that the P1 synthase is identical to the Pk synthase and that the phenotypic difference may be caused by two polymorphisms, -160A>G and -551_-550insC, found in the 5'-upstream regulatory region. Unfortunately, these authors were not able to show any functional effects of these two polymorphisms, at least not when tested in a fibroblast cell line. As they commented themselves, a haematopoietic cell line, preferentially even an erythropoietic one, would have been the optimal choice for the challenge but was technically difficult. Iwamura et al. also demonstrated the presence of cryptic and intracellular P1 antigen in cells from individuals known to have the P2 red blood cell phenotype. Whether this surprising finding is method-dependent or may differ between populations is currently unclear.
Our data indicate that an individual's P1/P2 status is not due to the -551_-550insC;-160G haplotype, contrary to the previously published indication [15]. However, it is striking that all (except one P2k) samples with the P2 phenotype tested here have the same -551_-550insC;-160G haplotype as all ten P2 samples tested by Iwamura et al. This clearly argues that the genetic linkage between the A4GALT locus and P2 status must be relatively strong. The results shown in Figure 3 indicate that none of the 16 polymorphic markers investigated can predict the serologically determined P1/P2 status. It is also interesting to note the high frequency with which SNPs occur in the A4GALT gene, as opposed to the gene coding for the next glycosyltransferase in this biosynthetic pathway, the P synthase, in which we have found no variations at all. The only exceptions are rare mutations causing the P1k and P2k phenotypes [8,9]. Despite the apparent variability in the A4GALT gene, the current study suggests that a limited number of haplotypes, with the A-1 and B-2 haplotypes being the predominant ones in the Swedish population, may constitute genetic clusters within which further variation has arisen (as judged by the lack of homogeneity within each group, see Table 3), still without obvious correlation to the blood group phenotype.
However, all this does not differentiate between the one-gene theory and the possibility of a tightly coupled independent locus responsible for P1 antigen expression. Since there are no apparent α4GalT homologues in this genetic region this may imply that a hypothetical second closely linked gene would give rise to either a regulatory molecule modifying the acceptor specificity of the Pk-synthesizing α4GalT (analogous to lactose synthetase [16]) or a chaperone type of molecule to make a fraction of the α4GalT molecules more suitably located/positioned for P1 synthesis. The finding of intracellular P1 antigen in P2 individuals [15], would tend to support the latter possibility. Chaperones have been shown to be involved in processes related to glycosyltransferase action [17] but it is somewhat difficult (although not impossible) to imagine an α4GalT-specific chaperone to be the solution for this long-standing enigma.
Conclusion
The study of potential regulatory regions surrounding the Pk coding sequence revealed nine previously unreported polymorphisms but none of them correlated with the P1/P2 red blood cell phenotypes. Two polymorphisms, -551_-550insC and-160A>G, suggested to cause the P2 phenotype in Japanese individuals [15] were found in homozygous form also in P1 samples in this study and since the P2 is the null phenotype of this blood group system, it is therefore very unlikely that these mutations cause the P2 phenotype.
Methods
Blood samples and DNA preparation
Samples with the P1 (n = 58) and P2 (n = 20) phenotypes were chosen from our in-house panel of test erythrocytes. The majority of the donors are of Swedish origin but a few are of Asian or African descent. Three P1k, three P2k and 20 p samples, genetically characterized in our laboratory, were also included for screening purposes [8,9,12,18]. The erythrocyte phenotype was determined by standard serological techniques.
DNA was prepared from EDTA blood using a simple salting-out method for small volumes modified from Miller et al. [19], or Qiagen QIAmp Blood Extraction kit (Qiagen GmbH, Hilden, Germany). The DNA was dissolved in H2O at a concentration of 100 ng/μl.
Screening for the -551_-550insC and -160A>G polymorphisms by PCR-ASP
All oligonucleotide primers used in the study were synthesized by DNA Technology ApS (Aarhus, Denmark) and the sequences are shown in Table 4. PCR with allele-specific primers designed to detect the polymorphisms at -551_-550insC and -160A>T, described to cause the P2 phenotype [15] was performed. For all heterozygous samples double allele-specific amplification, -551_-550insC;-160G and -550T;160A, were performed. The primer combinations used are listed in Table 5 and the locations in Figure 4. Primers were mixed with 100 ng of genomic DNA, 2 nmol of each dNTP, 2% glycerol, 1% cresol red and 0.5 U of AmpliTaq Gold (Perkin Elmer/Roche Molecular Systems, Branchburg, NJ, USA) in 10 × PCR buffer with 15 mM MgCl2. The final reaction volume was 11 μl. Thermocycling was undertaken in GeneAmp PCR system 2400/2700 (Perkin-Elmer/Cetus, Norwalk, CT, USA) under PCR conditions described in Table 5.
Table 4 Oligonucleotide primers used in this study.
Primer name Nucleotide sequence (5'→3') Function
Pk-5'-(-1056)-F† ACAGCCTGTGATGGGAATGAC a, b, e
Pk-5'-(-740)-R† TTGAGTGCTGACGCCCATCC b
Pk-5'-(-834)-F TGGGCACCCATTGAGTGCCA b, e
Pk-5'-(-550)-R ACCTCGCCCCATCTTCACAC e
Pk-5'-(550T)-F GAACAAATTACCAATAGCAATATGT e
Pk-5'-(-550insC)-R CCTCGCCCCATCTTCACAGC e
Pk-5'-(550insC)-F AACAAATTACCAATAGCAATATGCT e
Pk-5'-(-477)-F GCGGCGTTAAGGATACAGCAA b
Pk-5'-(-418)-R CTGATCCCACCGCCTCCTG b
Pk-5'-(-235)-F GCGCTCCCTACCTGTTGGC b
Pk-5'-(-131)-F GGACCGGGACCCGCAGGG a, b
Pk-5'-(-160Gmis)-R CCCGGTCCCCAGAGCACC e
Pk-5'-(-160Amis)-R CCCGGTCCCCAGAGCACT e
Pk-5'-(-160G)-R CCGGTCCCCAGAGCCCTC e
Pk-5'-(-160A)-R CCCGGTCCCCAGAGCCCTT e
Pk-int1-35-R CGTCCCCCGCAACATCGGC b
Pk-int1-160-R GCACAAATGTCGCCTCCAGAA a, b
Pk-ex2-74_75insC-F AGGTCGGCTGCTGAGCCCA e
Pk-int2-R GGGTGCAACCTGATTGCTAAG e
Pk-109G-F TTCACGTTTTTCGTCTCCATCG e
Pk-109A-F TTCACGTTTTTCGTCTCCATCA e
Pk-987G-F CACGCGGTTCGAGGCCACG e
Pk-987A-F CACGCGGTTCGAGGCCACA e
Pk-1120-R GGAAGGGCGGCCCAGTGC e
Pk-1006-F CCAGGGCACTGCTGGCCC c, d
Pk-1253-F GGACAGTGTCCTGTCTCGAG d
Pk-1697G-R CCTGTCTGAGGGAAGGGGC e
Pk-1697A-R CCCTGTCTGAGGGAAGGGGT e
Pk-1791-R TTATTCTATTGATTATTCTCCTGTG d
Pk-1881-R CCCCGTCAGAAGAATGGAGC c, d
HGH-F TGCCTTCCCAACCATTCCCTTA f
HGH-R CCACTCACGGATTTCTGTTGTGTTTC f
JK-781-L-F GCATGCTGCCATAGGATCATTGC f
JK-943-L-R GAGCCAGGAGGTGGGTTTGCC f
MO-21 GGTGAGAGAAGGAGGGTGAG f
MO-31 CCAGCACCCCGGCCAGCA f
† F = Forward. R = Reverse. The underlined nucleotide is mismatched for better specificity.
a. used for amplification of the 5' regulatory region.
b. used for sequencing of the 5' regulatory region.
c. used for amplification of the 3'-UTR.
d. used for sequencing the 3'-UTR.
e. used for PCR-ASP.
f. used as internal control.
Table 5 Oligonucleotide primer combinations and PCR conditions (previously not described) used for PCR-ASP in this study.
Primer combination Specific primer (pmol) Control primer DMSO (%) Annealing temperature (°C) Extension time (sec)
HGH-F/R (pmol) JK-L-F/R (pmol) MO-21/31 (pmol)
Pk-5'-(-1056)-F 5 1 64 40
Pk-5'-(-550)-R 5
Pk-5'-(-1056)F 5 1 64 40
Pk-5'-(-550insC)-R 5
Pk-5'-(-834)-F 7.5 0.5 4 62 60
Pk-5'-(-160A-mis)-R 7.5
Pk-5'-(-834)-F 5 0.5 3 62 60
Pk-5'-(-160G-mis)-R 5
Pk-5'(-550insC)-F 10 0.75 64 40
Pk-5'-(-160G)-R 10 0.75
Pk-5'(-550T)-F 7.5 0.5 63 40
Pk-5'-(-160A)-R 7.5 0.5
Pk-109A-F or Pk-109G-F 6 0.4 2 65 90
Pk-1697A-R or Pk-1697G-R* 6
Pk-987A-F 5 0.4 66 60
Pk-1697A-R 5
Pk-987G-F 5 0.4 66 60
Pk-1697G-R 5
Pk-109A-F or Pk-109G-F 7.5 0.5 64 60
Pk-1120-R**
Denaturation was carried out at 96°C for 7 min followed by 35 cycles at 94°C for 30 s. The annealing time was 30 s.
* Four double allele-specific amplifications were performed.
** These combinations were used for allele-specific amplifications for sequencing purposes.
Figure 4 Schematic representation of the primers used to identify the polymorphisms at -551_-550insC and -160A>T, described to cause the P2 phenotype. The figure also shows combinations and positions for the primers used to detect a linkage phase of SNPs in the reading frame with a polymorphism in the 3'-UTR. For size reference, see Figure 2.
Amplification of the 5'-, 3'- and coding regions of the Pk gene for DNA sequencing
The coding region, exon 3, in the Pk gene was amplified with the primer pair Pk-(-140)-F and Pk-1120-R in the Expand High Fidelity PCR System (Roche Molecular Systems, Pleasanton, CA, USA) and sequenced as previously described [12].
The 5'-regulatory region of the Pk gene was amplified with primers Pk-5'-(-1056)-F and Pk-int1-160-R and Pk-5'-(-131)-F and Pk-int1-160-R. Amplification was performed in a reaction volume of 22 μL with four pmol of each primer, 2 nmol of each dNTP, 100 ng of genomic DNA, GC-rich enzyme mix (0.5 U per reaction), GC-rich resolution solution and buffer with a final MgCl2 concentration of 1.5 mM (GC-rich PCR System, Roche Diagnostics GmbH, Mannheim, Germany). Thermocycling was undertaken in GeneAmp PCR system 2400/2700 (Perkin-Elmer/Cetus): Initial denaturation at 96°C for 7 min followed by 10 cycles at 94°C for 30 s, 62°C for 30 s and 72°C for 1 min and then 25 cycles at 94°C for 30 s, 60°C for 30 s and 72°C for 1 min.
For the 3'-region of the Pk gene 5 pmol of primers Pk-(1006)-F and Pk-(1881)-R were mixed with 100 ng of genomic DNA, 2 nmol of each dNTP, 2% glycerol, 1% cresol red and 0.5 U of AmpliTaq Gold (Perkin Elmer/Roche Molecular Systems) in 10 × PCR buffer with 15 mM MgCl2. The final reaction volume was 11 μl. PCR was run at 96°C for 7 min followed by 35 cycles at 94°C for 30 s, 64°C for 30 s and 72°C for 1 min.
PCR products were excised from 3% agarose gels (Seakem, FMC Bioproducts, Rockland, ME, USA) stained with ethidium bromide (0.56 mg/l gel, Sigma Chemicals, St. Louis, MO, USA) following high-voltage electrophoresis and purified using Qiaquick gel extraction kit (Qiagen). The Big Dye Terminator Cycle Sequencing kit (Applied Biosystems, Foster City, CA, USA) and an ABI PRISM 310 Genetic Analyser (Applied Biosystems) were used for direct DNA sequencing with capillary electrophoresis and automated fluorescence-based detection according to the manufacturer's instructions. Besides the PCR primers, internal primers were used as sequencing primers, see Table 4. To avoid detection of artefacts, sequencing was performed on both strands and using independently obtained fragments.
Detection of an insertion in exon 2 and linkage of SNPs in the reading frame with a polymorphism in the 3'-UTR using PCR-ASP
PCR-ASP for a previously found insertion, 75dupC in exon 2 (then believed to be exon 1), was also performed as described [12]. PCR-ASP was performed to investigate if the polymorphisms at nt. 109, 987 and 1697 were present in an allele-specific pattern. The reaction mixtures comprised 100 ng of genomic DNA, 2 nmol of each dNTP, 2% glycerol, 1% cresol red and 0.5 U of AmpliTaq Gold (Perkin Elmer/Roche Molecular Systems) in 10 × PCR buffer with 15 mM MgCl2. The final reaction volume was 11 μl. Primer combinations and PCR conditions are described in Figure 4 and Table 5.
List of abbreviations
ASP, allele specific primer
bp, base pairs
ins, insertion
del, deletion
kb, kilo bases
PCR, polymerase chain reaction
SNP, single nucleotide polymorphism
UTR, untranslated region
ISBT, International Society of Blood Transfusion.
Authors' contributions
ÅH carried out the experimental studies and participated in the discussion and preparation of the manuscript. AC participated in the discussion and preparation of the manuscript. MLO contributed to the design and coordination of the study and participated in preparation of the manuscript. All authors read and approved the final manuscript.
Acknowledgements
Sources of support: This work was supported in part by the Swedish Research Council (project no. K2005-71X-14251), the Medical Faculty at Lund University, governmental ALF research grants to Lund University Hospital, the Inga & John Main Foundation for Medical Research and the Lund University Hospital Donation Funds.
Part of this study has previously been reported in abstract form at the 28th Congress of the International Society of Blood Transfusion held in Edinburgh, UK, on 11–15 July 2004 [20].
==== Refs
Daniels G Human Blood Groups 2002 2nd Oxford, UK, Blackwell Scientific
Daniels GL Fletcher A Garratty G Henry S Jorgensen J Judd WJ Levene C Lomas-Francis C Moulds JJ Moulds JM Moulds M Overbeeke M Reid ME Rouger P Scott M Sistonen P Smart E Tani Y Wendel S Zelinski T Blood group terminology 2004: from the International Society of Blood Transfusion committee on terminology for red cell surface antigens Vox Sang 2004 87 304 316 15585029 10.1111/j.1423-0410.2004.00564.x
Dunstan RA Status of major red cell blood group antigens on neutrophils, lymphocytes and monocytes Br J Haematol 1986 62 301 309 3511947
Dunstan RA Simpson MB Rosse WF Presence of P blood group antigens on human platelets Am J Clin Pathol 1985 83 731 735 4003342
Spitalnik PF Spitalnik SL The P blood group system: biochemical, serological, and clinical aspects Transfus Med Rev 1995 9 110 122 7795329
Steffensen R Carlier K Wiels J Levery SB Stroud M Cedergren B Nilsson SB Bennett EP Jersild C Clausen H Cloning and expression of the histo-blood group Pk UDP-galactose: Gal-beta1-4Glc-beta1-Cer alpha1,4-galactosyltransferase. Molecular genetic basis of the p phenotype J Biol Chem 2000 275 16723 16729 10747952 10.1074/jbc.M000728200
Okajima T Nakamura Y Uchikawa M Haslam DB Numata SI Furukawa K Urano T Furukawa K Expression cloning of human globoside synthase cDNAs. Identification of beta3Gal-T3 as UDP-N-acetylgalactosamine:globotriaosylceramide beta1,3-N-acetylgalactosaminyltransferase J Biol Chem 2000 275 40498 40503 10993897 10.1074/jbc.M006902200
Hellberg Å Poole J Olsson ML Molecular basis of the globoside-deficient Pk blood group phenotype. Identification of four inactivating mutations in the UDP-N- acetylgalactosamine:globotriaosylceramide 3-beta-N-acetylgalactosaminyltransferase gene J Biol Chem 2002 277 29455 29459 12023287 10.1074/jbc.M203047200
Hellberg Å Ringressi A Yahalom V Safwenberg J Reid ME Olsson ML Genetic heterogeneity at the glycosyltransferase loci underlying the GLOB blood group system and collection Br J Haematol 2004 125 528 536 15142124 10.1111/j.1365-2141.2004.04930.x
Furukawa K Iwamura K Uchikawa M Sojka BN Wiels J Okajima T Urano T Furukawa K Molecular basis for the p phenotype. Identification of distinct and multiple mutations in the alpha1,4-galactosyltransferase gene in Swedish and Japanese individuals J Biol Chem 2000 275 37752 37756 10993874 10.1074/jbc.C000625200
Koda Y Soejima M Sato H Maeda Y Kimura H Three-base deletion and one-base insertion of the alpha(1,4)galactosyltransferase gene responsible for the p phenotype Transfusion 2002 42 48 51 11896312 10.1046/j.1537-2995.2002.00014.x
Hellberg Å Steffensen R Yahalom V Sojka BN Heier HE Levene C Poole J Olsson ML Additional molecular bases of the clinically important p blood group phenotype Transfusion 2003 43 899 907 12823750 10.1046/j.1537-2995.2003.00425.x
Naiki M Marcus DM An immunochemical study of the human blood group P1, P, and PK glycosphingolipid antigens Biochemistry 1975 14 4837 4841 1182122 10.1021/bi00693a010
Graham HA Williams AN A genetic model for the inheritance of the P, P1 and Pk antigens Immunol Commun 1980 9 191 201 6772545
Iwamura K Furukawa K Uchikawa M Sojka BN Kojima Y Wiels J Shiku H Urano T Furukawa K The blood group P1 synthase gene is identical to the Gb3/CD77 synthase gene. A clue to the solution of the P1/P2/p puzzle J Biol Chem 2003 278 44429 44438 12888565 10.1074/jbc.M301609200
Hill RL Brew K Lactose synthetase Adv Enzymol Relat Areas Mol Biol 1975 43 411 490 812340
Ju T Cummings RD A unique molecular chaperone Cosmc required for activity of the mammalian core 1 beta 3-galactosyltransferase Proc Natl Acad Sci U S A 2002 99 16613 16618 12464682 10.1073/pnas.262438199
Hellberg Å McGrath C Reid ME Olsson ML A novel missense mutation in the A4GALT gene of Amish individuals with the p phenotype. Tranfusion 2003 43 99A
Miller SA Dykes DD Polesky HF A simple salting out procedure for extracting DNA from human nucleated cells Nucleic Acids Res 1988 16 1215 3344216
Hellberg Å Chester MA Olsson ML Sixteen polymorphic sites in the 5´/3´-UTR of the Pk gene do not correlate with P1/P2 phenotypes Vox Sanguinis 2004 87 P18.11
Cedergren B Population studies in northern Sweden. IV. Frequency of the blood type p Hereditas 1973 73 27 30 4805484
|
16212661
|
PMC1282566
|
CC BY
|
2021-01-04 16:30:39
|
no
|
BMC Genet. 2005 Oct 7; 6:49
|
utf-8
|
BMC Genet
| 2,005 |
10.1186/1471-2156-6-49
|
oa_comm
|
==== Front
Front ZoolFrontiers in Zoology1742-9994BioMed Central London 1742-9994-2-161624202210.1186/1742-9994-2-16ReviewThe importance of immune gene variability (MHC) in evolutionary ecology and conservation Sommer Simone [email protected] Animal Ecology & Conservation, Biocentre Grindel, University of Hamburg, Martin-Luther-King-Platz 3, D-20146 Hamburg, Germany2005 20 10 2005 2 16 16 9 3 2005 20 10 2005 Copyright © 2005 Sommer; licensee BioMed Central Ltd.2005Sommer; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Genetic studies have typically inferred the effects of human impact by documenting patterns of genetic differentiation and levels of genetic diversity among potentially isolated populations using selective neutral markers such as mitochondrial control region sequences, microsatellites or single nucleotide polymorphism (SNPs). However, evolutionary relevant and adaptive processes within and between populations can only be reflected by coding genes. In vertebrates, growing evidence suggests that genetic diversity is particularly important at the level of the major histocompatibility complex (MHC). MHC variants influence many important biological traits, including immune recognition, susceptibility to infectious and autoimmune diseases, individual odours, mating preferences, kin recognition, cooperation and pregnancy outcome. These diverse functions and characteristics place genes of the MHC among the best candidates for studies of mechanisms and significance of molecular adaptation in vertebrates. MHC variability is believed to be maintained by pathogen-driven selection, mediated either through heterozygote advantage or frequency-dependent selection. Up to now, most of our knowledge has derived from studies in humans or from model organisms under experimental, laboratory conditions. Empirical support for selective mechanisms in free-ranging animal populations in their natural environment is rare. In this review, I first introduce general information about the structure and function of MHC genes, as well as current hypotheses and concepts concerning the role of selection in the maintenance of MHC polymorphism. The evolutionary forces acting on the genetic diversity in coding and non-coding markers are compared. Then, I summarise empirical support for the functional importance of MHC variability in parasite resistance with emphasis on the evidence derived from free-ranging animal populations investigated in their natural habitat. Finally, I discuss the importance of adaptive genetic variability with respect to human impact and conservation, and implications for future studies.
==== Body
Introduction
Many natural populations are threatened not only by a dramatic reduction in total area of available habitat but also by increasing habitat fragmentation and degradation leading to declining population sizes and barriers to gene flow if exchange of individuals between subpopulations is restricted [1-3]. Small populations often suffer from reduction of genetic diversity due to genetic drift and inbreeding effects [4-6]. Negative effects such as increased rates of allelic loss, fixation of deleterious alleles and decreased average individual heterozygosity relative to the overall population were observed by both, theoretical and empirical studies [7,8]. The loss of genetic variation can lead to short-term reduction of fitness components such as survival, reproductive output, growth rates and to impaired ability to adapt to long-term changes in the environment [7,9-13]. An increasing number of studies indicates that host genetic diversity plays an important role in buffering populations against pathogens and widespread epidemics [6,14-20]. Study of the genetic effects of population fragmentation is therefore of central importance for conservation biology [21].
Genetic studies of wild animals often employ neutral markers such as mitochondrial d-loop DNA (mtDNA), microsatellites or single nucleotide polymorphism (SNPs) to estimate the amount of variation present in individuals and populations [22-24]. While these markers are very informative for phylogenetic reconstructions and population history (bottleneck effects), for molecular clocks, to examine dispersal patterns of individuals (gene flow) and to classify individuals by relatedness and paternity analyses [25-28], the variation at neutral loci cannot provide direct information on selective processes involving the interaction of individuals with their environment or on the capacity for future adaptive changes [29,30]. However, these are issues of particular relevance in evolutionary ecology and conservation [31,32]. In addition, recent research in a variety of taxa and situations has revealed that evolution often occurs on contemporary timescales, often within decades (summarised in [32]). In some cases, the time span between the separation of populations might even be too short to leave a signal at neutral loci so that differences between populations are only detectable at genes under selection [33], such as those of the highly variable major histocompatibility complex (MHC). Contrary to neutral markers, MHC variability reflects evolutionary relevant and adaptive processes within and between populations and is very suitable to investigate a wide range of open questions in evolutionary ecology and conservation. The comparison with neutral markers allows the construction of null hypotheses concerning the diversity at selectively relevant genes and conclusions on the relevance of MHC polymorphism. One might argue that many recent studies report that individual heterozygosity at apparently neutral microsatellite markers is correlated with key components of individual fitness such as survival [34], fecundity [35], disease resistance [14,36] and lifetime reproductive success [37]. However, null results are likely to be underrepresented in the literature because of publication bias in favour of significant correlations [38]. A recent review and meta-analysis of both published and unpublished studies of the association between neutral marker heterozygosity and traits or components of individual fitness reported that associations were common, yet typically weak [39]. A correlation between individual heterozygosity at neutral genetic markers and components of individual fitness can arise in different ways, with the effects of inbreeding depression due to a genome-wide reduction in genetic variability (including fitness-relevant loci) and linkage disequilibrium to loci under selection being the most likely explanations [[6,38,40], see also [41]].
In this review, I first introduce general information about the structure and function of MHC genes, as well as current hypotheses and concepts concerning the role of selection in the maintenance of MHC polymorphism. The evolutionary forces acting on the genetic diversity in coding and non-coding markers are compared. Then, I summarise empirical support for the functional importance of MHC variability in parasite resistance with emphasis on the evidence derived from free-ranging animal populations investigated in their natural habitat. Finally, I discuss the importance of adaptive genetic variability with respect to human impact and conservation, and implications for future studies.
Major histocompatibility complex (MHC): structure, function and selection mechanisms
Structure and function
The MHC consists of a group of closely linked genes that constitute the most important genetic component of the mammalian immune system [42]. The MHC encodes cell-surface glycoproteins that bind antigens derived from pathogens or parasites and present them to T-lymphocytes which trigger the appropriate immune response. Two major groups of MHC genes can be distinguished. MHC class I genes play an essential role in the immune defence against intracellular pathogens by binding peptides mainly derived from viral proteins and cancer infected cells. They are expressed on the surface of all nucleated somatic cells. In contrast, MHC class II genes are predominantly involved in monitoring the extracellular environment by presenting peptides mainly derived from parasites to the T-cells (e.g. bacteria, nematodes, cestodes) [43,44]. They are primarily expressed on antigen-presenting cells of the immune system, such as B cells and macrophages. Within class II genes, most research in mammals focuses on the second exon of DRB genes because these loci code for parts of the functionally important antigen binding sites (ABS) [45]. Alternatively, the β-chain in general is used if loci assignment is not possible due to missing information (e.g. in teleost, [46,47]). The class II region genes are closely linked in humans and all other mammals examined, and variants at these genes are generally in strong linkage equilibrium [48]. Thus, the pattern observed for DRB loci should be a good indicator of the genetic variation in other class II genes and even some other less closely linked genes in the MHC [49-51].
Genes within the MHC involved in antigen presentation constitute the most polymorphic loci known in vertebrates [52,53]. The variability of the MHC-molecules is correlated with the diversity of the T-lymphocyte receptors which in turn determine the disease and parasite resistance of an organism and thus may influence the long-term survival probability of populations [54-57]. The antigen binding sites show high levels of variation not only in the number of alleles but also in the extent of sequence variation between alleles [58]. Under neutrality theory, the rate of synonymous nucleotide substitution (ds) is predicted to be larger (ds > dn) than the rate of non-synonymous substitution (dn) because non-synonymous substitutions change the amino acid composition and are thereby likely to be deleterious [59,60]. However, several studies demonstrate that the ABS display more non-synonymous than synonymous substitutions (dn > ds) ([61,62], reviewed by [19]). This cannot be explained by a higher mutation rate in this specific region [58,61,62]. The emerging general view is that the determinant role in shaping patterns of nucleotide diversity in MHC genes is balancing selection [19,59,60,63]. Balancing selection results not only in the maintenance of large numbers of alleles in populations, but also in greatly enhanced persistence of allelic diversity over extremely long time periods relative to neutral genetic variation [64], an observation termed 'trans-species evolution of polymorphism' [42]. The subsequent alteration in ABS allows binding of a diverse array of antigens [61,62,65].
Selection mechanisms
Two main types of balancing selection ('heterozygote advantage hypothesis' and 'frequency-dependence selection') have been suggested as important in retaining high levels of genetic diversity at the MHC in humans and vertebrates (reviewed by [19,64,66-68]).
In evaluating the evolutionary potential of 'heterozygote advantage' mechanism [69] a clear distinction between 'dominance' (heterozygote advantage in a broad sense) and 'overdominance' (heterozygote superiority) is necessary. The term 'dominance' refers to heterozygotes that are as resistant as the most resistant homozygote (if the allele A is associated with resistance, then the genotype AB is as resistant as AA (AB = AA)). In this case the heterozygote advantage could be due to masking of susceptible alleles. Whereas there is some support for this selection mechanism among experimental infection studies using mainly congenic mice it is clearly not sufficient to maintain high MHC variability [68,70,71]. 'Overdominance' seems to be the more efficient 'heterozygote advantage' mechanism promoting MHC diversity. In this case, heterozygotes are expected to have higher fitness than either parental homozygotes especially if confronted with multiple species or strains of pathogens or parasites (the genotype AB has a higher fitness than AA (AB>AA) and BB (AB>BB) [72]). The assumption is based on the theoretical background that heterozygous individuals should be able to detect and present a wider range of pathogen-driven antigens due to a larger number of different MHC molecules, hence increasing the relative fitness of MHC heterozygotes compared with homozygotes [60,73]. Thereby, two different 'overdominance' models have been suggested: a) 'symmetric overdominance' or 'symmetric balancing selection' [74], whereby all heterozygotes derive a similar selective advantage to homozygotes (= all heterozygous are selectively equivalent), and b) 'divergent allele advantage' [75]. In the later it is speculated that heterozygotes carrying more divergent allelic sequences have a selective advantage relative to individuals carrying relatively similar alleles by presenting a broader spectrum of antigens to the immune system. To the best of my knowledge, the 'divergent allele advantage'-hypothesis has never been applied in infectious disease studies but to explain the persistence of highly divergent MHC alleles over millions of years [75,76]. Richman and colleagues [77] used a theoretical model to confirm Wakeland's contention that MHC alleles are more divergent than expected under a model of balanced genetic polymorphism assuming selective equivalence of different alleles (but see [78-80]). Application of this model to MHC class IIb gene sequence data of deer mice (Peromyscus maniculatus) provided more support for the 'divergent allele advantage' model than for the 'symmetric overdominance' model for the maintenance of MHC polymorphism [77]. Thereby it is important to note that the analysis are based on assumptions of the coalescent models in which no gene conversion is allowed and therefore conclusions should be taken with care if such a mutational process is suspected [19].
The second mechanism, 'frequency-dependent selection', occurs when an allele or genotype is favoured at one frequency, but disadvantaged at another frequency [73,81,82]. Host-parasite dynamics are considered as an coevolutionary arms race. Pathogens adapt to infect the most common host genotype, leaving rare genotypes least infected [83]. If alleles are favoured when they are rare, but selected against when they are common, a balanced polymorphism results. Thus, the 'frequency-dependent selection' hypothesis is also described as 'rare-allele advantage hypothesis', 'Red Queen hypothesis' or 'moving-target hypothesis' [84-87]. The hypothesis assumes the following details. Rare (e.g. new) MHC alleles that are more resistant to parasites cause an advantage to the host, spread through the population and become common. This increases selection on parasites to evade recognition by these common alleles. As the parasite antigenicity changes, the relative fitness of the common host genotypes decreases and provides a selective advantage to other rare alleles. The time-lag nature of these antagonistic coevolutionary responses could lead to a cycling of fitness values of different alleles/genotypes in both hosts and pathogens, and result in the maintenance of high genetic diversity. As a consequence of these processes, pathogen-driven selection varies over time and may differ among habitats/environments within the range of a species, such that one host MHC-allele is favoured at a certain time in one environment and selected against in another. This should lead to varying spatiotemporal selection directions in space and time ('diversifying selection in space and time') [88,89,91]. So far, only one study investigated variation in MHC frequencies over time in a natural population to test the assumptions of the frequency-dependent model. Westerdahl and colleagues [92] compared the temporal changes in allele frequencies of 23 class I alleles and 23 neutral microsatellites of Great reed warblers (Acrocephalus arundinaceus) in nine consecutive cohorts. The MHC alleles showed on average slightly higher variation in temporal fluctuations compared to the microsatellite alleles. The frequency of two specific class I alleles varied more between cohorts than expected from random, whereas none of the neutral markers showed fluctuations exceeding the expectation from stochastic variation. The authors suggested that the variation in MHC allele frequencies between cohorts is not a result of demographic events, but rather an effect of selection favouring different MHC alleles in different years. However, Westerdahl and colleagues [92] did not include investigations of parasites or pathogens dynamics for explaining this pattern.
In addition, reproductive mechanisms such as disassortative mating and maternal-foetal interactions have been suggested as alternative or complementary mechanisms maintaining MHC diversity (summarised by [68,87,93-97]). MHC dissimilar mating preferences might act to increase offspring heterozygosity ('good-genes as heterozygosity hypothesis' [98]), to provide offspring with a moving target of MHC alleles as protection against pathogens which rapidly adopt to the parental genotypes ('rare-allele advantage hypothesis', 'Red Queen hypothesis', 'moving-target hypotheses' [73,81,82,87], to avoid inbreeding or genetic incompatibility ('genetic compatibility hypothesis' [99]) or to achieve an optimal MHC diversity in offspring with respect to parasite resistance ('allele counting hypothesis' [46,47] but see [91]).
The actual cue used in MHC-based mate choice is thought to be based on odour which allows to distinguish MHC-identity (summarised by [87,100-102]). Peptides/MHC complexes that are not retained at the cell surface but instead are released into the extracellular space might appear in the urine and other body secretions and be used for interindividual communication [103,104]. In mammals, the vomeronasal organ is essential in odour-based social recognition by detecting pheromones and other chemosignals that carry information about gender, sexual and social status, dominance hierarchies, and individualities, but it has been difficult to define the molecular nature of these chemosignals. Recent studies provided evidence that MHC class I peptides serve as chemosensory signals in the vomeronasal organ by which individual MHC genotype diversity can be used as a relatedness marker and may influence social behaviour [105].
These diverse functions and characteristics place genes of the MHC among the best candidates for studies of mechanisms and significance of molecular adaptation in vertebrates [19,52,93].
Evolutionary forces acting on the genetic diversity in coding and non-coding markers
The maintenance of genetic variation in natural populations in neutral parts of the genome under the non-selective evolutionary forces such as genetic drift and inbreeding depend not only on the number of individuals constituting a population, but also on the particular life history, the dispersal patterns (gene flow) and the breeding system of the species under study [106,107]. In contrast, the ability of natural populations to maintain genetic variation in functional genes depends on the selection pressures involved. Balancing selection is thought to counteract the effects of genetic drift and to retard the rate of fixation of alleles [58].
Evidence for selection maintaining high MHC diversity despite restricted variability in non-coding markers
There is increasing evidence for high MHC diversity due to balancing selection in species with otherwise restricted diversity in non-coding markers. For example, the San Nicolas Island fox (Urocyon littoralis dickeyi) is the most monomorphic sexually reproducing animal population yet reported with respect to variation in neutral genetic markers. No variation has been discovered in supposedly neutral hypervariable microsatellite loci and multilocus fingerprints, for which the probability of genetic identity is commonly <1 in several millions. Such low levels of variation imply lower resistance to pathogens, reduced fitness, and problems in distinguishing kin from non-kin. However, high MHC diversity is probably still maintained in this population by balancing selection. It is assumed that periodic selection has rescued genetic variation at the MHC and, potentially other fitness-related genes ([108] but see also [90]). Another example was found in Hawaiian honeycreepers (Vestiaria coccinea) [109]. Natural selection has maintained variation within the MHC while mitochondrial d-loop sequences and cytochrome b sequences were invariant and allozymes revealed low variability probably due to a genetic bottleneck. Moreover, in fragmented Malagasy gray mouse lemur (Microcebus murinus) populations, the number of DRB-alleles and the gene diversity were still high [110] but microsatellite and mitochondrial marker showed very low levels of polymorphism [111]. In the same study area, also the introduced black rat (Rattus rattus) revealed a similar pattern of genetic polymorphism: high levels of variability in the functional important MHC DRB marker [Sommer, unpublished data] in contrast to low mitochondrial d-loop variability (five haplotypes) [112].
These studies indicate that until a threshold level, genetic variation at the MHC might persist due to balancing selection despite low levels of variability shown by neutral markers. The results support the importance of balancing selection as a mechanism to maintain variation in natural populations and expose the difficulty of using neutral markers as surrogates for variation in fitness-related loci [108].
Processes leading to low variability in both coding and non-coding markers
The maintenance of polymorphism within populations is dependent on the product of selection intensity, mutation rate and effective population size [58,113,114]. Under certain circumstances strength of selection acting on MHC loci can be insufficient to maintain variation in small or fragmented populations for a long period of time. The effects of balancing selection and genetic drift on the genetic diversity of coding MHC class II (DQA) variability, neutral mitochondrial control region and microsatellite marker were recently investigated in 14 island and two mainland populations of the Australian bush rat, Rattus fuscipes [115,116]. Both neutral marker sets revealed high levels of genetic variability over-all but clear signs of genetic drift such as little to no diversity in the small island populations and extreme differentiation between the populations. In the MHC, higher levels of heterozygosity were observed on two of the islands than would be expected under neutrality, but genetic drift played a dominant role in the majority of island populations leading to a decrease in the number of MHC alleles.
Similarly, historical events such as bottlenecks and founder effects but also constraints of the mating system can be reflected in low numbers of MHC alleles (for example in an Asian lion population (Panthera leo persica) [117]; cheetahs (Aconyx jubatus) [118]; Malagasy giant jumping rats (Hypgeomys antimena) [119-121]; Malagasy western forest mouse (Macrotarsomys bastardi) [120]; common hamsters in the Netherlands (Cricetus cricetus) [122]; Scandinavian beavers (Castor fiber) [123]; Swedish moose (Alces alces) [124,125]; musk ox (Ovibos moschatus) [126]; Spanish ibex (Capra pyrenaica) [127]; island population of desert bighorn sheep (Ovis canadensis mexicana) [128]; Arabian oryx (Oryx leucoryx) [129]; South African bontebok (Damaliscus pygargus pygargus) [130]; Przewalski's horses (Equus przewalskii) [55]; Northern elephant seals (Mirounga angustirostris) [131], fin whales (Balaenoptera physlaus] [132], sei whales (Balaenoptera borealis) [132], and black robins (Petroica traversi) [133]). Under these circumstances, the power of genetic drift has been stronger than the power of selection. As predicted by theoretical models [135], the reduced MHC polymorphism is usually correlated with low genome-wide genetic variation [89]. For example, cheetahs (Aconyx jubatus) show low MHC diversity, which correlates with a genome-wide loss of diversity presumably due to a genetic bottleneck about 10,000 years ago [118]. Also Northern elephant seals (Mirounga angustirostris) which were hunted near to extinction in the 19th century lost most of the variability in allozymes, mitochondrial DNA, mini- and microsatellite loci and MHC class II loci [131,135].
Empirical support for the functional importance of MHC variability in pathogen and parasite resistance
Evidence for the functional importance of MHC variability and selective mechanisms derived from studies in humans or under experimental, laboratory conditions
While predictions of an association between MHC diversity and disease resistance are straightforward extensions of MHC theory, up to now, most of the empirical evidence has been derived from studies in humans or under experimental/laboratory conditions [19,67,136].
'MHC heterozygote advantage' [69] was indicated in humans by a slower progression to AIDS after HIV infection [137] and in a more effective clearance of hepatitis B viral infections [138]. In laboratory experiments, MHC-heterozygous mice showed reduced pathogenicity during bacterial and viral infection (streptococcus-induced lesions [139], Salmonella, Lysteria [70], Salmonella enterica, Theiler's virus [140]), an increased T-cell mediated immunity during lymphocytic choriomeningitis (LCM) infection [69] and they had a faster clearance rate of parasitic worms (Heligmosomoides polygyrus [141], Schistosoma mansoni [142]), than the average homozygote. Tumor incidence was lower and regression faster in heterozygous, rous sarcoma virus (RSV) infected chicken (Gallus domestica [143]). MHC class IIB heterozygotes had an increased survival rate in captive-raised fish, e.g. in Chinook salmon (Oncorhynchus tsawytscha) infected with a haematopoietic necrosis virus (HNV) [144] and in fluke-infected (Gyrodactilus turnbulli) Gila topminnows (Poeciliopis o. occidentalis) [54].
The 'frequency-dependent selection hypothesis' [81,82] is engaged by both mathematical models [73,145] and some empirical studies that show correlations between certain alleles and disease resistance in humans (e.g. malaria [146], Epstein-Barr-virus [147], hepatitis B [148], leprosy, tuberculosis [67], Heliobacter-infected gastric cancer [149]). In humans, a correlation was observed between some MHC class II haplotypes and the clinical severity of cestode infections (Echinococcus multilocularis) [150]. Certain MHC alleles also played a role in resistance/susceptibility to a fungal disease (Cryptococcus neoformans [151]), infections with gastrointestinal nematodes in lab mice (Trichinella spiralis [152,153], Nematospiroides dubius [154], Trichuris muris [155]) and in straightbred Scottish Blackface sheep (Ostertagia circumcincta [156,157]). Associations between resistance and MHC genotype was found in chicken suffering from infection with Marek's disease (a tumour disease caused by a herpes virus [158]). Experimental evidence for MHC-allele-specific resistance to Aeromonas salmonicida bacteria [57,159] and to the infectious salmon anaemia virus (ISAV) was found in captive-raised Atlantic salmon (Salmo salar [160]).
Evidence for the functional importance of MHC variability and selective mechanisms derived from studies in free-ranging animal populations in their natural environment
Whereas studies carried out under experimental or laboratory conditions can be better standardised to account for different parameters (e.g. in inbred congenic mice), they do not provide sufficient information to evaluate the ubiquity of pathogen-driven selective mechanisms acting in free-ranging animal populations in their natural habitat. Doing MHC research in wild vertebrates allows to test whether the results of studies on inbred congenic lab strains will hold in animals with a more diverse genetic background. Further, laboratory studies cannot reveal the effects of conditionally advantageous or deleterious alleles which will be discovered only in the presence of natural stress, such as spatially and temporally changes in climate, food availability, competition, and associated levels of parasitism [18,161]. Predicting the evolutionary potential of wild host populations in response to parasites requires at least a minimal understanding of the genetic basis for host resistance and heritability under field conditions, and the strength and mode of parasite-mediated selection [162]. Few studies have attempted to test for an association between MHC polymorphism and parasite resistance in wild populations under natural conditions [19]. Available information is summarised in Table 1.
Table 1 Evidence for pathogen-driven selection mechanisms in free-ranging vertebrate populations investigated in their natural environment.
Host species Host environment Country Infectious agent Heterozygote advantage Negative frequency-dependent selection Reference
Three-spined stickleback
(Gasterosteus aculeatus) Lakes and rivers Germany 14 species of macroparasites Supported in terms of a general diversity advantage; minimal parasitation at intermediate MHC class IIB diversity; population exposed to more diverse parasites had more different alleles. Not investigated [47]
Soay sheep
(Ovis aries) Large unmanaged population on an island Scotland Strongyle nematode Not supported; heterozygosity is not the critical factor determining mortality in lambs and yearlings. Common alleles (OLADRB 205, OLADRB 257) were associated with decreased lamb or yearling survivorship and a high incidence of parasitism; the rarer allele (OLADRB 263) with increased yearling survival. [56]
Gray mouse lemur
(Microcebus murinus) Littoral rain forest Madagascar Seventeen nematode species; separate data analysis for (most common) single and multiple infections. Not supported; heterozygosity was uncorrelated with infection status (being infected or not), the number of different nematodes per individual (NNI) as well as with the faecal egg counts (FEC, eggs/g faeces). The common allele Mimu-DRB*1 was more frequently found in infected individuals, in individuals with high number of different nematode species infections (NNI) and faecal egg counts (FEC); the rare alleles Mimu-DRB*6 and 10 were more prevalent in not infected individuals and in individuals with low NNI and FEC values. [174]
Yellow-necked mouse
(Apodemus flavicollis) Tree-dominated habitat Germany Eight nematode species; separate data analysis for (most common) single and multiple infections. Not supported; heterozygosity did neither influence the infection status (being infected or not), nor the number of different nematode infections (NNI) nor the individual faecal egg count (FEC, eggs/g faeces) values. Mice carrying allele Apfl-DRB*5 or the closely related allele Apfl-DRB*15 had an increased risk of being nematode infected and displayed higher FEC than individuals carrying other alleles; the allele Apfl-DRB*23 was associated with low FEC in separate analyses of the most common nematode. [173]
Hairy-footed gerbil
(Gerbillurus paeba) Dunefield of the Southern Kalahari Desert South Africa Two different cestode species, six different nematode species Not investigated Gepa-DRB*15 was only found in not infected mice. [172]
Striped mouse
(Rhabdomys pumilio) Dunefield of the Southern Kalahari Desert South Africa Eight different nematode species Supported; heterozygosity did influence the infection status (being infected or not) and the individual faecal egg count (FEC) value with higher values observed in homozygous individuals. The allele Rhpu-DRB*1 occurred more frequently in infected individuals and in individuals with high FEC values (high parasite load). In contrary, the allele Rhpu-DRB*8 occurred more often in individuals with low FEC values. [163]
Under field conditions, associations between MHC heterozygosity and resistance/susceptibility to parasite infections have only been found in the African striped mouse (Rhabdomys pumilio [163]) and in three-spined sticklebacks (Gasterosteus aculeatus [47,164]) which seem to possess up to six MHC class II loci. In the later, a modification of a simple heterozygote advantage was identified as within individual fish, intermediate, rather than maximal allele numbers were associated with minimal parasite load [47,164]. This is explained by the fact that MHC-genes are involved in the preservation of T-cells during thymic selection. At some point, increasing the number of MHC molecules expressed should cause a net loss of T-cells and therefore negatively affect the organism [165] (but see also Borghans and colleagues [166] who used a simulation approach which revealed that several hundred alleles would be required to cause such a net loss of T-cells). Different allele numbers can be produced by both heterozygosity at single loci and differences in MHC class II gene duplication numbers across haplotypes [167]. At the moment it is not clear whether or not this selection pattern of intermediate, rather than maximal allele numbers is confined to species with a relatively flexible genomic architecture such as sticklebacks and other teleosts with haplotype variation in their MHC locus duplication numbers, or whether it represents a more general feature that has been overlooked in previous studies [168]. In mammals, a flexible MHC genomic architecture, namely the appearance of multiple MHC class II DRB loci with variable loci numbers between individuals has been described in rhesus macaques (Macaca mulatta [169]) and in California sea lions (Zalophus californicus [170]). The later possess up to eight different DRB loci in variable configurations among individuals but with low levels of allelic variation per loci. Preliminary evidence suggested an association between a certain MHC genotype and urogenital cancer. In contrary to sticklebacks, no relationship between the total number of unique DRB genes and the presence of cancer has been identified [171]. A possible relationship between the number of MHC alleles and parasitic load in mammals was also investigated in hairy-footed gerbils (Gerbillurus paeba) which possess two functional DRB loci [172]. Here, individuals carrying three different MHC alleles had significantly higher faecal egg count values than individuals with four alleles [172]. This is in accord with the theoretical background which assumes that animals containing more MHC alleles than others should be able to recognise a larger spectrum of pathogen-derived antigens and consequently be infected by less parasite species and/or to be generally less intensively infected [69].
An association between certain MHC alleles and disease resistance or susceptibility was found in a free-ranging sheep population (Soay sheep, Ovis aries) where MHC variants appear to play a major role in protection against strongyle nematode invasion, the most prevalent gastrointestinal parasite found [56]. As expected by the assumptions of the 'negative frequency-dependent selection' ('rare-allele advantage hypothesis', 'Red Queen hypothesis', 'moving-target hypotheses') [73,81,82], the most common alleles OLADRB 205 and 257 (allele frequencies: 0.21–0.24) were associated with decreased lamb or yearling survivorship, whereas the rarer OLADRB 263 allele (allele frequency: 0.13) was associated with increased yearling survival (Table 1). Further evidence for the importance of certain MHC alleles and resistance or susceptibility to helminths was revealed in a common European rodent (yellow-necked mouse, Apodemus flavicollis [173]), in the two African rodent species mentioned before (Gerbillurus paeba [172], Rhabdomys pumilio [163]), and in a primate species (gray mouse lemur, Microcebus murinus [174]) (Table 1). Also in R. pumilio, it was the most common allele Rhpu-DRB*1 (allele frequency: 0.22) which occurred more frequently in infected individuals and in individuals with high faecal egg count values (indicating high parasite load) whereas the rare allele Rhpu-DRB*8 (allele frequency: 0.05) occurred more often in individuals with low FEC values (indicating low parasite load). Also in M. murinus, the common allele Mimu-DRB*1 (allele frequency: 0.33) was more frequently found in infected individuals and in individuals with a high number of different nematode species infections and faecal egg count values (eggs/g faeces) (indicating high parasite load), the rarer alleles Mimu-DRB*6 and *10 (allele frequencies: 0.11 and 0.06) were more prevalent in not infected individuals, in individuals with low number of different nematode species infections and faecal egg count values (indicating low parasite load). These examples demonstrate the frequency-dependence of selection between parasites and hosts in the form of a rare allele advantage in the host population.
Evaluating the relative importance of balancing selective mechanisms
Right now there is still much debate whether 'heterozygote advantage' or 'frequency dependent selection hypothesis' is most important for balancing selection [89]. Most studies investigating 'heterozygote advantage' compared the infectious disease outcomes of heterozygotes at a given MHC loci, as a group, to the outcomes of homozygotes at the same locus, as a group ('population heterozygote advantage' [70,175], examples see above) probably always due to restrictions in sample size. However, comparing the average performance of all heterozygotes against homozygotes, instead of using allele specific tests for 'overdominance', can not distinguish whether the observed advantage is due to 'dominance' or 'overdominance'. Grouping all homozygotes and all heterozygotes, respectively, circumvent tests of the original hypothesis namely the superiority of heterozygotes over either corresponding homozygote [69] (see paragraph 'Selection mechanisms' above) as this hypothesis is conditional on the alleles involved (and should be more precisely termed 'allele-specific overdominance' [175]). However, a theoretical model showed that under a very wide range of assumptions about the relationship between homozygote and heterozygote infectious risk, 'allele-specific overdominance' might be consistent with 'population heterozygote advantage', e.g. a 'population heterozygote advantage' might occur when the diversity of resistant alleles is sufficiently high and the diversity of susceptible alleles is sufficiently low [175]. But also the opposite might be true. Because of confounding effects of differences in frequencies of susceptible or resistant alleles, population level tests can, in a worse case, find a heterozygote advantage even when every heterozygote is at greater infection risk than either corresponding homozygote in allele-specific analyses [175]. Direct estimates of the allele-specific effects of heterozygosity relative to the corresponding homozygotes are rare. The most convincing experimental evidence for heterozygote advantage through 'allele-specific overdominance' derived from McClelland and colleagues [140] using co-infections with multiple pathogens in MHC-congenic mice with reciprocal resistance/susceptibility profiles (but the authors did not test for fitness consequences). In humans, the only studies that directly compare the outcomes of heterozygotes to those of homozygotes for the same alleles derived from investigations of autoimmune but not from infectious diseases (summarised by [70,175]).
As mentioned above allele-specific analyses were most often impossible due to restrictions in sample size. In humans, recently a new approach to circumvent this problem was proposed by classifying alleles to supertypes based on shared binding motifs [176,177]. Though it is clear that the highly polymorphic HLA genes play a crucial role in the immune response, their great diversity is a major obstacle in distinguishing HLA allele-specific effects and complicates the attribution of specific alleles with the outcome of diseases. Collecting samples of the size needed for definitive results is often not feasible. The biological relevance of a classification scheme based on functional binding specificities is supported by a growing body of evidence of cross-presentation of specific peptide-binding motifs by different HLA molecules. Trachtenberg and colleagues [176] investigated the usefulness of grouping HLA alleles to supertypes by their overlapping peptide binding specificities in explaining the association between HLA polymorphism and HIV disease progression. Their study indicated that HLA-supertypes are highly predictive of viral load. Consistent with the rare-allele advantage model the authors could show the advantage of a rare HLA supertype in progression of HIV [176]. After more intense studies of the implications and limits of supertypes in large human data sets, this classification approach of common functional traits may also provide tools for the MHC research in natural endangered populations, where high allelic diversity causes problems in obtaining sufficiently large statistical sample sizes. Caution must be taken not to miss the effects of new, rare alleles by clustering alleles in functional types.
As outlined before, two different 'overdominance' models have been suggested: a) 'symmetric overdominance' or 'symmetric balancing selection' [74], whereby all heterozygotes derive a similar selective advantage to homozygotes (= all heterozygous are selectively equivalent), and b) 'divergent allele advantage' [75]. As almost all heterozygote advantage studies were carried out on the population level, so far, no effort has been made to differentiate between these two 'overdominance' models in infectious disease studies. However, the 'divergent allele advantage' hypothesis has recently been considered in MHC-dependent mate choice studies [121,178]. In the African striped mouse (Rhabdomys pumilio) where associations between MHC heterozygosity and resistance/susceptibility to parasite infections have been found [163], no significant correlation between pairwise genetic distances of heterozygotes (a measurement for allelic divergence) and infection rates (faecal egg count: log-transformed number of eggs per gram faeces) was found (Froeschke & Sommer, unpublished data). Another point that need to be mentioned is that only a few studies in natural populations indicating correlative evidence for 'heterozygote advantage' combined MHC research with estimates of genome-wide diversity by using neutral markers and thus could rule out possible effects of genome-wide heterosis [47,137].
With respect to the 'frequency dependent selection hypothesis', evidence for pathogen-resistant/susceptible alleles/haplotypes is not equally available. So far, more alleles/haplotypes were found to be associated with susceptibility to disease [67] (but see Table 1). This bias could be simply due to over presentation of human studies, in which the emphasis has been on finding disease-allele associations [179]. But it is also in line with theoretical predictions of host-parasite coevolution. Susceptibility is expected to be more common, because fast evolution of the parasite is assumed to fuel the arms race between them and their hosts. For most pathogens it is valid to assume a higher evolutionary potential compared to that of the host, because generation times are usually much shorter or effective population sizes of pathogen populations are larger [18]. The human HLA-A11 allele, for example, confers resistance to infection with Epstein-Barr-virus only in populations where the allele is rare. In populations with high frequency of this allele, virus strains have fixed a mutation that prevents presentation of immunodominant epitopes by HLA-A11 molecules [147]. Also certain HLA alleles are associated with a slower progression of HIV if they are rare and have no advantage if they are common (summarised by [176]). This shows that a fast-evolving pathogen is able to adapt to host defence.
Evaluating the relative importance of both balancing selective mechanisms, so far, more evidence is available for the importance of specific MHC-alleles in parasite resistance or susceptibility. It is conceivable that a rare allele may have a high fitness and at the same time a constant advantage for heterozygotes. Both hypotheses may be in accord with each other and are not mutually exclusive. But as most studies deal with single viral, bacterial or parasitic agents it was suggested that studies combining two or more pathogens may increase the amount of evidence for heterozygote advantage [57,136,140] (but see [173,174]). De Boer and colleagues [180] studied the degree of MHC polymorphism arising when 'heterozygote advantage' is the only selection pressure by using mathematical models. The simulations revealed that 'heterozygote advantage' on its own is not sufficient to explain the high population diversity of the MHC. This would require that the fitness contributions of all alleles would be unrealistically similar. 'Heterozygote advantage' in pathogen resistance could, however, promote mating preferences for MHC-dissimilarity, which in turn drive high allelic diversity [52,68,70]. This could explain why MHC-heterozygous males have attributes important during sexual selection such as an increased antler development and body mass in white deer [181] and sexually attractive odour in stickleback fish [46]. In contrast, a study on sexual selection in pheasants (Phasianus colchicus) found that females prefer males with larger spurs, and that this sexually selected trait is associated with a particular MHC allele [182,183]. The overall view is emerging that although 'heterozygote advantage' is clearly an important selection pressure additional frequency-dependent selection pressure is required. A theoretical model by Hedrick [89] indicated that the selective force from pathogens, which vary in space and time ('diversifying selection in space and time'), could maintain the genetic polymorphism in MHC genes. Since evolving pathogens mainly evade presentation by the most common MHC alleles in the host population, they provide a selective pressure for a large variety of rare alleles. Host-parasite coevolution would be sufficient to explain the large degree of MHC polymorphism [145].
In ongoing studies assessing the evolutionary genetic details of vertebrate host-parasite relationships and evidence for frequency-dependent parasite-driven selection four patterns ought to be evident: (1) parasitism reduces host fitness, (2) MHC alleles differ in susceptibility, (3) alleles frequencies change according to (2), and (4) in the longer term dynamics should encompass frequency-dependent allele frequency fluctuations [17]. Whereas (1) and (2) indicate the potential for selection and have been shown in recent studies in wildlife populations (e.g. [184], this review), so far evidence for (3) and (4) is limited. Ongoing investigations of the parasite-driven selection mechanisms under natural conditions should focus on temporal variation of pathogens, host fitness attributes and allele frequencies to test whether allele frequencies change accordingly in a cycling pattern.
Functional differences of amino acid variation in the antigen binding sites
There is increasing evidence that pathogen escape from MHC-dependent immune system recognition may involve changes in only a few amino acids so that small binding-motif differences can lead to large differences in protection. Common mechanisms include changes in pathogen antigens (epitopes) that prevent binding (1) to the MHC-encoded cell surface glycoprotein or (2) to the T-cell receptor. (3) A third mechanism is molecular mimicry of host proteins that prevent T-cell receptor binding (T-cells that recognise host proteins are destroyed during thymic selection). For instance, a one-amino-acid difference in the antigen-binding region of the DRB*1302 allele abrogates its protection to malaria (summarised by [185,186]). In Malagasy mouse lemurs (Microcebus murinus), MHC-alleles associated with gastrointestinal nematode susceptibility (Mimu-DRB*1, *6 and *10) have unique amino acid motifs in the antigen binding sites (ABS) [174]. Mimu-DRB*1 associated with high parasite load differs from all other alleles by three unique amino acids, all of them located within the functional important ABS (aspartic acid in position 70, glutamic acid in position 71, lysine in position 74). Two of these ABS are mutated in Mimu-DRB*6 and *10 currently associated with low parasite load: the allele Mimu-DRB*6 has a unique motif at position 74 (glycine) and Mimu-DRB*10 at position 71 (methionine). In addition, only Mimu-DRB*6 and *10 possess the amino acid arginine located next to the ABS in position 78 [174] (position numbers after [62]). This indicates the functional differences of certain amino acids in the ABS and thus the influence of different amino acid compositions on parasite resistance.
So far, the molecular details of the interactions between helminth parasites and the intestinal components of the immune system are not as well understood as for viral or bacterial infections. However, huge progress was made in understanding the cellular and molecular mechanisms in the immune regulation by gastrointestinal helminth parasites in recent years. The recognition of gastrointestinal parasites and their antigens, and the initiation of the immune response occur in specialised lymph nodes in the epithelium of the gut wall, the so called Peyer's patches. In these Peyer's patches all cell types necessary for antigen presentation to CD4+ T-cells including MHC class II molecules are present. This activates a range of interacting processes against the parasite culminating in an inflammatory reaction in the intestinal mucosa and different effector mechanisms against the invading parasite (summarised in [186-191]).
Importance of MHC variability in conservation
Importance of adaptive genetic variability with respect to human impact
Human impact (e.g. habitat fragmentation, degradation, isolation, urbanisation, pollution) has diverse impacts on the ecology and genetics of both, vertebrate and parasite populations. It often causes a loss of genetic variation leading to short-term reduction of fitness components, and to an impaired ability to adapt to changing environments which in turn influences evolutionary outcomes [5,6,12,18,192]. Habitat degradation and climatic conditions are also crucial parameters in terms of distribution, transmission and developmental success of parasites and pathogens [18,192,193]. Such changes may have significant implications for outbreak patterns of pest species, the conservation of rare mammal species and their ecological functions, as well as associated veterinary and medical consequences for wildlife, lifestock and humans [194]. Rapid evolution (on the order of decades or shorter) has been supported by numerous examples from host-parasite systems, and it is now clear that pathogens can cause major shifts in the genetic composition of their hosts on short timescales [18,162,195]. Detectable changes in allele frequencies can occur between subsequent generations and can be a sensitive indicator for demographic changes in some species [196].
The effects of pollution on the MHC was investigated in the estuary killifish (Fundulus heteroclitus) [33]. Populations experiencing strong differences in antigenic challenges (PCB-contaminated versus unpolluted site) show significant differences in amino acid substitution patterns in a highly variable MHC class II B locus. However, whether MHC population profile differences represent direct effects of chemical toxicants or parasite-mediated selection need to be investigated [33]. The only study including an environmental variable such as habitat fragmentation in the analysis of associations of MHC-constitution and parasite burden was carried out in a subdivided mouse lemur population [174]. The work indicated that variation in MHC-allele frequencies in the fragments were linked to parasite load as certain alleles which differed in a few amino acids in the ABS from other alleles (see above) were associated to parasite resistance or susceptibility. Female mouse lemurs inhabiting the fragment with the highest parasite load had a lower fat deposition in the tail (important during the dry season) and therefore lower survival rates than populations of the three other fragments. In addition, the population size declined dramatically in recent years [197]. However, to clearly separate whether the higher parasite load in the respective fragment is due to the MHC-constitution of individuals inhabiting this fragment or due to other ecological factors associated with fragment size or degradation needs further investigations [174]. Nevertheless, the study suggests that the MHC-constitution might influence the long-term survival of small fragmented animal populations and indicates the functional importance of maintenance of MHC variability in declining or fragmented animal populations.
More studies in free ranging animal populations with respect to human impact are needed to allow more general conclusions on the importance of adaptive genetic variability in conservation. According to the theoretical background, temporal and spatial variation in the parasitic fauna will cause shifts of selective advantage of certain MHC-alleles in different areas changing over time. This should result in habitat- and climate-specific amino acid substitution patterns in the functional important ABS in relation to local pathogen-driven selective pressures. So far, empirical evidence for 'diversifying selection in space and time' is limited. Studies on the interaction between environmental conditions and the expression of genetic covariation (the so called genotype-environment interaction) might be an important avenue for future work. Genotype-environment interactions have commonly been found in live history traits when multiple environments were considered reflecting the fact that genes influencing a trait in one environment may not be important in a different one [198]. In this context, host and parasite movement among habitat fragments could be crucial to both parasite persistence, and the spread and maintenance of resistance alleles and thus to allow ongoing coevolutionary processes. The role of metapopulation dynamics in maintaining the diversity of host resistance genes can be a matter of concern in conservation genetics aiming at the preservation of both current patterns and ongoing processes. As contemporary evolution is influenced by complex interactions among population size, genetic variation, strength of selection, and gene flow, the overall goal in conservation genetics – maintenance of short-term local adaptations and preservation of long-term adaptive potential – might be a challenging task [32].
Relevance of MHC polymorphism for individual fitness and long-term persistence
Genetic variation at MHC loci is thought to be important for resistance against pathogens, thereby increasing individual fitness and thus the long-term survival of endangered species [60,73]. Several studies have reported decreased pathogen resistance among MHC homozygotes, or an increase in pathogen susceptibility in inbred individuals in general. However, a direct link between pathogen-mediated population decline and low MHC variation has been difficult to demonstrate in natural populations [49]. Recent studies indicated that although MHC allele numbers are low in many bottlenecked species most of them still indicate a high degree of divergence between alleles. Table 2 summarises the number of functional important MHC class II DRB exon 2 alleles and sequence diversity in some free-ranging vertebrate populations investigated in their natural habitat. The comparison indicates that also species with a low number of different MHC alleles, such as the critically endangered Malagasy Giant Jumping Rat (Hypogeomys antimena, 5 alleles) whose geographic range was recently restricted to less than 20,000 ha, still have high levels of nucleotide and amino acid divergence between MHC DRB-alleles while mitochondrial d-loop sequences revealed very low variability [199-201]. A similar picture was revealed in the Przewalski's horse (Equus przewalskii, 6 alleles [55]), in the Arabian oryx (Oryx leucoryx, 3 alleles [129]) and in the South African bontebok (Damaliscus pygargus pygargus, 6 alleles [130]) (Table 2). Considering the nonABS, the ratio between non-synonymous (dn) and synonymous (ds) substitutions was significantly smaller (dn < ds) than unity in some species (Fig. 1) which indicates purifying selection acting on these codons depending on their respective function [202]. In contrary, comparisons of non-synonymous (dn) and synonymous (ds) substitution rates in parts coding for the functional important ABS revealed a significantly higher rate of substitutions (dn > ds) which change the amino acid constitution in the ABS and thus increase the divergence between alleles in all species irrespective of the number of MHC alleles still present (Fig. 1).
Table 2 Number of MHC class II DRB exon 2 alleles (ca. 200 bp) and sequence diversity in free-ranging vertebrate populations investigated in their natural environment. In addition, for comparison of variability levels of species with a low number of MHC alleles two studies of captive-bred populations are included. * DRB-locus is duplicated. N = sample size.
Species Order Country Nr of alleles (N) Nr (%) of variable nucleotide positions Nr (%) of variable amino acid positions Nr (%) of amino acid differences between alleles References
Microcebus murinus primates Madagascar 14 (228) 71 (41.5) 31 (54.4) 5 (8.8) – 25 (43.9) [110,174]
Microcebus berthae primates Madagascar 9 (42) 46 (26.9) 24 (42.1) 3 (5.3) – 19 (33.3) Sommer et al., unpublished data
Apodemus sylvaticus rodentia Germany 38 (119) 71 (32.7) 38 (52.8) 2 (2.7) – 28 (38.8) [206]
Apodemus flavicollis rodentia Germany 27 (146) 49 (22.6) 28 (38.9) 1 (1.4) – 21 (29.2) [173]
Leopoldamys sabanus rodentia Borneo 28* (49) 85 (49.7) 39 (68.4) 4 (7.0) – 25 (43.9) Lenz et al., unpublished data
Gerbillurus paeba rodentia South Africa 34* (40) 68 (39.8) 33 (57.9) 1 (1.8) – 19 (33.3) [172]
Rhabdomys pumilio rodentia South Africa 20 (58) 43 (25.1) 23 (40.4) 1 (1.8) – 14 (24.6) [163]
Rattus rattus rodentia Madagascar 13 (58) 72 (40.7) 33 (55.9) 3 (5.1) – 26 (44.1) Sommer, unpublished data
Hypogeomys antimena rodentia Madagascar 5 (229) 37 (17.1) 19 (26.4) 6 (8.3) – 21 (29.2) [201]
Equus przewalskii perissodactyla captive-bred 6* (14) 52 (20.8) 29 (34.9) 1 (1.2) – 22 (26.5) [55]
Oryx leucoryx artiodactyla captive-bred 3 (57) 35 (14.8) 21 (26.9) 13 (16.7) – 17 (21.8) [129]
Damaliscus pygargus pygargus artiodactyla South Africa 6 (45) 21 (8.4) 14 (16.9) 1 (1.2) – 13 (15.7) [130]
Figure 1 Differences between the rates of non-synonymous (dn) and synonymous (ds) substitutions (= dn-ds, amino acid sequence-changing substitution rate) in antigen-binding sites (ABS, shaded bars) and nonantigen-binding sites (nonABS, black bars) of MHC class II DRB exon 2 alleles of the species included in Table 2. Abbreviations follow the MHC nomenclature [42], the first two letters of the genus and the species name are used.
The studies indicate that selection processes are able to maintain MHC polymorphism also under unfavourable conditions at least for a certain time which in turn might suggest that levels of variability in species with low numbers of MHC alleles might be sufficient to prevent immediate pathogen-mediated population decline. However, in such populations adaptive processes to changing conditions might be limited. An intriguing question is still what happens if new pathogens arise which differ from commonly encountered diseases by the respective populations. It is reasonable to assume that the maintenance and even more renewal of variation in functional important parts of the MHC such as in the antigen binding sites, either from mutation, recombination or immigration from other populations is an important genetic component in the cascade leading to an appropriate immune response, when combating new or coevolving virulent pathogens. It was suggested that the extremely low genetic variability in cheetahs (Aconyx jubatus) as a consequence of inbreeding depression due to a substantial bottleneck about 10,000 ago limits adaptation processes to temporary pathogens which might explain the high disease susceptibility nowadays [118,203]. As in other mainly endangered species, the proposed association cannot be tested as there are no genetically variable outbred cheetah population to compare with [21]. Samples from preserved bones or from museum specimens in the case of recently presumed bottlenecks would be necessary to directly assess historical levels of MHC variation and to evaluate the relative role of demographic changes in determining existing levels of genetic diversity at the MHC [127].
Concerning the role of MHC in conservation genetics, the potential functional role of background genes (non-MHC genes) in disease resistance should not be ignored. Human studies indicate that background genes might also play an important role in determining pathogen and parasite resistance, either by themselves or in a epistatic manner with MHC-genes (summarised by [179,204]. Many of the regulatory genes show high polymorphism and, for example, variation in the tumor necrosis factor gene promotor, cytokines such as interleucin receptor, γ-interferon receptor vitamin and D receptor has been associated to infectious diseases (summarised by [187,191,204]. Although evidence is accumulating that the MHC is one of the main factors controlling resistance to diseases [176,205] conservation genetics should focus on the preservation of both, MHC and genome-wide diversity. Thereby, how much MHC diversity is required to ensure long-term population viability remains a fundamental question in conservation genetics [133] and can only be investigated close follow-ups of the genetic and health status of declining populations.
Conclusion
The diverse functions and characteristics place genes of the MHC among the best candidates for studies of mechanisms and significance of molecular adaptation in vertebrates. In contrary to neutral markers, MHC variability reflects evolutionary relevant and adaptive processes within and between populations and is very suitable to investigate a wide range of open questions in evolutionary ecology and conservation.
The selective effects from different pathogens appear to be the major driving force in the maintenance of MHC variation. Evidence of balancing selection at MHC genes has been found at different temporal scales. Selection in the distant past has been documented as an excess of nonsynonymous to synonymous substitutions, and as trans-species polymorphism. Selection in the recent past has been determined by excess heterozygosity compared to neutral theory expectations, differences in FST-values compared to neutral theory, or excess linkage equilibrium. Selection in the current generation has been identified by measuring deviations from Hardy-Weinberg or random mating proportions, survival differences between homozygotes and heterozygotes, and correlations of disease resistance with MHC-allele or genotype [90]. Changes in certain amino acids in the functional important MHC-coded antigen binding sites and thus the amino acid compositions influence functional differences in pathogen and parasite resistance. It is reasonable to assume that the maintenance and even more renewal of variation in functional important parts of the MHC such as in the antigen binding sites, either from mutation, recombination or immigration from other populations is an important genetic component in the cascade leading to an appropriate immune response, when combating new or coevolving virulent pathogens and might be important in conservation genetics. Studies indicate the functional importance of MHC variability in pathogen and parasite resistance not only in humans or in model organisms under experimental, laboratory conditions where most of our current knowledge derived from, but also in wild animal populations investigated in their natural environment. Only field studies in free-ranging animal populations can reveal the effects of conditionally advantageous or deleterious alleles in the presence of natural stress (e.g. spatially and temporally changes in climate, food availability, and competition), associated levels of parasitism, and thus the ubiquity of pathogen-driven selective mechanisms and the importance of MHC diversity across taxa. The combination with an experimental approach under standardized laboratory conditions is needed to prove the causal relationships behind correlations observed in the field.
Right now it is not quite clear whether 'heterozygote advantage' or 'frequency-dependent selection hypothesis' is most important for balancing selection. Most studies investigating 'heterozygote advantage' compared the infectious disease outcomes of heterozygotes at a given MHC loci, as a group, to the outcomes of homozygotes at the same locus, as a group ('population heterozygote advantage'). Comparing the average performance of all heterozygotes against homozygotes, instead of using allele specific tests for 'overdominance' (= 'allele-specific overdominance') circumvent tests of the original hypothesis namely the superiority of heterozygotes over either corresponding homozygote as this hypothesis is conditional on the alleles involved. Allele-specific analyses were most often impossible due to restrictions in sample size. In humans, recently a new approach to circumvent this problem was proposed by classifying alleles to supertypes based on shared binding motifs [176,177]. After more intense studies of the implications and limits of supertypes in large human data sets, this classification approach of common functional traits may also provide tools for the MHC research in natural endangered populations, where high allelic diversity causes problems in obtaining sufficiently large statistical sample sizes. Caution must be taken not to miss the effects of new, rare alleles by clustering alleles in functional types. So far, more evidence is available for the importance of specific MHC-alleles in parasite resistance or susceptibility. It is conceivable that a rare allele may have a high fitness and at the same time a constant advantage for heterozygotes thus both modes of balancing selection may act synergistically to enhance the maintenance of polymorphism.
Ongoing investigations of the parasite-driven selection mechanisms under natural conditions should focus on temporal variation of pathogens, host fitness attributes and allele frequencies to test whether allele frequencies change accordingly in a cycling pattern. Assessing the immunogenetic status of a population relative to another experiencing different suites of antigenic challenges will help to increase our knowledge on the importance of adaptive genetic variability in free ranging animal populations with respect to human impact and the role of the MHC in evolutionary ecology and conservation.
Acknowledgements
I would like to thank the German Zoological Society (DZG), the study groups of ecology and evolutionary biology and especially B Koenig for the opportunity to present this paper at the 97th Annual Meeting in Rostock. I am grateful to two anonymous referees for very constructive comments and suggestions on a former version of this manuscript. I thank JU Ganzhorn for continual support and encouragement and my diploma and PhD students for their contributions and stimulating discussions on the functional importance of MHC variability. This work was supported by the German Research Foundation (DFG) and by the German Federal Ministry of Education and Research (BMBF).
==== Refs
Wahlberg N Moilanen A Hanski I Predicting the occurrence of endangered species in fragmented landscapes Science 1996 273 1536 1538
Meffe GK Carroll CG contributors Principles of Conservation Biology 1997 2 Sunderland, Massachusetts, Sinauer Associates
Peacock MM Smith AT The effects of habitat fragmentation on dispersal patterns, mating behavior and genetic variation in a pica (Ochotona princeps) metapopulation Oecol 1997 112 524 533 10.1007/s004420050341
Harrison S Hastings A Genetic and evolutionary consequences of a metapopulation structure Trends Ecol Evol 1996 11 180 183 10.1016/0169-5347(96)20008-4
Dudash MR Fenster CB Clarke G Inbreeding and outbreeding depression Genetics, Demography and Viability of Fragmented Populations 2000 Cambridge: Cambridge University Press
Keller L Waller D Inbreeding effects in wild populations Trends Ecol Evol 2002 17 230 241 10.1016/S0169-5347(02)02489-8
Primack RB Essentials of Conservation Biology 1993 Sunderland, Massachusetts, Sinauer Associates
Saccheri I Kuussaari M Kankare M Vikman P Fortelius W Hanski I Inbreeding and extinction in a butterfly metapopulation Nature 1998 392 491 494 10.1038/33136
Lande R Genetics and demography in biological conservation Science 1988 241 1455 1460 3420403
Brewer BA Lacy RC Foster ML Alaks G Inbreeding depression in insular and central populations of Peromyscus mice J Hered 1990 81 257 266 2273239
Lacy RC Importance of genetic variation to the viability of mammalian populations J Mamm 1997 78 320 335
Frankham R Ralls K Inbreeding leads to extinction Nature 1998 392 441 442 10.1038/33022
Biilsma R Bundgaard J Boerema AC Does inbreeding affect the extinction risk of small populations?: predictions from Drosophila J Evol Biol 2000 13 502 514 10.1046/j.1420-9101.2000.00177.x
Coltman DW Pilkington JG Smith JA Pemberton JM Parasite-mediated selection against inbred Soay sheep in a free-living, island population Evol 1999 53 1259 1267
Meagher S Genetic diversity and Capillaria hepatica (Nematoda) prevalence in Michigan deer mouse populations Evol 1999 53 1318 1324
Cassinello J Gomendio M Roldan ERS Relationship between coefficient of inbreeding and parasite burden in endangered gazelles Conserv Biol 2001 15 1171 1174 10.1046/j.1523-1739.2001.0150041171.x
Little TJ The evolutionary significance of parasitism: do parasite-driven genetic dynamics occur ex silico? J Evol Biol 2002 15 1 9 10.1046/j.1420-9101.2002.00366.x
Altizer S Harvell D Friedle E Rapid evolutionary dynamics and disease threats to biodiversity Trends Ecol Evol 2003 18 589 596 10.1016/j.tree.2003.08.013
Bernatchez L Landry C MHC studies in nonmodel vertebrates: what have we learned about natural selection in 15 years? J Evol Biol 2003 16 363 377 14635837 10.1046/j.1420-9101.2003.00531.x
Spielman D Brook BW Briscoe DA Frankham R Does inbreeding and loss of genetic diversity decrease disease resistance? Conserv Genet 2004 5 439 448 10.1023/B:COGE.0000041030.76598.cd
Frankham R Ballou JD Briscoe DA Introduction to Conservation Genetics 2002 Cambridge, Cambrige University Press 12126990
Avise JC Phylogeography 2000 Cambridge MA, Harvard University Press
Sunnucks P Efficient genetic markers for population biology Trends Ecol Evol 2000 15 199 203 10782134 10.1016/S0169-5347(00)01825-5
Lowe A Harris S Ashton P Ecological Genetics: Design, Analysis, and Application 2004 Oxford, UK, Blackwell Publishing Ltd
Blouin MS Parsons M Lacaille V Lotz S Use of microsatellite loci to classify individuals by relatedness Mol Ecol 1996 5 393 401 8688959 10.1046/j.1365-294X.1996.00094.x
Balloux F Lugon-Moulin N The estimation of population differentiation with microsatellite markers Mol Ecol 2002 11 155 165 11856418 10.1046/j.0962-1083.2001.01436.x
Brumfield RT Beerli P Nickerson DA Edwards SV The utility of single nucleotide polymorphisms in inferences of population history Trends Ecol Evol 2003 18 249 256 10.1016/S0169-5347(03)00018-1
Morin PA Luikart G Wayne RK the SNP workshop group SNPs in ecology, evolution and conservation Trends Ecol Evol 2004 19 208 216 10.1016/j.tree.2004.01.009
Meyers LA Bull JJ Fighting change with change: adaptive variation in an uncertain world Trends Ecol Evol 2002 17 551 557 10.1016/S0169-5347(02)02633-2
Van Tienderen PH de Haan AA van der Linden G Vosman B Biodiversity assessment using markers for ecologically important traits Trends Ecol Evol 2002 17 577 582 10.1016/S0169-5347(02)02624-1
Crandall KA Bininda-Emonds ORP Mace GM Wayne RK Considering evolutionary processes in evolutionary biology Trends Ecol Evol 2000 15 290 295 10856956 10.1016/S0169-5347(00)01876-0
Stockwell CA Hendry AP Kinnison MT Contemporary evolution meets conservation biology Trends Ecol Evol 2003 18 94 101 10.1016/S0169-5347(02)00044-7
Cohen S Strong positive selection and habitat-specific amino acid substitution patterns in MHC from an estuary fish under intense pollution stress Mol Biol Evol 2002 19 1870 1880 12411596
Coulson T Albon S Slate J Pemberton J Microsatellite loci reveal sex-dependent responses to inbreeding and outbreeding in red deer calves Evol 1999 53 1951 1960
Amos W Wilmer J Fullard K The influence of parental relatedness on reproductive success Proc R Soc Lond B 2001 268 2021 2027 10.1098/rspb.2001.1751
Acevedo-Whitehouse K Gulland F Greig D Amos W Disease susceptibility in California sea lions Nature 2003 422 35 12621424 10.1038/422035a
Slate J Kruuk L Marshall T Pemberton J Clutton-Brook T Inbreeding depression influences lifetime breeding success in a wild population of red deer (Cervus elaphus) Proc R Soc Lond B 2000 267 1657 1662 10.1098/rspb.2000.1192
Hansson B Westerberg L On the correlation between heterozygosity and fitness in natural populations Mol Ecol 2002 11 2467 2474 12453232 10.1046/j.1365-294X.2002.01644.x
Coltman DW Slate J Microsatellite measures of inbreeding: a meta-analysis Evol 2003 57 971 983
Balloux F Amos W Coulson T Does heterozygosity estimate inbreeding in real populations? Mol Ecol 2004 13 3021 3031 15367117 10.1111/j.1365-294X.2004.02318.x
DeWoody YD DeWoody JA On the estimation of genome-wide heterozygosity using molecular markers J Hered 2005 96 85 88 15618305 10.1093/jhered/esi017
Klein J Natural History of the Major Histocompatibility Complex 1986 New York, Wiley & Son
Klein J Horejsi V Immunology 1997 Oxford, Blackwell Science
Dengjel J Schoor O Fischer R Reich M Kraus M Autophagy promotes MHC class II presentation of peptides from intracellular source proteins Proc Nat Acad Sci USA 2005 102 7922 7927 15894616 10.1073/pnas.0501190102
Ohta T On the pattern of polymorphisms at major histocompatibility complex loci J Mol Evol 1998 46 633 638 9608045
Reusch TB Häberli MA Aeschlimann PB Milinski M Female sticklebacks count alleles in a strategy of sexual selection explaining MHC polymorphism Nature 2001 414 300 302 11713527 10.1038/35104547
Wegner KM Kalbe M Kurtz J Reusch TBH Milinski M Parasite selection for immunogenetic optimality Science 2003 301 1343 12958352 10.1126/science.1088293
Marsh SGE Parham P Barber LD The HLA Facts Book 2000 London, Academic Press
Gutierrez-Espeleta GA Hedrick PW Kalinowski ST Garrigan D Boyce WM Is the decline of desert bighorn sheep from infectious disease the result of low MHC variation? Heredity 2001 86 439 450 11520344 10.1046/j.1365-2540.2001.00853.x
Stenzel A Lu T Koch WA Hampe J Guenther SM De La Vega FM Krawczak M Schreiber S Patterns of linkage disequilibrium in the MHC region on human chromosome 6 p Hum Genet 2004 114 377 385 14740295 10.1007/s00439-003-1075-5
Kelley J Walter L Trowsdale J Comparative genomics of major histocompatibility complexes Immunogenetics 2005 56 683 695 15605248 10.1007/s00251-004-0717-7
Hedrick PW Evolutionary genetics at the major histocompatibility complex Am Nat 1994 143 945 964 10.1086/285643
Hughes A Hughes M Natural selection on the peptide-binding regions of major histocompatibility complex molecules Immunogenetics 1995 42 233 243 7672817 10.1007/BF00176440
Hedrick PW Kim TJ Parker KM Parasite resistance and genetic variation in the endangered Gila topminow Anim Cons 2001 4 103 109 10.1017/S1367943001001135
Hedrick PW Parker KM Miller EL Miller PS Major histocompatiblity complex variation in the endangered Przewalski's horse Genetics 2001 152 1701 1710 10430594
Paterson S Wilson K Pemberton JM Major histocompatibility complex variation associated with juvenile survival and parasite resistance in a large unmanaged ungulate population (Ovis aries L.) Evol 1998 95 3714 3719
Langefors A Lohm J Grahn M Andersen O von Schantz T Association between major histocombatibility complex class IIB alleles and resistance to Aeromonas salmonicida in Atlantic salmon Proc R Soc Lond B 2001 268 479 485 10.1098/rspb.2000.1378
Hughes AL Yeager M Natural selection at major histocompatibility complex loci of vertebrates Ann Rev Genet 1998 32 415 434 9928486 10.1146/annurev.genet.32.1.415
Hughes AL Nei M Pattern of nucleotide substitution at major histocompatibility complex class I loci reveals overdominant selection Nature 1988 335 167 170 3412472 10.1038/335167a0
Hughes AL Nei M Nucleotide substitution at major histocompatibility complex class II loci: Evidence for overdominant selection Proc Natl Acad Sci USA 1989 86 948 962 2536938
Brown JH Jardetzky TS Saper MA Samraoui B Bjorkman PJ Wiley DC A hypothetical model of foreign antigen binding site of class II histocompatibility molecules Nature 1988 332 845 850 3258651 10.1038/332845a0
Brown JH Jardetzky TS Gorga JC Stern LJ Urban RG Strominger JL Wiley DC Three-dimensional structure of the human class II histocompatibility antigen HLA-DR1 Nature 1993 364 33 39 8316295 10.1038/364033a0
Bergström T Gyllensten U Evolution of Mhc class II polymorphism: the rise and fall of class II gene function in primates Immunol Rev 1995 143 13 31 7558074
Richman A Evolution of balanced genetic polymorphism Mol Ecol 2000 9 1953 1963 11123608 10.1046/j.1365-294X.2000.01125.x
Janeway CA Travers P Immunologie 2002 Heidelberg, Berlin, Oxford, Spektrum Akademischer Verlag GmbH
Hedrick PW Balancing selection and the MHC Genetica 1999 104 207 214 10386384 10.1023/A:1026494212540
Jeffery KJ Bangham CR Do infectious diseases drive MHC diversity? Microb Infect 2000 2 1335 41 10.1016/S1286-4579(00)01287-9
Penn DJ The Scent of Genetic Compatibility: Sexual Selection and the Major Histocompatibility Complex Ethol 2002 108 1 21 10.1046/j.1439-0310.2002.00768.x
Doherty PC Zinkernagel RM Enhanced immunological surveillance in mice heterozygous at the H-2 gene complex Nature 1975 256 50 52 1079575 10.1038/256050a0
Penn DJ Damjanovich K Potts WK MHC heterozygosity confers a selective advantage against multi-strain infections Proc Natl Acad Sci 2002 99 11260 11264 12177415 10.1073/pnas.162006499
Gould SJ Hildreth JEK Booth AM The evolution of alloimmunity and the genesis of adaptive immunity Quart Rev Immunol 2004 79 359 382
Hughes AL Nei M Maintenance of MHC polymorphism Nature 1992 355 402 403 1734277 10.1038/355402b0
Takahata N Nei M Allelic genealogy under overdominant and frequency-dependent selection and polymorphism of major histocompatibility complex loci Genetics 1990 124 967 978 2323559
Takahata N A simple genealogical structure of strongly balanced allelic lines and trans-species evolution of polymorphism Proc Natl Acad Sci USA 1990 87 2419 2423 2320564
Wakeland EK Boehme S She JX Lu CC McIndoe RA Cheng I Ye Y Potts WK Ancestral polymorphism of MHC class II genes: divergent allele advantage Immunol Res 1990 9 115 122 2189934
She JX Wakeland EK Boehme S The generation and maintenance of MHC class II gene polymorphism in rodents Immunol Rev 1992 113 207 226 2180807
Richman AD Herrera LG Nash D MHC class II beta sequence diversity in the deer mouse (Peromyscus maniculatus): implications for models of balancing selection Mol Ecol 2001 10 2765 2773 11903890
Takahata N Satta Y Klein J Polymorphism and balancing selection at major histocompatibility complex loci Genetics 1992 130 925 938 1582567
Takahata N Satta Y Selection, convergence, and intragenic recombination in HLA diversity Genetica 1998 102–103 157 169 10.1023/A:1017029613342
Satta Y Effects of intra-locus recombination of HLA polymorphism Hereditas 1997 127 105 112 9420476 10.1111/j.1601-5223.1997.00105.x
Clarke B Kirby DR Maintenance of histocompatibility polymorphisms Nature 1966 211 999 1000 6007869
Bodmer W Evolutionary significance of the HLA-system Nature 1972 237 139 145 4113158 10.1038/237139a0
Lively CM Dybdahl MF Parasite adaptation to locally common host genotypes Nature 2000 405 679 681 10864323 10.1038/35015069
Jaenike J A hypothesis to account for sex within populations Evol Theory 1978 3 191 194
Ladle RJ Parasites and sex: catching the Red Queen Trends Ecol Evol 1992 7 405 408 10.1016/0169-5347(92)90021-3
Ebert D Hamilton WD Sex against virulence: the coevolution of parasitic diseases Trends Ecol Evol 1996 11 79 82 10.1016/0169-5347(96)81047-0
Penn DJ Potts WK The evolution of mating preferences and major histocompatibility complex genes Am Nat 1999 153 145 163 10.1086/303166
Nevo E Beiles A Selection for class II MHC heterozygosity by parasites in subterranean mole rats Experientia 1992 48 512 515 1351001 10.1007/BF01928177
Hedrick PW Pathogen resistance and genetic variation at MHC loci Evol 2002 56 1902 1908
Hedrick PW Foxy MHC selection story Hered 2004 93 237 238 10.1038/sj.hdy.6800539
Hedrick PW Comment on 'parasite selection for immunogenetic optimality" Science 2004 303 957 14963312 10.1126/science.1092163
Westerdahl H Hansson B Bensch S Hasselquist D Between-year variation of MHC allele frequencies in great reed warblers: selection or drift? J Evol Biol 2004 17 485 492 15149391 10.1111/j.1420-9101.2004.00711.x
Potts WK Wakeland EK Evolution of MHC genetic diversity: a tale of incest, pestilence and sexual preference Trends Genet 1993 9 408 412 8122307 10.1016/0168-9525(93)90103-O
Edwards SV Potts WK Smith TB and Wayne RK Polymorphism of genes in the major histocompatibility complex (MHC): implications for conservation genetics Molecular Genetic Approaches in Conservation 1996 New York, Oxford University Press 214 237
Edwards S Hedrick PW Evolution and ecology of MHC molecules: from genomics to sexual selection Trends Ecol Evol 1998 13 305 311 10.1016/S0169-5347(98)01416-5
Jordan WC Bruford MW New perspectives on mate choice and the MHC Hered 81 127 133
Penn DJ Fisher S Sniffing out genetic compatibility Biologist 2004
Zuk M Reproductive strategies and disease susceptibility: an evolutionary viewpoint Parasitol Today 1990 6 231 233 15463350 10.1016/0169-4758(90)90202-F
Brown JL Eklund A Kin recognition and the major histocompatibility complex: an integrative review Am Nat 1994 143 435 461 10.1086/285612
Penn DJ Potts WK How do major histocompatibility complex genes influence odor and mating preferences? Adv Immunol 1998 69 411 435 9646849
Yamazaki K Singer A Beauchamp GK Origin, functions and chemistry of H-2 regulated odorants Genetica 1998 104 235 240 10386388 10.1023/A:1026433925744
Eggert F Muller-Ruchholtz W Ferstl R Olfactory cues associated with the major histocompatibility complex Genetica 1999 104 191 197 10386382 10.1023/A:1026402531196
Singh PM Brown RE Roser B MHC antigens in urine as olfactory recognition cues Nature 1987 327 161 164 3574476 10.1038/327161a0
Singh P The present status of the 'carrier hypotheses' for chemosensory recognition of genetoc individuality Genetica 1998 104 231 233 10386387 10.1023/A:1026475118901
Leinders-Zufall T Brennan P Widmayer P Chandramani P Maul-Pavicic A MHC class I peptides as chemosensory signals in the vomeronasal organ Science 2004 306 1033 1037 15528444 10.1126/science.1102818
Sugg DW Chesser RK Dobson FS Hoogland JL Population genetics meets behavioural ecology Trends Ecol Evol 1996 11 338 342 10.1016/0169-5347(96)20050-3
Balloux F Goudet J Perrin N Breeding system and genetic variance in the monogamous, semi-social shrew, Crocidura russula Evol 1998 52 1230 1235
Aguilar A Roemer G Debenham S Binns M Garcelon D Wayne RK High MHC diversity maintained by balancing selection in an otherwise genetically monomorphic mammal Proc Natl Acad Sci USA 2004 101 3490 3494 14990802 10.1073/pnas.0306582101
Jarvi SI Tarr CL McIntosh CE Atkinson CT Fleischer RC Natural selection of the major histocompatibility complex (MHC) in Hawaiian honeycreepers (Drepanidinae) Mol Ecol 2004 13 2157 2168 15245391 10.1111/j.1365-294X.2004.02228.x
Schad J Sommer S Ganzhorn JU MHC variability of a small lemur in the littoral forest fragments of southeastern Madagascar Conserv Genet 2004 5 299 309 10.1023/B:COGE.0000031137.50239.d3
Hapke A Population genetics and differentiation of species of the genus of mouse lemurs, Microcebus (E. Geoffroy St. Hilaire, 1828) and dwarf lemurs, Cheirogaleus (E. Geoffrey St. Hilaire, 1812) in southeastern Madagascar PhD Thesis 2004 University Hamburg, Germany
Hingston M Goodman SM Ganzhorn JU Sommer S Reconstruction of the colonization of southern Madagascar by introduced Rattus rattus J Biogeography 2005 32 1549 1559 10.1111/j.1365-2699.2005.01311.x
Robertson A Selection for heterozygotes in small populations Genetics 1962 47 1291 1300 13974344
Nevo E Kirzher V Beiles A Karol A Selection versus random drift: long-term polymorphism persistence in small populations (evidence and modelling) Phil Trans R Soc Lond B 1997 352 381 389 10.1098/rstb.1997.0028
Seddon JM Baverstock PR Variation on islands: major histocombatibility complex (Mhc) polymorphism in population of the Australian bush rat Mol Ecol 1999 8 2071 2079 10632858 10.1046/j.1365-294x.1999.00822.x
Hinten G Harriss F Rossetto M Baverstock PR Genetic variation and island biogeography: microsatellite and mitochondrial DNA variation in island populations of the Australian bush rat, Rattus fuscipes greyii Cons Genet 2004 4 759 778 10.1023/B:COGE.0000006113.58749.ac
Yuhki N O'Brien SJ DNA variation at the mammalian major histocompatibility complex reflects genomic diversity and population history Proc Natl Acad Sci 1990 87 836 840 1967831
O'Brien SJ Wildt DE Goldman D Merril CR Bush M The cheetah is depauperate in genetic variation Science 1985 221 459 462 17755482
Sommer S Tichy H Major histocompatibility complex (MHC) class II polymorphism and paternity in the monogamous Hypogeomys antinema, the endangered, largest endemic Malagasy rodent Mol Ecol 1999 8 1259 1272 10447867 10.1046/j.1365-294X.1999.00687.x
Sommer S Schwab D Ganzhorn JU MHC diversity of endemic Malagasy rodents in relation to range contraction and social system Behav Ecol Sociobiol 2002 51 214 221 10.1007/s00265-001-0432-4
Sommer S MHC-dependent mate choice in a monogamous rodent Behav Ecol Sociobiol 2005 58 181 189 10.1007/s00265-005-0909-7
Smulders MJM Snoek LB Booy G Vosman B Complete loss of MHC genetic diversity in the Common hamster (Cricetus cricetus) population in the Netherlands. Consequences of conservation strategies Conserv Genet 2003 4 441 451 10.1023/A:1024767114707
Ellegren H Hartman G Johansson M Andersson L Major histocompatibility complex monomorphism and low levels of DNA fingerprinting variability in a reintroduced and rapidly expanding population of beavers Proc Natl Acad Sci, USA 1993 90 8150 8153 8367476
Mikko S Andersson L Low major histocompatiblity complex class II diversity in European and North American moose Proc Natl Acad Sci USA 1995 92 4259 4263 7753793
Ellegren H Mikko S Wallin K Andersson L Limited polymorphism at major histocompatibility complex (MHC) loci in the Swedish moose A. alces Mol Ecol 1996 5 3 9 9147694
Mikko S Roed K Schmutz S Andersson L Monomorphism and polymorphism at MHC DRB loci in domestic and wild ruminants Immunol Rev 1999 167 169 178 10319259
Amills M Jiménez N Jordana J Riccardi A Low diversity in the major histocompatibility complex class II DRB1 gene of the Spanish ibex (Capra pyrenaica) Hered 2004 93 266 272 10.1038/sj.hdy.6800499
Hedrick PW Gutierrez-Espeleta GA Lee RN Founder effects in an island population of bighorn sheep Mol Ecol 2001 10 851 857 11348494 10.1046/j.1365-294X.2001.01243.x
Hedrick PW Parker KM Gutiérrez-Espeleta GA Rattink A Lievers K Major histocompatibility complex variation in the Arabian oryx Evol 2000 54 2145 2151
Van der Walt JM Nel LH Hoelzel AR Characterization of major histocompatiblity complex DRB diversity in the endemic South African antilope Damaliscus pygargus : a comparison in two subspecies with different demographic histories Mol Ecol 2001 10 1679 1688 11472536 10.1046/j.0962-1083.2001.01321.x
Hoelzel AR Stephens JC O'Brien SJ Molecular genetic diversity and evolution at the MHC DQB locus in four species of pinnipeds Mol Biol Evol 1999 16 611 618 10335654
Trowsdale J Groves V Arnason A Limited MHC polymorphism in whales Immunogenetics 1989 41 19 24 2562799
Miller HC Lambert DM Genetic drift outweighs balancing selection in shaping post-bottleneck major histocompatiblity complex variation in New Zealand robins (Petroicidae) Mol Ecol 2004 13 3709 3721 15548285 10.1111/j.1365-294X.2004.02368.x
Nei M Maruyama T Chakraborty R The bottleneck effect and genetic variability in populations Evol 1975 29 1 10
Weber DS Stewart BS Schienman J Lehman N Major histocompatibility complex variation at three class II loci in the northern elephant seal Mol Ecol 2004 13 711 718 14871373 10.1111/j.1365-294X.2004.02095.x
Wegner KM Kalbe M Schaschl H Reusch TBH Parasites and individual major histocompatiblity diversity – an optimal choice Microb Infect 2004 6 1110 1116 10.1016/j.micinf.2004.05.025
Carrington M Nelson GW Martin MP Kissner T Vlahov D Goedert JJ Kaslow R Buchbinder S Hoots K O'Brien SJ HLA and HIV-1: heterozygote advantage and B*35-Cw*04 disadvantage Science 1999 283 1748 1752 10073943 10.1126/science.283.5408.1748
Thursz MR Thomas HC Greenwood BM Hill AV Heterozygote advantage for HLA class-II type in hepatitis B virus infection Nat Genet 1997 17 11 12 9288086 10.1038/ng0997-11
Chen CY Cohen SA Zaleski MB Albini B Genetic control of streptococcus-induced hepatic granulomatous lesions in mice Immunogenetics 1992 36 28 32 1587550 10.1007/BF00209289
McClelland EE Penn DJ Potts WK Major histocompatibility complex heterozygote superiority during coinfection Infect Immun 2003 71 2079 2086 12654829 10.1128/IAI.71.4.2079-2086.2003
Behnke JM Wahid FN Immunological relationships during primary infection with Heligmosomoides polygyrus (Nematospiroides dubius): H-2 linked genes determine worm survival Parasitol 1991 103 157 164
Sher A Hieny S James S Mechanisms of protective immunity against S. mansoni infection in mice vaccinated with irradiated cercariae: VI. Influence of the major histocompatibility complex Parasit Immunol 1984 6 319 328
Senseney HL Briles WE Abplanalp H Taylor RL Allelic complementation between MHC haplotypes B-Q and B-17 increases regression of Rous sarcomas Poultry Sci 2000 79 1736 1740 11194035
Arkush KD Giese AR Mendonca HL McBirde AM Marty GD Hedrick PW Resistance to three pathogens in the endangered winter-run chinook salomon (Oncorhynchus tshawytscha): effects of inbreeding and major histocompatiblity complex genotypes Can J Fish Aquat Sci 2002 59 966 975 10.1139/f02-066
Borghans JAM Beltman JB De Boer RJ MHC polymorphism under host-pathogen coevolution Immunogenetics 2004 55 732 739 14722687 10.1007/s00251-003-0630-5
Hill AVS Allsopp CEM Kwiatkowski D Anstey NM Tumasi P Rowe PA Bennett S Brewster D McMichael AJ Greenwood BM Common West-African HLA antigens are associated with protection from severe malaria Nature 1991 352 595 600 1865923 10.1038/352595a0
Decamposlima PO Gavioli R Zhang QJ Wallace LE Dolcetti R Rowe M Rickinson AB Masucci MG HLA-A11 epitope loss isolates of Epstein-Barr-virus from a highly A11+ population Science 2003 260 98 100
Thursz MR Kwiatkowski D Allsopp CE Association between an MHC class II allele and clearance of hepatitis B virus in the Gambia New Engl J Med 1995 332 1065 1069 7898524 10.1056/NEJM199504203321604
Li Z Chen D Zhang C Li Y Cao B Ning T Zhao Y You W Ke Y HLA polymorphisms are associated with Heliobacter pylori infected gastric cancer in a high risk population, China Immunogenetics 2005 56 781 787 15650879 10.1007/s00251-004-0723-9
Godot V Harraga S Beurton I Resistance/susceptibility to Echinococcus multilocularis infection and cytocine profile in humans. II. Influence of the HLA B8, DR3, DQ2 haplotype Clin Exp Immunol 2000 121 491 498 10971516 10.1046/j.1365-2249.2000.01309.x
McClelland EE Granger DL Potts WK Major histocompatiblity complex-dependent susceptibility to Cryptococcus neoformans in mice Infect Immun 2003 71 4815 4817 12874366 10.1128/IAI.71.8.4815-4817.2003
Wassom DL Brooks BO Babisch JG David CS A gene mapping between the S and D regions of the H-2 complex influences resistance to Trichinella spiralis infections in mice J Immunogenetics 1983 10 371 378 6417245
Wassom DL Krco CJ David CS I-E expression and susceptibility to parasite infection Immunol Today 1987 8 39 43 10.1016/0167-5699(87)90236-2
Enriquez FJ Brooks BO Cypess RH David CS Wassom DL Nematospiroides dubius : two H-2-linked genes influence levels of resistance to infection in mice Exp Parasitol 1988 67 221 226 3142789 10.1016/0014-4894(88)90069-0
Else KJ Wakelin D Wassom DL Hauda KM The influence of genes mapping within the major histocompatibility complex on resistance to Trichuris muris infections in mice Parasitol 1990 101 61 67
Schwaiger FW Gostomski D Stear M Duncan JL McKellar QA Epplen JT Buitkamp J An ovine major histocompatibility complex DRB1 allele is associated with low faecal egg count following natural, predominantly Ostertagia circumcincta infection Int J Parasitol 1995 25 815 822 7558567 10.1016/0020-7519(94)00216-B
Buitkamp J Filmether P Stear MJ Epplen JT Class I and class II major histocompatiblity complex alleles are associated with faecal egg counts following natural, predominantly Ostertagia circumcincta infection Parasitol Res 1996 82 693 696 8897503 10.1007/s004360050187
Briles WE Stone HA Cole RK Mareks-disease – effects of B-histocompatiblity alloalleles in resistant and susceptible chicken Science 1977 195 193 195 831269
Lohm J Grahn M Langefors A Andersen O Storset A von Schantz T Experimental evidence for major histocompatibility complex-allele-specific resistance to a bacterial infection Proc R Soc Lond B 2002 269 2029 2033 10.1098/rspb.2002.2114
Grimholt U Larsen S Nordmo R Midtlyng P Kjoeglum S Storset A Saebo S Stet RJM MHC polymorphism and disease resistance in Atlantic salmon (Salmo salar); facing pathogens with single expressed major histocompatibilty class I and class II loci Immunogenetics 2003 55 210 219 12811427 10.1007/s00251-003-0567-8
Meagher S Penn DJ Potts WK Male – male competition magnifies inbreeding depression in wild house mice Proc Natl Acad Sci USA 2000 97 3324 3329 10716731 10.1073/pnas.060284797
Little TJ Ebert D Temporal patterns of genetic variation for resistance and infectivity in a Daphnia-microparasite system Evol 2001 55 1146 1152
Froeschke G Sommer S MHC Class II DRB constitution and parasite load in the striped mouse, Rhabdomys pumilio, in the Southern Kalahari Mol Biol Evol 2005 22 1254 1259 15703235 10.1093/molbev/msi112
Wegner KM Reusch TBH Kalbe M Multiple infections drive major histocompatibility complex polymorphism in the wild J Evol Biol 2003 16 224 232 14635861 10.1046/j.1420-9101.2003.00519.x
Lawlor DA Zemmour J Ennis PD Parham P Evolution of class-I MHC genes and proteins: from natural selection to thymic selection Ann Rev Immunol 1990 8 23 63 2188663 10.1146/annurev.iy.08.040190.000323
Borghans JAM Noest AJ De Boer RJ Thymic selection does not limit the individual MHC diversity Europ J Immunol 2003 33 3353 3358 10.1002/eji.200324365
Malaga-Trillo E Zaleska-Rutczynska Z McAndrew B Vincek V Figuera F Sültmann H Klein J Linkage relationships and haplotype polymorphism among cichlid MHC class II B loci Genetics 1998 149 1527 1537 9649539
Reusch TBH Schaschl H Wegner KM Recent duplication and inter-locus gene conversion in major histocompatibility class II genes in a teleost, the three-spined stickleback Immunogenetics 2004 56 427 437 15322775 10.1007/s00251-004-0704-z
Doxiadis GG Otting N de Groot NG Bontrop RE Differential evolutionary MHC class II strategies in humans and rhesus macaques: relevance for biomedical studies Immunol Rev 2001 183 76 85 11782248 10.1034/j.1600-065x.2001.1830106.x
Bowen L Aldridge BM Stott JL Gulland F Woo J Class II multiformity generated by variable MHC-DRB region configurations in the California sea lion (Zalophus californianus) Immunogenetics 2004 56 12 27 14997355 10.1007/s00251-004-0655-4
Bowen L Aldridge BM DeLong R Melin S Buckles EL An immunogenetic basis for the high prevalence of urogenital cancer in a free-ranging population of California seas lions (Zalophus californianus) Immunogenetics 2005 56 846 848 15650877 10.1007/s00251-004-0757-z
Harf R Sommer S Association between MHC Class II DRB alleles and parasite load in the hairy-footed gerbil, Gerbillurus paeba, in the Southern Kalahari Mol Ecol 2005 14 85 91 15643953 10.1111/j.1365-294X.2004.02402.x
Meyer-Lucht Y Sommer S MHC diversity and the association to nematode parasitism in the yellow necked mouse (Apodemus flavicollis) Mol Ecol 2005 14 2233 2243 15910340 10.1111/j.1365-294X.2005.02557.x
Schad J Ganzhorn JU Sommer S MHC constitution and parasite burden in the Malagasy mouse lemur, Microcebus murinus Evol 2005 59 439 450
Lipsitch M Bergstrom CT Antia R Effect of human leukocyte antigen heterozygosity on infectious disease outcome: the need for allele-specific measures BMC Med Gen 2003 4 2 10.1186/1471-2350-4-2
Trachtenberg E Korber B Sollars C Kepler TB Advantage of rare HLA supertypes in HIV disease progression Nature Medicine 2003 9 928 935 12819779 10.1038/nm893
Lund O Nielsen M Kesmir C Petersen AG Lundegaard C Definition of supertypes for HLA molecules using clustering of specificity matrices Immunogenetics 2004 55 797 810 14963618 10.1007/s00251-004-0647-4
Landry C Garant D Duchesne P Bernatchez L 'Good genes as heterozygosity': the major histocompatibility complex and mate choice in Atlantic salmon (Salmo salar) Proc R Soc Lond B 2001 268 1279 1285 10.1098/rspb.2001.1659
Apanius V Penn D Slev PR Ruff LR The nature of selection on the major histocompatibility complex Crit Rev Immunol 1997 17 179 224 9094452
De Boer RJ Borghans JAM Van Boven M Kesmir C Weissing FJ Heterozygote advantage fails to explain the high degree of polymorphism of the MHC Immunogenetics 2004 55 725 731 14722686 10.1007/s00251-003-0629-y
Ditchkoff SS Lochmiller RL Masters RE Hoofer SR Van den Bussche RA Major-histocompatibility-complex-associated variation in secondary sexual traits of white-tailed deer (Odocoileus virginianus): evidence for good-genes advertisement Evol 2001 55 616 625
Von Schantz T Göransson G Andersson G Fröberg I Grahn M Helgí A Wittzell H Female choice selects for a viability-based male trait in pheasants Nature 1989 337 166 169 2911350 10.1038/337166a0
Von Schantz T Wittzell H Goransson G Grahn M Persson K MHC genotype and male ornamentation: genetic evidence for the Hamilton-Zuk model Proc R Soc 1996 263 265 271
Tompkins DM Begon M Parasites can regulate wildlife populations Parasitol 1999 15 311 313
Frank SA Immunology and the Evolution of Infectious Disease 2002 Princeton, Princeton University Press
Summers K McKeon S Sellars J Parasitic exploitation as an engine of diversity Biol Rev 2003 78 639 675 14700394 10.1017/S146479310300616X
Behnke JM Iraki F Menge D Baker RL Gibson J Wakelin D Chasing the genes that control resistance to gastrointestinal nematodes J Helminth 2003 77 99 109 12756063 10.1079/JOH2003174
Garside P Kennedy MW Wakelin D Lawrence CE Immunopathology of intestinal helminth infection Parasite Immunology 2000 22 605 612 11123752 10.1046/j.1365-3024.2000.00344.x
Macdonald TT The mucosal immune system Parasite Immunology 2003 25 235 246 12969442 10.1046/j.1365-3024.2003.00632.x
Maizels RM Yazdanbakhsh M Immune regulation by helminth parasites: cellular and molecular mechanisms Nat Rev 2003 3 733 744
Quinnell RJ Genetics of susceptibility to human helminth infection Int J 2003 33 1219 1231
Mitchell SE Rogers ES Little TJ Read AF Host-parasite and genotype-by-environment interactions: temperature modifies potential for selection by a sterilizing pathogen Evol 2005 59 70 80
Woolhouse MEJ Haydon DT Antia R Emerging pathogens: the epidemiology and evolution of species jumps Trends Ecol Evol 2005 20 238 244 16701375 10.1016/j.tree.2005.02.009
Poulin R The functional importance of parasites in animal communities: many roles at many levels? Int J Parasitol 1999 29 903 914 10480727 10.1016/S0020-7519(99)00045-4
Dybdahl MF Lively CM Host parasite coevolution: evidence for rare advantage and time-lagged selection in a natural population Evol 1998 52 1057 1066
Luikart G Cornuet JM Allendorf FW Temporal changes in allele frequencies provide estimates of population bottleneck size Conserv Biol 1999 13 523 530 10.1046/j.1523-1739.1999.98133.x
Rüdel N Consequences of degradation and fragmentation of Malagasy littoral rain forests on gray mouse lemur populations (Microcebus murinus) Diploma Thesis 2004 University Hamburg, Germany
Sgrò CM Hoffmann AA Genetic correlations, tradeoffs and environmental variation Hered 2004 93 241 248 10.1038/sj.hdy.6800532
Sommer S Hommen U Modelling the effects of life-history traits and changing ecological conditions on the population dynamics and persistence of the endangered Malagasy giant jumping rat (Hypogeomys antimena) Anim Cons 2000 4 333 343 10.1017/S1367943000001050
Sommer S Toto Volahy A Seal US A population and habitat viability assessment for the highly endangered Giant Jumping Rat (Hypogeomys antimena), the largest endemic rodent of Madagascar Anim Cons 2002 5 263 273
Sommer S Effects of habitat fragmentation and changes of dispersal behaviour after a recent population decline on the genetic variability of non-coding and coding DNA of a monogamous Malagasy rodent Mol Ecol 2003 12 2845 2851 12969486 10.1046/j.1365-294X.2003.01906.x
Nei M Gojobori T Simple methods for estimating the number of synonymous and non-synonymous nucleotide substitutions Mol Biol Evol 1986 3 418 426 3444411
O'Brien SJ Evermann JF Interactive influence of infectious disease and genetic diversity in natural populations Trends Ecol Evol 1988 3 254 259 10.1016/0169-5347(88)90058-4
Hill AVS The immunogenetics of human infectious diseases Ann Rev Immunol 1998 16 593 617 9597143 10.1146/annurev.immunol.16.1.593
Bontrop R Watkins DI MHC polymorphism: AIDS susceptibility in non-human primates Trends Immunol 2005 26 227 233 15797514 10.1016/j.it.2005.02.003
Musolf K Meyer-Lucht Y Sommer S Evolution of MHC-DRB class II polymorphism in the genus Apodemus and a comparison of DRB sequences within the family Muridae (Mammalia: Rodentia) Immunogenetics 2004 56 420 426 15351919 10.1007/s00251-004-0715-9
|
16242022
|
PMC1282567
|
CC BY
|
2021-01-04 16:38:34
|
no
|
Front Zool. 2005 Oct 20; 2:16
|
utf-8
|
Front Zool
| 2,005 |
10.1186/1742-9994-2-16
|
oa_comm
|
==== Front
Front ZoolFrontiers in Zoology1742-9994BioMed Central London 1742-9994-2-161624202210.1186/1742-9994-2-16ReviewThe importance of immune gene variability (MHC) in evolutionary ecology and conservation Sommer Simone [email protected] Animal Ecology & Conservation, Biocentre Grindel, University of Hamburg, Martin-Luther-King-Platz 3, D-20146 Hamburg, Germany2005 20 10 2005 2 16 16 9 3 2005 20 10 2005 Copyright © 2005 Sommer; licensee BioMed Central Ltd.2005Sommer; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Genetic studies have typically inferred the effects of human impact by documenting patterns of genetic differentiation and levels of genetic diversity among potentially isolated populations using selective neutral markers such as mitochondrial control region sequences, microsatellites or single nucleotide polymorphism (SNPs). However, evolutionary relevant and adaptive processes within and between populations can only be reflected by coding genes. In vertebrates, growing evidence suggests that genetic diversity is particularly important at the level of the major histocompatibility complex (MHC). MHC variants influence many important biological traits, including immune recognition, susceptibility to infectious and autoimmune diseases, individual odours, mating preferences, kin recognition, cooperation and pregnancy outcome. These diverse functions and characteristics place genes of the MHC among the best candidates for studies of mechanisms and significance of molecular adaptation in vertebrates. MHC variability is believed to be maintained by pathogen-driven selection, mediated either through heterozygote advantage or frequency-dependent selection. Up to now, most of our knowledge has derived from studies in humans or from model organisms under experimental, laboratory conditions. Empirical support for selective mechanisms in free-ranging animal populations in their natural environment is rare. In this review, I first introduce general information about the structure and function of MHC genes, as well as current hypotheses and concepts concerning the role of selection in the maintenance of MHC polymorphism. The evolutionary forces acting on the genetic diversity in coding and non-coding markers are compared. Then, I summarise empirical support for the functional importance of MHC variability in parasite resistance with emphasis on the evidence derived from free-ranging animal populations investigated in their natural habitat. Finally, I discuss the importance of adaptive genetic variability with respect to human impact and conservation, and implications for future studies.
==== Body
Introduction
Many natural populations are threatened not only by a dramatic reduction in total area of available habitat but also by increasing habitat fragmentation and degradation leading to declining population sizes and barriers to gene flow if exchange of individuals between subpopulations is restricted [1-3]. Small populations often suffer from reduction of genetic diversity due to genetic drift and inbreeding effects [4-6]. Negative effects such as increased rates of allelic loss, fixation of deleterious alleles and decreased average individual heterozygosity relative to the overall population were observed by both, theoretical and empirical studies [7,8]. The loss of genetic variation can lead to short-term reduction of fitness components such as survival, reproductive output, growth rates and to impaired ability to adapt to long-term changes in the environment [7,9-13]. An increasing number of studies indicates that host genetic diversity plays an important role in buffering populations against pathogens and widespread epidemics [6,14-20]. Study of the genetic effects of population fragmentation is therefore of central importance for conservation biology [21].
Genetic studies of wild animals often employ neutral markers such as mitochondrial d-loop DNA (mtDNA), microsatellites or single nucleotide polymorphism (SNPs) to estimate the amount of variation present in individuals and populations [22-24]. While these markers are very informative for phylogenetic reconstructions and population history (bottleneck effects), for molecular clocks, to examine dispersal patterns of individuals (gene flow) and to classify individuals by relatedness and paternity analyses [25-28], the variation at neutral loci cannot provide direct information on selective processes involving the interaction of individuals with their environment or on the capacity for future adaptive changes [29,30]. However, these are issues of particular relevance in evolutionary ecology and conservation [31,32]. In addition, recent research in a variety of taxa and situations has revealed that evolution often occurs on contemporary timescales, often within decades (summarised in [32]). In some cases, the time span between the separation of populations might even be too short to leave a signal at neutral loci so that differences between populations are only detectable at genes under selection [33], such as those of the highly variable major histocompatibility complex (MHC). Contrary to neutral markers, MHC variability reflects evolutionary relevant and adaptive processes within and between populations and is very suitable to investigate a wide range of open questions in evolutionary ecology and conservation. The comparison with neutral markers allows the construction of null hypotheses concerning the diversity at selectively relevant genes and conclusions on the relevance of MHC polymorphism. One might argue that many recent studies report that individual heterozygosity at apparently neutral microsatellite markers is correlated with key components of individual fitness such as survival [34], fecundity [35], disease resistance [14,36] and lifetime reproductive success [37]. However, null results are likely to be underrepresented in the literature because of publication bias in favour of significant correlations [38]. A recent review and meta-analysis of both published and unpublished studies of the association between neutral marker heterozygosity and traits or components of individual fitness reported that associations were common, yet typically weak [39]. A correlation between individual heterozygosity at neutral genetic markers and components of individual fitness can arise in different ways, with the effects of inbreeding depression due to a genome-wide reduction in genetic variability (including fitness-relevant loci) and linkage disequilibrium to loci under selection being the most likely explanations [[6,38,40], see also [41]].
In this review, I first introduce general information about the structure and function of MHC genes, as well as current hypotheses and concepts concerning the role of selection in the maintenance of MHC polymorphism. The evolutionary forces acting on the genetic diversity in coding and non-coding markers are compared. Then, I summarise empirical support for the functional importance of MHC variability in parasite resistance with emphasis on the evidence derived from free-ranging animal populations investigated in their natural habitat. Finally, I discuss the importance of adaptive genetic variability with respect to human impact and conservation, and implications for future studies.
Major histocompatibility complex (MHC): structure, function and selection mechanisms
Structure and function
The MHC consists of a group of closely linked genes that constitute the most important genetic component of the mammalian immune system [42]. The MHC encodes cell-surface glycoproteins that bind antigens derived from pathogens or parasites and present them to T-lymphocytes which trigger the appropriate immune response. Two major groups of MHC genes can be distinguished. MHC class I genes play an essential role in the immune defence against intracellular pathogens by binding peptides mainly derived from viral proteins and cancer infected cells. They are expressed on the surface of all nucleated somatic cells. In contrast, MHC class II genes are predominantly involved in monitoring the extracellular environment by presenting peptides mainly derived from parasites to the T-cells (e.g. bacteria, nematodes, cestodes) [43,44]. They are primarily expressed on antigen-presenting cells of the immune system, such as B cells and macrophages. Within class II genes, most research in mammals focuses on the second exon of DRB genes because these loci code for parts of the functionally important antigen binding sites (ABS) [45]. Alternatively, the β-chain in general is used if loci assignment is not possible due to missing information (e.g. in teleost, [46,47]). The class II region genes are closely linked in humans and all other mammals examined, and variants at these genes are generally in strong linkage equilibrium [48]. Thus, the pattern observed for DRB loci should be a good indicator of the genetic variation in other class II genes and even some other less closely linked genes in the MHC [49-51].
Genes within the MHC involved in antigen presentation constitute the most polymorphic loci known in vertebrates [52,53]. The variability of the MHC-molecules is correlated with the diversity of the T-lymphocyte receptors which in turn determine the disease and parasite resistance of an organism and thus may influence the long-term survival probability of populations [54-57]. The antigen binding sites show high levels of variation not only in the number of alleles but also in the extent of sequence variation between alleles [58]. Under neutrality theory, the rate of synonymous nucleotide substitution (ds) is predicted to be larger (ds > dn) than the rate of non-synonymous substitution (dn) because non-synonymous substitutions change the amino acid composition and are thereby likely to be deleterious [59,60]. However, several studies demonstrate that the ABS display more non-synonymous than synonymous substitutions (dn > ds) ([61,62], reviewed by [19]). This cannot be explained by a higher mutation rate in this specific region [58,61,62]. The emerging general view is that the determinant role in shaping patterns of nucleotide diversity in MHC genes is balancing selection [19,59,60,63]. Balancing selection results not only in the maintenance of large numbers of alleles in populations, but also in greatly enhanced persistence of allelic diversity over extremely long time periods relative to neutral genetic variation [64], an observation termed 'trans-species evolution of polymorphism' [42]. The subsequent alteration in ABS allows binding of a diverse array of antigens [61,62,65].
Selection mechanisms
Two main types of balancing selection ('heterozygote advantage hypothesis' and 'frequency-dependence selection') have been suggested as important in retaining high levels of genetic diversity at the MHC in humans and vertebrates (reviewed by [19,64,66-68]).
In evaluating the evolutionary potential of 'heterozygote advantage' mechanism [69] a clear distinction between 'dominance' (heterozygote advantage in a broad sense) and 'overdominance' (heterozygote superiority) is necessary. The term 'dominance' refers to heterozygotes that are as resistant as the most resistant homozygote (if the allele A is associated with resistance, then the genotype AB is as resistant as AA (AB = AA)). In this case the heterozygote advantage could be due to masking of susceptible alleles. Whereas there is some support for this selection mechanism among experimental infection studies using mainly congenic mice it is clearly not sufficient to maintain high MHC variability [68,70,71]. 'Overdominance' seems to be the more efficient 'heterozygote advantage' mechanism promoting MHC diversity. In this case, heterozygotes are expected to have higher fitness than either parental homozygotes especially if confronted with multiple species or strains of pathogens or parasites (the genotype AB has a higher fitness than AA (AB>AA) and BB (AB>BB) [72]). The assumption is based on the theoretical background that heterozygous individuals should be able to detect and present a wider range of pathogen-driven antigens due to a larger number of different MHC molecules, hence increasing the relative fitness of MHC heterozygotes compared with homozygotes [60,73]. Thereby, two different 'overdominance' models have been suggested: a) 'symmetric overdominance' or 'symmetric balancing selection' [74], whereby all heterozygotes derive a similar selective advantage to homozygotes (= all heterozygous are selectively equivalent), and b) 'divergent allele advantage' [75]. In the later it is speculated that heterozygotes carrying more divergent allelic sequences have a selective advantage relative to individuals carrying relatively similar alleles by presenting a broader spectrum of antigens to the immune system. To the best of my knowledge, the 'divergent allele advantage'-hypothesis has never been applied in infectious disease studies but to explain the persistence of highly divergent MHC alleles over millions of years [75,76]. Richman and colleagues [77] used a theoretical model to confirm Wakeland's contention that MHC alleles are more divergent than expected under a model of balanced genetic polymorphism assuming selective equivalence of different alleles (but see [78-80]). Application of this model to MHC class IIb gene sequence data of deer mice (Peromyscus maniculatus) provided more support for the 'divergent allele advantage' model than for the 'symmetric overdominance' model for the maintenance of MHC polymorphism [77]. Thereby it is important to note that the analysis are based on assumptions of the coalescent models in which no gene conversion is allowed and therefore conclusions should be taken with care if such a mutational process is suspected [19].
The second mechanism, 'frequency-dependent selection', occurs when an allele or genotype is favoured at one frequency, but disadvantaged at another frequency [73,81,82]. Host-parasite dynamics are considered as an coevolutionary arms race. Pathogens adapt to infect the most common host genotype, leaving rare genotypes least infected [83]. If alleles are favoured when they are rare, but selected against when they are common, a balanced polymorphism results. Thus, the 'frequency-dependent selection' hypothesis is also described as 'rare-allele advantage hypothesis', 'Red Queen hypothesis' or 'moving-target hypothesis' [84-87]. The hypothesis assumes the following details. Rare (e.g. new) MHC alleles that are more resistant to parasites cause an advantage to the host, spread through the population and become common. This increases selection on parasites to evade recognition by these common alleles. As the parasite antigenicity changes, the relative fitness of the common host genotypes decreases and provides a selective advantage to other rare alleles. The time-lag nature of these antagonistic coevolutionary responses could lead to a cycling of fitness values of different alleles/genotypes in both hosts and pathogens, and result in the maintenance of high genetic diversity. As a consequence of these processes, pathogen-driven selection varies over time and may differ among habitats/environments within the range of a species, such that one host MHC-allele is favoured at a certain time in one environment and selected against in another. This should lead to varying spatiotemporal selection directions in space and time ('diversifying selection in space and time') [88,89,91]. So far, only one study investigated variation in MHC frequencies over time in a natural population to test the assumptions of the frequency-dependent model. Westerdahl and colleagues [92] compared the temporal changes in allele frequencies of 23 class I alleles and 23 neutral microsatellites of Great reed warblers (Acrocephalus arundinaceus) in nine consecutive cohorts. The MHC alleles showed on average slightly higher variation in temporal fluctuations compared to the microsatellite alleles. The frequency of two specific class I alleles varied more between cohorts than expected from random, whereas none of the neutral markers showed fluctuations exceeding the expectation from stochastic variation. The authors suggested that the variation in MHC allele frequencies between cohorts is not a result of demographic events, but rather an effect of selection favouring different MHC alleles in different years. However, Westerdahl and colleagues [92] did not include investigations of parasites or pathogens dynamics for explaining this pattern.
In addition, reproductive mechanisms such as disassortative mating and maternal-foetal interactions have been suggested as alternative or complementary mechanisms maintaining MHC diversity (summarised by [68,87,93-97]). MHC dissimilar mating preferences might act to increase offspring heterozygosity ('good-genes as heterozygosity hypothesis' [98]), to provide offspring with a moving target of MHC alleles as protection against pathogens which rapidly adopt to the parental genotypes ('rare-allele advantage hypothesis', 'Red Queen hypothesis', 'moving-target hypotheses' [73,81,82,87], to avoid inbreeding or genetic incompatibility ('genetic compatibility hypothesis' [99]) or to achieve an optimal MHC diversity in offspring with respect to parasite resistance ('allele counting hypothesis' [46,47] but see [91]).
The actual cue used in MHC-based mate choice is thought to be based on odour which allows to distinguish MHC-identity (summarised by [87,100-102]). Peptides/MHC complexes that are not retained at the cell surface but instead are released into the extracellular space might appear in the urine and other body secretions and be used for interindividual communication [103,104]. In mammals, the vomeronasal organ is essential in odour-based social recognition by detecting pheromones and other chemosignals that carry information about gender, sexual and social status, dominance hierarchies, and individualities, but it has been difficult to define the molecular nature of these chemosignals. Recent studies provided evidence that MHC class I peptides serve as chemosensory signals in the vomeronasal organ by which individual MHC genotype diversity can be used as a relatedness marker and may influence social behaviour [105].
These diverse functions and characteristics place genes of the MHC among the best candidates for studies of mechanisms and significance of molecular adaptation in vertebrates [19,52,93].
Evolutionary forces acting on the genetic diversity in coding and non-coding markers
The maintenance of genetic variation in natural populations in neutral parts of the genome under the non-selective evolutionary forces such as genetic drift and inbreeding depend not only on the number of individuals constituting a population, but also on the particular life history, the dispersal patterns (gene flow) and the breeding system of the species under study [106,107]. In contrast, the ability of natural populations to maintain genetic variation in functional genes depends on the selection pressures involved. Balancing selection is thought to counteract the effects of genetic drift and to retard the rate of fixation of alleles [58].
Evidence for selection maintaining high MHC diversity despite restricted variability in non-coding markers
There is increasing evidence for high MHC diversity due to balancing selection in species with otherwise restricted diversity in non-coding markers. For example, the San Nicolas Island fox (Urocyon littoralis dickeyi) is the most monomorphic sexually reproducing animal population yet reported with respect to variation in neutral genetic markers. No variation has been discovered in supposedly neutral hypervariable microsatellite loci and multilocus fingerprints, for which the probability of genetic identity is commonly <1 in several millions. Such low levels of variation imply lower resistance to pathogens, reduced fitness, and problems in distinguishing kin from non-kin. However, high MHC diversity is probably still maintained in this population by balancing selection. It is assumed that periodic selection has rescued genetic variation at the MHC and, potentially other fitness-related genes ([108] but see also [90]). Another example was found in Hawaiian honeycreepers (Vestiaria coccinea) [109]. Natural selection has maintained variation within the MHC while mitochondrial d-loop sequences and cytochrome b sequences were invariant and allozymes revealed low variability probably due to a genetic bottleneck. Moreover, in fragmented Malagasy gray mouse lemur (Microcebus murinus) populations, the number of DRB-alleles and the gene diversity were still high [110] but microsatellite and mitochondrial marker showed very low levels of polymorphism [111]. In the same study area, also the introduced black rat (Rattus rattus) revealed a similar pattern of genetic polymorphism: high levels of variability in the functional important MHC DRB marker [Sommer, unpublished data] in contrast to low mitochondrial d-loop variability (five haplotypes) [112].
These studies indicate that until a threshold level, genetic variation at the MHC might persist due to balancing selection despite low levels of variability shown by neutral markers. The results support the importance of balancing selection as a mechanism to maintain variation in natural populations and expose the difficulty of using neutral markers as surrogates for variation in fitness-related loci [108].
Processes leading to low variability in both coding and non-coding markers
The maintenance of polymorphism within populations is dependent on the product of selection intensity, mutation rate and effective population size [58,113,114]. Under certain circumstances strength of selection acting on MHC loci can be insufficient to maintain variation in small or fragmented populations for a long period of time. The effects of balancing selection and genetic drift on the genetic diversity of coding MHC class II (DQA) variability, neutral mitochondrial control region and microsatellite marker were recently investigated in 14 island and two mainland populations of the Australian bush rat, Rattus fuscipes [115,116]. Both neutral marker sets revealed high levels of genetic variability over-all but clear signs of genetic drift such as little to no diversity in the small island populations and extreme differentiation between the populations. In the MHC, higher levels of heterozygosity were observed on two of the islands than would be expected under neutrality, but genetic drift played a dominant role in the majority of island populations leading to a decrease in the number of MHC alleles.
Similarly, historical events such as bottlenecks and founder effects but also constraints of the mating system can be reflected in low numbers of MHC alleles (for example in an Asian lion population (Panthera leo persica) [117]; cheetahs (Aconyx jubatus) [118]; Malagasy giant jumping rats (Hypgeomys antimena) [119-121]; Malagasy western forest mouse (Macrotarsomys bastardi) [120]; common hamsters in the Netherlands (Cricetus cricetus) [122]; Scandinavian beavers (Castor fiber) [123]; Swedish moose (Alces alces) [124,125]; musk ox (Ovibos moschatus) [126]; Spanish ibex (Capra pyrenaica) [127]; island population of desert bighorn sheep (Ovis canadensis mexicana) [128]; Arabian oryx (Oryx leucoryx) [129]; South African bontebok (Damaliscus pygargus pygargus) [130]; Przewalski's horses (Equus przewalskii) [55]; Northern elephant seals (Mirounga angustirostris) [131], fin whales (Balaenoptera physlaus] [132], sei whales (Balaenoptera borealis) [132], and black robins (Petroica traversi) [133]). Under these circumstances, the power of genetic drift has been stronger than the power of selection. As predicted by theoretical models [135], the reduced MHC polymorphism is usually correlated with low genome-wide genetic variation [89]. For example, cheetahs (Aconyx jubatus) show low MHC diversity, which correlates with a genome-wide loss of diversity presumably due to a genetic bottleneck about 10,000 years ago [118]. Also Northern elephant seals (Mirounga angustirostris) which were hunted near to extinction in the 19th century lost most of the variability in allozymes, mitochondrial DNA, mini- and microsatellite loci and MHC class II loci [131,135].
Empirical support for the functional importance of MHC variability in pathogen and parasite resistance
Evidence for the functional importance of MHC variability and selective mechanisms derived from studies in humans or under experimental, laboratory conditions
While predictions of an association between MHC diversity and disease resistance are straightforward extensions of MHC theory, up to now, most of the empirical evidence has been derived from studies in humans or under experimental/laboratory conditions [19,67,136].
'MHC heterozygote advantage' [69] was indicated in humans by a slower progression to AIDS after HIV infection [137] and in a more effective clearance of hepatitis B viral infections [138]. In laboratory experiments, MHC-heterozygous mice showed reduced pathogenicity during bacterial and viral infection (streptococcus-induced lesions [139], Salmonella, Lysteria [70], Salmonella enterica, Theiler's virus [140]), an increased T-cell mediated immunity during lymphocytic choriomeningitis (LCM) infection [69] and they had a faster clearance rate of parasitic worms (Heligmosomoides polygyrus [141], Schistosoma mansoni [142]), than the average homozygote. Tumor incidence was lower and regression faster in heterozygous, rous sarcoma virus (RSV) infected chicken (Gallus domestica [143]). MHC class IIB heterozygotes had an increased survival rate in captive-raised fish, e.g. in Chinook salmon (Oncorhynchus tsawytscha) infected with a haematopoietic necrosis virus (HNV) [144] and in fluke-infected (Gyrodactilus turnbulli) Gila topminnows (Poeciliopis o. occidentalis) [54].
The 'frequency-dependent selection hypothesis' [81,82] is engaged by both mathematical models [73,145] and some empirical studies that show correlations between certain alleles and disease resistance in humans (e.g. malaria [146], Epstein-Barr-virus [147], hepatitis B [148], leprosy, tuberculosis [67], Heliobacter-infected gastric cancer [149]). In humans, a correlation was observed between some MHC class II haplotypes and the clinical severity of cestode infections (Echinococcus multilocularis) [150]. Certain MHC alleles also played a role in resistance/susceptibility to a fungal disease (Cryptococcus neoformans [151]), infections with gastrointestinal nematodes in lab mice (Trichinella spiralis [152,153], Nematospiroides dubius [154], Trichuris muris [155]) and in straightbred Scottish Blackface sheep (Ostertagia circumcincta [156,157]). Associations between resistance and MHC genotype was found in chicken suffering from infection with Marek's disease (a tumour disease caused by a herpes virus [158]). Experimental evidence for MHC-allele-specific resistance to Aeromonas salmonicida bacteria [57,159] and to the infectious salmon anaemia virus (ISAV) was found in captive-raised Atlantic salmon (Salmo salar [160]).
Evidence for the functional importance of MHC variability and selective mechanisms derived from studies in free-ranging animal populations in their natural environment
Whereas studies carried out under experimental or laboratory conditions can be better standardised to account for different parameters (e.g. in inbred congenic mice), they do not provide sufficient information to evaluate the ubiquity of pathogen-driven selective mechanisms acting in free-ranging animal populations in their natural habitat. Doing MHC research in wild vertebrates allows to test whether the results of studies on inbred congenic lab strains will hold in animals with a more diverse genetic background. Further, laboratory studies cannot reveal the effects of conditionally advantageous or deleterious alleles which will be discovered only in the presence of natural stress, such as spatially and temporally changes in climate, food availability, competition, and associated levels of parasitism [18,161]. Predicting the evolutionary potential of wild host populations in response to parasites requires at least a minimal understanding of the genetic basis for host resistance and heritability under field conditions, and the strength and mode of parasite-mediated selection [162]. Few studies have attempted to test for an association between MHC polymorphism and parasite resistance in wild populations under natural conditions [19]. Available information is summarised in Table 1.
Table 1 Evidence for pathogen-driven selection mechanisms in free-ranging vertebrate populations investigated in their natural environment.
Host species Host environment Country Infectious agent Heterozygote advantage Negative frequency-dependent selection Reference
Three-spined stickleback
(Gasterosteus aculeatus) Lakes and rivers Germany 14 species of macroparasites Supported in terms of a general diversity advantage; minimal parasitation at intermediate MHC class IIB diversity; population exposed to more diverse parasites had more different alleles. Not investigated [47]
Soay sheep
(Ovis aries) Large unmanaged population on an island Scotland Strongyle nematode Not supported; heterozygosity is not the critical factor determining mortality in lambs and yearlings. Common alleles (OLADRB 205, OLADRB 257) were associated with decreased lamb or yearling survivorship and a high incidence of parasitism; the rarer allele (OLADRB 263) with increased yearling survival. [56]
Gray mouse lemur
(Microcebus murinus) Littoral rain forest Madagascar Seventeen nematode species; separate data analysis for (most common) single and multiple infections. Not supported; heterozygosity was uncorrelated with infection status (being infected or not), the number of different nematodes per individual (NNI) as well as with the faecal egg counts (FEC, eggs/g faeces). The common allele Mimu-DRB*1 was more frequently found in infected individuals, in individuals with high number of different nematode species infections (NNI) and faecal egg counts (FEC); the rare alleles Mimu-DRB*6 and 10 were more prevalent in not infected individuals and in individuals with low NNI and FEC values. [174]
Yellow-necked mouse
(Apodemus flavicollis) Tree-dominated habitat Germany Eight nematode species; separate data analysis for (most common) single and multiple infections. Not supported; heterozygosity did neither influence the infection status (being infected or not), nor the number of different nematode infections (NNI) nor the individual faecal egg count (FEC, eggs/g faeces) values. Mice carrying allele Apfl-DRB*5 or the closely related allele Apfl-DRB*15 had an increased risk of being nematode infected and displayed higher FEC than individuals carrying other alleles; the allele Apfl-DRB*23 was associated with low FEC in separate analyses of the most common nematode. [173]
Hairy-footed gerbil
(Gerbillurus paeba) Dunefield of the Southern Kalahari Desert South Africa Two different cestode species, six different nematode species Not investigated Gepa-DRB*15 was only found in not infected mice. [172]
Striped mouse
(Rhabdomys pumilio) Dunefield of the Southern Kalahari Desert South Africa Eight different nematode species Supported; heterozygosity did influence the infection status (being infected or not) and the individual faecal egg count (FEC) value with higher values observed in homozygous individuals. The allele Rhpu-DRB*1 occurred more frequently in infected individuals and in individuals with high FEC values (high parasite load). In contrary, the allele Rhpu-DRB*8 occurred more often in individuals with low FEC values. [163]
Under field conditions, associations between MHC heterozygosity and resistance/susceptibility to parasite infections have only been found in the African striped mouse (Rhabdomys pumilio [163]) and in three-spined sticklebacks (Gasterosteus aculeatus [47,164]) which seem to possess up to six MHC class II loci. In the later, a modification of a simple heterozygote advantage was identified as within individual fish, intermediate, rather than maximal allele numbers were associated with minimal parasite load [47,164]. This is explained by the fact that MHC-genes are involved in the preservation of T-cells during thymic selection. At some point, increasing the number of MHC molecules expressed should cause a net loss of T-cells and therefore negatively affect the organism [165] (but see also Borghans and colleagues [166] who used a simulation approach which revealed that several hundred alleles would be required to cause such a net loss of T-cells). Different allele numbers can be produced by both heterozygosity at single loci and differences in MHC class II gene duplication numbers across haplotypes [167]. At the moment it is not clear whether or not this selection pattern of intermediate, rather than maximal allele numbers is confined to species with a relatively flexible genomic architecture such as sticklebacks and other teleosts with haplotype variation in their MHC locus duplication numbers, or whether it represents a more general feature that has been overlooked in previous studies [168]. In mammals, a flexible MHC genomic architecture, namely the appearance of multiple MHC class II DRB loci with variable loci numbers between individuals has been described in rhesus macaques (Macaca mulatta [169]) and in California sea lions (Zalophus californicus [170]). The later possess up to eight different DRB loci in variable configurations among individuals but with low levels of allelic variation per loci. Preliminary evidence suggested an association between a certain MHC genotype and urogenital cancer. In contrary to sticklebacks, no relationship between the total number of unique DRB genes and the presence of cancer has been identified [171]. A possible relationship between the number of MHC alleles and parasitic load in mammals was also investigated in hairy-footed gerbils (Gerbillurus paeba) which possess two functional DRB loci [172]. Here, individuals carrying three different MHC alleles had significantly higher faecal egg count values than individuals with four alleles [172]. This is in accord with the theoretical background which assumes that animals containing more MHC alleles than others should be able to recognise a larger spectrum of pathogen-derived antigens and consequently be infected by less parasite species and/or to be generally less intensively infected [69].
An association between certain MHC alleles and disease resistance or susceptibility was found in a free-ranging sheep population (Soay sheep, Ovis aries) where MHC variants appear to play a major role in protection against strongyle nematode invasion, the most prevalent gastrointestinal parasite found [56]. As expected by the assumptions of the 'negative frequency-dependent selection' ('rare-allele advantage hypothesis', 'Red Queen hypothesis', 'moving-target hypotheses') [73,81,82], the most common alleles OLADRB 205 and 257 (allele frequencies: 0.21–0.24) were associated with decreased lamb or yearling survivorship, whereas the rarer OLADRB 263 allele (allele frequency: 0.13) was associated with increased yearling survival (Table 1). Further evidence for the importance of certain MHC alleles and resistance or susceptibility to helminths was revealed in a common European rodent (yellow-necked mouse, Apodemus flavicollis [173]), in the two African rodent species mentioned before (Gerbillurus paeba [172], Rhabdomys pumilio [163]), and in a primate species (gray mouse lemur, Microcebus murinus [174]) (Table 1). Also in R. pumilio, it was the most common allele Rhpu-DRB*1 (allele frequency: 0.22) which occurred more frequently in infected individuals and in individuals with high faecal egg count values (indicating high parasite load) whereas the rare allele Rhpu-DRB*8 (allele frequency: 0.05) occurred more often in individuals with low FEC values (indicating low parasite load). Also in M. murinus, the common allele Mimu-DRB*1 (allele frequency: 0.33) was more frequently found in infected individuals and in individuals with a high number of different nematode species infections and faecal egg count values (eggs/g faeces) (indicating high parasite load), the rarer alleles Mimu-DRB*6 and *10 (allele frequencies: 0.11 and 0.06) were more prevalent in not infected individuals, in individuals with low number of different nematode species infections and faecal egg count values (indicating low parasite load). These examples demonstrate the frequency-dependence of selection between parasites and hosts in the form of a rare allele advantage in the host population.
Evaluating the relative importance of balancing selective mechanisms
Right now there is still much debate whether 'heterozygote advantage' or 'frequency dependent selection hypothesis' is most important for balancing selection [89]. Most studies investigating 'heterozygote advantage' compared the infectious disease outcomes of heterozygotes at a given MHC loci, as a group, to the outcomes of homozygotes at the same locus, as a group ('population heterozygote advantage' [70,175], examples see above) probably always due to restrictions in sample size. However, comparing the average performance of all heterozygotes against homozygotes, instead of using allele specific tests for 'overdominance', can not distinguish whether the observed advantage is due to 'dominance' or 'overdominance'. Grouping all homozygotes and all heterozygotes, respectively, circumvent tests of the original hypothesis namely the superiority of heterozygotes over either corresponding homozygote [69] (see paragraph 'Selection mechanisms' above) as this hypothesis is conditional on the alleles involved (and should be more precisely termed 'allele-specific overdominance' [175]). However, a theoretical model showed that under a very wide range of assumptions about the relationship between homozygote and heterozygote infectious risk, 'allele-specific overdominance' might be consistent with 'population heterozygote advantage', e.g. a 'population heterozygote advantage' might occur when the diversity of resistant alleles is sufficiently high and the diversity of susceptible alleles is sufficiently low [175]. But also the opposite might be true. Because of confounding effects of differences in frequencies of susceptible or resistant alleles, population level tests can, in a worse case, find a heterozygote advantage even when every heterozygote is at greater infection risk than either corresponding homozygote in allele-specific analyses [175]. Direct estimates of the allele-specific effects of heterozygosity relative to the corresponding homozygotes are rare. The most convincing experimental evidence for heterozygote advantage through 'allele-specific overdominance' derived from McClelland and colleagues [140] using co-infections with multiple pathogens in MHC-congenic mice with reciprocal resistance/susceptibility profiles (but the authors did not test for fitness consequences). In humans, the only studies that directly compare the outcomes of heterozygotes to those of homozygotes for the same alleles derived from investigations of autoimmune but not from infectious diseases (summarised by [70,175]).
As mentioned above allele-specific analyses were most often impossible due to restrictions in sample size. In humans, recently a new approach to circumvent this problem was proposed by classifying alleles to supertypes based on shared binding motifs [176,177]. Though it is clear that the highly polymorphic HLA genes play a crucial role in the immune response, their great diversity is a major obstacle in distinguishing HLA allele-specific effects and complicates the attribution of specific alleles with the outcome of diseases. Collecting samples of the size needed for definitive results is often not feasible. The biological relevance of a classification scheme based on functional binding specificities is supported by a growing body of evidence of cross-presentation of specific peptide-binding motifs by different HLA molecules. Trachtenberg and colleagues [176] investigated the usefulness of grouping HLA alleles to supertypes by their overlapping peptide binding specificities in explaining the association between HLA polymorphism and HIV disease progression. Their study indicated that HLA-supertypes are highly predictive of viral load. Consistent with the rare-allele advantage model the authors could show the advantage of a rare HLA supertype in progression of HIV [176]. After more intense studies of the implications and limits of supertypes in large human data sets, this classification approach of common functional traits may also provide tools for the MHC research in natural endangered populations, where high allelic diversity causes problems in obtaining sufficiently large statistical sample sizes. Caution must be taken not to miss the effects of new, rare alleles by clustering alleles in functional types.
As outlined before, two different 'overdominance' models have been suggested: a) 'symmetric overdominance' or 'symmetric balancing selection' [74], whereby all heterozygotes derive a similar selective advantage to homozygotes (= all heterozygous are selectively equivalent), and b) 'divergent allele advantage' [75]. As almost all heterozygote advantage studies were carried out on the population level, so far, no effort has been made to differentiate between these two 'overdominance' models in infectious disease studies. However, the 'divergent allele advantage' hypothesis has recently been considered in MHC-dependent mate choice studies [121,178]. In the African striped mouse (Rhabdomys pumilio) where associations between MHC heterozygosity and resistance/susceptibility to parasite infections have been found [163], no significant correlation between pairwise genetic distances of heterozygotes (a measurement for allelic divergence) and infection rates (faecal egg count: log-transformed number of eggs per gram faeces) was found (Froeschke & Sommer, unpublished data). Another point that need to be mentioned is that only a few studies in natural populations indicating correlative evidence for 'heterozygote advantage' combined MHC research with estimates of genome-wide diversity by using neutral markers and thus could rule out possible effects of genome-wide heterosis [47,137].
With respect to the 'frequency dependent selection hypothesis', evidence for pathogen-resistant/susceptible alleles/haplotypes is not equally available. So far, more alleles/haplotypes were found to be associated with susceptibility to disease [67] (but see Table 1). This bias could be simply due to over presentation of human studies, in which the emphasis has been on finding disease-allele associations [179]. But it is also in line with theoretical predictions of host-parasite coevolution. Susceptibility is expected to be more common, because fast evolution of the parasite is assumed to fuel the arms race between them and their hosts. For most pathogens it is valid to assume a higher evolutionary potential compared to that of the host, because generation times are usually much shorter or effective population sizes of pathogen populations are larger [18]. The human HLA-A11 allele, for example, confers resistance to infection with Epstein-Barr-virus only in populations where the allele is rare. In populations with high frequency of this allele, virus strains have fixed a mutation that prevents presentation of immunodominant epitopes by HLA-A11 molecules [147]. Also certain HLA alleles are associated with a slower progression of HIV if they are rare and have no advantage if they are common (summarised by [176]). This shows that a fast-evolving pathogen is able to adapt to host defence.
Evaluating the relative importance of both balancing selective mechanisms, so far, more evidence is available for the importance of specific MHC-alleles in parasite resistance or susceptibility. It is conceivable that a rare allele may have a high fitness and at the same time a constant advantage for heterozygotes. Both hypotheses may be in accord with each other and are not mutually exclusive. But as most studies deal with single viral, bacterial or parasitic agents it was suggested that studies combining two or more pathogens may increase the amount of evidence for heterozygote advantage [57,136,140] (but see [173,174]). De Boer and colleagues [180] studied the degree of MHC polymorphism arising when 'heterozygote advantage' is the only selection pressure by using mathematical models. The simulations revealed that 'heterozygote advantage' on its own is not sufficient to explain the high population diversity of the MHC. This would require that the fitness contributions of all alleles would be unrealistically similar. 'Heterozygote advantage' in pathogen resistance could, however, promote mating preferences for MHC-dissimilarity, which in turn drive high allelic diversity [52,68,70]. This could explain why MHC-heterozygous males have attributes important during sexual selection such as an increased antler development and body mass in white deer [181] and sexually attractive odour in stickleback fish [46]. In contrast, a study on sexual selection in pheasants (Phasianus colchicus) found that females prefer males with larger spurs, and that this sexually selected trait is associated with a particular MHC allele [182,183]. The overall view is emerging that although 'heterozygote advantage' is clearly an important selection pressure additional frequency-dependent selection pressure is required. A theoretical model by Hedrick [89] indicated that the selective force from pathogens, which vary in space and time ('diversifying selection in space and time'), could maintain the genetic polymorphism in MHC genes. Since evolving pathogens mainly evade presentation by the most common MHC alleles in the host population, they provide a selective pressure for a large variety of rare alleles. Host-parasite coevolution would be sufficient to explain the large degree of MHC polymorphism [145].
In ongoing studies assessing the evolutionary genetic details of vertebrate host-parasite relationships and evidence for frequency-dependent parasite-driven selection four patterns ought to be evident: (1) parasitism reduces host fitness, (2) MHC alleles differ in susceptibility, (3) alleles frequencies change according to (2), and (4) in the longer term dynamics should encompass frequency-dependent allele frequency fluctuations [17]. Whereas (1) and (2) indicate the potential for selection and have been shown in recent studies in wildlife populations (e.g. [184], this review), so far evidence for (3) and (4) is limited. Ongoing investigations of the parasite-driven selection mechanisms under natural conditions should focus on temporal variation of pathogens, host fitness attributes and allele frequencies to test whether allele frequencies change accordingly in a cycling pattern.
Functional differences of amino acid variation in the antigen binding sites
There is increasing evidence that pathogen escape from MHC-dependent immune system recognition may involve changes in only a few amino acids so that small binding-motif differences can lead to large differences in protection. Common mechanisms include changes in pathogen antigens (epitopes) that prevent binding (1) to the MHC-encoded cell surface glycoprotein or (2) to the T-cell receptor. (3) A third mechanism is molecular mimicry of host proteins that prevent T-cell receptor binding (T-cells that recognise host proteins are destroyed during thymic selection). For instance, a one-amino-acid difference in the antigen-binding region of the DRB*1302 allele abrogates its protection to malaria (summarised by [185,186]). In Malagasy mouse lemurs (Microcebus murinus), MHC-alleles associated with gastrointestinal nematode susceptibility (Mimu-DRB*1, *6 and *10) have unique amino acid motifs in the antigen binding sites (ABS) [174]. Mimu-DRB*1 associated with high parasite load differs from all other alleles by three unique amino acids, all of them located within the functional important ABS (aspartic acid in position 70, glutamic acid in position 71, lysine in position 74). Two of these ABS are mutated in Mimu-DRB*6 and *10 currently associated with low parasite load: the allele Mimu-DRB*6 has a unique motif at position 74 (glycine) and Mimu-DRB*10 at position 71 (methionine). In addition, only Mimu-DRB*6 and *10 possess the amino acid arginine located next to the ABS in position 78 [174] (position numbers after [62]). This indicates the functional differences of certain amino acids in the ABS and thus the influence of different amino acid compositions on parasite resistance.
So far, the molecular details of the interactions between helminth parasites and the intestinal components of the immune system are not as well understood as for viral or bacterial infections. However, huge progress was made in understanding the cellular and molecular mechanisms in the immune regulation by gastrointestinal helminth parasites in recent years. The recognition of gastrointestinal parasites and their antigens, and the initiation of the immune response occur in specialised lymph nodes in the epithelium of the gut wall, the so called Peyer's patches. In these Peyer's patches all cell types necessary for antigen presentation to CD4+ T-cells including MHC class II molecules are present. This activates a range of interacting processes against the parasite culminating in an inflammatory reaction in the intestinal mucosa and different effector mechanisms against the invading parasite (summarised in [186-191]).
Importance of MHC variability in conservation
Importance of adaptive genetic variability with respect to human impact
Human impact (e.g. habitat fragmentation, degradation, isolation, urbanisation, pollution) has diverse impacts on the ecology and genetics of both, vertebrate and parasite populations. It often causes a loss of genetic variation leading to short-term reduction of fitness components, and to an impaired ability to adapt to changing environments which in turn influences evolutionary outcomes [5,6,12,18,192]. Habitat degradation and climatic conditions are also crucial parameters in terms of distribution, transmission and developmental success of parasites and pathogens [18,192,193]. Such changes may have significant implications for outbreak patterns of pest species, the conservation of rare mammal species and their ecological functions, as well as associated veterinary and medical consequences for wildlife, lifestock and humans [194]. Rapid evolution (on the order of decades or shorter) has been supported by numerous examples from host-parasite systems, and it is now clear that pathogens can cause major shifts in the genetic composition of their hosts on short timescales [18,162,195]. Detectable changes in allele frequencies can occur between subsequent generations and can be a sensitive indicator for demographic changes in some species [196].
The effects of pollution on the MHC was investigated in the estuary killifish (Fundulus heteroclitus) [33]. Populations experiencing strong differences in antigenic challenges (PCB-contaminated versus unpolluted site) show significant differences in amino acid substitution patterns in a highly variable MHC class II B locus. However, whether MHC population profile differences represent direct effects of chemical toxicants or parasite-mediated selection need to be investigated [33]. The only study including an environmental variable such as habitat fragmentation in the analysis of associations of MHC-constitution and parasite burden was carried out in a subdivided mouse lemur population [174]. The work indicated that variation in MHC-allele frequencies in the fragments were linked to parasite load as certain alleles which differed in a few amino acids in the ABS from other alleles (see above) were associated to parasite resistance or susceptibility. Female mouse lemurs inhabiting the fragment with the highest parasite load had a lower fat deposition in the tail (important during the dry season) and therefore lower survival rates than populations of the three other fragments. In addition, the population size declined dramatically in recent years [197]. However, to clearly separate whether the higher parasite load in the respective fragment is due to the MHC-constitution of individuals inhabiting this fragment or due to other ecological factors associated with fragment size or degradation needs further investigations [174]. Nevertheless, the study suggests that the MHC-constitution might influence the long-term survival of small fragmented animal populations and indicates the functional importance of maintenance of MHC variability in declining or fragmented animal populations.
More studies in free ranging animal populations with respect to human impact are needed to allow more general conclusions on the importance of adaptive genetic variability in conservation. According to the theoretical background, temporal and spatial variation in the parasitic fauna will cause shifts of selective advantage of certain MHC-alleles in different areas changing over time. This should result in habitat- and climate-specific amino acid substitution patterns in the functional important ABS in relation to local pathogen-driven selective pressures. So far, empirical evidence for 'diversifying selection in space and time' is limited. Studies on the interaction between environmental conditions and the expression of genetic covariation (the so called genotype-environment interaction) might be an important avenue for future work. Genotype-environment interactions have commonly been found in live history traits when multiple environments were considered reflecting the fact that genes influencing a trait in one environment may not be important in a different one [198]. In this context, host and parasite movement among habitat fragments could be crucial to both parasite persistence, and the spread and maintenance of resistance alleles and thus to allow ongoing coevolutionary processes. The role of metapopulation dynamics in maintaining the diversity of host resistance genes can be a matter of concern in conservation genetics aiming at the preservation of both current patterns and ongoing processes. As contemporary evolution is influenced by complex interactions among population size, genetic variation, strength of selection, and gene flow, the overall goal in conservation genetics – maintenance of short-term local adaptations and preservation of long-term adaptive potential – might be a challenging task [32].
Relevance of MHC polymorphism for individual fitness and long-term persistence
Genetic variation at MHC loci is thought to be important for resistance against pathogens, thereby increasing individual fitness and thus the long-term survival of endangered species [60,73]. Several studies have reported decreased pathogen resistance among MHC homozygotes, or an increase in pathogen susceptibility in inbred individuals in general. However, a direct link between pathogen-mediated population decline and low MHC variation has been difficult to demonstrate in natural populations [49]. Recent studies indicated that although MHC allele numbers are low in many bottlenecked species most of them still indicate a high degree of divergence between alleles. Table 2 summarises the number of functional important MHC class II DRB exon 2 alleles and sequence diversity in some free-ranging vertebrate populations investigated in their natural habitat. The comparison indicates that also species with a low number of different MHC alleles, such as the critically endangered Malagasy Giant Jumping Rat (Hypogeomys antimena, 5 alleles) whose geographic range was recently restricted to less than 20,000 ha, still have high levels of nucleotide and amino acid divergence between MHC DRB-alleles while mitochondrial d-loop sequences revealed very low variability [199-201]. A similar picture was revealed in the Przewalski's horse (Equus przewalskii, 6 alleles [55]), in the Arabian oryx (Oryx leucoryx, 3 alleles [129]) and in the South African bontebok (Damaliscus pygargus pygargus, 6 alleles [130]) (Table 2). Considering the nonABS, the ratio between non-synonymous (dn) and synonymous (ds) substitutions was significantly smaller (dn < ds) than unity in some species (Fig. 1) which indicates purifying selection acting on these codons depending on their respective function [202]. In contrary, comparisons of non-synonymous (dn) and synonymous (ds) substitution rates in parts coding for the functional important ABS revealed a significantly higher rate of substitutions (dn > ds) which change the amino acid constitution in the ABS and thus increase the divergence between alleles in all species irrespective of the number of MHC alleles still present (Fig. 1).
Table 2 Number of MHC class II DRB exon 2 alleles (ca. 200 bp) and sequence diversity in free-ranging vertebrate populations investigated in their natural environment. In addition, for comparison of variability levels of species with a low number of MHC alleles two studies of captive-bred populations are included. * DRB-locus is duplicated. N = sample size.
Species Order Country Nr of alleles (N) Nr (%) of variable nucleotide positions Nr (%) of variable amino acid positions Nr (%) of amino acid differences between alleles References
Microcebus murinus primates Madagascar 14 (228) 71 (41.5) 31 (54.4) 5 (8.8) – 25 (43.9) [110,174]
Microcebus berthae primates Madagascar 9 (42) 46 (26.9) 24 (42.1) 3 (5.3) – 19 (33.3) Sommer et al., unpublished data
Apodemus sylvaticus rodentia Germany 38 (119) 71 (32.7) 38 (52.8) 2 (2.7) – 28 (38.8) [206]
Apodemus flavicollis rodentia Germany 27 (146) 49 (22.6) 28 (38.9) 1 (1.4) – 21 (29.2) [173]
Leopoldamys sabanus rodentia Borneo 28* (49) 85 (49.7) 39 (68.4) 4 (7.0) – 25 (43.9) Lenz et al., unpublished data
Gerbillurus paeba rodentia South Africa 34* (40) 68 (39.8) 33 (57.9) 1 (1.8) – 19 (33.3) [172]
Rhabdomys pumilio rodentia South Africa 20 (58) 43 (25.1) 23 (40.4) 1 (1.8) – 14 (24.6) [163]
Rattus rattus rodentia Madagascar 13 (58) 72 (40.7) 33 (55.9) 3 (5.1) – 26 (44.1) Sommer, unpublished data
Hypogeomys antimena rodentia Madagascar 5 (229) 37 (17.1) 19 (26.4) 6 (8.3) – 21 (29.2) [201]
Equus przewalskii perissodactyla captive-bred 6* (14) 52 (20.8) 29 (34.9) 1 (1.2) – 22 (26.5) [55]
Oryx leucoryx artiodactyla captive-bred 3 (57) 35 (14.8) 21 (26.9) 13 (16.7) – 17 (21.8) [129]
Damaliscus pygargus pygargus artiodactyla South Africa 6 (45) 21 (8.4) 14 (16.9) 1 (1.2) – 13 (15.7) [130]
Figure 1 Differences between the rates of non-synonymous (dn) and synonymous (ds) substitutions (= dn-ds, amino acid sequence-changing substitution rate) in antigen-binding sites (ABS, shaded bars) and nonantigen-binding sites (nonABS, black bars) of MHC class II DRB exon 2 alleles of the species included in Table 2. Abbreviations follow the MHC nomenclature [42], the first two letters of the genus and the species name are used.
The studies indicate that selection processes are able to maintain MHC polymorphism also under unfavourable conditions at least for a certain time which in turn might suggest that levels of variability in species with low numbers of MHC alleles might be sufficient to prevent immediate pathogen-mediated population decline. However, in such populations adaptive processes to changing conditions might be limited. An intriguing question is still what happens if new pathogens arise which differ from commonly encountered diseases by the respective populations. It is reasonable to assume that the maintenance and even more renewal of variation in functional important parts of the MHC such as in the antigen binding sites, either from mutation, recombination or immigration from other populations is an important genetic component in the cascade leading to an appropriate immune response, when combating new or coevolving virulent pathogens. It was suggested that the extremely low genetic variability in cheetahs (Aconyx jubatus) as a consequence of inbreeding depression due to a substantial bottleneck about 10,000 ago limits adaptation processes to temporary pathogens which might explain the high disease susceptibility nowadays [118,203]. As in other mainly endangered species, the proposed association cannot be tested as there are no genetically variable outbred cheetah population to compare with [21]. Samples from preserved bones or from museum specimens in the case of recently presumed bottlenecks would be necessary to directly assess historical levels of MHC variation and to evaluate the relative role of demographic changes in determining existing levels of genetic diversity at the MHC [127].
Concerning the role of MHC in conservation genetics, the potential functional role of background genes (non-MHC genes) in disease resistance should not be ignored. Human studies indicate that background genes might also play an important role in determining pathogen and parasite resistance, either by themselves or in a epistatic manner with MHC-genes (summarised by [179,204]. Many of the regulatory genes show high polymorphism and, for example, variation in the tumor necrosis factor gene promotor, cytokines such as interleucin receptor, γ-interferon receptor vitamin and D receptor has been associated to infectious diseases (summarised by [187,191,204]. Although evidence is accumulating that the MHC is one of the main factors controlling resistance to diseases [176,205] conservation genetics should focus on the preservation of both, MHC and genome-wide diversity. Thereby, how much MHC diversity is required to ensure long-term population viability remains a fundamental question in conservation genetics [133] and can only be investigated close follow-ups of the genetic and health status of declining populations.
Conclusion
The diverse functions and characteristics place genes of the MHC among the best candidates for studies of mechanisms and significance of molecular adaptation in vertebrates. In contrary to neutral markers, MHC variability reflects evolutionary relevant and adaptive processes within and between populations and is very suitable to investigate a wide range of open questions in evolutionary ecology and conservation.
The selective effects from different pathogens appear to be the major driving force in the maintenance of MHC variation. Evidence of balancing selection at MHC genes has been found at different temporal scales. Selection in the distant past has been documented as an excess of nonsynonymous to synonymous substitutions, and as trans-species polymorphism. Selection in the recent past has been determined by excess heterozygosity compared to neutral theory expectations, differences in FST-values compared to neutral theory, or excess linkage equilibrium. Selection in the current generation has been identified by measuring deviations from Hardy-Weinberg or random mating proportions, survival differences between homozygotes and heterozygotes, and correlations of disease resistance with MHC-allele or genotype [90]. Changes in certain amino acids in the functional important MHC-coded antigen binding sites and thus the amino acid compositions influence functional differences in pathogen and parasite resistance. It is reasonable to assume that the maintenance and even more renewal of variation in functional important parts of the MHC such as in the antigen binding sites, either from mutation, recombination or immigration from other populations is an important genetic component in the cascade leading to an appropriate immune response, when combating new or coevolving virulent pathogens and might be important in conservation genetics. Studies indicate the functional importance of MHC variability in pathogen and parasite resistance not only in humans or in model organisms under experimental, laboratory conditions where most of our current knowledge derived from, but also in wild animal populations investigated in their natural environment. Only field studies in free-ranging animal populations can reveal the effects of conditionally advantageous or deleterious alleles in the presence of natural stress (e.g. spatially and temporally changes in climate, food availability, and competition), associated levels of parasitism, and thus the ubiquity of pathogen-driven selective mechanisms and the importance of MHC diversity across taxa. The combination with an experimental approach under standardized laboratory conditions is needed to prove the causal relationships behind correlations observed in the field.
Right now it is not quite clear whether 'heterozygote advantage' or 'frequency-dependent selection hypothesis' is most important for balancing selection. Most studies investigating 'heterozygote advantage' compared the infectious disease outcomes of heterozygotes at a given MHC loci, as a group, to the outcomes of homozygotes at the same locus, as a group ('population heterozygote advantage'). Comparing the average performance of all heterozygotes against homozygotes, instead of using allele specific tests for 'overdominance' (= 'allele-specific overdominance') circumvent tests of the original hypothesis namely the superiority of heterozygotes over either corresponding homozygote as this hypothesis is conditional on the alleles involved. Allele-specific analyses were most often impossible due to restrictions in sample size. In humans, recently a new approach to circumvent this problem was proposed by classifying alleles to supertypes based on shared binding motifs [176,177]. After more intense studies of the implications and limits of supertypes in large human data sets, this classification approach of common functional traits may also provide tools for the MHC research in natural endangered populations, where high allelic diversity causes problems in obtaining sufficiently large statistical sample sizes. Caution must be taken not to miss the effects of new, rare alleles by clustering alleles in functional types. So far, more evidence is available for the importance of specific MHC-alleles in parasite resistance or susceptibility. It is conceivable that a rare allele may have a high fitness and at the same time a constant advantage for heterozygotes thus both modes of balancing selection may act synergistically to enhance the maintenance of polymorphism.
Ongoing investigations of the parasite-driven selection mechanisms under natural conditions should focus on temporal variation of pathogens, host fitness attributes and allele frequencies to test whether allele frequencies change accordingly in a cycling pattern. Assessing the immunogenetic status of a population relative to another experiencing different suites of antigenic challenges will help to increase our knowledge on the importance of adaptive genetic variability in free ranging animal populations with respect to human impact and the role of the MHC in evolutionary ecology and conservation.
Acknowledgements
I would like to thank the German Zoological Society (DZG), the study groups of ecology and evolutionary biology and especially B Koenig for the opportunity to present this paper at the 97th Annual Meeting in Rostock. I am grateful to two anonymous referees for very constructive comments and suggestions on a former version of this manuscript. I thank JU Ganzhorn for continual support and encouragement and my diploma and PhD students for their contributions and stimulating discussions on the functional importance of MHC variability. This work was supported by the German Research Foundation (DFG) and by the German Federal Ministry of Education and Research (BMBF).
==== Refs
Wahlberg N Moilanen A Hanski I Predicting the occurrence of endangered species in fragmented landscapes Science 1996 273 1536 1538
Meffe GK Carroll CG contributors Principles of Conservation Biology 1997 2 Sunderland, Massachusetts, Sinauer Associates
Peacock MM Smith AT The effects of habitat fragmentation on dispersal patterns, mating behavior and genetic variation in a pica (Ochotona princeps) metapopulation Oecol 1997 112 524 533 10.1007/s004420050341
Harrison S Hastings A Genetic and evolutionary consequences of a metapopulation structure Trends Ecol Evol 1996 11 180 183 10.1016/0169-5347(96)20008-4
Dudash MR Fenster CB Clarke G Inbreeding and outbreeding depression Genetics, Demography and Viability of Fragmented Populations 2000 Cambridge: Cambridge University Press
Keller L Waller D Inbreeding effects in wild populations Trends Ecol Evol 2002 17 230 241 10.1016/S0169-5347(02)02489-8
Primack RB Essentials of Conservation Biology 1993 Sunderland, Massachusetts, Sinauer Associates
Saccheri I Kuussaari M Kankare M Vikman P Fortelius W Hanski I Inbreeding and extinction in a butterfly metapopulation Nature 1998 392 491 494 10.1038/33136
Lande R Genetics and demography in biological conservation Science 1988 241 1455 1460 3420403
Brewer BA Lacy RC Foster ML Alaks G Inbreeding depression in insular and central populations of Peromyscus mice J Hered 1990 81 257 266 2273239
Lacy RC Importance of genetic variation to the viability of mammalian populations J Mamm 1997 78 320 335
Frankham R Ralls K Inbreeding leads to extinction Nature 1998 392 441 442 10.1038/33022
Biilsma R Bundgaard J Boerema AC Does inbreeding affect the extinction risk of small populations?: predictions from Drosophila J Evol Biol 2000 13 502 514 10.1046/j.1420-9101.2000.00177.x
Coltman DW Pilkington JG Smith JA Pemberton JM Parasite-mediated selection against inbred Soay sheep in a free-living, island population Evol 1999 53 1259 1267
Meagher S Genetic diversity and Capillaria hepatica (Nematoda) prevalence in Michigan deer mouse populations Evol 1999 53 1318 1324
Cassinello J Gomendio M Roldan ERS Relationship between coefficient of inbreeding and parasite burden in endangered gazelles Conserv Biol 2001 15 1171 1174 10.1046/j.1523-1739.2001.0150041171.x
Little TJ The evolutionary significance of parasitism: do parasite-driven genetic dynamics occur ex silico? J Evol Biol 2002 15 1 9 10.1046/j.1420-9101.2002.00366.x
Altizer S Harvell D Friedle E Rapid evolutionary dynamics and disease threats to biodiversity Trends Ecol Evol 2003 18 589 596 10.1016/j.tree.2003.08.013
Bernatchez L Landry C MHC studies in nonmodel vertebrates: what have we learned about natural selection in 15 years? J Evol Biol 2003 16 363 377 14635837 10.1046/j.1420-9101.2003.00531.x
Spielman D Brook BW Briscoe DA Frankham R Does inbreeding and loss of genetic diversity decrease disease resistance? Conserv Genet 2004 5 439 448 10.1023/B:COGE.0000041030.76598.cd
Frankham R Ballou JD Briscoe DA Introduction to Conservation Genetics 2002 Cambridge, Cambrige University Press 12126990
Avise JC Phylogeography 2000 Cambridge MA, Harvard University Press
Sunnucks P Efficient genetic markers for population biology Trends Ecol Evol 2000 15 199 203 10782134 10.1016/S0169-5347(00)01825-5
Lowe A Harris S Ashton P Ecological Genetics: Design, Analysis, and Application 2004 Oxford, UK, Blackwell Publishing Ltd
Blouin MS Parsons M Lacaille V Lotz S Use of microsatellite loci to classify individuals by relatedness Mol Ecol 1996 5 393 401 8688959 10.1046/j.1365-294X.1996.00094.x
Balloux F Lugon-Moulin N The estimation of population differentiation with microsatellite markers Mol Ecol 2002 11 155 165 11856418 10.1046/j.0962-1083.2001.01436.x
Brumfield RT Beerli P Nickerson DA Edwards SV The utility of single nucleotide polymorphisms in inferences of population history Trends Ecol Evol 2003 18 249 256 10.1016/S0169-5347(03)00018-1
Morin PA Luikart G Wayne RK the SNP workshop group SNPs in ecology, evolution and conservation Trends Ecol Evol 2004 19 208 216 10.1016/j.tree.2004.01.009
Meyers LA Bull JJ Fighting change with change: adaptive variation in an uncertain world Trends Ecol Evol 2002 17 551 557 10.1016/S0169-5347(02)02633-2
Van Tienderen PH de Haan AA van der Linden G Vosman B Biodiversity assessment using markers for ecologically important traits Trends Ecol Evol 2002 17 577 582 10.1016/S0169-5347(02)02624-1
Crandall KA Bininda-Emonds ORP Mace GM Wayne RK Considering evolutionary processes in evolutionary biology Trends Ecol Evol 2000 15 290 295 10856956 10.1016/S0169-5347(00)01876-0
Stockwell CA Hendry AP Kinnison MT Contemporary evolution meets conservation biology Trends Ecol Evol 2003 18 94 101 10.1016/S0169-5347(02)00044-7
Cohen S Strong positive selection and habitat-specific amino acid substitution patterns in MHC from an estuary fish under intense pollution stress Mol Biol Evol 2002 19 1870 1880 12411596
Coulson T Albon S Slate J Pemberton J Microsatellite loci reveal sex-dependent responses to inbreeding and outbreeding in red deer calves Evol 1999 53 1951 1960
Amos W Wilmer J Fullard K The influence of parental relatedness on reproductive success Proc R Soc Lond B 2001 268 2021 2027 10.1098/rspb.2001.1751
Acevedo-Whitehouse K Gulland F Greig D Amos W Disease susceptibility in California sea lions Nature 2003 422 35 12621424 10.1038/422035a
Slate J Kruuk L Marshall T Pemberton J Clutton-Brook T Inbreeding depression influences lifetime breeding success in a wild population of red deer (Cervus elaphus) Proc R Soc Lond B 2000 267 1657 1662 10.1098/rspb.2000.1192
Hansson B Westerberg L On the correlation between heterozygosity and fitness in natural populations Mol Ecol 2002 11 2467 2474 12453232 10.1046/j.1365-294X.2002.01644.x
Coltman DW Slate J Microsatellite measures of inbreeding: a meta-analysis Evol 2003 57 971 983
Balloux F Amos W Coulson T Does heterozygosity estimate inbreeding in real populations? Mol Ecol 2004 13 3021 3031 15367117 10.1111/j.1365-294X.2004.02318.x
DeWoody YD DeWoody JA On the estimation of genome-wide heterozygosity using molecular markers J Hered 2005 96 85 88 15618305 10.1093/jhered/esi017
Klein J Natural History of the Major Histocompatibility Complex 1986 New York, Wiley & Son
Klein J Horejsi V Immunology 1997 Oxford, Blackwell Science
Dengjel J Schoor O Fischer R Reich M Kraus M Autophagy promotes MHC class II presentation of peptides from intracellular source proteins Proc Nat Acad Sci USA 2005 102 7922 7927 15894616 10.1073/pnas.0501190102
Ohta T On the pattern of polymorphisms at major histocompatibility complex loci J Mol Evol 1998 46 633 638 9608045
Reusch TB Häberli MA Aeschlimann PB Milinski M Female sticklebacks count alleles in a strategy of sexual selection explaining MHC polymorphism Nature 2001 414 300 302 11713527 10.1038/35104547
Wegner KM Kalbe M Kurtz J Reusch TBH Milinski M Parasite selection for immunogenetic optimality Science 2003 301 1343 12958352 10.1126/science.1088293
Marsh SGE Parham P Barber LD The HLA Facts Book 2000 London, Academic Press
Gutierrez-Espeleta GA Hedrick PW Kalinowski ST Garrigan D Boyce WM Is the decline of desert bighorn sheep from infectious disease the result of low MHC variation? Heredity 2001 86 439 450 11520344 10.1046/j.1365-2540.2001.00853.x
Stenzel A Lu T Koch WA Hampe J Guenther SM De La Vega FM Krawczak M Schreiber S Patterns of linkage disequilibrium in the MHC region on human chromosome 6 p Hum Genet 2004 114 377 385 14740295 10.1007/s00439-003-1075-5
Kelley J Walter L Trowsdale J Comparative genomics of major histocompatibility complexes Immunogenetics 2005 56 683 695 15605248 10.1007/s00251-004-0717-7
Hedrick PW Evolutionary genetics at the major histocompatibility complex Am Nat 1994 143 945 964 10.1086/285643
Hughes A Hughes M Natural selection on the peptide-binding regions of major histocompatibility complex molecules Immunogenetics 1995 42 233 243 7672817 10.1007/BF00176440
Hedrick PW Kim TJ Parker KM Parasite resistance and genetic variation in the endangered Gila topminow Anim Cons 2001 4 103 109 10.1017/S1367943001001135
Hedrick PW Parker KM Miller EL Miller PS Major histocompatiblity complex variation in the endangered Przewalski's horse Genetics 2001 152 1701 1710 10430594
Paterson S Wilson K Pemberton JM Major histocompatibility complex variation associated with juvenile survival and parasite resistance in a large unmanaged ungulate population (Ovis aries L.) Evol 1998 95 3714 3719
Langefors A Lohm J Grahn M Andersen O von Schantz T Association between major histocombatibility complex class IIB alleles and resistance to Aeromonas salmonicida in Atlantic salmon Proc R Soc Lond B 2001 268 479 485 10.1098/rspb.2000.1378
Hughes AL Yeager M Natural selection at major histocompatibility complex loci of vertebrates Ann Rev Genet 1998 32 415 434 9928486 10.1146/annurev.genet.32.1.415
Hughes AL Nei M Pattern of nucleotide substitution at major histocompatibility complex class I loci reveals overdominant selection Nature 1988 335 167 170 3412472 10.1038/335167a0
Hughes AL Nei M Nucleotide substitution at major histocompatibility complex class II loci: Evidence for overdominant selection Proc Natl Acad Sci USA 1989 86 948 962 2536938
Brown JH Jardetzky TS Saper MA Samraoui B Bjorkman PJ Wiley DC A hypothetical model of foreign antigen binding site of class II histocompatibility molecules Nature 1988 332 845 850 3258651 10.1038/332845a0
Brown JH Jardetzky TS Gorga JC Stern LJ Urban RG Strominger JL Wiley DC Three-dimensional structure of the human class II histocompatibility antigen HLA-DR1 Nature 1993 364 33 39 8316295 10.1038/364033a0
Bergström T Gyllensten U Evolution of Mhc class II polymorphism: the rise and fall of class II gene function in primates Immunol Rev 1995 143 13 31 7558074
Richman A Evolution of balanced genetic polymorphism Mol Ecol 2000 9 1953 1963 11123608 10.1046/j.1365-294X.2000.01125.x
Janeway CA Travers P Immunologie 2002 Heidelberg, Berlin, Oxford, Spektrum Akademischer Verlag GmbH
Hedrick PW Balancing selection and the MHC Genetica 1999 104 207 214 10386384 10.1023/A:1026494212540
Jeffery KJ Bangham CR Do infectious diseases drive MHC diversity? Microb Infect 2000 2 1335 41 10.1016/S1286-4579(00)01287-9
Penn DJ The Scent of Genetic Compatibility: Sexual Selection and the Major Histocompatibility Complex Ethol 2002 108 1 21 10.1046/j.1439-0310.2002.00768.x
Doherty PC Zinkernagel RM Enhanced immunological surveillance in mice heterozygous at the H-2 gene complex Nature 1975 256 50 52 1079575 10.1038/256050a0
Penn DJ Damjanovich K Potts WK MHC heterozygosity confers a selective advantage against multi-strain infections Proc Natl Acad Sci 2002 99 11260 11264 12177415 10.1073/pnas.162006499
Gould SJ Hildreth JEK Booth AM The evolution of alloimmunity and the genesis of adaptive immunity Quart Rev Immunol 2004 79 359 382
Hughes AL Nei M Maintenance of MHC polymorphism Nature 1992 355 402 403 1734277 10.1038/355402b0
Takahata N Nei M Allelic genealogy under overdominant and frequency-dependent selection and polymorphism of major histocompatibility complex loci Genetics 1990 124 967 978 2323559
Takahata N A simple genealogical structure of strongly balanced allelic lines and trans-species evolution of polymorphism Proc Natl Acad Sci USA 1990 87 2419 2423 2320564
Wakeland EK Boehme S She JX Lu CC McIndoe RA Cheng I Ye Y Potts WK Ancestral polymorphism of MHC class II genes: divergent allele advantage Immunol Res 1990 9 115 122 2189934
She JX Wakeland EK Boehme S The generation and maintenance of MHC class II gene polymorphism in rodents Immunol Rev 1992 113 207 226 2180807
Richman AD Herrera LG Nash D MHC class II beta sequence diversity in the deer mouse (Peromyscus maniculatus): implications for models of balancing selection Mol Ecol 2001 10 2765 2773 11903890
Takahata N Satta Y Klein J Polymorphism and balancing selection at major histocompatibility complex loci Genetics 1992 130 925 938 1582567
Takahata N Satta Y Selection, convergence, and intragenic recombination in HLA diversity Genetica 1998 102–103 157 169 10.1023/A:1017029613342
Satta Y Effects of intra-locus recombination of HLA polymorphism Hereditas 1997 127 105 112 9420476 10.1111/j.1601-5223.1997.00105.x
Clarke B Kirby DR Maintenance of histocompatibility polymorphisms Nature 1966 211 999 1000 6007869
Bodmer W Evolutionary significance of the HLA-system Nature 1972 237 139 145 4113158 10.1038/237139a0
Lively CM Dybdahl MF Parasite adaptation to locally common host genotypes Nature 2000 405 679 681 10864323 10.1038/35015069
Jaenike J A hypothesis to account for sex within populations Evol Theory 1978 3 191 194
Ladle RJ Parasites and sex: catching the Red Queen Trends Ecol Evol 1992 7 405 408 10.1016/0169-5347(92)90021-3
Ebert D Hamilton WD Sex against virulence: the coevolution of parasitic diseases Trends Ecol Evol 1996 11 79 82 10.1016/0169-5347(96)81047-0
Penn DJ Potts WK The evolution of mating preferences and major histocompatibility complex genes Am Nat 1999 153 145 163 10.1086/303166
Nevo E Beiles A Selection for class II MHC heterozygosity by parasites in subterranean mole rats Experientia 1992 48 512 515 1351001 10.1007/BF01928177
Hedrick PW Pathogen resistance and genetic variation at MHC loci Evol 2002 56 1902 1908
Hedrick PW Foxy MHC selection story Hered 2004 93 237 238 10.1038/sj.hdy.6800539
Hedrick PW Comment on 'parasite selection for immunogenetic optimality" Science 2004 303 957 14963312 10.1126/science.1092163
Westerdahl H Hansson B Bensch S Hasselquist D Between-year variation of MHC allele frequencies in great reed warblers: selection or drift? J Evol Biol 2004 17 485 492 15149391 10.1111/j.1420-9101.2004.00711.x
Potts WK Wakeland EK Evolution of MHC genetic diversity: a tale of incest, pestilence and sexual preference Trends Genet 1993 9 408 412 8122307 10.1016/0168-9525(93)90103-O
Edwards SV Potts WK Smith TB and Wayne RK Polymorphism of genes in the major histocompatibility complex (MHC): implications for conservation genetics Molecular Genetic Approaches in Conservation 1996 New York, Oxford University Press 214 237
Edwards S Hedrick PW Evolution and ecology of MHC molecules: from genomics to sexual selection Trends Ecol Evol 1998 13 305 311 10.1016/S0169-5347(98)01416-5
Jordan WC Bruford MW New perspectives on mate choice and the MHC Hered 81 127 133
Penn DJ Fisher S Sniffing out genetic compatibility Biologist 2004
Zuk M Reproductive strategies and disease susceptibility: an evolutionary viewpoint Parasitol Today 1990 6 231 233 15463350 10.1016/0169-4758(90)90202-F
Brown JL Eklund A Kin recognition and the major histocompatibility complex: an integrative review Am Nat 1994 143 435 461 10.1086/285612
Penn DJ Potts WK How do major histocompatibility complex genes influence odor and mating preferences? Adv Immunol 1998 69 411 435 9646849
Yamazaki K Singer A Beauchamp GK Origin, functions and chemistry of H-2 regulated odorants Genetica 1998 104 235 240 10386388 10.1023/A:1026433925744
Eggert F Muller-Ruchholtz W Ferstl R Olfactory cues associated with the major histocompatibility complex Genetica 1999 104 191 197 10386382 10.1023/A:1026402531196
Singh PM Brown RE Roser B MHC antigens in urine as olfactory recognition cues Nature 1987 327 161 164 3574476 10.1038/327161a0
Singh P The present status of the 'carrier hypotheses' for chemosensory recognition of genetoc individuality Genetica 1998 104 231 233 10386387 10.1023/A:1026475118901
Leinders-Zufall T Brennan P Widmayer P Chandramani P Maul-Pavicic A MHC class I peptides as chemosensory signals in the vomeronasal organ Science 2004 306 1033 1037 15528444 10.1126/science.1102818
Sugg DW Chesser RK Dobson FS Hoogland JL Population genetics meets behavioural ecology Trends Ecol Evol 1996 11 338 342 10.1016/0169-5347(96)20050-3
Balloux F Goudet J Perrin N Breeding system and genetic variance in the monogamous, semi-social shrew, Crocidura russula Evol 1998 52 1230 1235
Aguilar A Roemer G Debenham S Binns M Garcelon D Wayne RK High MHC diversity maintained by balancing selection in an otherwise genetically monomorphic mammal Proc Natl Acad Sci USA 2004 101 3490 3494 14990802 10.1073/pnas.0306582101
Jarvi SI Tarr CL McIntosh CE Atkinson CT Fleischer RC Natural selection of the major histocompatibility complex (MHC) in Hawaiian honeycreepers (Drepanidinae) Mol Ecol 2004 13 2157 2168 15245391 10.1111/j.1365-294X.2004.02228.x
Schad J Sommer S Ganzhorn JU MHC variability of a small lemur in the littoral forest fragments of southeastern Madagascar Conserv Genet 2004 5 299 309 10.1023/B:COGE.0000031137.50239.d3
Hapke A Population genetics and differentiation of species of the genus of mouse lemurs, Microcebus (E. Geoffroy St. Hilaire, 1828) and dwarf lemurs, Cheirogaleus (E. Geoffrey St. Hilaire, 1812) in southeastern Madagascar PhD Thesis 2004 University Hamburg, Germany
Hingston M Goodman SM Ganzhorn JU Sommer S Reconstruction of the colonization of southern Madagascar by introduced Rattus rattus J Biogeography 2005 32 1549 1559 10.1111/j.1365-2699.2005.01311.x
Robertson A Selection for heterozygotes in small populations Genetics 1962 47 1291 1300 13974344
Nevo E Kirzher V Beiles A Karol A Selection versus random drift: long-term polymorphism persistence in small populations (evidence and modelling) Phil Trans R Soc Lond B 1997 352 381 389 10.1098/rstb.1997.0028
Seddon JM Baverstock PR Variation on islands: major histocombatibility complex (Mhc) polymorphism in population of the Australian bush rat Mol Ecol 1999 8 2071 2079 10632858 10.1046/j.1365-294x.1999.00822.x
Hinten G Harriss F Rossetto M Baverstock PR Genetic variation and island biogeography: microsatellite and mitochondrial DNA variation in island populations of the Australian bush rat, Rattus fuscipes greyii Cons Genet 2004 4 759 778 10.1023/B:COGE.0000006113.58749.ac
Yuhki N O'Brien SJ DNA variation at the mammalian major histocompatibility complex reflects genomic diversity and population history Proc Natl Acad Sci 1990 87 836 840 1967831
O'Brien SJ Wildt DE Goldman D Merril CR Bush M The cheetah is depauperate in genetic variation Science 1985 221 459 462 17755482
Sommer S Tichy H Major histocompatibility complex (MHC) class II polymorphism and paternity in the monogamous Hypogeomys antinema, the endangered, largest endemic Malagasy rodent Mol Ecol 1999 8 1259 1272 10447867 10.1046/j.1365-294X.1999.00687.x
Sommer S Schwab D Ganzhorn JU MHC diversity of endemic Malagasy rodents in relation to range contraction and social system Behav Ecol Sociobiol 2002 51 214 221 10.1007/s00265-001-0432-4
Sommer S MHC-dependent mate choice in a monogamous rodent Behav Ecol Sociobiol 2005 58 181 189 10.1007/s00265-005-0909-7
Smulders MJM Snoek LB Booy G Vosman B Complete loss of MHC genetic diversity in the Common hamster (Cricetus cricetus) population in the Netherlands. Consequences of conservation strategies Conserv Genet 2003 4 441 451 10.1023/A:1024767114707
Ellegren H Hartman G Johansson M Andersson L Major histocompatibility complex monomorphism and low levels of DNA fingerprinting variability in a reintroduced and rapidly expanding population of beavers Proc Natl Acad Sci, USA 1993 90 8150 8153 8367476
Mikko S Andersson L Low major histocompatiblity complex class II diversity in European and North American moose Proc Natl Acad Sci USA 1995 92 4259 4263 7753793
Ellegren H Mikko S Wallin K Andersson L Limited polymorphism at major histocompatibility complex (MHC) loci in the Swedish moose A. alces Mol Ecol 1996 5 3 9 9147694
Mikko S Roed K Schmutz S Andersson L Monomorphism and polymorphism at MHC DRB loci in domestic and wild ruminants Immunol Rev 1999 167 169 178 10319259
Amills M Jiménez N Jordana J Riccardi A Low diversity in the major histocompatibility complex class II DRB1 gene of the Spanish ibex (Capra pyrenaica) Hered 2004 93 266 272 10.1038/sj.hdy.6800499
Hedrick PW Gutierrez-Espeleta GA Lee RN Founder effects in an island population of bighorn sheep Mol Ecol 2001 10 851 857 11348494 10.1046/j.1365-294X.2001.01243.x
Hedrick PW Parker KM Gutiérrez-Espeleta GA Rattink A Lievers K Major histocompatibility complex variation in the Arabian oryx Evol 2000 54 2145 2151
Van der Walt JM Nel LH Hoelzel AR Characterization of major histocompatiblity complex DRB diversity in the endemic South African antilope Damaliscus pygargus : a comparison in two subspecies with different demographic histories Mol Ecol 2001 10 1679 1688 11472536 10.1046/j.0962-1083.2001.01321.x
Hoelzel AR Stephens JC O'Brien SJ Molecular genetic diversity and evolution at the MHC DQB locus in four species of pinnipeds Mol Biol Evol 1999 16 611 618 10335654
Trowsdale J Groves V Arnason A Limited MHC polymorphism in whales Immunogenetics 1989 41 19 24 2562799
Miller HC Lambert DM Genetic drift outweighs balancing selection in shaping post-bottleneck major histocompatiblity complex variation in New Zealand robins (Petroicidae) Mol Ecol 2004 13 3709 3721 15548285 10.1111/j.1365-294X.2004.02368.x
Nei M Maruyama T Chakraborty R The bottleneck effect and genetic variability in populations Evol 1975 29 1 10
Weber DS Stewart BS Schienman J Lehman N Major histocompatibility complex variation at three class II loci in the northern elephant seal Mol Ecol 2004 13 711 718 14871373 10.1111/j.1365-294X.2004.02095.x
Wegner KM Kalbe M Schaschl H Reusch TBH Parasites and individual major histocompatiblity diversity – an optimal choice Microb Infect 2004 6 1110 1116 10.1016/j.micinf.2004.05.025
Carrington M Nelson GW Martin MP Kissner T Vlahov D Goedert JJ Kaslow R Buchbinder S Hoots K O'Brien SJ HLA and HIV-1: heterozygote advantage and B*35-Cw*04 disadvantage Science 1999 283 1748 1752 10073943 10.1126/science.283.5408.1748
Thursz MR Thomas HC Greenwood BM Hill AV Heterozygote advantage for HLA class-II type in hepatitis B virus infection Nat Genet 1997 17 11 12 9288086 10.1038/ng0997-11
Chen CY Cohen SA Zaleski MB Albini B Genetic control of streptococcus-induced hepatic granulomatous lesions in mice Immunogenetics 1992 36 28 32 1587550 10.1007/BF00209289
McClelland EE Penn DJ Potts WK Major histocompatibility complex heterozygote superiority during coinfection Infect Immun 2003 71 2079 2086 12654829 10.1128/IAI.71.4.2079-2086.2003
Behnke JM Wahid FN Immunological relationships during primary infection with Heligmosomoides polygyrus (Nematospiroides dubius): H-2 linked genes determine worm survival Parasitol 1991 103 157 164
Sher A Hieny S James S Mechanisms of protective immunity against S. mansoni infection in mice vaccinated with irradiated cercariae: VI. Influence of the major histocompatibility complex Parasit Immunol 1984 6 319 328
Senseney HL Briles WE Abplanalp H Taylor RL Allelic complementation between MHC haplotypes B-Q and B-17 increases regression of Rous sarcomas Poultry Sci 2000 79 1736 1740 11194035
Arkush KD Giese AR Mendonca HL McBirde AM Marty GD Hedrick PW Resistance to three pathogens in the endangered winter-run chinook salomon (Oncorhynchus tshawytscha): effects of inbreeding and major histocompatiblity complex genotypes Can J Fish Aquat Sci 2002 59 966 975 10.1139/f02-066
Borghans JAM Beltman JB De Boer RJ MHC polymorphism under host-pathogen coevolution Immunogenetics 2004 55 732 739 14722687 10.1007/s00251-003-0630-5
Hill AVS Allsopp CEM Kwiatkowski D Anstey NM Tumasi P Rowe PA Bennett S Brewster D McMichael AJ Greenwood BM Common West-African HLA antigens are associated with protection from severe malaria Nature 1991 352 595 600 1865923 10.1038/352595a0
Decamposlima PO Gavioli R Zhang QJ Wallace LE Dolcetti R Rowe M Rickinson AB Masucci MG HLA-A11 epitope loss isolates of Epstein-Barr-virus from a highly A11+ population Science 2003 260 98 100
Thursz MR Kwiatkowski D Allsopp CE Association between an MHC class II allele and clearance of hepatitis B virus in the Gambia New Engl J Med 1995 332 1065 1069 7898524 10.1056/NEJM199504203321604
Li Z Chen D Zhang C Li Y Cao B Ning T Zhao Y You W Ke Y HLA polymorphisms are associated with Heliobacter pylori infected gastric cancer in a high risk population, China Immunogenetics 2005 56 781 787 15650879 10.1007/s00251-004-0723-9
Godot V Harraga S Beurton I Resistance/susceptibility to Echinococcus multilocularis infection and cytocine profile in humans. II. Influence of the HLA B8, DR3, DQ2 haplotype Clin Exp Immunol 2000 121 491 498 10971516 10.1046/j.1365-2249.2000.01309.x
McClelland EE Granger DL Potts WK Major histocompatiblity complex-dependent susceptibility to Cryptococcus neoformans in mice Infect Immun 2003 71 4815 4817 12874366 10.1128/IAI.71.8.4815-4817.2003
Wassom DL Brooks BO Babisch JG David CS A gene mapping between the S and D regions of the H-2 complex influences resistance to Trichinella spiralis infections in mice J Immunogenetics 1983 10 371 378 6417245
Wassom DL Krco CJ David CS I-E expression and susceptibility to parasite infection Immunol Today 1987 8 39 43 10.1016/0167-5699(87)90236-2
Enriquez FJ Brooks BO Cypess RH David CS Wassom DL Nematospiroides dubius : two H-2-linked genes influence levels of resistance to infection in mice Exp Parasitol 1988 67 221 226 3142789 10.1016/0014-4894(88)90069-0
Else KJ Wakelin D Wassom DL Hauda KM The influence of genes mapping within the major histocompatibility complex on resistance to Trichuris muris infections in mice Parasitol 1990 101 61 67
Schwaiger FW Gostomski D Stear M Duncan JL McKellar QA Epplen JT Buitkamp J An ovine major histocompatibility complex DRB1 allele is associated with low faecal egg count following natural, predominantly Ostertagia circumcincta infection Int J Parasitol 1995 25 815 822 7558567 10.1016/0020-7519(94)00216-B
Buitkamp J Filmether P Stear MJ Epplen JT Class I and class II major histocompatiblity complex alleles are associated with faecal egg counts following natural, predominantly Ostertagia circumcincta infection Parasitol Res 1996 82 693 696 8897503 10.1007/s004360050187
Briles WE Stone HA Cole RK Mareks-disease – effects of B-histocompatiblity alloalleles in resistant and susceptible chicken Science 1977 195 193 195 831269
Lohm J Grahn M Langefors A Andersen O Storset A von Schantz T Experimental evidence for major histocompatibility complex-allele-specific resistance to a bacterial infection Proc R Soc Lond B 2002 269 2029 2033 10.1098/rspb.2002.2114
Grimholt U Larsen S Nordmo R Midtlyng P Kjoeglum S Storset A Saebo S Stet RJM MHC polymorphism and disease resistance in Atlantic salmon (Salmo salar); facing pathogens with single expressed major histocompatibilty class I and class II loci Immunogenetics 2003 55 210 219 12811427 10.1007/s00251-003-0567-8
Meagher S Penn DJ Potts WK Male – male competition magnifies inbreeding depression in wild house mice Proc Natl Acad Sci USA 2000 97 3324 3329 10716731 10.1073/pnas.060284797
Little TJ Ebert D Temporal patterns of genetic variation for resistance and infectivity in a Daphnia-microparasite system Evol 2001 55 1146 1152
Froeschke G Sommer S MHC Class II DRB constitution and parasite load in the striped mouse, Rhabdomys pumilio, in the Southern Kalahari Mol Biol Evol 2005 22 1254 1259 15703235 10.1093/molbev/msi112
Wegner KM Reusch TBH Kalbe M Multiple infections drive major histocompatibility complex polymorphism in the wild J Evol Biol 2003 16 224 232 14635861 10.1046/j.1420-9101.2003.00519.x
Lawlor DA Zemmour J Ennis PD Parham P Evolution of class-I MHC genes and proteins: from natural selection to thymic selection Ann Rev Immunol 1990 8 23 63 2188663 10.1146/annurev.iy.08.040190.000323
Borghans JAM Noest AJ De Boer RJ Thymic selection does not limit the individual MHC diversity Europ J Immunol 2003 33 3353 3358 10.1002/eji.200324365
Malaga-Trillo E Zaleska-Rutczynska Z McAndrew B Vincek V Figuera F Sültmann H Klein J Linkage relationships and haplotype polymorphism among cichlid MHC class II B loci Genetics 1998 149 1527 1537 9649539
Reusch TBH Schaschl H Wegner KM Recent duplication and inter-locus gene conversion in major histocompatibility class II genes in a teleost, the three-spined stickleback Immunogenetics 2004 56 427 437 15322775 10.1007/s00251-004-0704-z
Doxiadis GG Otting N de Groot NG Bontrop RE Differential evolutionary MHC class II strategies in humans and rhesus macaques: relevance for biomedical studies Immunol Rev 2001 183 76 85 11782248 10.1034/j.1600-065x.2001.1830106.x
Bowen L Aldridge BM Stott JL Gulland F Woo J Class II multiformity generated by variable MHC-DRB region configurations in the California sea lion (Zalophus californianus) Immunogenetics 2004 56 12 27 14997355 10.1007/s00251-004-0655-4
Bowen L Aldridge BM DeLong R Melin S Buckles EL An immunogenetic basis for the high prevalence of urogenital cancer in a free-ranging population of California seas lions (Zalophus californianus) Immunogenetics 2005 56 846 848 15650877 10.1007/s00251-004-0757-z
Harf R Sommer S Association between MHC Class II DRB alleles and parasite load in the hairy-footed gerbil, Gerbillurus paeba, in the Southern Kalahari Mol Ecol 2005 14 85 91 15643953 10.1111/j.1365-294X.2004.02402.x
Meyer-Lucht Y Sommer S MHC diversity and the association to nematode parasitism in the yellow necked mouse (Apodemus flavicollis) Mol Ecol 2005 14 2233 2243 15910340 10.1111/j.1365-294X.2005.02557.x
Schad J Ganzhorn JU Sommer S MHC constitution and parasite burden in the Malagasy mouse lemur, Microcebus murinus Evol 2005 59 439 450
Lipsitch M Bergstrom CT Antia R Effect of human leukocyte antigen heterozygosity on infectious disease outcome: the need for allele-specific measures BMC Med Gen 2003 4 2 10.1186/1471-2350-4-2
Trachtenberg E Korber B Sollars C Kepler TB Advantage of rare HLA supertypes in HIV disease progression Nature Medicine 2003 9 928 935 12819779 10.1038/nm893
Lund O Nielsen M Kesmir C Petersen AG Lundegaard C Definition of supertypes for HLA molecules using clustering of specificity matrices Immunogenetics 2004 55 797 810 14963618 10.1007/s00251-004-0647-4
Landry C Garant D Duchesne P Bernatchez L 'Good genes as heterozygosity': the major histocompatibility complex and mate choice in Atlantic salmon (Salmo salar) Proc R Soc Lond B 2001 268 1279 1285 10.1098/rspb.2001.1659
Apanius V Penn D Slev PR Ruff LR The nature of selection on the major histocompatibility complex Crit Rev Immunol 1997 17 179 224 9094452
De Boer RJ Borghans JAM Van Boven M Kesmir C Weissing FJ Heterozygote advantage fails to explain the high degree of polymorphism of the MHC Immunogenetics 2004 55 725 731 14722686 10.1007/s00251-003-0629-y
Ditchkoff SS Lochmiller RL Masters RE Hoofer SR Van den Bussche RA Major-histocompatibility-complex-associated variation in secondary sexual traits of white-tailed deer (Odocoileus virginianus): evidence for good-genes advertisement Evol 2001 55 616 625
Von Schantz T Göransson G Andersson G Fröberg I Grahn M Helgí A Wittzell H Female choice selects for a viability-based male trait in pheasants Nature 1989 337 166 169 2911350 10.1038/337166a0
Von Schantz T Wittzell H Goransson G Grahn M Persson K MHC genotype and male ornamentation: genetic evidence for the Hamilton-Zuk model Proc R Soc 1996 263 265 271
Tompkins DM Begon M Parasites can regulate wildlife populations Parasitol 1999 15 311 313
Frank SA Immunology and the Evolution of Infectious Disease 2002 Princeton, Princeton University Press
Summers K McKeon S Sellars J Parasitic exploitation as an engine of diversity Biol Rev 2003 78 639 675 14700394 10.1017/S146479310300616X
Behnke JM Iraki F Menge D Baker RL Gibson J Wakelin D Chasing the genes that control resistance to gastrointestinal nematodes J Helminth 2003 77 99 109 12756063 10.1079/JOH2003174
Garside P Kennedy MW Wakelin D Lawrence CE Immunopathology of intestinal helminth infection Parasite Immunology 2000 22 605 612 11123752 10.1046/j.1365-3024.2000.00344.x
Macdonald TT The mucosal immune system Parasite Immunology 2003 25 235 246 12969442 10.1046/j.1365-3024.2003.00632.x
Maizels RM Yazdanbakhsh M Immune regulation by helminth parasites: cellular and molecular mechanisms Nat Rev 2003 3 733 744
Quinnell RJ Genetics of susceptibility to human helminth infection Int J 2003 33 1219 1231
Mitchell SE Rogers ES Little TJ Read AF Host-parasite and genotype-by-environment interactions: temperature modifies potential for selection by a sterilizing pathogen Evol 2005 59 70 80
Woolhouse MEJ Haydon DT Antia R Emerging pathogens: the epidemiology and evolution of species jumps Trends Ecol Evol 2005 20 238 244 16701375 10.1016/j.tree.2005.02.009
Poulin R The functional importance of parasites in animal communities: many roles at many levels? Int J Parasitol 1999 29 903 914 10480727 10.1016/S0020-7519(99)00045-4
Dybdahl MF Lively CM Host parasite coevolution: evidence for rare advantage and time-lagged selection in a natural population Evol 1998 52 1057 1066
Luikart G Cornuet JM Allendorf FW Temporal changes in allele frequencies provide estimates of population bottleneck size Conserv Biol 1999 13 523 530 10.1046/j.1523-1739.1999.98133.x
Rüdel N Consequences of degradation and fragmentation of Malagasy littoral rain forests on gray mouse lemur populations (Microcebus murinus) Diploma Thesis 2004 University Hamburg, Germany
Sgrò CM Hoffmann AA Genetic correlations, tradeoffs and environmental variation Hered 2004 93 241 248 10.1038/sj.hdy.6800532
Sommer S Hommen U Modelling the effects of life-history traits and changing ecological conditions on the population dynamics and persistence of the endangered Malagasy giant jumping rat (Hypogeomys antimena) Anim Cons 2000 4 333 343 10.1017/S1367943000001050
Sommer S Toto Volahy A Seal US A population and habitat viability assessment for the highly endangered Giant Jumping Rat (Hypogeomys antimena), the largest endemic rodent of Madagascar Anim Cons 2002 5 263 273
Sommer S Effects of habitat fragmentation and changes of dispersal behaviour after a recent population decline on the genetic variability of non-coding and coding DNA of a monogamous Malagasy rodent Mol Ecol 2003 12 2845 2851 12969486 10.1046/j.1365-294X.2003.01906.x
Nei M Gojobori T Simple methods for estimating the number of synonymous and non-synonymous nucleotide substitutions Mol Biol Evol 1986 3 418 426 3444411
O'Brien SJ Evermann JF Interactive influence of infectious disease and genetic diversity in natural populations Trends Ecol Evol 1988 3 254 259 10.1016/0169-5347(88)90058-4
Hill AVS The immunogenetics of human infectious diseases Ann Rev Immunol 1998 16 593 617 9597143 10.1146/annurev.immunol.16.1.593
Bontrop R Watkins DI MHC polymorphism: AIDS susceptibility in non-human primates Trends Immunol 2005 26 227 233 15797514 10.1016/j.it.2005.02.003
Musolf K Meyer-Lucht Y Sommer S Evolution of MHC-DRB class II polymorphism in the genus Apodemus and a comparison of DRB sequences within the family Muridae (Mammalia: Rodentia) Immunogenetics 2004 56 420 426 15351919 10.1007/s00251-004-0715-9
|
16253139
|
PMC1282568
|
CC BY
|
2021-01-04 16:36:39
|
no
|
Retrovirology. 2005 Oct 27; 2:65
|
latin-1
|
Retrovirology
| 2,005 |
10.1186/1742-4690-2-65
|
oa_comm
|
==== Front
Aust New Zealand Health PolicyAustralia and New Zealand Health Policy1743-8462BioMed Central London 1743-8462-2-231621267410.1186/1743-8462-2-23ReviewAustralian health policy on access to medical care for refugees and asylum seekers Correa-Velez Ignacio [email protected] Sandra M [email protected] Sara J [email protected] Refugee Health Research Centre, Faculty of Health Sciences, La Trobe University, VICTORIA 3086 Australia2 Advocacy Coordinator, Extractive Industries, Oxfam Australia2005 9 10 2005 2 23 23 18 4 2005 9 10 2005 Copyright © 2005 Correa-Velez et al; licensee BioMed Central Ltd.2005Correa-Velez et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Since the tightening of Australian policy for protection visa applicants began in the 1990s, access to health care has been increasingly restricted to asylum seekers on a range of different visa types. This paper summarises those legislative changes and discusses their implications for health policy relating to refugees and asylum seekers in Australia. Of particular concern are asylum seekers on Bridging Visas with no work rights and no access to Medicare. The paper examines several key questions: What is the current state of play, in terms of health screening and medical care policies, for asylum seekers and refugees? Relatedly, how has current policy changed from that of the past? How does Australia compare with other countries in relation to health policy for asylum seekers and refugees? These questions are addressed with the aim of providing a clear description of the current situation concerning Australian health policy on access to medical care for asylum seekers and refugees. Issues concerning lack of access to appropriate health care and related services are raised, ethical and practical issues are explored, and current policy gaps are investigated.
==== Body
Introduction
Australian health care policy regarding entitlements to medical care for refugees and asylum seekers is complex. On the one hand, health care policy for refugees entering Australia on the offshore humanitarian program is comprehensive, entitling refugees to Medicare, early health assessment, specialised torture and trauma services, and access to the same services as other Australians. On the other hand, health care policy for refugees and asylum seekers who have entered Australia in an unauthorised manner, and who are on a range of visa types, is fragmented. About 40% of asylum seekers living in the community have no rights to access medical care [1].
Health policy for refugee and asylum seekers is directly tied to immigration policy and visa types and this in turn is a complex and rapidly changing field of play. In this paper, we review current health care policies for refugees and asylum seekers in Australia with a focus on those areas of policy gaps that result in a lack of access to medical and health care for some, and less optimal access to care for others. We begin by briefly discussing the definitions of who is deemed to be a refugee and who is deemed to be an asylum seeker, for health care policies vary within and between these two categories accordingly. Second, we describe in some detail, the current health policy for asylum seekers and refugees within Australia's onshore program compared to health policy for refugees who have come to Australia through the offshore program. Third, we provide a broad comparison of Australian health care policies for refugees and asylum seekers to those of the United Kingdom, Canada and New Zealand, countries with comparable public health systems. Finally, we discuss the current areas where policy on access to health care is in conflict between States and Territories and the Commonwealth and where Australian policy may be in breach of various human rights conventions.
Who is a refugee or asylum seeker?
Under the United Nations 1951 Convention and 1967 Protocol relating to the Status of Refugees (the Refugee Convention), a refugee is a person who '...owing to a well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group, or political opinion, is outside the country of his nationality, and is unable to or, owing to such fear, is unwilling to avail himself of the protection of that country' [2]. An asylum seeker is a person who has fled their own country and has sought sanctuary in a second state [3]. They then apply to be recognized as a bona fide refugee and to receive legal protection and material assistance that that status implies [4].
Although the 1951 Refugee Convention does not deal specifically with asylum seekers, two of its Articles are particularly relevant to the issues of access to health care considered in this paper. First, Article 33 (refoulement) states that no refugee shall be expelled or returned to where his/her life is threatened. Second, Article 31 prohibits punishment or penalties for entry to a state when they come from a territory where life or freedom is threatened. Other aspects of asylum law are particularly relevant to the issue of health policy discussed here. It is important to note that it is not illegal to seek asylum in any country and that the basic provisions related to humane treatment and basic rights apply to asylum seekers. Additionally, International Human Rights Law recognizes the right of all individuals to an adequate standard of living [5].
At the end of 2004 there were an estimated 19.2 million asylum seekers, refugees and other people of concern to the UNHCR [6]. In the same year, 676,000 first instance or appeal applications for asylum were submitted in 143 countries; only 3,276 asylum applications were lodged in Australia, compared to over 500,000 in other industrialized countries [6]. In relation to refugee intake, Australia is one of 16 countries currently participating in the UNHCR-facilitated resettlement program [7]. In 2004, 84,809 refugees were resettled in these countries of which 13,030 were resettled in Australia [8].
Australian health policy for refugees and asylum seekers: Humanitarian program, health screening and health care access
In order to understand health policies for refugees and asylum seekers, it is important to briefly describe Australia's humanitarian program and the current legislation on visa status and protection. This is because entitlements to health care vary by visa status, and legislation concerning visa status is in a constant state of flux.
Australia's humanitarian program
Australia's humanitarian program for refugees and others with humanitarian needs includes two components [9]: offshore resettlement for people overseas, and onshore protection for those who are already in Australia, arrived on temporary visas (e.g. visitor or student visas) or in an unauthorised manner, and are seeking Australia's protection. These two components and their visa categories are shown in Table 1.
Table 1 Australia's immigration humanitarian program
Visa categories and subcategories
Permanent offshore humanitarian visa categories Temporary offshore humanitarian visa categories
Offshore resettlement • Refugee: for people who are subject to persecution in their home country and who are in need of resettlement
•Special Humanitarian Program (SHP): for people who are outside their home country and are subject to substantial discrimination amounting to gross violation of human rights in their home country • Secondary Movement Relocation: temporary humanitarian visa (THV) to people who have moved from a safe first country of asylum to another country before applying to enter Australia (5-year visa)
• Secondary Movement Offshore Entry: temporary visa to people who arrived without authorisation in Australia at a place outside Australia's migration zone and have moved from a safe first country of asylum (3-year visa)
Form of arrival and visas granted
Authorised arrivals Unauthorised arrivals
Onshore protection Most authorised arrivals who subsequently apply for protection receive a bridging visa. The bridging visa allows them to remain in the community while the Protection Visa application is processed. Those who are found to be refugees are granted Permanent Protection Visa (PPV). Unauthorised arrivals are placed in immigration detention until granted a Protection Visa or removed from Australia. Those who are found to be refugees are granted a 3-year Temporary Protection Visa (TPV).
Sources: [9, 48]
Policy on health screening for offshore and onshore applicants
Australia's health policy for humanitarian entrants begins pre-arrival. Those who apply under the offshore resettlement program must satisfy the health requirement [10] specified in the Migration Regulations. This health requirement, which is set by the Department of Immigration and Multicultural and Indigenous Affairs (DIMIA) on advice from the Commonwealth Department of Health and Ageing, is designed to minimise public health risks to the Australian community, regulate public expenditure on health and community services, and maintain access to health and other services for Australian residents. In general, the health assessment involves a medical examination, a radiological examination to test for Tuberculosis (children under 11 are generally exempt), and HIV/AIDS testing (for all applicants aged 15 or older). For some applicants screening for Hepatitis B is mandatory. Other tests may be requested by a Medical Officer of the Commonwealth.
For those who apply under the onshore protection program, the health screening varies depending on their specific circumstances. In general, those who arrive in Australia on a 3-month temporary visa – or less – are not required to have formal health examinations prior to their arrival (some exceptions apply [11]). Those arriving on a greater than 3 months temporary visa may be required to undergo formal medical and radiological examinations prior to arrival, depending on the level of health risk of their country of origin, their age, and the purpose of their stay (e.g. likely to enter a hospital, health care area, classroom, preschool or childcare centres).
All unauthorised arrivals who are applying for Australia's protection undergo health screening soon after their arrival at immigration detention centres. "Where any serious communicable diseases are suspected or confirmed, formal notification procedures are followed with Commonwealth and State/Territory health authorities" [12] (p. 2). In general, those who make an onshore application for protection, whether arriving on an authorised or unauthorised manner, are required to undergo medical and radiological examinations before the granting of the visa.
According to DIMIA, the only health condition that formally precludes the grant of a visa is active or untreated tuberculosis (TB) [10]. Those applicants whose TB has been treated or those with a previous but currently non-active TB, are required to make a Health Undertaking. In other words, they are required to contact the Health Undertaking Service on arrival in Australia, and to report to State or Territory health authorities for follow-up assessment [10]. The Health Undertaking also applies for a pregnant applicant who has not had the chest X-ray as part of the standard health examination (although this is not commonly extended to persons from high risk TB countries).
All other health conditions are assessed on a case by case basis taking into account the risk to public health, the estimated costs, and the resource use impact on the Australian community. The final authority as to whether the health requirement is met rests with the Medical Officer of the Commonwealth (MOC). The visa processing officers are required to accept the opinion of the MOC. However, where there are compelling factors, the Minister's delegate processing officer may waive the health requirement for refugees and other humanitarian visa applicants. If the applicant or any member of their family fails to pass the health requirement, the entire family group can be denied a visa. They may however, be referred to other resettlement countries such as the United States, who have different health screening guidelines and selection criteria. For those applying under the onshore protection program, they may be denied a protection visa should they fail the health requirement under the same above considerations.
Policy on medical care for offshore and onshore applicants
While Australian policies on resettlement of humanitarian refugees are arguably among the best, compared to other United Nations High Commissioner for Refugees (UNHCR) resettlement countries, its policies on the treatment of onshore protection applicants have been strongly criticised by human rights organisations, scholars, government members and others for their denial of basic human rights guaranteed under the 1951 Refugee Convention [2]. A brief description of the medical care entitlements granted to different categories of offshore humanitarian entrants and onshore protection applicants is given in Table 2.
Table 2 Health entitlements commonly granted to refugees and asylum seekers in Australia
Humanitarian program Circumstances Entitlements
Offshore resettlement Refugees who hold permanent offshore humanitarian visas Same eligibility for Medicare and Health Care Card, including Pharmaceutical Benefits Scheme, as other permanent residents. Eligible for Early Health Assessment and Intervention Program and torture/trauma services.
Refugees who hold temporary offshore humanitarian visas (THV) Able to gain access to Medicare and Health Care Cards, including Pharmaceutical Benefits Scheme; eligible for referral to the Early Health Assessment and Intervention Program and torture/trauma services.
Onshoore protection Authorised arrivals
Authorised arrivals who have been found to be refugees and are granted permanent protection visa (PPV) Same eligibility for Medicare and Health Care Card, including Pharmaceutical Benefits Scheme, as other permanent residents.
Authorised arrivals who applied for protection within 45 days of their arrival in Australia and are awaiting primary decision on their application Eligible for Medicare a
Authorised arrivals who have been in Australia for 45 days or more before they applied for protection No access to Medicare a
Authorised arrivals who have appealed or are about to appeal to the Refugee Review Tribunal or the Administrative Appeals Tribunal after their primary protection application has been refused No access to Medicare a
Authorised arrivals who are appealing to the Minister of Immigration after being found not to be refugees at the review stage No access to Medicare a
Unauthorised arrivals
Unauthorised arrivals who have been found to be refugees and are granted temporary protection visa (TPV) Access to Medicare benefits and Health Care Card; eligibility for torture and trauma counselling
Unauthorised arrivals who are applying for protection and are in immigration detention Access to health care through health professionals contracted by the private company in charge of the detention centres
Unauthorised arrivals who have been in mandatory detention, are subsequently released into the community, and have an outstanding visa application No access to Medicare
Others
Asylum seekers who hold or held a TPV or a THV, have had their application for further protection finally refused, have exhausted all legal options to remain in Australia and are making arrangements for departure (return pending visa) Eligible for Medicare access through work rights, Health Care Card, including Pharmaceutical Benefits Scheme, torture/trauma counselling, Maternity Allowance, maternity immunisation
Asylum seekers who have been held in detention, do not have any outstanding visa applications or litigation, who cannot currently reasonably be removed from Australia, and who agree in writing to cooperate with their removal from Australia when advised that they must leave (removal pending bridging visa) b Eligible for Medicare access through work rights, Health Care Card, including Pharmaceutical Benefits Scheme, torture/trauma counselling, Maternity Allowance, maternity immunisation
Sources: [15, 49-54]
a Under certain circumstances, these individuals may be eligible for the Asylum Seeker Assistance Scheme (ASAS) [21]
b It can apply to both unauthorised arrivals held in detention or authorised arrivals who are in detention after breaching their visa conditions
Offshore resettlement program
On arrival to Australia, all refugees and special humanitarian program (SHP) entrants under the offshore resettlement program receive all the entitlements granted to Australian permanent residents, including access to social security benefits (Centrelink), Medicare, education and training (including 510 hours of English language lessons), and employment services. In addition, they receive settlement support through the Integrated Humanitarian Settlement Strategy (IHSS), which is carried out by agencies responsible for a range of settlement services. The IHSS includes [13]: initial information and orientation assistance on the essential services available, accommodation support, household formation support, early health assessment and intervention, and community support. The early health assessment takes places within the first 12 months of arrival, and involves information on available health services, physical health and psychological/psychosocial assessments, and referral to other health services where required (including torture/trauma services). A current policy gap however, relates to early health screening for infectious diseases.
In the 1970's and 80's, all new arrivals under the offshore program were offered health screening and assessments which included immunization updates, screening for parasitic and infectious diseases. This was more easily accomplished as new arrivals spent the first few weeks within a migrant hostel. However, in the 1990's Commonwealth policy shifted so that new arrivals were settled directly into the community along with the expectation that early health assessments would be carried out by local general practitioners. Thus, there is at present no comprehensive policy on health screening for new arrivals. Further, there is considerable debate about the merits of bringing in compulsory health screening or whether this is better carried out within a more holistic approach allowing people to access screening throughout the resettlement period [14].
Onshore protection program
Our discussion on Australia's policy on medical care within the onshore protection program begins with those asylum seekers in detention centres. The immigration detention policy, introduced in 1992, has been maintained with bipartisan support in Parliament [15]. The Commonwealth, not the States and Territories, is responsible for the health care of detainees. The private company, Global Solutions Limited (Australia) Pty Ltd, contracted by the Commonwealth to manage the detention centres, is responsible for providing healthcare to all detainees. It employs nurses, general practitioners and psychologists [16]. According to the Commonwealth Government, detained asylum seekers have '24 hour medical services, dental services, culturally responsive physical and psychological health services' [15]. Inquiries carried out by the Human Rights and Equal Opportunity Commission (HREOC) into the detention centres, however, report serious concerns in terms of the adequacy and quality of health care services available for the detainees, particularly the failure to diagnose and treat torture/trauma survivors [17,18].
Although attention is often focussed on Australia's detention policies, Australian policy can also be criticised for infringing the rights of asylum seekers through denial of access to appropriate health care and related services [19]. The three case studies below illustrate the complexities of current policies relating to the provision of health care for asylum seekers and Temporary Protection Visa (TPV) holders in Australia (NB: names have been changed).
Case 1
Ms Jayawardene and her husband have been living in Australia as asylum seekers for the past year. They are currently on Bridging Visa E, which allows them to live in the community but gives them no work rights, no entitlement to Medicare and no access to any social security benefits. Ms Jayawardene has recently given birth. She had no prenatal care. Only after a caseworker from a community based organisation (CBO) advocated on her behalf, was she hospitalised for the birth. She was, however, discharged early because she was unable to pay the fees. The hospital only agreed to waive the fees after much negotiation, carried out on behalf of the couple by the CBO.
Case 2
Mr Hassan sought asylum in Australia two years ago, after having spent several years living in a number of different countries. Mr Hassan spent one year in mandatory detention before he was deemed a refugee and granted a three-year Temporary Protection Visa (TPV), which confers on him a range of entitlements, including Medicare. He recently consulted a local general practitioner to seek treatment for a stomach ailment, but was unable to properly communicate about his symptoms due to his lack of English language skills. Although Mr Hassan is entitled to Medicare benefits, he is not entitled to other settlement services, such as adequate English-language tuition or fee-free interpreting services. The lack of access to an interpreter created a barrier to Mr Hassan receiving a proper health assessment.
Case 3
Mr Ahmed has a serious chronic heart condition that requires medication. He waited six months for a primary decision on his application for protection before he qualified for the Asylum Seeker Assistance Scheme (ASAS)-funded by DIMIA- which assists with basic needs, including health care costs. Two months ago, however, Mr Ahmed lost access to ASAS because he appealed a negative Refugee Review Tribunal (RRT) decision on his application for refugee status. He is now unable to pay for essential medication and relies on charitable assistance to provide his medication.
These three case studies illustrate a number of key areas where access to health care is either denied or inadequately provided to people residing in Australia, and raise a number of questions about health policies for onshore protection applicants and TPV holders.
Community based onshore protection applicants
Access to health care for asylum seekers living in the community depends on two elements: the type of bridging visa they hold and the particular stage of their application [20].
Until 1997, most onshore protection applicants were granted a bridging visa after their original visa (e.g. a visitor visa or student visa) had expired. With some exemptions, this bridging visa conferred work rights, which consequently entitled most holders of the visa to Medicare, the Australian Government health insurance scheme. On 1 July 1997, however, the government introduced work rights regulations for asylum seekers who applied for a protection visa (PV) on or after that date. According to these regulations, 'a bridging visa with work rights may be granted to people who have been in Australia for fewer than 45 days in the 12 months before they lodge a PV application [21]'. In other words, those asylum seekers who have been in Australia for 45 days or more in the 12 months before they make the PV application are not permitted to work, and therefore, do not have access to Medicare. It has been reported that this requirement has resulted in approximately 40 per cent of community based asylum seekers being denied Medicare [1]. Usually, these asylum seekers are granted a Bridging Visa E.
From 1 January 2001, PV applicants who 'have ever applied (on- or off-shore) for a parent visa, irrespective of whether their application is on-hand, finally determined or withdrawn, have no access to Medicare [21]'. For individual asylum seekers with no work rights, the 'no work' condition may only be changed if DIMIA has not made a primary decision within six months of the lodgement of their application and the applicant demonstrates a compelling need to work [21]. In some circumstances, work rights and Medicare may also be available to asylum seekers if they are the spouse, child or parent of an Australian citizen or permanent resident.
In an attempt to fill this welfare gap, DIMIA administers the Asylum Seeker Assistance Scheme (ASAS) through contractual arrangements with the Australian Red Cross. Operating since 1993, ASAS was designed to financially assist asylum seekers who were unable to meet their most basic needs (e.g. food, accommodation, health care). ASAS recipients can receive assistance with health care costs and referral to counselling services.
Initially, all asylum seekers were entitled to ASAS at the primary and review stages of their application, if a decision had not been made within six months. The eligibility criteria for ASAS, however, have been gradually restricted. Currently, to be eligible for ASAS, asylum seekers must be in financial hardship and: have lodged a valid PV application for which a primary decision has not been made within six months; not be in detention and must hold a bridging or other visa; not have been released from detention on an undertaking of support; not be eligible for either Commonwealth or overseas government income support; and not be a spouse, de facto or sponsored fiancé(e) of a permanent resident [21]. There are some exemptions to the above criteria, such as: unaccompanied minors or elderly persons (over 65 years); parents with children under 18 years of age; women with high risk pregnancies; and persons who are unable to work as a result of a disability, illness or torture/trauma [21].
ASAS may also be extended to RRT applicants who are unable to meet their basic needs and who have no adequate support [21]. There have been a few cases in which DIMIA has used its discretionary powers to continue providing certain asylum seekers, who are experiencing exceptional welfare circumstances, with 'special payments' while their cases are at the post-RRT stage [19]. However, once the RRT makes a decision on the application, most asylum seekers cease to be eligible for ASAS [21].
In March 2005, the Minister for Immigration and Multicultural and Indigenous Affairs announced the creation of the Removal Pending Bridging Visa (RPBV), intended to 'provide greater ability to manage the cases of long term detainees who are awaiting removal' [22]. The visa allows a relatively small number of asylum seekers – those who have exhausted all legal appeals, but who cannot be reasonably removed from Australia, and who are willing to agree in writing that they will leave Australia when instructed to do so by the government – to enter the community. These asylum seekers are eligible for Medicare, through work rights, and have access to Health Care Cards, including the Pharmaceutical Benefits Scheme, maternity care and torture/trauma counselling.
More recently (21 June 2005) the Federal Government introduced further changes to detention which will include, among others, the release from detention of children and their families into the community [23]. Under these arrangements, DIMIA is funding the Red Cross to develop a national system that will provide material and health care support for those asylum seekers released into the community [24].
These recent changes, including the introduction of the RPBV, add another significant layer of complexity to an already complicated system. Questions raised by the RPBV visa and its potential implications for asylum seekers are discussed later in the paper.
Temporary Protection Visa holders
According to current migration regulations, refugees on TPV have access to Medicare, are eligible for referral to the Early Health Assessment and Intervention Program (EHAI), and for torture/trauma counselling [25]. TPV holders, however, face serious challenges when accessing health care services. For instance, those over the age of 18 are not eligible for government-funded English language classes. Low levels of English language skills cause social isolation, unemployment (or low paid employment and therefore inability to afford medications), higher incidence of occupational health and safety issues, and obvious difficulties when accessing health care services. TPV holders are not allowed to access the federally funded Telephone Interpreting Service or the various health interpreting services designed to assist health workers who treat people from non-English Speaking backgrounds. In addition, amputees on TPV are not able to access the Commonwealth Rehabilitation Service [26].
In early July 2004, the government introduced new measures for TPV and Temporary Humanitarian Visa (THV) holders [27]. Briefly, these changes include a reintegration assistance package for TPV and THV holders who are prepared to return to their countries, the introduction of a new 'Return Pending Visa', which will allow those TPV holders whose applications for further protection has been rejected to stay in Australia for 18 months while they prepare for departure; and changes to enable TPV and THV holders to apply for a range of non-humanitarian onshore visas to live permanently in Australia. These recent developments, however, do not represent major changes in health care access and other entitlements to TPV holders.
Despite the presence or absence of health policies relating to entitlements to medical care, there are a range of barriers to accessing health care services, common to most refugees, independent of their visa status. Some of the barriers that have been identified are:
• Long waiting times particularly in using Emergency Departments of public hospitals,
• Cost of services, especially for specialist health care and in relation to public dental health services,
• Lack of information and confusion about the health system, particularly the difference between public and private and entitlements,
• Lack of interpreters and female physicians, particularly in rural areas,
• Absence of bulk billing services in rural areas,
• Instances of discrimination,
• Other settlement needs taking precedence, particularly in cases where refugees are employed in casual or temporary work with no leave provisions,
• Lack of specialist care, particularly in regional areas [28].
These barriers are of particular concern as most refugees arrive in Australia in poor health and are likely to face particular health challenges in the resettlement period. These health challenges stem from previous experiences of torture and trauma and from having lived in poor social and economic prior to arrival, all of which impact on their well-being during the resettlement period [14].
A comparison of health policy for refugees and asylum seekers across selected industrialised countries
Australia is one of sixteen countries who accept refugees for resettlement under the UNHCR resettlement program [29]. Full entitlement to Medicare and additional entitlements to health care for special needs, including oral health and mental health, are also provided by the Commonwealth, State and Territory governments [21]. In fact, Australian health care policy for newly arrived humanitarian entrants is comprehensive and one of the best compared to other resettlement countries [30]. When it comes to onshore protection applicants, however, at the time of this writing Australia compares poorly to other resettlement countries.
While entitlements to social benefits and health care vary in the other resettlement countries, at the time of this writing, none of these countries deny asylum seekers the right to basic medical care. Indeed, a brief review of the refugee and asylum seeker health policies of the United Kingdom (UK), Canada and New Zealand indicates that Australian policy is comparatively lacking in this respect (See Table 3).
Table 3 Country comparison of health entitlements for refugees and asylum seekers according to status
Australia New Zealand Canada UK
Refugee Medicare NZ Health System coverage Provincial Health Cover: Canada's Federal Health system (comprised of Federal and Provincial contributions) includes comprehensive health cover, including hospital, physician, surgical-dental and specialist cover. NHS Coverage
Asylum Seeker in detention Health care through the private company in charge of the detention centres a N/A: Asylum Seekers rarely held in detention for longer than 48 hours b No overarching coverage: individuals are assessed on a case-by-case basis b National Health Service (NHS) Coverage. Coverage includes: primary & secondary care, free prescriptions, dental services, coverage of travel costs to/from hospital b
Asylum Seeker awaiting primary decision of refugee status Depends on when visa applied for: If application submitted within 45 days of arrival, then individuals have access to Medicare but no translating services, Early Health Assessment and Intervention, and torture/trauma counselling,
If application submitted after 45 days of arrival, then no Medicare access NZ Health System coverage. Coverage includes all services, such as: Primary & secondary care, co-payment of pharmaceuticals, specialist referral and coverage, cost offsets for 'frequent users' of medical services, hospital & accident cover, dental, mental, maternity and sexual health care. Interim Federal Health (IFH) Program coverage. Coverage includes essential health services for the treatment & prevention of serious medical conditions, essential prescription medications, contraception, prenatal care, obstetrical care, Immigration Medical Exam, emergency dental service. NHS coverage
Asylum Seeker appealing a negative Refugee Review Tribunal (or equivalent) outcome No access to Medicare c NZ Health System coverage NB: may take some time to receive Community Services Card, necessary for accessing a General Practitioner (GP) IFH Program Coverage NHS Coverage
Refused Asylum Seeker who has exhausted all appeals No access to Medicare d NZ Health System coverage NB: may take some time to receive Community Services Card, necessary for accessing a GP IFH Program Coverage Primary & urgent care only e
Sources: [14, 31-33, 36, 37, 49, 55-60]
a Detention is mandatory for all 'unauthorised arrivals'.
b Detention is not mandatory for 'unauthorised arrivals'.
c Under certain circumstances, these individuals may be eligible for ASAS. See: DIMIA, 2003.
d The majority of individuals in this circumstance will not have Medicare access. A small number of individuals living in the community on 'Removal Pending Bridging Visas' will, however, have access.
See: Table 2 for further explanation.
e Asylum Seekers who are deemed 'hard cases' maintain NHS coverage until a decision has been reached.
In Canada, for example, asylum seekers – including those facing deportation – are entitled to primary and emergency care, essential health services for the treatment and prevention of serious medical conditions, essential prescription medications, and prenatal and obstetrical care, among others. Refugees and asylum seekers may also receive secondary and mental health care with prior approval [31,32].
The UK offers similar coverage for refugees and asylum seekers, providing access to a broad range of National Health Service (NHS) benefits, including primary and secondary care, optical and dental care, free prescriptions and coverage of travel costs to/from hospital [33]. It is important to note that recent changes to UK law suggest a tightening in the NHS' willingness to provide such broad services across all asylum seeker categories [34]. Failed asylum seekers who have exhausted all rights of appeal are now only eligible for urgent care at no cost [35]. Despite this reduction in the range of available entitlements, the UK's policy still offers greater medical coverage than that available to many asylum seekers living in the community in Australia.
New Zealand, Australia's closest resettlement neighbour, provides refugees with the same health services as residents through the Publicly Funded Health and Disabilities (PFHD) Service. Asylum seekers with applications pending also have access to these services through a Community Services Card [36]. Other asylum seekers, for example those who are appealing a negative Refugee Status Appeal Authority decision, also have access to PFHD services [37].
Grey areas: complexities of the Australian health care policy for refugees and asylum seekers
The sheer range of diverse and complicated refugee and asylum seeker visa types exemplifies the complexity of Australian health care and health policy for refugees and asylum seekers. This complexity often leads to confusion among refugees, asylum seekers, community workers and health care practitioners alike. Additionally, current health care policy presents numerous grey areas. In particular, problems exist around gaps between the legal and practical applications of these policies; lack of policy coordination between State and Commonwealth governments; asylum seekers' ability to successfully access necessary health care; decisions about granting ASAS; lack of publicly accessible government data; and implications of the new RPBV.
Significant gaps between legal and practical policy implementation and the lack of coordination between State and Commonwealth governments may be best exemplified by the issue of access to public hospitals for asylum seekers. To begin, it is important to distinguish between legal restrictions and de facto restrictions. Disparities between being eligible to access health care and being able to access health care epitomise these gaps. In other words, asylum seekers with no Medicare can, theoretically, access public hospitals but may not be able to do so because they fear that their lack of income will leave them unable to pay hospital fees. There is growing evidence suggesting that Medicare ineligible asylum seekers have been turned away from hospitals, have not completed their required medical treatment, or have been asked to pay outstanding hospital bills and this is of significant concern [20].
Although some State governments have attempted to improve asylum seekers' access to health care and welfare [38], the Catholic Commission for Justice Development and Peace reports that State governments have failed 'to provide clear instructions to their departments and agencies to protect the human rights of asylum seekers in the areas of housing, health, transport and education' [39]. Additionally, State and Territory governments do not have clear policies concerning Medicare ineligible asylum seekers. In New South Wales (NSW), for example, an assurance of payment is required before treatment will be provided. If, however, that assurance is not available, then patients will 'receive only the minimum and necessary medical care to stabilise their condition' [1].
Strategies and practices for the provision of care to this population also vary widely across public health care services. While a few services provide ease of access to asylum seekers, many CBOs report that the majority deny access or attempt full fee recovery after providing the services. Commonly, access to these services, including waiving of fees, is dependent on long term advocacy from CBOs [19].
Other factors, such as English language skills and ability to access transport, also influence asylum seekers' ability to access successfully health care, thereby creating even greater gaps between policy and practice. As mentioned previously, current Commonwealth policy does not provide asylum seekers on Bridging Visas and refugees on TPV access to English language tuition or fee-free interpreting. Studies from the UK indicate that misunderstandings and poor communication between medical practitioners and asylum seekers operate as barriers to appropriate health care [1]. Clearly, this is also the case for refugees and asylum seekers within Australia, who do not have access to fee-free interpreting services.
Transport is often another major barrier to refugees and asylum seekers attempting to access health care with many lacking the money necessary for public transport or taxis. Could they access a vehicle, many are ineligible for drivers' licences, unless they can read, write and understand English sufficiently to pass the exam.
A key barrier towards improving health policy for refugees and asylum seekers is the difficulty in obtaining clear information from relevant Commonwealth Departments. For example, in relation to ASAS eligibility, it is unclear how DIMIA applies its discretionary powers to extend eligibility to ASAS or similar 'special payments' to some asylum seekers whose cases are at the post-RRT stage. Similarly there is a lack of publicly available government data on several key issues regarding asylum seeker health policy. First, it is unclear how many asylum seekers are living in the community on Bridging Visa E. This makes it difficult for CBOs to assess levels of need and asylum seeker populations. Second, there is currently no procedure for recording asylum seekers' access to health care which has lead to a lack of knowledge to inform policy and practice. Third, there are no available data for analysing the numbers of refugees and asylum seekers using public hospital services and, finally, there are no data on what they are being treated for. These issues, in turn, make it virtually impossible to estimate rates of and reasons for admission. Thus, we are left with case studies and documentation carried out by the already strapped CBOs working in the sector.
The recent introduction of RPBV only complicates these grey areas further. In particular, the RPBV creates disparities between bridging visas, presents significant human rights issues and has the potential to lead to further mental health issues for these asylum seekers. First, the RPBV's provision of Medicare raises significant issues around the levels of health care access for asylum seekers on other types of Bridging Visas. For example, what is the rationale behind the decision to grant Medicare rights to this Bridging Visa, when existing Bridging Visas offer no similar rights? Why has the provision of greater rights under this Bridging Visa not translated to other, similar visa categories? That is, if Medicare has been deemed necessary for these Bridging Visa holders, why has this not been extended to other Bridging Visa holders?
The RPBV also raises great concern about potential denial of human rights. Currently, the RPBV requires asylum seekers to agree in writing that they will cooperate with their removal from Australia when the government deems it is safe to do so. Thus, there is the very real potential for the promise of release from detention to lead to asylum seekers relinquishing their legal rights and future opportunities for visas. The written agreement may also facilitate involuntary return – individuals will have no say about the safety of their return, once the government deems it should happen [40].
The RPBV may also have significant effects on the mental health of asylum seekers. There is a growing body of evidence on the negative mental health impacts of detention on asylum seekers [41,42]. Releasing these already traumatised individuals into the community without immediate and ongoing access to counselling, and with no definite return date, no guaranteed visa term and no rights of appeal, could lead to further mental health issues, such as depression, feelings of isolation and anxiety.
Filling the Gaps: the need to adopt minimum standards
Current Australian health policy on access to medical care for refugees and asylum seekers has two faces. On the one hand, health care access for resettling refugees is comprehensive and one of the best compared to other resettlement countries. On the other hand, health policy for asylum seekers is less than adequate to ensure a minimum standard of health care. Moreover, the gaps between Commonwealth policy and State and Territory policy has produced a climate of confusion especially in regards to who pays.
As Taylor argues, current Australian policy for asylum seekers is 'insufficient to assure them of an adequate standard of living', and in breach of the International Covenant on Economic, Social and Cultural Rights (ICESCR) [43]. The current policy context may also leave Australia in breach of other 'international legal conventions and recommendations regarding its obligations toward asylum seekers living in its territory' [19] (p.16). Expanding on the human rights perspective, Dwyer argues that refugee and asylum seeker health care is also an issue of social justice and social responsibility [44]. In order to meet its basic human rights, social justice and social responsibility requirements, therefore, Australian health policy for asylum seekers needs significant and immediate change.
There are two key areas where policy reforms are urgently required. First, Commonwealth and State and Territory health care policy for asylum seekers needs to be coordinated to remove the current gaps which result in little or no access to health care for a proportion of the asylum seeker population [14]. Second, serious consideration should be given to extending medical care to all asylum seekers residing on Australia soil, regardless of their visa status. This medical care should include at least primary medical and psychological health care and care that cannot be postponed [45]. While there has been no rigorous economic analysis of the potential costs of extending Medicare to the entirety of this population, a small study indicates that the costs would be about A$ 2.9 million per year for NSW [46]. A similar analysis conducted by one of the authors (I C-V) estimated that the expected health care costs for asylum seekers living in the Victorian community would be about A$1.5 million per year. This figure represents about 0.009% of the total annual health care expenditure in Victoria in 2000–2001 [47]. These two initiatives would go a long way towards health policy reform for this vulnerable, high need but numerically small population residing in Australia.
List of abbreviations
ASAS: Asylum Seeker Assistance Scheme
CBO: Community based organisation
DIMIA: Department of Immigration and Multicultural and Indigenous Affairs
EHAI: Early Health Assessment and Intervention Program
HIV: Human Immunodeficiency Virus
HREOC: Human Rights and Equal Opportunity Commission
ICESCR: International Covenant on Economic, Social and Cultural Rights
IFH: Interim Federal Health
IHSS: Integrated Humanitarian Settlement Strategy
MOC: Medical Officer of the Commonwealth
NHS: National Health Service
PFHD: Publicly Funded Health and Disabilities
PPV: Permanent Protection Visa
PV: Protection visa
RPBV: Removal Pending Bridging Visa
RRT: Refugee Review Tribunal
SHP: Special Humanitarian Program
TB: Tuberculosis
THV: Temporary Humanitarian Visa
TPV: Temporary Protection Visa
UNHCR: United Nations High Commissioner for Refugees
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
All authors contributed equally to this paper.
Acknowledgements
The authors would like to thank: Paris Aristotle, Mardi Stow, Jenny Mitchell, Jo Szwarc, and Robyn Sampson for their helpful comments on initial drafts of the paper; the anonymous reviewers for their valuable comments.
==== Refs
Harris MF Telfer BL The health needs of asylum seekers living in the community eMJA 2001 175 589 5592 11837853
UNHCR Convention Relating to the Status of Refugees 1951 Geneva, UNHCR
UNHCR UNHCR Guidelines on Detention of Asylum Seekers
UNHCR Refugees by Numbers 2004 2005 Geneva, United Nations High Commissioner for Refugees
UNHCR Global Consultations on International Protection: Report of the Meetings within the Framework of the Standing Committee (Third Track).
UNHCR 2004 Global Refugee Trends 2005 Geneva, Population and Geographical Data Section, Division of Operational Support, United Nations High Commissioner for Refugees
UNHCR Protecting Refugees: New resettlement programmes 2005 Geneva, United Nations High Commissioner for Refugees
U.S. Committee for Refugees and Immigrants World Refugee Survey 2005 2005 Washington, U.S. Committee for Refugees and Immigrants
DIMIA Fact Sheet 60: Australia's Refugee and Humanitarian Program 2005 Canberra, Department of Immigration and Multicultural and Indigenous Affairs
DIMIA Fact Sheet 22: The Health Requirement 2004 Canberra, Department of Immigration and Multicultural and Indigenous Affairs
DIMIA Health requirement for temporary entry to Australia 2005 Canberra, Department of Immigration and Multicultural and Indigenous Affairs
DIMIA Fact Sheet 75: Processing unlawful boat arrivals 2004 Canberra, Department of Immigration and Multicultural and Indigenous Affairs
DIMIA Australia's support for humanitarian entrants 2003 Canberra, Department of Immigration and Multicultural and Indigenous Affairs
VFST Towards a health strategy for refugees and asylum seekers in Victoria: Report to the Victorian Department of Human Services 2004 Melbourne, Victorian Foundation for Survivors of Torture
DIMIA Fact Sheet 82: Immigration Detention 2004 Canberra, Department of Immigration and Multicultural and Indigenous Affairs
Smith MM Asylum seekers in Australia eMJA 2001 175 587 5589 11837852
HREOC Those who've come across the seas: Detention of unauthorised arrivals Report 1998 Sydney, Human Rights and Equal Opportunity Commission
HREOC Immigration Detention: Human Rigths Commissioner's 1998-1999 Review 1999 Canberra, Human Rights and Equal Opportunity Commission
McNevin A Seeking Safety, Not Charity: A report in support of work-rights for asylum-seekers living in the community on Bridging Visa E 2005 Melbourne, Network of Asylum Seeker Agencies Victoria
Asylum Seeker Project Hotham Mission Welfare issues and immigration outcomes for asylum seekers on Bridging Visa E Research and evaluation paper 2003 Melbourne,
DIMIA Fact Sheet 62: Assistance for asylum seekers in Australia 2003 Canberra, Department of Immigration and Multicultural and Indigenous Affairs
Minister for Immigration and Multicultural and Indigenous Affairs Broader Powers for Immigration Minister to Manage Long Term Detainees
Commonwealth of Australia Migration Amendment (Detention Arrangements) Bill 2005 - Explanatory Memorandum 2005 Canberra, The Parliament of the Commonwealth of Australia
Aristotle P Changes to detention arrangements 2005 Melbourne, Victorian Foundation for Survivors of Torture
DIMIA Temporary protection visas Fact Sheet
Pickering S Gard M Richardson R "We're working with people here" : The impact of the TPV regime on refugee settlement service provision in NSW Report 2003
DIMIA New measures for temporary protection and temporary humanitarian visa holders Report 2004 Canberra, Department of Immigration and Multicultural and Indigenous Affairs
Marston G Temporary protection permanent uncertainty : The experience of refugees living on temporary protection visas Report 2003 Melbourne, Centre for Applied Social Research RMIT University
UNHCR New Resettlement Programs
UNHCR Resettlement Handbook 2004 Geneva, Department of Internal Portection, United Nations High Commissioner for Refugees
Citizenship and Immigration Canada Interim Federal Health Program Information for Health Professionals
Citizenship and Immigration Canada Interim Federal Health Program
National Health Service UK Health Services Meeting the Needs of Refugees and Asylum Seekers in the UK: An information pack for health care workers
Hinsliff G Hutton Announces Crackdown on Health Cheats The Observer 2003 London,
Statutory Instrument 2004, No. 614: The National Health Service (Changes to Overseas Visitors) (Amendment) Regulations 2004 2004
Minister of Health Refugee Health Care: A Handbook for Porfessionals
Department of Labour Refugee Voices: Interim Report: A Journey Towards Resettlemetn
Minister for Community Services Brack's Government Helps Victoria's Asylum Seekers 2004 Melbourne, Victorian Government
Schwartz L Government visa policies a 'sin', say rights advocates The Age 2004 Melbourne,
Asylum Seeker Resource Centre Proposed Release of Long Term Detainees Does Not Go Far Enough
Steel Z Silove DM The mental health implications of detaining asylum seekers eMJA 2001 175 596 5599 11837855
Sultan A O'Sullivan K Psychological disturbances in asylum seekers held in long term detention: A participant-observer account eMJA 2001 175 593 5596 11837854
Taylor S Do on-shore asylum seekers have economic and social right? Dealing with the moral contradiction of liberal democracy Melbourne Journal of International Law 2000 1 70 796
Dwyer J Illegal immigrants, health care, and social responsibility Hasting Center Report 2004 34 34 41
Council of the European Union Laying down minimum standards for the reception of asylum seekers 2003 Brussels, Council of the European Union
Kardamanidis K Tangible and intangible costs of the 'no work - no Medicare' policy for community based asylum seekers in New South Wales Australia School of Public Health 2004 Sydney, University of Sydney
AIHW Health system expenditure on disease and injury in Australia, 2000-01 2004 Canberra, Australian Institute of Health and Welfare
DIMIA Fact Sheet 65: New Humanitarian Visa System 2002 Canberra, Department of Immigration and Multicultural and Indigenous Affairs
DIMIA Form 1024i: Bridging Visas 2003 Canberra, Department of Immigration and Multicultural and Indigenous Affairs
Australian Attorney-General's Department Migration Regulations 1994 2004 Canberra, SCALEplus, Commonwealth of Australia
DIMIA Seeking asylum within Australia Fact Sheet
Brotherhood of St Laurence Seeking Asylum: Living with Fear, Uncertainty and Exclusion 2002 Melbourne, BSL
DIMIA Fact Sheet 64c: Return Pending Visa 2004 Canberra, Department of Immigration and Multicultural and Indigenous Affairs
DIMIA Fact Sheet 85: Removal Pending Bridging Visa 2005 Canberra, Department of Immigration and Multicultural and Indigenous Affairs
|
16212674
|
PMC1282570
|
CC BY
|
2021-01-04 16:38:27
|
no
|
Aust New Zealand Health Policy. 2005 Oct 9; 2:23
|
utf-8
|
Aust New Zealand Health Policy
| 2,005 |
10.1186/1743-8462-2-23
|
oa_comm
|
==== Front
Behav Brain FunctBehavioral and brain functions : BBF1744-9081BioMed Central London 1744-9081-1-191623232210.1186/1744-9081-1-19ResearchCOMT genetic variation confers risk for psychotic and affective disorders: a case control study Funke Birgit [email protected] Anil K [email protected] Christine T [email protected] Alex M [email protected] Stephen L [email protected] Todd [email protected] Pamela [email protected] John M [email protected] Raju [email protected] Harvard Partners Center for Genetics and Genomics, Boston, USA2 Psychiatry Research, The Zucker Hillside Hospital, Glen Oaks, NY, USA3 Channing Laboratory, Brigham and Women's Hospital, Boston, USA2005 18 10 2005 1 19 19 25 8 2005 18 10 2005 Copyright © 2005 Funke et al; licensee BioMed Central Ltd.2005Funke et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
Variation in the COMT gene has been implicated in a number of psychiatric disorders, including psychotic, affective and anxiety disorders. The majority of these studies have focused on the functional Val108/158Met polymorphism and yielded conflicting results, with limited studies examining the relationship between other polymorphisms, or haplotypes, and psychiatric illness. We hypothesized that COMT variation may confer a general risk for psychiatric disorders and have genotyped four COMT variants (Val158Met, rs737865, rs165599, and a SNP in the P2 promoter [-278A/G; rs2097603]) in 394 Caucasian cases and 467 controls. Cases included patients with schizophrenia (n = 196), schizoaffective disorder (n = 62), bipolar disorder (n = 82), major depression (n = 30), and patients diagnosed with either psychotic disorder NOS or depressive disorder NOS (n = 24).
Results
SNP rs2097603, the Val/Met variant and SNP rs165599 were significantly associated (p = 0.004; p = 0.05; p = 0.035) with a broad "all affected" diagnosis. Haplotype analysis revealed a potentially protective G-A-A-A haplotype haplotype (-278A/G; rs737865; Val108/158Met; rs165599), which was significantly underrepresented in this group (p = 0.0033) and contained the opposite alleles of the risk haplotype previously described by Shifman et al. Analysis of diagnostic subgroups within the "all affecteds group" showed an association of COMT in patients with psychotic disorders as well as in cases with affective illness although the associated variants differed. The protective haplotype remained significantly underrepresented in most of these subgroups.
Conclusion
Our results support the view that COMT variation provides a weak general predisposition to neuropsychiatric disease including psychotic and affective disorders.
==== Body
Background
The role of Catechol-O-Methyltransferase (COMT) in dopamine metabolism has led to investigation of its variants in the etiology of numerous psychiatric disorders including psychotic, affective and anxiety disorders. The largest body of work exists for schizophrenia and bipolar disorder because 1. Imbalance of dopamine is thought be key to the pathogenesis of psychosis [1,2], 2. COMT is located in the region on chromosome 22q11 commonly deleted in velo-cardio-facial/DiGeorge syndrome (VCFS/DGS) whose phenotypic spectrum includes severe psychiatric disease that has been described as schizophrenia by some [3-5] and bipolar disorder by others [6] and 3. Genetic variation in COMT has been implicated in prefrontal cortical function [7,8], which is commonly impaired in both disorders [9]. In addition to schizophrenia and bipolar disorder, evidence for a contribution of COMT variants exists for panic disorder [10,11], attention deficit hyperactivity disorder [12], obsessive compulsive disorder [13], phobic anxiety [14] and anorexia nervosa [15].
Most studies focused on a common functional SNP (Val108/158Met) because the Methionine-containing variant shows a significant reduction in enzyme activity [16-18]. However, despite their large number, these studies have generated controversial and confusing results: For schizophrenia, initial studies have reported an association of the A (Met) allele [19-21]. However, current evidence favors an association of the of the G (Val) allele [22-27]. Similarly, for bipolar disorder, several studies reported an association of Met-COMT [[28-31], reviewed in: [32]]; but a recent study by Shifman et al. showed evidence of an association of Val-COMT [33]. As with schizophrenia and bipolar disorder, the associations remain controversial for other psychiatric illnesses including ADHD, OCD, anorexia nervosa, and anxiety disorder [34-37]. In all cases, likely contributing factors are small sample sizes and/or diagnostic differences as the absence of objective biomarkers in psychiatric disorders potentially hampers consistent classification of disease [38,39].
Although Shifman et al. saw an association of the Valine allele in their cohort, its moderate effect combined with highly significant p-values for two SNPs located in intron 1 and the 3'UTR (rs737865 and rs165599) has led to the hypothesis that the Val/Met variant may not contribute to disease but may simply be in strong LD with the actual, as of yet unidentified pre-disposing variant [40]. A recent study by Handoko et al. supports this view [41].
The COMT gene is transcribed from two promoters resulting in a cytoplasmatic form (soluble; S-COMT, transcribed from P1) and a membrane bound form (MB-COMT, transcribed from P2) [42]. Although both variants are widely expressed at varying levels, MB-COMT appears to be the predominant from in brain [42,43]. It has therefore been suggested that disease pre-disposing variant/s may be located in the P2 promoter, acting in cis to alter COMT protein levels via enhancement or suppression of transcription [44]. Several recent studies have investigated the effect of previously associated variants on COMT expression levels [16,45-47]. Interestingly, some showed reduced expression levels of Valine-coding COMT mRNAs [45,47]. Although this contrasts with the higher enzyme activity of Val-COMT, the net result of these two effects seems to be a 40% higher enzyme activity in human dorsolateral prefrontal cortex samples homozygous for Val-COMT [16]. A variant located in the P2 promoter (-278A/G) showed a small effect on enzyme activity, suggesting that it may indeed influence brain dopamine levels [16].
Taken together, current evidence suggests that COMT variants may provide a weak predisposition to a variety of psychiatric conditions via alteration of dopamine levels in the prefrontal cortex (supported by the association of the same haplotype with both, schizophrenia and bipolar disorder) [33]. Expression of specific disorders may require the presence of additional predisposing variants in susceptibility genes specific for these pathologies. Therefore, we have tested the relationship between COMT variation and psychiatric illness in a large cohort of Caucasian patients. We included subjects with a range of psychiatric diseases, and genotyped 4 SNPs including the Val/Met polymorphism, the P2 promoter SNP (-278A/G, rs2097603) as well as SNPs rs737865 and rs165599 to test the hypothesis that COMT genetic variation is associated with the risk for psychiatric illness. We analyzed the relationship between four-marker haplotypes, as well as the individual SNPs. Moreover, we conducted exploratory analyses of specific diagnostic subgroups within the cohort to assess the relationship between COMT and various psychiatric diagnoses.
Methods
Study Subjects
The sample was comprised of 394 US-Caucasian cases (mean age = 39.4 years; 35.7% females/64.3% males). Diagnostic categories included schizophrenia (n = 196), schizoaffective disorder (n = 62), bipolar disorder (n = 82), major depression (n = 30), psychotic disorder NOS (19) and depressive disorder NOS (5). The control group consisted of 467 Caucasian individuals (mean age = 39.3 years, 51.8% females/48.2% males). Cases were recruited from the inpatient and outpatient clinical services of the Zucker Hillside Hospital, a division of the North Shore – Long Island Jewish Health System, where patients are screened for potential recruitment into research studies by the Clinical Assessment and Training Unit (CAT) of the NIH-funded Hillside Hospital Intervention Research Center. The CAT monitors the inpatient and outpatient hospital census daily and conducts preliminary screening and recruitment functions. Inclusion criteria for screening for this study included a clinical diagnosis of a psychotic disorder, no active substance abuse, and ability to provide informed consent. After obtaining written informed consent, each subject was assessed with the Structured Clinical Interview for DSM-IV Axis I Disorders (SCID: version 2.0, 8/98), administered by trained raters. Standardized diagnostic assessments were supplemented with clinical information obtained by review of medical records and interviews with family informants when possible, and all diagnostic information was compiled into a narrative case summary. Information on the onset and course of Axis I illness, presence of Axis II pathology, presence of Axis III diagnoses, and a brief description of the subject's psychosocial and occupational functioning during the course of illness was presented to a consensus diagnostic committee consisting of a minimum of three senior faculty with DSM-IV diagnostic experience, as well as other faculty and trainees with SCID experience. All available information was used to arrive at a consensus DSM-IV diagnosis. Healthy controls were ascertained and recruited by the Zucker Hillside Hospital Normal Control Program. Potential participants were recruited via local newspaper advertisements, flyers and community internet resources and underwent initial telephone screening to assess eligibility criteria. Subjects meeting eligibility criteria were administered the SCID – NP to rule out the presence of an Axis I psychiatric disorder, urine toxicology screen for drugs of and a family history of psychiatric disorder assessment. Exclusion criteria included current or past: Axis I psychiatric disorder, psychotropic drug treatment, substance abuse, first-degree family member with an Axis I psychiatric disorder, or inability to provide written informed consent. A subset of the control cohort was collected through the Massachusetts General Hospital Clinical Research Program in conjunction with the Harvard-Partners Center for Genetics and Genomics in Boston, MA. This subset comprised disease free subjects over the age of 18. For the purposes of the study, "disease" was defined as current or past diagnoses made by a medical care provider that required medication or other forms of treatment/therapy. With the exception of orthopedic procedures, appendectomy, or those that were trauma related, surgery was considered an exclusionary criterion. Subjects who took prescription medications or regularly used over the counter medications were also excluded. These subjects completed a structured family and medical questionnaire that detailed current and past history of psychiatric illness and pharmacological or psychotherapeutic psychiatric treatment. All responses to the self-report form were confirmed by clinical interview by physicians. Physical exams were completed at the study visit. This study was approved by the Institutional Review Board at NSLIJHS and Partners Healthcare.
Genotyping
All samples were genotyped using the ABI PRISM 7900HT Sequence Detection System (Applied Biosystems, Foster City, CA). The 5' nuclease assay (TaqMan®) was used to distinguish the two alleles of a gene. PCR amplification was carried out on 5–20 ng DNA using 1 × TaqMan® universal PCR master mix (No Amp-erase UNG), 900 nM forward and reverse primers, 200 nM of the FAM labeled probe and 200 nM of the VIC labeled probe in a 5 ul reaction volume. Amplification conditions on an AB 9700 dual plate thermal cycle (Applied Biosystems, Foster City, CA) were as follows: 1 cycle of 95°C for 10 min, followed by 50 cycles of 92°C for 15 s and 58°C for 1 min. TaqMan® primers and probes were designed using the Primer Express® Oligo Design software v2.0 (ABI PRISM) or using the ABI Assays-By-Design service.
Statistical Analyses
Tests of Hardy-Weinberg equilibrium (HWE) were performed at each SNP locus [48]. Application of the test for HWE was restricted to the control samples from each ethnic group as a means of identification of genotyping problems. Single SNP analyses: Statistical inference for single SNP associations was based on Chi-square test statistics. For each SNP an allelic association test and the Cochran-Armitage trend test (test for additive allelic effects) were performed [49]. Haplotype analyses: Haplotype associations were explored using score tests that account for linkage phase ambiguity [50]. The score tests, derived from generalized linear models, are used for global tests of association, as well as haplotype-specific tests. In addition, a permutation algorithm was applied in the testing framework to find the maximum of the haplotype-specific score statistics and its associated p-value. The haplo.stats program implements the methods of Schaid et al. [50] and was used for these analyses. Haplotypes were imputed and frequencies estimated using the EM-algorithm-based estimation facility in haplo.stats.
Results
Single SNP analyses
We tested four COMT SNPs (-278 A/G, rs737865, Val108/158Met and rs165599) for association with diagnosis in our sample of 394 US-Caucasian cases and 467 controls. All SNPs were in HWE (data not shown). The results of the single SNP analysis are presented in Table 1. Three of the four SNPs were significantly associated with our broad "all affecteds" diagnosis: -278 A/G (p = 0.004; OR = 1.34), Val108/158Met (p = 0.05; OR = 1.21) and rs165599 (p = 0.035; OR = 1.25). When the cases were restricted to patients with schizophrenia or schizoaffective disorder (n = 258), only the promoter polymorphism (-278 A/G) remained significant (p = 0.015; OR = 1.33). This SNP also remained significant when only patients with schizophrenia were included (n = 196; p = 0.011, OR = 1.39). None of the SNPs yielded significant p-values in the group of patients with schizoaffective disorder (n = 62; data not shown). In the set of patients who were diagnosed with an affective disorder (n = 112; 82 bipolar disorder, 30 major depressive disorder), significant p-values were obtained for the G (Val) allele of the Val/Met polymorphism (p = 0.018; OR = 1.43) and the G allele of rs165599 (p = 0.039; OR = 1.38). The A allele of SNP -278A/G showed a trend for significance in this group (p = 0.062; OR = 1.34). When broken down into subcategories, -278A/G was marginally associated in the major depressive group (n = 30; p = 0.046; OR = 1.79) and Val158Met showed a trend towards association in the group of bipolar patients (n = 82; p = 0.058, OR = 1.39). Overall evidence for association was most robust for the promoter polymorphism -278 A/G.
Table 1 Allele frequencies
All affecteds* SCZ + SA
SNP ID (dbSNP No) Alleles (major/minor) * MAF Controls MAF Cases p-value Trend p-value OR (95%CI) MAF Controls MAF Cases p-value Trend p-value OR (95%CI)
n = 467 n = 394 n = 467 n = 258
rs2097063 (287A/G) A/G 0.414 0.346 0.004 0.006 1.34 (1.10, 1.64) 0.414 0.347 0.015 0.019 1.33 (1.06, 1.67)
rs737865 T/C 0.309 0.332 0.315 0.321 1.11 (0.90, 1.37) 0.309 0.329 0.438 0.448 1.10 (0.87, 1.39)
rs4680 (Val/Met) A/G 0.475 0.524 0.050 0.048 1.21 (1.00, 1.47) 0.475 0.504 0.304 0.299 1.12 (0.90, 1.40)
rs165599 A/G 0.326 0.376 0.035 0.040 1.25 (1.02, 1.53) 0.326 0.364 0.160 0.175 1.18 (0.94, 1.49)
SCZ Affective disorder*
SNP ID (dbSNP No) Alleles (major/minor) * MAF Controls MAF Cases p-value Trend p-value OR (95%CI) MAF Controls MAF Cases p-value Trend p-value OR (95%CI)
n = 467 n = 196 n = 467 n = 112
rs2097063 (287A/G) A/G 0.414 0.338 0.011 0.014 1.39 (1.08, 1.79) 0.414 0.346 0.062 0.067 1.34 (0.98, 1.82)
rs737865 T/C 0.309 0.327 0.511 0.518 1.09 (0.84, 1.41) 0.309 0.329 0.560 0.556 1.10 (0.80, 1.50)
rs4680 (Val/Met) A/G 0.475 0.500 0.419 0.414 1.10 (0.87, 1.41) 0.475 0.565 0.018 0.015 1.43 (1.06, 1.93)
rs165599 A/G 0.326 0.362 0.221 0.240 1.17 (0.91, 1.51) 0.326 0.400 0.039 0.043 1.38 (1.02, 1.87)
Major depressive disorder BP
SNP ID (dbSNP No) Alleles (major/minor) * MAF Controls MAF Cases p-value Trend p-value OR (95%CI) MAF Controls MAF Cases p-value Trend p-value OR (95%CI)
n = 467 n = 30 n = 467 n = 82
rs2097063 (287A/G) A/G 0.414 0.283 0.046 0.049 1.79 (1.01, 3.18) 0.414 0.369 0.280 0.290 1.21 (0.86, 1.71)
rs737865 T/C 0.309 0.333 0.688 0.689 1.12 (0.64, 1.95) 0.309 0.327 0.638 0.634 1.09 (0.76, 1.56)
rs4680 (Val/Met) A/G 0.475 0.586 0.101 0.093 1.56 (0.91, 2.68) 0.475 0.557 0.058 0.050 1.39 (0.99, 1.95)
rs165599 A/G 0.326 0.414 0.169 0.184 1.46 (0.85, 2.51) 0.326 0.395 0.088 0.095 1.35 (0.96, 1.91)
* All Affecteds: SCZ (schizophrenia), SA (schizoaffective disorder), BP (bipolar disorder), Major depressive disorder, psychotic disorder NOS, depressive disorder NOS
* Affective Disorder: Bipolar disorder and major depressive disorder
Marker-to-Marker Linkage Disequilibrium
Linkage disequilibrium (LD) between the four SNPs was assessed in cases and controls via Lewontin's D' statistic (Table 2). The markers span 28.5 kb on genomic DNA. In general, LD was higher between markers located in the 5' region of the gene. SNP -287 A/G is located only 2 kb from SNP rs737865 and is in complete LD with it. LD was also high for marker pair rs737865 – Val/Met. SNP rs165599 showed modest LD with the Val/Met polymorphism. This pattern is consistent with previous reports [40,45]. In contrast, a recent study by Chen et al. found only modest LD between rs737865 and the Val/Met polymorphism [22]. With the exception of the Val/Met – 737865 marker pair, D' values were similar in the schizophrenia/schizoaffective disorder group and the affective disorder group. In the latter, the Val/Met was in complete LD with rs737865, which may have contributed to the differences in the statistical difference observed for this SNP.
Table 2 Linkage Disequilibrium (D')
Controls rs737865 Val/Met rs165599
-287A/G 1 0.57 0.37
rs737865 0.72 0.23
Val/Met 0.67
Affected rs737865 Val/Met rs165599
-287A/G 1 0.51 0.27
rs737865 0.85 0.21
Val/Met 0.65
SCZ/SA rs737865 Val/Met rs165599
-287A/G 1 0.53 0.32
rs737865 0.76 0.28
Val/Met 0.64
Affective Disorder rs737865 Val/Met rs165599
-287A/G 1 0.53 0.29
rs737865 1 0.23
Val/Met 0.52
Haplotype analyses
Shifman et al. described a highly significant association of a three-site haplotype consisting of the G allele of SNPs rs737865, Val/Met and rs165599. This haplotype was associated with schizophrenia and bipolar disorder [33,40]. We constructed four-marker haplotypes for the Caucasian group including -278 A/G and the three SNPs present on the Shifman haplotype. Ten haplotypes had estimated frequencies above 0.025 and were included in the association testing (Table 3). This large number is most likely due to the low LD of rs165599 with the other three markers. One haplotype (G-A-A-A) was significantly underrepresented in cases and could be protective (p = 0.0033; maximum haplotype specific p = 0.026). Haplotype frequencies of the "opposite" haplotype (A-G-G-G), which encompasses the "Shifman haplotype did not differ significantly in the all affecteds group (Table 3), or any of the subgroups (data not shown). In contrast, the potentially protective G-A-A-A haplotype remained significantly underrepresented in most subgroups (Table 4).
Table 3 Haplotypes observed in the "all affecteds" group
287A/G (A/G) rs737865 (A/G) Val/Met (A/G) rs165599 (A/G) Frequency (cases) Frequency (controls) Haplotype-specific p-values
G A A A 0.2226 0.2809 0.0033
A G G G 0.1763 0.1489 0.1212
A A A A 0.1629 0.1446 0.2784
A G G A 0.1261 0.1191 0.7400
A A G G 0.0829 0.0807 0.3639
G A G G 0.0548 0.0385 0.3524
A A G A 0.0538 0.0386 0.2169
G A A G 0.0424 0.0462 0.8213
G A G A 0.0282 0.0490 0.0884
A G A A 0.0300 0.0418 0.2253
Table 4 Frequency of the G-A-A-A haplotype
cases controls Haplotype-specific p-values
All 0.2226 0.2809 0.0033
SCZ+SA 0.2282 0.2809 0.0191
SCZ 0.2297 0.2809 0.0361
SA 0.2255 0.2809 0.1641
BP + MD 0.2188 0.2809 0.0343
BP 0.2383 0.2809 0.1886
MD 0.1554 0.2809 0.0378
Discussion
The COMT gene has been extensively studied as a candidate gene for a variety of psychiatric disorders including schizophrenia, bipolar disorder, and other psychiatric conditions because of 1. Its known function in dopamine metabolism 2. The presence of a common functional nonsynonymous SNP in exon 4 (Val/Met), which alters enzyme activity and 3. Its location in the region commonly deleted in VCFS/DGS, which is associated with severe psychiatric disease often diagnosed as schizophrenia. However, no consistent picture has yet emerged, which might in part be due to small sample size of many studies.
We hypothesized that variation at the COMT locus confers a general basic risk for developing neuropsychiatric disease and therefore genotyped several variants in a heterogeneous group of patients. Several studies support this hypothesis: Results from family, twin, linkage and association studies show an overlapping genetic etiology of schizophrenia and bipolar disorder [reviewed in: [51]]. Several "overlap genes" including G72/G30 [52-55], Neuregulin [56,57] and DISC1 [58] have now been associated with both schizophrenia and bipolar disorder.
In addition, family studies have also shown evidence for an overlap between the genetic etiologies of schizophrenia and major depressive disorder: For example, the incidence of major affective disorder (bipolar and unipolar) was increased in relatives of probands with schizophrenia or schizoaffective disorder [59]. Data from another study supports the view that there could be a familial relationship between the predispositions to schizophrenia and to major depression as SCZ probands had an increased familial risk for unipolar major depressive disorder [60].
The SNPs genotyped in our sample included the well-known Val/Met variant, two SNPs that were highly associated in a previous study (rs737865 and rs165599 [40]) as well as -287A/G, a polymorphism in the P2 promoter. Three SNPs (-278A/G, Val/Met and rs165599 A/G) were associated in this broad diagnostic group with -278 A/G showing the most significant p-values. Haplotype revealed the existence of a potentially protective haplotype (G-A-A-A), which was significantly underrepresented in our all affecteds category and remained significant in most analyzed subgroups. The haplotype consisting of the "opposite" alleles of the risk haplotype described by Shifman et al. [33,40] was not significantly overrepresented in our cases.
Our results add to the growing number of studies showing an association of the Valine allele of the Val108/158 Met polymorphism, the only proven functional COMT variant so far. Interestingly, a comparison of risk haplotypes identified in previous studies showed that, even though the individual haplotypes vary with respect to the combination of alleles, all contain the Valine allele, indicating that it may be much older than their division [Figure 1 and references [22,24,27,33,40]].
Figure 1 Risk haplotypes identified in different studies and populations are shown. SNP IDs are indicated on top and the respective alleles are boxed underneath. "x" denotes a SNP that was not tested. * indicates a haplotype, which also included rs9265(A) and rs165849(A), both located distal to rs165599 in the ARVCF gene.
In addition to the heterogeneous "all affecteds" category, we separately analyzed cases with psychotic and affective illness to analyze the contribution of the four variants to disease in these subgroups. In the psychotic group (schizophrenia and schizoaffective disorder), only the promoter variant (-287A/G) remained significant. The other two SNPs associated in the all affecteds group were not significant although the allele frequencies still showed an overrepresentation of the same alleles. In the affective disorder group, the association was driven by the 3' portion of the gene (Val-COMT and rs165599 A/G). The promoter polymorphism showed a trend for association in this group. We suggest that one or a combination of the following factors might have contributed to these results: First, it is conceivable that COMT harbors more than one functional variant. Although these may collectively confer a general risk for neuropsychiatric disease, the magnitude of the effect may vary depending on diagnosis. Support for this view comes from a recent family-based study reporting two separate and interacting effects within a haplotype spanning rs737865-Val/Met-rs165599 [41]. Second, the lack of association of some variants in the patients with affective disorder and not in the SCZ/SA group may be related to the limitations inherent in a diagnostic system based upon clinical phenomena, which may not correctly reflect the underlying biology that predisposes to illness. Therefore, in the absence of biological markers, a strict separation of our cohort into schizophrenic and affective disorder patients may be too stringent [38,39]. Third, we noted that while the affective disorder group was well matched with regard to gender (49% male, 51% female), the schizophrenia/schizoaffective group showed an excess of males (67%). This imbalance may have contributed to the lack of association of Val/Met and rs165599 as schizophrenia may have gender-specific differences [61,62] and previous studies suggested a stronger association of the G (Val) allele or a Val-containing haplotype in females [25,40]. It has also been suggested that the Val/Met variant does not contribute to disease but is merely in high LD with the actual functional variant [40,41]. We observed different D' values for the Val/Met – rs737865 marker pair in the analyzed patient subgroups which may have contributed to the differences in the statistical significance of the Val/Met SNP.
Finally, one could argue that our results were influenced by hidden population substructure, leading to false positive results. However, while this is a possible limitation of our study, empirical data obtained from US-Caucasian, African American and European populations suggest that carefully matched studies of moderate size are unlikely to contain significant stratification levels [63,64]. In particular, a recent study by Tang et al. showed that self-identified ethnicity is the major determinant of genetic structure in the United States [65].
Conclusion
Our study supports a general role of the COMT gene in the genetic etiology of neuropsychiatric disease including psychotic and affective disorders. Further studies are needed to unambiguously determine the nature and specific contribution of the functional variant/s.
List of Abbreviations used
SCZ (schizophrenia); BP (Bipolar disorder), SA (schizoaffective disorder), ADHD (Attention deficit hyperactivity disorder), OCD (obsessive compulsive disorder).
Competing interests
The author(s) declare that they have no competing interests.
Acknowledgements
This project was supported by grants HD034980-09 to R.K. and MH001760 to A.K.M. We thank Dr. Bernice Morrow for providing services to generate lymphoblastoid cell lines and DNA and the Genotyping Core of the Harvard Partners Center for Genetics and Genomics for genotyping the samples.
==== Refs
Baumeister AA Francis JL Historical development of the dopamine hypothesis of schizophrenia J Hist Neurosci 2002 11 265 277 12481477 10.1076/jhin.11.3.265.10391
Harrison PJ Weinberger DR Schizophrenia genes, gene expression, and neuropathology: on the matter of their convergence Mol Psychiatry 2004
Murphy KC Jones LA Owen MJ High rates of schizophrenia in adults with velo-cardio-facial syndrome Arch Gen Psychiatry 1999 56 940 945 10530637 10.1001/archpsyc.56.10.940
Shprintzen RJ Goldberg R Golding-Kushner KJ Marion RW Late-onset psychosis in the velo-cardio-facial syndrome Am J Med Genet 1992 42 141 142 1308357 10.1002/ajmg.1320420131
Bassett AS Chow EW AbdelMalik P Gheorghiu M Husted J Weksberg R The schizophrenia phenotype in 22q11 deletion syndrome Am J Psychiatry 2003 160 1580 1586 12944331 10.1176/appi.ajp.160.9.1580
Papolos DF Faedda GL Veit S Goldberg R Morrow B Kucherlapati R Shprintzen RJ Bipolar spectrum disorders in patients diagnosed with velo-cardio-facial syndrome: does a hemizygous deletion of chromosome 22q11 result in bipolar affective disorder? Am J Psychiatry 1996 153 1541 1547 8942449
Egan MF Goldberg TE Kolachana BS Callicott JH Mazzanti CM Straub RE Goldman D Weinberger DR Effect of COMT Val108/158 Met genotype on frontal lobe function and risk for schizophrenia Proc Natl Acad Sci U S A 2001 98 6917 6922 11381111 10.1073/pnas.111134598
Malhotra AK Kestler LJ Mazzanti C Bates JA Goldberg T Goldman D A functional polymorphism in the COMT gene and performance on a test of prefrontal cognition Am J Psychiatry 2002 159 652 654 11925305 10.1176/appi.ajp.159.4.652
Zalla T Joyce C Szoke A Schurhoff F Pillon B Komano O Perez-Diaz F Bellivier F Alter C Dubois B Rouillon F Houde O Leboyer M Executive dysfunctions as potential markers of familial vulnerability to bipolar disorder and schizophrenia Psychiatry Res 2004 121 207 217 14675740 10.1016/S0165-1781(03)00252-X
Woo JM Yoon KS Choi YH Oh KS Lee YS Yu BH The association between panic disorder and the L/L genotype of catechol-O-methyltransferase J Psychiatr Res 2004 38 365 370 15203287 10.1016/j.jpsychires.2004.01.001
Domschke K Freitag CM Kuhlenbaumer G Schirmacher A Sand P Nyhuis P Jacob C Fritze J Franke P Rietschel M Garritsen HS Fimmers R Nothen MM Lesch KP Stogbauer F Deckert J Association of the functional V158M catechol-O-methyl-transferase polymorphism with panic disorder in women Int J Neuropsychopharmacol 2004 7 183 188 15009906 10.1017/S146114570400416X
Qian Q Wang Y Zhou R Li J Wang B Glatt S Faraone SV Family-based and case-control association studies of catechol-O-methyltransferase in attention deficit hyperactivity disorder suggest genetic sexual dimorphism Am J Med Genet 2003 118B 103 109 10.1002/ajmg.b.10064
Alsobrook JP Zohar AH Leboyer M Chabane N Ebstein RP Pauls DL Association between the COMT locus and obsessive-compulsive disorder in females but not males Am J Med Genet 2002 114 116 120 11840516 10.1002/ajmg.10040
McGrath M Kawachi I Ascherio A Colditz GA Hunter DJ De Vivo I Association between catechol-O-methyltransferase and phobic anxiety Am J Psychiatry 2004 161 1703 1705 15337664 10.1176/appi.ajp.161.9.1703
Frisch A Laufer N Danziger Y Michaelovsky E Leor S Carel C Stein D Fenig S Mimouni M Apter A Weizman A Association of anorexia nervosa with the high activity allele of the COMT gene: a family-based study in Israeli patients Mol Psychiatry 2001 6 243 245 11317231 10.1038/sj.mp.4000830
Chen J Lipska BK Halim N Ma QD Matsumoto M Melhem S Kolachana BS Hyde TM Herman MM Apud J Egan MF Kleinman JE Weinberger DR Functional Analysis of Genetic Variation in Catechol-O-Methyltransferase (COMT): Effects on mRNA, Protein, and Enzyme Activity in Postmortem Human Brain Am J Hum Genet 2004 75 807 821 15457404 10.1086/425589
Shield AJ Thomae BA Eckloff BW Wieben ED Weinshilboum RM Human catechol O-methyltransferase genetic variation: gene resequencing and functional characterization of variant allozymes Mol Psychiatry 2004 9 151 160 14966473 10.1038/sj.mp.4001386
Scanlon PD Raymond FA Weinshilboum RM Catechol-O-methyltransferase: thermolabile enzyme in erythrocytes of subjects homozygous for allele for low activity Science 1979 203 63 65 758679
Kotler M Barak P Cohen H Averbuch IE Grinshpoon A Gritsenko I Nemanov L Ebstein RP Homicidal behavior in schizophrenia associated with a genetic polymorphism determining low catechol O-methyltransferase (COMT) activity Am J Med Genet 1999 88 628 633 10581481 10.1002/(SICI)1096-8628(19991215)88:6<628::AID-AJMG10>3.0.CO;2-E
Park TW Yoon KS Kim JH Park WY Hirvonen A Kang D Functional catechol-O-methyltransferase gene polymorphism and susceptibility to schizophrenia Eur Neuropsychopharmacol 2002 12 299 303 12126868 10.1016/S0924-977X(02)00030-5
Ohmori O Shinkai T Kojima H Terao T Suzuki T Mita T Abe K Association study of a functional catechol-O-methyltransferase gene polymorphism in Japanese schizophrenics Neurosci Lett 1998 243 109 112 9535125 10.1016/S0304-3940(98)00100-1
Chen X Wang X O'Neill AF Walsh D Kendler KS Variants in the catechol-o-methyltransferase (COMT) gene are associated with schizophrenia in Irish high-density families Mol Psychiatry 2004
Wonodi I Stine OC Mitchell BD Buchanan RW Thaker GK Association between Val108/158 Met polymorphism of the COMT gene and schizophrenia Am J Med Genet 2003 120B 47 50 10.1002/ajmg.b.20037
Li T Ball D Zhao J Murray RM Liu X Sham PC Collier DA Family-based linkage disequilibrium mapping using SNP marker haplotypes: application to a potential locus for schizophrenia at chromosome 22q11 Mol Psychiatry 2000 5 452 10889560 10.1038/sj.mp.4000752
Kremer I Pinto M Murad I Muhaheed M Bannoura I Muller DJ Schulze TG Reshef A Blanaru M Gathas S Goichman R Rietschel M Dobrusin M Bachner-Melman R Nemanov L Belmaker RH Maier W Ebstein RP Family-based and case-control study of catechol-O-methyltransferase in schizophrenia among Palestinian Arabs Am J Med Genet 2003 119B 35 39 10.1002/ajmg.b.20008
Glatt SJ Faraone SV Tsuang MT Association between a functional catechol O-methyltransferase gene polymorphism and schizophrenia: meta-analysis of case-control and family-based studies Am J Psychiatry 2003 160 469 476 12611827 10.1176/appi.ajp.160.3.469
Sanders AR Rusu I Duan J Molen JE Hou C Schwab SG Wildenauer DB Martinez M Gejman PV Haplotypic association spanning the 22q11.21 genes COMT and ARVCF with schizophrenia Mol Psychiatry 2004 15007391
Mynett-Johnson LA Murphy VE Claffey E Shields DC McKeon P Preliminary evidence of an association between bipolar disorder in females and the catechol-O-methyltransferase gene Psychiatr Genet 1998 8 221 225 9861640
Papolos DF Veit S Faedda GL Saito T Lachman HM Ultra-ultra rapid cycling bipolar disorder is associated with the low activity catecholamine-O-methyltransferase allele Mol Psychiatry 1998 3 346 349 9702745 10.1038/sj.mp.4000410
Kirov G Murphy KC Arranz MJ Jones I McCandles F Kunugi H Murray RM McGuffin P Collier DA Owen MJ Craddock N Low activity allele of catechol-O-methyltransferase gene associated with rapid cycling bipolar disorder Mol Psychiatry 1998 3 342 345 9702744 10.1038/sj.mp.4000385
Li T Vallada H Curtis D Arranz M Xu K Cai G Deng H Liu J Murray R Liu X Collier DA Catechol-O-methyltransferase Val158Met polymorphism: frequency analysis in Han Chinese subjects and allelic association of the low activity allele with bipolar affective disorder Pharmacogenetics 1997 7 349 353 9352569
Craddock N Dave S Greening J Association studies of bipolar disorder Bipolar Disord 2001 3 284 298 11843779 10.1034/j.1399-5618.2001.30604.x
Shifman S Bronstein M Sternfeld M Pisante A Weizman A Reznik I Spivak B Grisaru N Karp L Schiffer R Kotler M Strous RD Swartz-Vanetik M Knobler HY Shinar E Yakir B Zak NB Darvasi A COMT: a common susceptibility gene in bipolar disorder and schizophrenia Am J Med Genet 2004 128B 61 64 10.1002/ajmg.b.30032
Azzam A Mathews CA Meta-analysis of the association between the catecholamine-O-methyl-transferase gene and obsessive-compulsive disorder Am J Med Genet 2003 123B 64 69 10.1002/ajmg.b.20013
Ohara K Nagai M Suzuki Y Ochiai M No association between anxiety disorders and catechol-O-methyltransferase polymorphism Psychiatry Res 1998 80 145 148 9754693 10.1016/S0165-1781(98)00062-6
Hawi Z Millar N Daly G Fitzgerald M Gill M No association between catechol-O-methyltransferase (COMT) gene polymorphism and attention deficit hyperactivity disorder (ADHD) in an Irish sample Am J Med Genet 2000 96 282 284 10898900 10.1002/1096-8628(20000612)96:3<282::AID-AJMG9>3.0.CO;2-R
Gabrovsek M Brecelj-Anderluh M Bellodi L Cellini E Di Bella D Estivill X Fernandez-Aranda F Freeman B Geller F Gratacos M Haigh R Hebebrand J Hinney A Holliday J Hu X Karwautz A Nacmias B Ribases M Remschmidt H Komel R Sorbi S Tomori M Treasure J Wagner G Zhao J Collier DA Combined family trio and case-control analysis of the COMT Val158Met polymorphism in European patients with anorexia nervosa Am J Med Genet B Neuropsychiatr Genet 2004 124 68 72 14681918 10.1002/ajmg.b.20085
Potash JB Willour VL Chiu YF Simpson SG MacKinnon DF Pearlson GD DePaulo JRJ McInnis MG The familial aggregation of psychotic symptoms in bipolar disorder pedigrees Am J Psychiatry 2001 158 1258 1264 11481160 10.1176/appi.ajp.158.8.1258
Helmuth L In sickness or in health? Science 2003 302 808 810 14593163 10.1126/science.302.5646.808
Shifman S Bronstein M Sternfeld M Pisante-Shalom A Lev-Lehman E Weizman A Reznik I Spivak B Grisaru N Karp L Schiffer R Kotler M Strous RD Swartz-Vanetik M Knobler HY Shinar E Beckmann JS Yakir B Risch N Zak NB Darvasi A A highly significant association between a COMT haplotype and schizophrenia Am J Hum Genet 2002 71 1296 1302 12402217 10.1086/344514
Handoko HY Nyholt DR Hayward NK Nertney DA Hannah DE Windus LC McCormack CM Smith HJ Filippich C James MR Mowry BJ Separate and interacting effects within the catechol-O-methyltransferase (COMT) are associated with schizophrenia Mol Psychiatry 2004
Tenhunen J Salminen M Lundstrom K Kiviluoto T Savolainen R Ulmanen I Genomic organization of the human catechol O-methyltransferase gene and its expression from two distinct promoters Eur J Biochem 1994 223 1049 1059 8055944 10.1111/j.1432-1033.1994.tb19083.x
Reenila I Mannisto PT Catecholamine metabolism in the brain by membrane-bound and soluble catechol-o-methyltransferase (COMT) estimated by enzyme kinetic values Med Hypotheses 2001 57 628 632 11735324 10.1054/mehy.2001.1430
Palmatier MA Pakstis AJ Speed W Paschou P Goldman D Odunsi A Okonofua F Kajuna S Karoma N Kungulilo S Grigorenko E Zhukova OV Bonne-Tamir B Lu RB Parnas J Kidd JR DeMille MM Kidd KK COMT haplotypes suggest P2 promoter region relevance for schizophrenia Mol Psychiatry 2004 9 859 870 15098000 10.1038/sj.mp.4001496
Bray NJ Buckland PR Williams NM Williams HJ Norton N Owen MJ O'Donovan MC A haplotype implicated in schizophrenia susceptibility is associated with reduced COMT expression in human brain Am J Hum Genet 2003 73 152 161 12802784 10.1086/376578
Tunbridge E Burnet PW Sodhi MS Harrison PJ Catechol-o-methyltransferase (COMT) and proline dehydrogenase (PRODH) mRNAs in the dorsolateral prefrontal cortex in schizophrenia, bipolar disorder, and major depression Synapse 2004 51 112 118 14618678 10.1002/syn.10286
Zhu G Lipsky RH Xu K Ali S Hyde T Kleinman J Akhtar LA Mash DC Goldman D Differential expression of human COMT alleles in brain and lymphoblasts detected by RT-coupled 5' nuclease assay Psychopharmacology (Berl) 2004
Hill WG Estimation of linkage disequilibrium in randomly mating populations Heredity 1974 33 229 239 4531429
Sasieni PD From genotypes to genes: doubling the sample size Biometrics 1997 53 1253 1261 9423247
Schaid DJ Rowland CM Tines DE Jacobson RM Poland GA Score tests for association between traits and haplotypes when linkage phase is ambiguous Am J Hum Genet 2002 70 425 434 11791212 10.1086/338688
Craddock N Owen MJ The beginning of the end for the Kraepelinian dichotomy Br J Psychiatry 2005 186 364 366 15863738 10.1192/bjp.186.5.364
Schumacher J Jamra RA Freudenberg J Becker T Ohlraun S Otte AC Tullius M Kovalenko S Bogaert AV Maier W Rietschel M Propping P Nothen MM Cichon S Examination of G72 and D-amino-acid oxidase as genetic risk factors for schizophrenia and bipolar affective disorder Mol Psychiatry 2004 9 203 207 14966479 10.1038/sj.mp.4001421
Chumakov I Blumenfeld M Guerassimenko O Cavarec L Palicio M Abderrahim H Bougueleret L Barry C Tanaka H La Rosa P Puech A Tahri N Cohen-Akenine A Delabrosse S Lissarrague S Picard FP Maurice K Essioux L Millasseau P Grel P Debailleul V Simon AM Caterina D Dufaure I Malekzadeh K Belova M Luan JJ Bouillot M Sambucy JL Primas G Saumier M Boubkiri N Martin-Saumier S Nasroune M Peixoto H Delaye A Pinchot V Bastucci M Guillou S Chevillon M Sainz-Fuertes R Meguenni S Aurich-Costa J Cherif D Gimalac A Van Duijn C Gauvreau D Ouellette G Fortier I Raelson J Sherbatich T Riazanskaia N Rogaev E Raeymaekers P Aerssens J Konings F Luyten W Macciardi F Sham PC Straub RE Weinberger DR Cohen N Cohen D Genetic and physiological data implicating the new human gene G72 and the gene for D-amino acid oxidase in schizophrenia Proc Natl Acad Sci U S A 2002 99 13675 13680 12364586 10.1073/pnas.182412499
Hattori E Liu C Badner JA Bonner TI Christian SL Maheshwari M Detera-Wadleigh SD Gibbs RA Gershon ES Polymorphisms at the G72/G30 gene locus, on 13q33, are associated with bipolar disorder in two independent pedigree series Am J Hum Genet 2003 72 1131 1140 12647258 10.1086/374822
Chen YS Akula N Detera-Wadleigh SD Schulze TG Thomas J Potash JB DePaulo JR McInnis MG Cox NJ McMahon FJ Findings in an independent sample support an association between bipolar affective disorder and the G72/G30 locus on chromosome 13q33 Mol Psychiatry 2004 9 87 92; image 5 14699445 10.1038/sj.mp.4001453
Green EK Raybould R Macgregor S Gordon-Smith K Heron J Hyde S Grozeva D Hamshere M Williams N Owen MJ O'Donovan MC Jones L Jones I Kirov G Craddock N Operation of the schizophrenia susceptibility gene, neuregulin 1, across traditional diagnostic boundaries to increase risk for bipolar disorder Arch Gen Psychiatry 2005 62 642 648 15939841 10.1001/archpsyc.62.6.642
Stefansson H Sigurdsson E Steinthorsdottir V Bjornsdottir S Sigmundsson T Ghosh S Brynjolfsson J Gunnarsdottir S Ivarsson O Chou TT Hjaltason O Birgisdottir B Jonsson H Gudnadottir VG Gudmundsdottir E Bjornsson A Ingvarsson B Ingason A Sigfusson S Hardardottir H Harvey RP Lai D Zhou M Brunner D Mutel V Gonzalo A Lemke G Sainz J Johannesson G Andresson T Gudbjartsson D Manolescu A Frigge ML Gurney ME Kong A Gulcher JR Petursson H Stefansson K Neuregulin 1 and susceptibility to schizophrenia Am J Hum Genet 2002 71 877 892 12145742 10.1086/342734
Hodgkinson CA Goldman D Jaeger J Persaud S Kane JM Lipsky RH Malhotra AK Disrupted in schizophrenia 1 (DISC1): association with schizophrenia, schizoaffective disorder, and bipolar disorder Am J Hum Genet 2004 75 862 872 15386212 10.1086/425586
Gershon ES DeLisi LE Hamovit J Nurnberger JIJ Maxwell ME Schreiber J Dauphinais D Dingman CW Guroff JJ A controlled family study of chronic psychoses. Schizophrenia and schizoaffective disorder Arch Gen Psychiatry 1988 45 328 336 3355320
Maier W Lichtermann D Minges J Hallmayer J Heun R Benkert O Levinson DF Continuity and discontinuity of affective disorders and schizophrenia. Results of a controlled family study Arch Gen Psychiatry 1993 50 871 883 8215813
Leung A Chue P Sex differences in schizophrenia, a review of the literature Acta Psychiatr Scand Suppl 2000 401 3 38 10887978 10.1111/j.0065-1591.2000.0ap25.x
Roy MA Maziade M Labbe A Merette C Male gender is associated with deficit schizophrenia: a meta-analysis Schizophr Res 2001 47 141 147 11278131 10.1016/S0920-9964(99)00231-5
Pankow JS Province MA Hunt SC Arnett DK Regarding "Testing for population subdivision and association in four case-control studies" Am J Hum Genet 2002 71 1478 1480 12515254 10.1086/344582
Ardlie KG Lunetta KL Seielstad M Testing for population subdivision and association in four case-control studies Am J Hum Genet 2002 71 304 311 12096349 10.1086/341719
Tang H Quertermous T Rodriguez B Kardia SL Zhu X Brown A Pankow JS Province MA Hunt SC Boerwinkle E Schork NJ Risch NJ Genetic structure, self-identified race/ethnicity, and confounding in case-control association studies Am J Hum Genet 2005 76 268 275 15625622 10.1086/427888
|
16232322
|
PMC1282571
|
CC BY
|
2021-01-04 16:39:20
|
no
|
Behav Brain Funct. 2005 Oct 18; 1:19
|
utf-8
|
Behav Brain Funct
| 2,005 |
10.1186/1744-9081-1-19
|
oa_comm
|
==== Front
BMC BiotechnolBMC Biotechnology1472-6750BioMed Central London 1472-6750-5-281621265410.1186/1472-6750-5-28Research ArticleAvidin related protein 2 shows unique structural and functional features among the avidin protein family Hytönen Vesa P [email protected]äättä Juha AE [email protected] Heidi [email protected] Katrin K [email protected]örhä Jarno [email protected] Tuomas [email protected] Thomas KM [email protected] Mark S [email protected] Tiina A [email protected] Markku S [email protected] Tomi T [email protected] NanoScience Center, Department of Biological and Environmental Science, P.O. Box 35 (YAB), FI-40014 University of Jyväskylä, Finland2 Department of Biochemistry and Pharmacy, Åbo Akademi University, Tykistökatu 6 A, FI-20520, Turku, Finland2005 7 10 2005 5 28 28 5 7 2005 7 10 2005 Copyright © 2005 Hytönen et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
The chicken avidin gene family consists of avidin and several avidin related genes (AVRs). Of these gene products, avidin is the best characterized and is known for its extremely high affinity for D-biotin, a property that is utilized in numerous modern life science applications. Recently, the AVR genes have been expressed as recombinant proteins, which have shown different biotin-binding properties as compared to avidin.
Results
In the present study, we have employed multiple biochemical methods to better understand the structure-function relationship of AVR proteins focusing on AVR2. Firstly, we have solved the high-resolution crystal structure of AVR2 in complex with a bound ligand, D-biotin. The AVR2 structure reveals an overall fold similar to the previously determined structures of avidin and AVR4. Major differences are seen, especially at the 1–3 subunit interface, which is stabilized mainly by polar interactions in the case of AVR2 but by hydrophobic interactions in the case of AVR4 and avidin, and in the vicinity of the biotin binding pocket. Secondly, mutagenesis, competitive dissociation analysis and differential scanning calorimetry were used to compare and study the biotin-binding properties as well as the thermal stability of AVRs and avidin. These analyses pinpointed the importance of residue 109 for biotin binding and stability of AVRs. The I109K mutation increased the biotin-binding affinity of AVR2, whereas the K109I mutation decreased the biotin-binding affinity of AVR4. Furthermore, the thermal stability of AVR2(I109K) increased in comparison to the wild-type protein and the K109I mutation led to a decrease in the thermal stability of AVR4.
Conclusion
Altogether, this study broadens our understanding of the structural features determining the ligand-binding affinities and stability as well as the molecular evolution within the protein family. This novel information can be applied to further develop and improve the tools already widely used in avidin-biotin technology.
==== Body
Background
Avidin from eukaryotic chicken together with streptavidin from prokaryotic Streptomyces avidinii share an unique property not seen in any other known proteins, an extremely high affinity (Kd ≈ 10-15 M) to its natural ligand, the water-soluble vitamin D-biotin [1,2]. High affinity and stability of the free and complex forms of (strept)avidin and biotin [1], the easy attachment of biotin to various target molecules [3] and the non-obtrusive chemical nature of biotin are currently exploited in numerous (strept)avidin-biotin based life science applications [4].
Avidin is postulated to exist throughout the oviparous vertebrates [5-7] and has long been known to be the operational biotin-harvester in chicken egg-white, comprising about 0.05% of the total protein [1]. Recently, avidin related genes (AVRs) highly similar to avidin (91–95% identity at the nucleotide level) have also been found in the chicken genome [8-10], suggesting that in addition to avidin AVRs may also play a role in biotin-harvesting. AVRs seem to have functional promoter regions [11] and are located in close vicinity of the avidin gene on the chicken male-sex chromosome Z [10,12]. Interestingly, the total number of AVRs is likely to vary between different individuals and even between different cells of the individual chicken [13]. The function of the AVR gene products is unknown, however, since the proteins encoded by them have not yet been isolated from chicken although mRNAs of AVR1, AVR2 and AVR3 are found during inflammation [11].
Since the avidin gene of chicken is the only cloned avidin gene within the vertebrates [14], structure-function and protein engineering studies have long been concentrated on it [15-24]. The three-dimensional structure of chicken avidin has already been determined, too [25,26]. The bacterial homolog of avidin, streptavidin, is also well characterized: the gene encoding steptavidin has been cloned [27], its structure solved [28] and several studies on the biochemical properties and protein engineering of streptavidin have been reported [17,29-34].
In order to characterize the proteins encoded by the AVRs, avidin related proteins (AVRs) have recently been produced in insect cells using a baculovirus expression system and have been demonstrated to be functional biotin-binding proteins like chicken avidin [35]. AVRs do, however, show unique features when compared to avidin. The AVRs differ from avidin, with respect to glycosylation and charge properties, and all AVRs except AVR2 contain an uneven number of cysteine residues in their sequence, which can form inter-subunit disulphide bridges in addition to the intra-subunit disulphide bridges also seen in avidin [35]. Interestingly, the biotin-binding affinities of AVRs have been reported to vary over a wide range of values, AVR4 being almost as efficient a biotin binder as avidin [35,36] and AVR2 showing the lowest affinity for biotin [35]. AVRs, like avidin, have been found to be very stable proteins, too; AVR4 has clearly higher thermal stability than avidin [35,36]. Recently, we have been able to produce avidin and some AVRs in Escherichia coli, too [37]. The 3D-structure of AVR4 has been recently determined [38], and we have been able to produce chimeric forms of avidin and AVR4, which retained their high biotin affinity and showed improved thermal stability [39].
In this study, we have determined the high-resolution X-ray crystal structure of AVR2 in complex with the natural ligand, D-biotin. By using site-directed mutagenesis and recombinant expression techniques combined with structural studies, we have been able to characterize some of the structural factors responsible for the varying biochemical properties of the members of the chicken avidin protein family. The results may be utilized in avidin protein engineering aiming to fine-tune the ligand binding properties and thermal stability of AVRs and their chimeric forms in experiments needed to expand the tools available in the area of avidin-biotin technology. This study provides insight into the molecular evolution within the avidin family, too.
Results
Production and mutagenesis of proteins
Several different AVRs, AVR2, AVR2(I109K), AVR4(K109I), AVR6, and a novel avidin mutant K111I, were efficiently expressed either in insect cells or using a bacterial expression system as summarised in Table 1 [37,40]. Since wild type AVR6 was found to form oligomers in solution via disulphide bridges (data not shown), Cys-58 of AVR6 was mutated to Ser based on the alignment of AVR6 with AVR2 and AVR4 (Figure 1). The resulting protein AVR6(C58S) (hereafter referred to AVR6), as well as all of the other studied proteins, were found to form avidin-like tetramers according to analysis by size exclusion chromatography (Table 2) and confirmed to be pure and homogenous using SDS-PAGE. The elution time of AVR2 was slightly different from that of avidin and of the other studied AVRs, which can partially be explained by the different charge properties of these proteins: AVR2 has a theoretical isoelectric point of 4.7 compared to 9.6 for AVR4, 6.9 for AVR6 and 9.7 for avidin. The effect of glycosylation is also clearly seen by the varying elution times in gel filtration analysis (Table 2) for AVR2-b produced in bacteria (no glycosylation) and from insect cells (two potential glycosylation sites; both utilized to some extent [35]).
Purification of AVR2 and AVR6 on the 2-iminobiotin column was inefficient and therefore biotin affinity chromatography was used to isolate these proteins. In order to confirm the quality of the protein and correct cleavage of the bacterial signal peptide used in the production of AVRs in E. coli, the molecular weight of AVR2-b was determined using ESI-MS. The Mr (14310.0 ± 0.3 Da) determined from the experimental data using four charge states correlates well with the expected Mr of 14307.8 Da. The final product from E. coli expression carries three additional residues (QTV) at the N-terminus as previously reported for avidin produced using the same method [37].
In a previous study, the K109I mutation of AVR2 was hypothesised to be at least partially responsible for the lower biotin-binding affinity of AVR2 when compared to other AVRs and avidin [35]. In order to validate this suggestion, we subjected AVRs to mutagenesis. The mutation I109K was introduced into AVR2 to increase its biotin-binding affinity. Likewise, the mutation Lys→Ile was introduced into both AVR4 and avidin in order to cross-validate the hypothesis and lower the biotin-binding affinity of AVR4 and avidin. Gel filtration analysis showed that all of the mutated proteins corresponded to tetramers (Table 2).
X-ray structure of AVR2
The X-ray structure of AVR2-b in complex with D-biotin was determined at 1.4 Å resolution. The structure determination statistics are summarized in Table 3. As expected based on the sequence alignment (Figure 1), AVR2 has an overall three-dimensional structure similar to those of avidin [PDB:1AVD] [26] and AVR4 [PDB: 1Y52] [38]. Each monomer in the homotetrameric protein has a β-barrel fold of eight β-strands with one end of the barrel adapted to bind D-biotin (Figure 1C).
Although AVR2 shares many structural features with avidin and AVR4, there are clear differences in their functional properties (Table 2, Figure 4). The most distinctive structural differences are found around the terminal carboxylate group of D-biotin. In avidin, atom O10B of the terminal carboxylate group of D-biotin is hydrogen bonded to the backbone nitrogen atoms of Ala-39 and Thr-40 of the L3,4 loop [26], whereas in AVR4 there is only one polar contact between the O10B atom and the backbone nitrogen atom of Asp-39 of the L3,4 loop [38]. In AVR2, the O10B atom of D-biotin forms an additional hydrogen bond to the NE2 atom of Gln-97. Moreover, in avidin, the O10A atom of the carboxylate group of biotin is hydrogen bonded to the OG atoms of Ser-73 and Ser-75, while in AVR2 and AVR4 there is only one hydrogen bond to the OG atom of Ser-71 (equivalent to Ser-73 in avidin).
A few differences are also found around the central aliphatic segment and the bicyclic ring system of D-biotin. Leucine in avidin (Leu-99) and AVR4 (Leu-97) is replaced by glutamine in AVR2 (Gln-97), and threonine in avidin (Thr-77) and AVR4 (Thr-75) is replaced by Ser-75 in AVR2. The side-chain atoms of Gln-97 of AVR2 not only interact with the valeryl segment of D-biotin, as in the case of the corresponding Leu in avidin and AVR4, but also with the carboxylate group of D-biotin (see above). The corresponding residue in AVR4 and avidin is leucine, but glutamine is also found aligned at the equivalent position in other AVRs. The presence of the polar head group of Gln-97 in AVR2 also affects the type and conformation of the neighbouring residues, such as Ser-73, Leu-99 and Arg-112 as well as Ile-109 from another subunit. These residues line the entrance of the biotin-binding pocket: i) Ser-73 of AVR2 has two alternative rotamers unlike in the X-ray structure of avidin and AVR4, ii) Leu-99 of AVR2 is replaced by Ser-101 in avidin (Ser-99 in AVR4), iii) Arg-112 of AVR2 and AVR4 form a salt bridge with Asp-39 from the L3,4 loop, but Arg-112 has a different conformation in AVR2 than in AVR4 (in avidin the corresponding ARG-114 does not form a salt bridge with any of the residues from the L3,4 loop) and iv) Ile-109 of AVR2 is substituted by Lys-111 in avidin and by Lys-109 in AVR4. In addition, Ile-109 of AVR2 is located next to Trp-108, a residue known to be important for biotin binding [17,30], but Ile-109 does not appear to alter the conformation of this residue significantly in comparison to avidin and AVR4. All of these differences are located at the open end of the biotin-binding pocket and are very likely responsible for the varying biotin binding properties of avidin, AVR2 and AVR4. Moreover, although D-biotin adopts a very similar conformation in avidin, AVR2 and AVR4, the biotin-binding network is not identical (Figure 2).
The 1–3 subunit interface (numbering according to Ref. [25]) is also markedly different in amino acid composition in AVR2 in comparison with avidin and AVR4. In avidin, the subunit contacts are established between the hydrophobic amino acid residues Met-96, Val-115 and Ile-117 [25,26], which are in contact with the same set of residues from a neighbouring subunit. In AVR2, only valine (Val-113 in the AVRs) is conserved, while Met-96 and Ile-117 of avidin are respectively substituted by the hydrophilic residues Lys-94 and Asn-115 in AVR2 (Figure 1A). In the crystal structure of AVR2, the side chain of Lys-94 can exist in two alternate rotamers, which are hydrogen bonded either to the side-chain oxygen atom of Asn-115 or to the main-chain oxygen of Val-113, both from a neighbouring subunit (Figure 3). Asn-115 is also in contact with Asn-115 from a neighbouring subunit. In AVR4, Ile-117 of avidin is substituted by Tyr-115, the latter interacting strongly with Tyr-115 of an adjacent subunit and through a hydrogen bond to Lys-92 [38].
Comparative analysis of avidin family proteins
Previously, it has been found that AVR6 forms intermonomeric disulphide bridges [35]. Gel filtration analysis of AVR6 revealed that these disulphide bonds are formed between tetramers, thus causing further oligomerization of the protein (not shown). Consequently, we introduced the C58S mutation into AVR6, which successfully blocked oligomerisation, and used this mutated protein form in the comparative analyses in the present study.
The overall charge of AVR2 (pI ≈ 5) is very different when compared to that of avidin and AVR4 (pI ≈ 10). The number of ionic bonds in avidin is seven per subunit, whereas three salt bridges are seen in AVR4 [38]. In the AVR2-biotin complex, four intra-subunit salt bridges are detected: Asp-39-Arg-112, Glu-89-Arg-120, Lys-92-Asp-117 and Arg-98-Asp-107.
Biotin dissociation analysis
Of the proteins studied, the fastest [3H]biotin dissociation rate was found with AVR2, while the slowest rate was measured for avidin (Figure 4). Ile-109 is found close to the biotin-binding site in AVR2, whereas all other proteins in avidin family [35] have lysine at the equivalent position. In order to test the effect of Ile-109 on the biotin dissociation rate, the AVR2(I109K) mutant was produced. The resulting mutant had a significantly slower dissociation rate than the wild-type protein. AVR6, in turn, showed a dissociation rate constant in between the values observed for the two AVR2 forms. The dissociation rate constant for AVR4, measured in a previous study [39], was somewhat higher when compared to that of avidin. When Lys-109 of AVR4 was mutated to isoleucine according to the sequence of AVR2, the rate of dissociation increased as expected, but biotin binding of the resulting protein was still stronger than for the mutated AVR2 form. Similarly, the analogous mutation K111I in avidin increased the biotin dissociation rate compared to the wild-type protein. Hence, the analyzed proteins can be sorted according to their biotin-binding affinities (as the biotin dissociation rate decreases, biotin binding strengthens): AVR2 < AVR6 < AVR2(I109K) < AVR4(K109I) < AVR4 < AVD(K111I) < avidin (Figure 4).
The biotin dissociation data, measured at various temperatures, were analysed using the global fit method described elsewhere [41]. The resulting dissociation rate temperature-dependency model was compared to the one previously measured for avidin [39]. The analysis revealed a different temperature-dependency for AVR2 in comparison to avidin (Figure 4B). The mutation I109K caused a shift in the temperature-dependency of AVR2-biotin dissociation, resulting in a model resembling that determined for AVR6. AVR4 has a similar temperature-dependency of the biotin dissociation rate as avidin, and the K109I mutation did not significantly change the temperature-dependency although it clearly increased the biotin dissociation rate constant (Figure 4B). The equivalent mutation K111I in avidin resulted in a nearly two-fold increase in the biotin dissociation rate over a temperature range of 40–60°C (Figure 4B).
Differential scanning calorimetry
The thermostability of avidin, AVR2, AVR4 and AVR6 were measured using DSC analysis (Table 2). In this analysis, AVR2 showed higher thermostability than avidin. The measured Tm (91.3°C) was between the values measured previously for avidin (83.5°C) and AVR4 (106.4°C) [36]. As expected, the thermal stability of AVR2 increased in the presence of biotin (Tm = 112.5°C), similarly as reported for avidin, AVR4 and streptavidin [22,36,42]. The I109K mutation significantly stabilised AVR2, resulting in a 6.3°C increase in Tm as compared to the wild-type protein. The reverse mutation, K109I in AVR4 and K111I in avidin, led to destabilisation of the proteins, resulting in a 1.9°C and 7.0°C decrease in the Tm, respectively. Interestingly, AVR6 showed slightly higher thermal stability (Tm = 87.7°C) than avidin (Tm = 83.5°C) in the absence of biotin, while in its presence the Tm of AVR6 (114.0°C) was raised significantly but remained lower than that measured for the avidin-biotin complex (Tm = 117.0°C).
Discussion
In the present study, we have used targeted mutagenesis and X-ray crystallography combined with the comparative analysis of thermal stability and ligand-binding kinetics to dissect the functional properties of the chicken avidin protein family. The high-resolution structure of AVR2, a close relative of avidin, provides new insights into the biotin-binding mechanism of the avidins and serves as a new source of knowledge for protein engineering studies, too.
In order to understand the observed differences in the biotin-binding affinities and thermal stabilities within the avidin protein family, the crystal structures of avidin [26], AVR2, and AVR4 [38] were compared, all in complex with D-biotin. Overall, these proteins share high structural similarity and their ligand-binding sites within the eight-strand β-barrel resemble each other. The most distinctive structural differences are found around the terminal carboxylate group and central valeryl segment of D-biotin. In the AVR2 structure, D-biotin is in contact with the L3,4 loop as in the case of avidin [26] and AVR4 [38], but also in contact with the side-chain atom of Gln-97 unlike the avidin or AVR4 complexes where leucine is found at the equivalent position. Glutamine is conserved in all of the AVRs except for AVR4 [35], resides 3.3 Å away from D-biotin in the AVR2 structure, and a hydrogen bond may form between Gln-97 and biotin even though the angle is not optimal. In addition to D-biotin, Gln-97 seems to form a hydrogen bond with Ser-73, which exists in two alternative conformations. The presence of Gln-97 in AVR2 probably affects the conformation of nearby Arg-112, which is slightly displaced with respect to the corresponding residue seen in AVR4 and avidin. Then again, the conformation of Arg-112 may be altered due to interactions with the AVR2-specific Ile-109, too. Ile-109 of AVR2 respectively corresponds to Lys-109 and Lys-111 in avidin and AVR4. In AVR2, Ile-109 resides close to Trp-108, which is known to be important for biotin binding [17,30], but the conformation of Trp-108 does not seem to be significantly affected by Ile-109. Yet another sequence difference, whereby Thr-77 of avidin and the corresponding threonine of AVR4 (residue 75 in the AVRs) is substituted to Ser in AVR2 does not appear to disrupt hydrogen bonding to the sulfur atom of D-biotin. However, this substitution enlarges the binding cavity around bound biotin (Figure 2) and hence contributes to the lower affinity of AVR2 for biotin. The biotin-binding network is not identical in avidin, AVR2 and AVR4 despite the similar conformation that D-biotin adopts in all of these structures. In general, the polar contacts with D-biotin seem to be much more variable than the hydrophobic ones, which are highly conserved, indicating their important role in the biotin-binding process. In line with this, the importance of hydrophobic residues for biotin-binding of streptavidin has been demonstrated experimentally [30]. Moreover, the interactions of avidin, AVR2 and AVR4 with the carboxylate group of D-biotin are clearly less conserved than the interactions with the central aliphatic valeryl segment and the bicyclic ring system of the tetrahydrothiophenic and ureido rings buried deeper within the biotin binding pocket [26,38,43].
Based on the temperature-dependence of the biotin dissociation rates and relative biotin dissociation rate constants (Figure 4), the order of the biotin binding affinities is as follows: AVR2 < AVR6 < AVR4 < avidin. These results are in line with the previous ligand-binding analyses performed for AVRs using an optical biosensor [35]. Furthermore, the presence of an isoleucine residue at sequence position 109 in AVR2 rather than lysine seems to be the most dominant difference affecting biotin binding in comparison to AVR6. However, this sequence variation does not explain the differences in the biotin-binding properties of AVR2 versus AVR4. This was confirmed by analysing the AVR4(K109I) mutant, which showed significantly stronger interactions with biotin when compared to wild-type AVR2. Moreover, the equivalent mutation K111I in avidin affected only slightly the dissociation rate constant of avidin. The temperature-dependency model suggests even slower dissociation rates at low temperatures for the mutant compared to wild-type avidin (Figure 4B). The different effects of the Lys→Ile mutation on avidin versus AVR4 may reflect differences at the L3,4 loop of avidin and AVRs. Although the AVR2-biotin dissociation rate was over 5000-fold higher at 20°C than that of avidin and thus showed significantly lower biotin-binding affinity than avidin, the thermal stability of AVR2 in the absence of biotin is higher (Tm = 91.3°C) than for avidin (Tm = 83.5°C). Higher thermal stability is notable and may have a functional role.
The biological role of AVRs is unclear; avidin is thought to work as a biotin-harvester in chicken egg-white, thus preventing growth of biotin-dependent organisms [1]. The lower biotin-binding affinity and higher stability of AVR2 raises the question if AVR2 has any biological role similar to avidin. The expression of avidin is induced in chicken during inflammation in various tissues and mRNAs of some AVRs, including AVR2, have been detected during inflammation [11]. This suggests that AVR2 (and the other AVRs) may play a role in inflammatory reactions.
The conformation of the L3,4 loop of AVR2 was found to strongly resemble that of AVR4 [38]. In avidin, this loop is disordered in the absence and ordered in the presence of D-biotin [25]. In contrast, the L3,4 loop of AVR4 was previously found to be in an nearly identical, fixed conformation both in the absence and presence of D-biotin [38]. The latter situation seems to be true for all AVR proteins, since the L3,4 loop per se, as well as the neighbouring sequences between the β3 and β5 strands, are highly conserved within the AVR family but quite different from avidin [35]. Recently, this region was transferred to avidin from AVR4 [39]. The resultant chimeric ChiAVD showed better thermal stability (Tm = 96.5°C) than avidin (Tm = 83.5°C) [39] and, interestingly, the observed stability of AVR2 (Tm = 91.3°C) and its mutant AVR2(I109K) (Tm = 97.6°C) was similar to that of ChiAVD. Hence, in addition to the 1–3 subunit interface, the region between the β3 and β5 strands of AVR2 is likely to affect the stability of the protein (Figure 3). This view is supported by our preliminary results of engineered dual chain avidins suggesting only slightly better stability for the AVR2-type 1–3 interface compared to the 1–3 interface of avidin (Hytönen et al. unpublished results).
The present study provides novel knowledge of the structural characteristics of AVR proteins. Avidin related proteins are considered as an individual branch in the evolutionary tree of avidins in chicken [44]. Based on the results of the current study (Figure 5), AVR2 and AVR6 seem to be functionally closely related to each other, supporting the previous phylogenetic analysis [44]. It seems that the critical difference, isoleucine at position 109 in AVR2 in comparison to lysine in avidin and the other AVRs, has arisen after the divergence of AVR4 and the rest of AVRs. This sequence difference also explains why AVR2 has the lowest observed biotin-binding affinity among the AVRs. All AVRs have a region between β3 and β5 strands, which is quite different from that in avidin [35]. In comparison with the other AVRs, this region in AVR4 shares a higher level of similarity with avidin. This likely correlates with the high, avidin-like biotin-affinity of AVR4 versus the lower biotin-affinity of the other AVRs [39].
Ligand binding to avidin and streptavidin can be considered as an extreme discovery of nature in the sense of affinity and free energy [45]. The structural complementarity between biotin and its binding site in (strept)avidin is almost perfect, which together with the numerous hydrogen bonds that are formed between (strept)avidin and biotin is the basis for the extraordinary tight binding [25,28]. Thus, it is not surprising to find that a small perturbation in this highly perfected system can reverberate as a major change in the biotin binding kinetics. It is known that the high biotin-binding affinity of (strept)avidin is dominated by extremely slow ligand dissociation rates, especially in the case of avidin.
Conclusion
The high-resolution structure of AVR2 combined with the ligand binding data broadens our understanding of the general principles of ligand-binding processes. Furthermore, the structural information can be employed as a basis to create improved tools for biotechnology. This was demonstrated in a previous study, where chimeric forms of avidin and AVR4 showed improved properties compared to the native proteins [39].
Methods
Production and mutagenesis of proteins
Proteins were expressed using the Bac-to-Bac baculovirus expression system in Sf9 insect cells in biotin-free media as previously reported [40]. Bacterial expression in BL-21(AI) (Invitrogen) was also used for protein expression as described in Ref. [36]. The proteins were isolated using affinity chromatography with an 2-iminobiotin or biotin matrix (Affiland S. A., Belgium) as described earlier [17]. Biotin was used as the capture ligand for AVR2, AVR2-b, AVR2(I109K) and AVR6-b, and for these proteins elution was achieved using acetic acid. The recombinant proteins investigated in this article are summarised in Table 1.
Crystallization and data collection
Minimal Screen 12 [46], a sparse matrix protein crystallization screen [47], was used to search for suitable conditions for crystallization of AVR2-b with the vapor diffusion hanging drop method at 22°C. An orthorhombic crystal with approximate dimensions of 0.15 × 0.1 × 0.1 mm was obtained within three weeks using equal volumes (1 μl) of sample solution containing 0.5 mg/ml protein in 50 mM NaPO4 (pH 7.0), 100 mM NaCl and well solution containing 0.1 M Na-citrate (pH 4.6) and 1.5 M NH4PO4. Before crystallization, the AVR2-b – biotin complex was prepared by adding biotin to the protein solution in a molar ratio of 5:1, respectively, followed by incubation at 4°C for 1.5 hours. For data collection, the AVR2-b crystal was cryoprotected with 20% glycerol (v/v) and 2 M lithium sulfate just prior to flash-freezing in a 100 K nitrogen stream (Oxford Cryostream). Diffraction data were collected from a single crystal at the ESRF beam line ID14-1, Grenoble at 100 K using an ADSC Q4R CCD detector. Data were processed with programs of the XDS program package [48]. Data collection statistics are summarized in Table 3.
Structure determination
The X-ray structure of AVR2-b was solved using the molecular replacement method and programs from the CCP4i suite [49]. The space group (P212121) of the AVR2-b crystal was ascertained by Amore [50] and molecular replacement was done with Molrep [51]. A tetramer (biological unit) composed of only main-chain atoms and based on a high resolution X-ray structure of avidin (Airenne, Hytönen et al. unpublished; [PDB:1VYO]) was used as a trial model. The best solution (correlation coefficient = 0.291) from molecular replacement was selected as the input for automatic model building with ARP/wARP [52]. After adding side chains separately for each monomer A to H using the guiSIDE mode of ARP/wARP, the model was refined with Refmac5 [53], and modified and rebuilt with O [54]. Solvent atoms were added to the model with the automatic procedure of ARP/wARP [55] and the ligand biotin was built with the ARP/wARP LigandBuild program [56]. Sulfate ions and glycerol molecules were built either manually in O or with the aid of the program Coot [57]. The AVR2-b structure was analyzed with the programs PROCHECK [58] and WHATIF [59]. Structure determination statistics are summarized in Table 3. The coordinates and structure factors of AVR2-b have been deposited in the Protein Data Bank with entry code [PDB:1WBI].
Biotin dissociation analysis
The dissociation rate constant of AVR2-b, AVR2(I109K), AVR4(K109I), AVR6-b and AVD(K111I)-b for [3H]biotin was measured at various temperatures as previously described [60]. [3H]Biotin was purchased from Amersham. The data were analysed by using the global fit approach as shown by Hyre et al. [41], in which the temperature dependence of the dissociation rate constant is modelled by the Eyring equation.
Differential scanning calorimetry
The thermal stability of AVR2, AVR2-b, AVR2(I109K), AVR4(K109I), AVR6-b and AVD(K111I) was studied using differential scanning calorimetry (DSC) as previously described [61]. The melting point of protein unfolding was determined from thermograms measured in a buffer containing 50 mM NaPO4 (pH 7.0) and 100 mM NaCl. Proteins were also analysed in the presence of biotin (three-fold molar excess of biotin per protein subunit).
Size exclusion chromatography
Gel filtration experiments were performed as described in Ref. [23] with a Superdex HR 10/30 column using 50 mM NaCO3 (pH 11.0), 150 mM NaCl as the liquid phase. The column was calibrated using IgG (158 kDa), BSA (68 kDa) and ovalbumin (44 kDa) as molecular weight standards.
Mass spectroscopy
The molecular weight of AVR2-b was measured with a Micromass LCT Electronspray ionization TOF Mass spectrometer essentially as described previously [37]. Samples were dialysed against water and diluted 1:1 with acetonitrile. The final protein concentration was 7 μM and the pH was adjusted using formic acid (0.2 %). Positive ions were detected using the default parameters (source temperature 100°C, desolvation temperature 120°C, RF lens voltage 750 V, extraction cone voltage 6 V, sample cone voltage 50 V, capillary voltage 3800 V) and the sample was injected at a rate of 20 μl/min.
Phylogeny inference
Avidin-related sequences were aligned using MALIGN [62,63]. One-thousand bootstrap variations [64] of the alignment were generated using SEQBOOT and distance matrices produced using a structure-based scoring matrix [62,63]. Trees were produced using NEIGHBOR, and the consensus tree produced using CONSENSE. SEQBOOT, NEIGHBOR and CONSENSE are programs from the PHYLIP package [65,66].
Miscellaneous methods
The multiple sequence alignment shown in Figure 1A was created using the program MALIGN [62,63] of Bodil [67] and edited with Corel Draw11. The protein representations in Figure 1, 2, 3 were made with the PyMOL Molecular Graphics System [68] and edited with the programs Gimp and/or Corel Draw11. Cavities were calculated with Surfnet [69] using 1.4 Å and 3.0 Å radii for minimum and maximum gap spheres, respectively. The electron density map shown in Figure 1C and 1D was calculated with programs of the CCP4i suite.
Authors' contributions
The biochemical studies of AVD and AVRs were mainly carried out by VPH and JAEM in the group of Prof. MSK at the University of Jyväskylä and Tampere. JAEM performed also the mutagenesis analyses. HK had an important role in crystallizing the AVR2, whereas KKH and TKMN carried out the DSC analyses. JH had a major role in the expression and purification of the studied AVD and AVRs. TK performed most of the 3H-biotin dissociation analyses and MSJ made the phylogenetic inference analysis. The structural analysis of AVR2 was made mainly by TTA in the Structural Bioinformatics Laboratory led by the group leaders MSJ and TAS.
Acknowledgements
We would like to thank Irene Helkala and Eila Korhonen for excellent technical assistance, Professor J. Peter Slotte for access to the calorimetry facilities and Professor Kari Rissanen for access to the mass spectrometry laboratory. We thank Drs. David Hyre and Olli H. Laitinen for helpful discussions. This study was supported by the ISB (National Graduate School in Informational and Structural Biology (V.P.H, H.K.)), grants from the Academy of Finland, the Sigrid Jusélius Foundation, and the Foundation of Åbo Akademi. This work was supported by ARK Therapeutics Oy, Kuopio, Finland. We acknowledge the European Synchrotron Radiation Facility for provision of synchrotron radiation facilities and we would like to thank the staff for assistance in using beamline ID14-1.
Figures and Tables
Figure 1 The X-ray structure of AVR2. (A) Multiple sequence alignment of the avidin family. The secondary structure elements are numbered according to the AVR2 structure. The residues mutated in this study (AVR2, Ile-109; AVR4, Lys-109; AVR6, Cys-58) are coloured green. Residue Cys-122 in AVR4, which was mutated to serine in a previous study [36], is indicated in blue. (B) Two tetramers found in the asymmetric unit of the AVR2-b crystal. (C) Monomer A of AVR2-b. The weighted difference Fo – Fc electron density map (blue), calculated in the absence of biotin, is drawn with a 1.5 Å radius around the atoms of D-biotin of the final structure of AVR2-b. Contours are shown at 3.0σ. The secondary structure elements of AVR2-b are numbered. (D) A close-up view of (C) focused on biotin.
Figure 2 Biotin-binding sites of avidin, AVR2 and AVR4. Stereo images of the biotin-binding sites of (A) AVR2-b, (B) AVR4 [PDB:1Y52] [38] and (C) avidin [PDB:1AVD] [26] are shown. The cavities (transparent) around the bound biotin molecules of subunit A are shown. The water molecules within the cavities are shown as red spheres. Hydrogen bonds between the atoms of D-biotin and the surrounding proteins are shown with dashed lines. Carbon atoms of residues from subunit 1 are coloured white; those from a neighbouring subunit 2 are indicated in cyan: residues 108 and 109 (110 and 111 in avidin).
Figure 3 The 1–3 subunit interface of AVR2, AVR4 and avidin. Stereo images of the 1–3 subunit interface of (A) AVR2-b, (B) AVR4 [PDB:1Y52] [38], and of (C) avidin expressed in E. coli [PDB:1VYO] (Airenne, Hytönen et al. unpublished) are shown. Lys-94 of AVR2 exists in two alternate conformations, which can form a hydrogen bond with the side-chain oxygen atom of Asn-115 or with the main-chain oxygen atom of Val-113, both from an adjacent subunit. Putative hydrogen bonds are shown by dashed lines. The carbon atoms of subunits 1 and 3 are coloured white and cyan, respectively.
Figure 4 Biotin dissociation analysis. (A) Temperature-dependence of biotin dissociation rates measured by the [3H]biotin dissociation assay. Radiobiotin dissociation rate constants measured for the proteins are plotted as a function of temperature. The models for dissociation rate constants obtained by global fit analysis [41] are shown by lines and the determined individual dissociation rates by symbols. The dissociation rate constants determined previously for avidin and AVR4-b are also shown [39]. (B) Relative dissociation rate constants of the AVRs and avidin mutant K111I compared to that of avidin. The individual dissociation rates as well as global fit analysis models of the dissociation rate are divided by the dissociation rate constants of avidin obtained from global fit analysis. The data are plotted using a logarithmic y-axis.
Figure 5 Summary of the biochemical properties and phylogeny of avidin (AVD) and the AVRs. The putative chicken biotin-binding proteins BBP-A and BBP-B [12] as well as streptavidin (STR) are included. The confidence levels of the branching of the phylogenetic tree were assessed using the bootstrap method [64]. The tree is unrooted and the branch lengths are not to scale. The values at each node represent the percentage of the 1000 trees where the species above the node are consistently found. Tm(-), heat-induced unfolding temperature without D-biotin (°C). Tm(+), heat-induced unfolding temperature in the presence of D-biotin (°C). Tr(-), transition temperature of oligomeric disassembly without D-biotin determined using SDS-PAGE assay (°C) [70]. Tr(+), transition temperature of oligomeric disassembly in the presence of D-biotin (°C). kdiss, biotin dissociation rate constant (× 10-6 s-1) obtained from global fitting [41] of the 3H-biotin dissociation data. I1–3, three residues important for the 1–3 subunit interface (equivalent to residues 94, 113 and 115 in AVR2). The residues in BBP-A and BBP-B are numbered according to avidin. N. D., not determined. aHytönen VP, unpublished data. Analysis performed on wt proteins produced in the baculovirus expression system as described in Ref. [35]. bFrom Ref. [70]. cFrom Ref. [36]. dFrom Ref. [71]. eFrom Ref. [72].
Table 1 The recombinant proteins of this study. The immobilised ligand used for affinity chromatography purification and the elution conditions are shown. Protein eluted using acetic acid was immediately dialyzed against 50 mM NaPO4 pH 7.0 + 100 mM NaCl.
Protein Affinity chromatography ligand Elution conditions Expression system Production yield (mg/l)a
AVR2 D-biotin 0.5 M acetic acid BEVSb 0.8
AVR2-b D-biotin 0.5 M acetic acid E. colic 4.9
AVR2(I109K) D-biotin 0.5 M acetic acid BEVS 9.0
AVR4d 2-iminobiotin 50 mM Na-Ac + 100 mM NaCl BEVS 6.1
AVR4-bd 2-iminobiotin 50 mM Na-Ac + 100 mM NaCl E. coli 21.9
AVR4(K109I)d 2-iminobiotin 50 mM Na-Ac + 100 mM NaCl BEVS 16.2
AVR4(K109I)-bd 2-iminobiotin 50 mM Na-Ac + 100 mM NaCl E. coli 6.7
AVR6-be D-biotin 2 M acetic acid E. coli 8.3
AVD(K111I) 2-iminobiotin 50 mM Na-Ac + 100 mM NaCl BEVS 11.5
AVD(K111I)-b 2-iminobiotin 50 mM Na-Ac + 100 mM NaCl E. coli 0.7
aAfter affinity chromatography per one liter of culture medium. The yields are calculated based on only a few protein purifications.
bProduced using a baculovirus expression system in insect cells as described previously [40].
cThe bacterial signal peptide from the protein OmpA was utilised in order to produce the protein in an active form in E. coli as described in [37].
dAVR4 is identical to AVR5. The protein carries the mutation C122S, which prevents oligomerisation via intermolecular disulphide bridges [36].
eCarries mutation C58S, which prevents oligomerisation via intermolecular disulphide bridges (this study).
Table 2 Structural properties of avidin and avidin related proteins. Elution times from FPLC gel filtration and calculated molecular weights of the proteins. Heat-induced unfolding temperatures of proteins from DSC analysis (average ± S.D).
Gel filtration DSC
Protein Elution time (min) Molecular mass (kDa) Tm (-biotin) (°C) Tm (+biotin) (°C) ΔTma (°C)
AVDb 31.5 46.5 83.5 ± 0.1 117.0 ± 0.7 33.5
AVD(K111I) 31.9 42.1 76.5 ± 0.1 116.0 ± 0.1 39.5
AVR2 30.1 66.6 91.3 112.5 21.2
AVR2(I109K) 30.0 68.2 97.6 ± 0.1 118.7 ± 0.1 21.1
AVR2-b 30.9 55.1 93.4 ± 0.4 112.5 19.1
AVR4c 31.9 42.1 106.4 ± 0.8 125.4 ± 0.8 19.0
AVR4(K109I) 31.3 49.8 104.5 ± 0.5 118.6 ± 1.2 14.1
AVR6-b 31.1 51.5 87.7 114.0 26.3
aΔTm is the change in Tm upon addition of a three-fold molar excess of D-biotin.
bDSC results for avidin are from Ref [61].
cDSC results for AVR4 are from Ref [36].
Table 3 Data collection and structure determination statistics for AVR2.
Data collectiona
Wavelength (Å) 0.934
Beamline ID14-1 (ESRF)
Detector ADSC Q4R CCD
Resolution (Å) 25 – 1.40 (1.50 – 1.40)
Unique observations 257876 (47505)
I/sigma 14.3 (3.2)
Rsymb (%) 7.2 (54.6)
Completeness 99.6 (99.0)
Redundancy 6.0 (5.9)
Refinement
Space group P212121
Unit cell:
a, b, c (Å) 97.7, 99.9, 135.2
α, β, γ (°) 90, 90, 90
Monomers (asymmetric unit) 8
Resolution (Å) 25 – 1.40
Rworkc (%) 17.4
Rfreec,d (%) 20.1
Protein atoms 7720
Heterogen. atoms 234
Solvent atoms 1168
R.m.s.d:
Bond lengths (Å) 0.015
Bond angles (°) 1.7
Ramachandran plot:
Residues in most favored regions 94.3%
Residues in additional allowed regions 5.7%
aThe numbers in parenthesis refer to the highest resolution bin.
b
c
dPerformed on 5% of the reflections.
==== Refs
Green NM Avidin Adv Prot Chem 1975 29 85 133
Green NM Avidin and streptavidin Method Enzymol 1990 184 51 67
Wilchek M Bayer EA Introduction to avidin-biotin technology Method Enzymol 1990 184 5 13
Wilchek M Bayer EA Foreword and introduction to the book (strept)avidin-biotin system Biomol Eng 1999 16 1 4 10796978 10.1016/S1050-3862(99)00032-7
Hertz R Sebrell WH Occurrence of avidin in the oviduct and secretions of the genital tract of several species Science 1942 96 257
Jones PDBMH Distribution of avidin Life Sci 1962 11 621 623 10.1016/0024-3205(62)90094-2
Korpela JK Kulomaa MS Elo HA Tuohimaa PJ Biotin-binding proteins in eggs of oviparous vertebrates Experientia 1981 37 1065 1066 7308390 10.1007/BF02085010
Keinänen RA Laukkanen ML Kulomaa MS Molecular cloning of three structurally related genes for chicken avidin J Steroid Biochem 1988 30 17 21 2838690 10.1016/0022-4731(88)90071-4
Keinänen RA Wallén MJ Kristo PA Laukkanen MO Toimela TA Helenius MA Kulomaa MS Molecular cloning and nucleotide sequence of chicken avidin-related genes 1-5 Eur J Biochem 1994 220 615 621 8125122 10.1111/j.1432-1033.1994.tb18663.x
Ahlroth MK Kola EH Ewald D Masabanda J Sazanov A Fries R Kulomaa MS Characterization and chromosomal localization of the chicken avidin gene family Anim Genet 2000 31 367 375 11167523 10.1046/j.1365-2052.2000.00681.x
Kunnas TA Wallén MJ Kulomaa MS Induction of chicken avidin and related mRNAs after bacterial infection Biochim Biophys Acta 1993 1216 441 445 8268225
Niskanen EA Hytönen VP Grapputo A Nordlund HR Kulomaa MS Laitinen OH Chicken genome analysis reveals novel genes encoding biotin-binding proteins related to avidin family BMC Genomics 2005 6 41 15777476 10.1186/1471-2164-6-41
Ahlroth MK Ahlroth P Kulomaa MS Copy-number fluctuation by unequal crossing-over in the chicken avidin gene family Biochem Bioph Res Co 2001 288 400 406 10.1006/bbrc.2001.5760
Wallén MJ Laukkanen MO Kulomaa MS Cloning and sequencing of the chicken egg-white avidin-encoding gene and its relationship with the avidin-related genes Avr1-Avr5 Gene 1995 161 205 209 7665080 10.1016/0378-1119(95)00187-B
Nardone E Rosano C Santambrogio P Curnis F Corti A Magni F Siccardi AG Paganelli G Losso R Apreda B Bolognesi M Sidoli A Arosio P Biochemical characterization and crystal structure of a recombinant hen avidin and its acidic mutant expressed in Escherichia coli Eur J Biochem 1998 256 453 460 9760187 10.1046/j.1432-1327.1998.2560453.x
Marttila AT Airenne KJ Laitinen OH Kulik T Bayer EA Wilchek M Kulomaa MS Engineering of chicken avidin: a progressive series of reduced charge mutants FEBS Lett 1998 441 313 317 9883906 10.1016/S0014-5793(98)01570-1
Laitinen OH Airenne KJ Marttila AT Kulik T Porkka E Bayer EA Wilchek M Kulomaa MS Mutation of a critical tryptophan to lysine in avidin or streptavidin may explain why sea urchin fibropellin adopts an avidin-like domain FEBS Lett 1999 461 52 58 10561495 10.1016/S0014-5793(99)01423-4
Marttila AT Laitinen OH Airenne KJ Kulik T Bayer EA Wilchek M Kulomaa MS Recombinant NeutraLite avidin: a non-glycosylated, acidic mutant of chicken avidin that exhibits high affinity for biotin and low non-specific binding properties FEBS Lett 2000 467 31 36 10664451 10.1016/S0014-5793(00)01119-4
Laitinen OH Marttila AT Airenne KJ Kulik T Livnah O Bayer EA Wilchek M Kulomaa MS Biotin induces tetramerization of a recombinant monomeric avidin. A model for protein-protein interactions J Biol Chem 2001 276 8219 8224 11076945 10.1074/jbc.M007930200
Laitinen OH Nordlund HR Hytönen VP Uotila ST Marttila AT Savolainen J Airenne KJ Livnah O Bayer EA Wilchek M Kulomaa MS Rational design of an active avidin monomer J Biol Chem 2003 278 4010 4014 12458212 10.1074/jbc.M205844200
Marttila AT Hytönen VP Laitinen OH Bayer EA Wilchek M Kulomaa MS Mutation of the important Tyr-33 residue of chicken avidin: functional and structural consequences Biochem J 2003 369 249 254 12358604 10.1042/BJ20020886
Nordlund HR Laitinen OH Uotila ST Nyholm T Hytönen VP Slotte JP Kulomaa MS Enhancing the thermal stability of avidin. Introduction of disulfide bridges between subunit interfaces J Biol Chem 2003 278 2479 2483 12446662 10.1074/jbc.M210721200
Nordlund HR Hytönen VP Laitinen OH Uotila ST Niskanen EA Savolainen J Porkka E Kulomaa MS Introduction of histidine residues into avidin subunit interfaces allows pH-dependent regulation of quaternary structure and biotin binding FEBS Lett 2003 555 449 454 14675754 10.1016/S0014-5793(03)01302-4
Nordlund HR Laitinen OH Hytönen VP Uotila ST Porkka E Kulomaa MS Construction of a dual chain pseudotetrameric chicken avidin by combining two circularly permuted avidins J Biol Chem 2004 279 36715 36719 15131113 10.1074/jbc.M403496200
Livnah O Bayer EA Wilchek M Sussman JL Three-dimensional structures of avidin and the avidin-biotin complex Proc Natl Acad Sci USA 1993 90 5076 5080 8506353
Pugliese L Coda A Malcovati M Bolognesi M Three-dimensional structure of the tetragonal crystal form of egg-white avidin in its functional complex with biotin at 2.7 Å resolution J Mol Biol 1993 231 698 710 8515446 10.1006/jmbi.1993.1321
Argarana CE Kuntz ID Birken S Axel R Cantor CR Molecular cloning and nucleotide sequence of the streptavidin gene Nucleic Acids Res 1986 14 1871 1882 3951999
Weber PC Ohlendorf DH Wendoloski JJ Salemme FR Structural origins of high-affinity biotin binding to streptavidin Science 1989 243 85 88 2911722
Sano T Pandori MW Chen X Smith CL Cantor CR Recombinant core streptavidins. A minimum-sized core streptavidin has enhanced structural stability and higher accessibility to biotinylated macromolecules J Biol Chem 1995 270 28204 28209 7499314 10.1074/jbc.270.47.28204
Chilkoti A Tan PH Stayton PS Site-directed mutagenesis studies of the high-affinity streptavidin-biotin complex: contributions of tryptophan residues 79, 108, and 120 Proc Natl Acad Sci USA 1995 92 1754 1758 7878054
Reznik GO Vajda S Smith CL Cantor CR Sano T Streptavidins with intersubunit crosslinks have enhanced stability Nat Biotechnol 1996 14 1007 1011 9631041 10.1038/nbt0896-1007
Stayton PS Nelson KE McDevitt TC Bulmus V Shimoboji T Ding Z Hoffman AS Smart and biofunctional streptavidin Biomol Eng 1999 16 93 99 10796990 10.1016/S1050-3862(99)00043-1
Qureshi MH Yeung JC Wu SC Wong SL Development and characterization of a series of soluble tetrameric and monomeric streptavidin muteins with differential biotin binding affinities J Biol Chem 2001 276 46422 46428 11584006 10.1074/jbc.M107398200
Pazy Y Raboy B Matto M Bayer EA Wilchek M Livnah O Structure-based rational design of streptavidin mutants with pseudo-catalytic activity J Biol Chem 2003 278 7131 7134 12493758 10.1074/jbc.M209983200
Laitinen OH Hytönen VP Ahlroth MK Pentikäinen OT Gallagher C Nordlund HR Ovod V Marttila AT Porkka E Heino S Johnson MS Airenne KJ Kulomaa MS Chicken avidin-related proteins show altered biotin-binding and physico-chemical properties as compared with avidin Biochem J 2002 363 609 617 11964162 10.1042/0264-6021:3630609
Hytönen VP Nyholm TK Pentikäinen OT Vaarno J Porkka EJ Nordlund HR Johnson MS Slotte JP Laitinen OH Kulomaa MS Chicken Avidin-related Protein 4/5 Shows Superior Thermal Stability when Compared with Avidin while Retaining High Affinity to Biotin J Biol Chem 2004 279 9337 9343 14660583 10.1074/jbc.M310989200
Hytönen VP Laitinen OH Airenne TT Kidron H Meltola NJ Porkka E Hörhä J Paldanius T Määttä JA Nordlund HR Johnson MS Salminen TA Airenne KJ Ylä-Herttuala S Kulomaa MS Efficient production of active chicken avidin using a bacterial signal peptide in Escherichia coli Biochem J 2004 384 385 390 15324300 10.1042/BJ20041114
Eisenberg-Domovich Y Hytönen VP Wilchek M Bayer EA Kulomaa MS Livnah O High-resolution crystal structure of an avidin-related protein: insight into high-affinity biotin binding and protein stability Acta Crystallogr D 2005 61 528 538 15858262 10.1107/S0907444905003914
Hytönen VP Määttä JA Nyholm TK Livnah O Eisenberg-Domovich Y Hyre D Nordlund HR Hörhä J Niskanen EA Paldanius T Kulomaa T Porkka EJ Stayton PS Laitinen OH Kulomaa MS Design and construction of highly stable, protease-resistant chimeric avidins J Biol Chem 2005 280 10228 10233 15649900 10.1074/jbc.M414196200
Airenne KJ Oker-Blom C Marjomäki VS Bayer EA Wilchek M Kulomaa MS Production of biologically active recombinant avidin in baculovirus-infected insect cells Prot Exp Pur 1997 9 100 108 10.1006/prep.1996.0660
Hyre DE Le Trong I Freitag S Stenkamp RE Stayton PS Ser45 plays an important role in managing both the equilibrium and transition state energetics of the streptavidin-biotin system Protein Sci 2000 9 878 885 10850797
Gonzalez M Argarana CE Fidelio GD Extremely high thermal stability of streptavidin and avidin upon biotin binding Biomol Eng 1999 16 67 72 10796986 10.1016/S1050-3862(99)00041-8
Livnah O Bayer A Wilchek M Sussman JL The structure of the complex between avidin and the dye, 2-(4'-hydroxyazobenzene) benzoic acid (HABA) FEBS 1993 328 165 168 10.1016/0014-5793(93)80986-5
Ahlroth MK Grapputo A Laitinen OH Kulomaa MS Sequence features and evolutionary mechanisms in the chicken avidin gene family Biochem Bioph Res Co 2001 285 734 741 10.1006/bbrc.2001.5163
Kuntz ID Chen K Sharp KA Kollman PA The maximal affinity of ligands Proc Natl Acad Sci USA 1999 96 9997 10002 10468550 10.1073/pnas.96.18.9997
Kimber MS Vallee F Houston S Necakov A Skarina T Evdokimova E Beasley S Christendat D Savchenko A Arrowsmith CH Vedadi M Gerstein M Edwards AM Data mining crystallization databases: knowledge-based approaches to optimize protein crystal screens Proteins 2003 51 562 568 12784215 10.1002/prot.10340
Jancarik J Scott WG Milligan DL Koshland DEJ Kim SH Crystallization and preliminary X-ray diffraction study of the ligand-binding domain of the bacterial chemotaxis-mediating aspartate receptor of Salmonella typhimurium J Mol Biol 1991 221 31 34 1656050 10.1016/0022-2836(91)80198-4
Kabsch W Automatic Processing of Rotation Diffraction Data from Crystals of Initially Unknown Symmetry and Cell Constants J Appl Crystallogr 1993 26 795 800 10.1107/S0021889893005588
Collaborative computational project number 4 The CCP4 suite: programs for protein crystallography Acta Crystallogr D Biol Crystallogr 1994 50 760 763 15299374 10.1107/S0907444994003112
Navaza J Amore - an automated package for molecular replacement Acta Crystallogr A 1994 50 157 163 10.1107/S0108767393007597
Vagin A Teplyakov A MOLREP: an automated program for molecular replacement J Appl Crystallogr 1997 30 1022 1025 10.1107/S0021889897006766
Perrakis A Antoniadou-Vyza E Tsitsa P Lamzin VS Wilson KS Hamodrakas SJ Molecular, crystal and solution structure of a beta-cyclodextrin complex with the bromide salt of 2-(3-dimethylaminopropyl)tricyclo[3.3.1.1(3,7)]decan-2-ol, a potent antimicrobial drug Carbohydr Res 1999 317 19 28 10498440 10.1016/S0008-6215(99)00021-X
Murshudov GN Vagin AA Dodson EJ Refinement of macromolecular structures by the maximum-likelihood method Acta Crystallogr D 1997 53 240 255 15299926 10.1107/S0907444996012255
Jones TA Zou JY Cowan SW Kjeldgaard Improved methods for building protein models in electron density maps and the location of errors in these models Acta Crystallogr A 1991 47 110 119 2025413 10.1107/S0108767390010224
Lamzin VS Automated refinement of protein models Acta Crystallogr D 1993 49 129 147 15299554 10.1107/S0907444992008886
Zwart PH Langer GG Lamzin VS Modelling bound ligands in protein crystal structures Acta Crystallogr D 2004 60 2230 2239 15572776 10.1107/S0907444904012995
Emsley P Cowtan K Coot: model-building tools for molecular graphics Acta Crystallogr D 2004 60 2126 2132 15572765 10.1107/S0907444904019158
Laskowski RA Macarthur MW Moss DS Thornton JM Procheck - a program to check the stereochemical quality of protein structures J Appl Crystallogr 1993 26 283 291 10.1107/S0021889892009944
Vriend G WHAT IF: a molecular modeling and drug design program J Mol Graph 1990 8 52 56 2268628 10.1016/0263-7855(90)80070-V
Klumb LA Chu V Stayton PS Energetic roles of hydrogen bonds at the ureido oxygen binding pocket in the streptavidin-biotin complex Biochemistry 1998 37 7657 7663 9601024 10.1021/bi9803123
Hytönen VP Laitinen OH Grapputo A Kettunen A Savolainen J Kalkkinen N Marttila AT Nordlund HR Nyholm TK Paganelli G Kulomaa MS Characterization of poultry egg-white avidins and their potential as a tool in pretargeting cancer treatment Biochem J 2003 372 219 225 12558501 10.1042/BJ20021531
Johnson MS Overington JP A structural basis for sequence comparisons. An evaluation of scoring methodologies J Mol Biol 1993 233 716 738 8411177 10.1006/jmbi.1993.1548
Johnson MS May AC Rodionov MA Overington JP Discrimination of common protein folds: application of protein structure to sequence/structure comparisons Method Enzymol 1996 266 575 598
Felsenstein J Confidence limits on phylogenies: an approach using the bootstrap Evolution 1985 39 783 791
Felsenstein J PHYLIP -- Phylogeny Inference Package (Version 3.2). Cladistics 1989 5 164 166
Felsenstein J PHYLIP (Phylogeny Inference Package) version 3.5c. Distributed by the author. 1993 Seattle, Department of Genetics, University of Washington
Lehtonen JV Still DJ Rantanen VV Ekholm J Björklund D Iftikhar Z Huhtala M Repo S Jussila A Jaakkola J Pentikainen OT Nyrönen T Salminen TA Gyllenberg M Johnson M BODIL: a molecular modeling environment for structure-function analysis and drug design. J Comput Aided Mol Des 2004 18 401 419 15663001 10.1007/s10822-004-3752-4
DeLano WL The PyMOL Molecular Graphics System 2002 , DeLano Scientific, San Carlos, CA, USA.
Laskowski RA SURFNET: a program for visualizing molecular surfaces, cavities, and intermolecular interactions J Mol Graph 1995 13 323-330 307-308 8603061 10.1016/0263-7855(95)00073-9
Bayer EA Ehrlich-Rogozinski S Wilchek M Sodium dodecyl sulfate-polyacrylamide gel electrophoretic method for assessing the quaternary state and comparative thermostability of avidin and streptavidin Electrophoresis 1996 17 1319 1324 8874057 10.1002/elps.1150170808
Gonzalez M Bagatolli LA Echabe I Arrondo JLR Argarana CE Cantor CR Fidelio GD Interaction of biotin with streptavidin. Thermostability and conformational changes upon binding J Biol Chem 1997 25 11288 11294
Waner MJ Navrotskaya I Bain A Oldham ED Mascotti DP Thermal and sodium dodecylsulfate induced transitions of streptavidin Biophys J 2004 87 2701 2713 15298874 10.1529/biophysj.104.047266
|
16212654
|
PMC1282572
|
CC BY
|
2021-01-04 16:02:58
|
no
|
BMC Biotechnol. 2005 Oct 7; 5:28
|
utf-8
|
BMC Biotechnol
| 2,005 |
10.1186/1472-6750-5-28
|
oa_comm
|
==== Front
BMC CancerBMC Cancer1471-2407BioMed Central London 1471-2407-5-1291621266210.1186/1471-2407-5-129Case ReportSerous borderline tumor of the fallopian tube presented as hematosalpinx: a case report Krasevic Maja [email protected] Teodora [email protected] Oleg [email protected] Neda [email protected] Department of Pathology, School of Medicine, University of Rijeka, Br. Branchetta 20, Rijeka, Croatia2 Department of Citology, Clinical Hospital Centar, Rijeka, Cambierieva 17/5, Croatia3 Department of Obstetrics and Gynecology, Clinical Hospital Centar, School of Medicine, University of Rijeka, Cambierieva 17/5, Rijeka, Croatia2005 7 10 2005 5 129 129 14 6 2005 7 10 2005 Copyright © 2005 Krasevic et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
Compared with their ovarian counterparts, serous borderline tumors of the fallopian tube are uncommon, with limited experience about their clinical behaviour. We present a case of serous borderline tumor of the fallopian tube with unusual presentation and summarise all the published cases to date.
Case presentation
A case of serous borderline tumor of the fallopian tube in a 34-year old patient is presented, incidentally found during routine gynecologic examination. At laparoscopy the tumor was unusualy presented as hematosalpinx and was treated by salpingectomy. Cell-cycle analysis of the tumor tissue revealed a diploid DNA content and a low S-phase fraction. There was no evidence of the disease during the follow-up period of 4.6 years.
Conclusion
The current case and review of the literature suggest salpingectomy as the optimal treatment for patients with serous borderline tumor of the fallopian tube.
==== Body
Background
Compared with their ovarian counterparts, serous borderline tumors (SBTs) of the fallopian tube are uncommon, with limited experience about their clinical behaviour.
Herein we report a case of SBT of the fallopian tube in a young woman that was unusualy presented as hematosalpinx during the laparoscopic operation.
Case presentation
A 34-year old multigravida was found to have right adnexal mass on her routine gynecologic examination. Her previous medical history was uneventfull and Pap smear was normal. Transvaginal ultrasonography identified a cystic mass adjacent to the right ovary. Serum CA 125 was 5.1 U/ml (reference range: < 35 U/ml). At laparoscopy a dilated fallopian tube with bluish discoloration was found. The contralateral fallopian tube, ovaries and uterus were unremarkable. Exploration of the abdomino-pelvic cavity revealed smooth and shiny peritoneal surphace. Obtained peritoneal and pelvic washing were negative. Fine needle aspiration of dilated part of the fallopian tube revealed a 4 ml of bloody content. Cytological findings were consistent with hematosalpinx. Right salpingectomy was performed without using endoscopic bag. The patient was followed up by means of ultrasonography and serum CA 125 for 4.6 years. During this period she had no evidence of the disease.
Grossly, a 7.0 cm long fallopian tube was irregularly dilated up to 4.4 cm in diameter at the ampulary region. The fimbriae were intact. The serosal surphace was smooth. Sections of the dilated part of the fallopian tube revealed a cystic tumor with focally yellow to tan, soft papillary excrescences protruding into the lumen (Fig. 1A) and foci of intracystic hemorrhage. On microscopic examination the papillae were covered by serous type of epithelium, displaying stratification and budding with focal nuclear atypia (Fig. 1B). Three types of cells were recognised; ciliated cells, hob-nail cells and mesothelium-like cells. Small foci of tumor tissue necroses and hemorrhage were noted. There was no invasion of the supportive stroma of the papillae or into the fallopian tube wall. Focus of endosalpingiosis within the adjacent mesosalpinx was found. DNA analysis determined by flow cytometry paraffin technique revealed DNA diploid tumor with low S-phase fraction of 6.5 %.
Conclusion
To the best of our knowledge, only 6 cases of SBTs of the fallopian tube have been previously reported in the literature. Clinico-pathologic features of these tumors, including the current case are summarized in Table 1. The first case [1] that was originally reported as adenocarcinoma, Alvaredo-Cabrero et al [2] considered it to be a tumor of borderline malignancy. Age of the patients ranged from 19 to 34 (average 29.2) years. Four tumors were found incidentaly [2-4], three were presented with abdominal pain [1,5,6]. Serum CA 125 was elevated in one case [5]. All the tumors were unilateral, measuring from 1.7 – 13 cm. Three of them were located at the fimbriated end, two at the ampulary region, and in one case the whole tube was enlarged. Grossly, all the tumors were cystic with internal papillation, except one that was solid [5]. Histologicaly the epithelium showed the morphology of borderline serous tumors, according to the criteria applied in the ovary.
SBTs were treated exclusively by surgery, most of them by unilateral salpingectomy. Complete staging/restaging was performed only in one case [5], where except salpingo-oophorectomy, contralateral ovarian biopsy, partial omentectomy and appendectomy were performed. Peritoneal and pelvic washing were obtained and multiple pelvic lymph nodes were sampled. Ten months later, contralateral ovary due to adenofibroma was removed and peritoneal and pelvic washing were obtained. The patient was without evidence of the disease during the follow-up period of 6 years.
Seidman and Kurman [7] reviewed 97 reports that included 4,129 patients with ovarian SBTs. They concluded that surgical-pathological stage and type of extraovarian disease (invasive versus noninvasive implants) were the most powerful prognostic indicators for these tumors. Survival rate for stage I SBTs was almost 100 % (99.5 %). Survival for advanced stage SBTs with noninvasive and invasive implants were 95.3 % and 66 % respectively.
In order to predict the behaviour of the SBT in current case, similary to the ovarian tumors, DNA ploidy was determined. In our case the tumor was diploid with low S-phase fraction of 6.3 %. DNA ploidy was found to be important prognostic factor in 321 patients with ovarian borderline tumors (BTs) [8]. Aneuploidy was associated with older age, more advanced disease and non-serous histologic type as well. However, although aneuploidy correlated with higher stage in a study of 42 ovarian BTs, ploidy/DNA index could not predict which borderline tumors would behave in a more aggressive fashion [9]. Analysis of 49 advanced stage ovarian SBTs revealed no correlation between DNA content and frequency of tumor recurrence [10]. No death occurred in a subset of patients with aneuploid tumors. The patients with aneuploid tumors were more likely to have received adjuvant chemotherapy, that could affect the results.
Four additional cases of BTs of the fallopian tube were reported. Three had mucinous type of epithelium [11-13] and one was endometrioid borderline tumour [2]. The mean age of patients with mucinous borderline tumors (MBT) was 54 years. Presented symptoms included constipation [11] and pelvic mass [11,13]. MBT was bilateral in one case [13]. Two cases were associated with pseudomyxoma peritonei [11,12]. Follow-up informations were not available in these cases.
The unique case of borderline endometrioid tumor was reported by Alvaredo-Cabrero et al in a serie of 20 cases of tumors of the fimbriated end of the fallopian tube [2]. The tumor was borderline adenofibroma and contained large number of endometrioid-type glands with focal villoglandular pattern separated by a fibromatous stroma. No follow-up information was available.
In conclusion, confinement to the tube as well as normal DNA content of the tumor may indicate a benign clinical behaviour of the SBTs of fallopian tube, suggesting unilateral salpingectomy as the optimal treatment for these young patients. However, long term follow-up and more cases are needed to make definitive conclusions.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
TS carried out the cytologic examination and participated in the design of the study.
OP and NSS have performed the operation and participated in drafting the manuscript.
MK carried out the pathohistological examination and participated in writing the manuscript.
All authors read and approved the final manuscript.
Pre-publication history
The pre-publication history for this paper can be accessed here:
Figures and Tables
Figure 1 Composite figure of serous borderline tumor of the fallopian tube. A) Papillary tumor protruding into the lumen of the fallopian tube (low magnification). There is no invasion into the wall of the tube (top). B) Stratification and budding of the epithelium with focal nuclear atypia (high magnification).
Table 1 Serous borderline tumors of the fallopian tube: clinical and pathological features of 7 cases.
Case no. (reference no.) Original diagnosis Age (yr) Symptoms Procedure Localization Gross finding Follow-up
1. (1) papillary cystadenocarcinoma 19 lower-guadrant pain, coul-de-sac mass partial salpingectomy fimbriated end 4 cm, cystic with papillations 18 mo, WED
2. (3) borderline cystadenofibroma 32 IF during an early pregnancy salpingectomy ampulary region 2.5 cm, cystic with papillations WED after unspecified time
3. (5) borderline papillary serous tumor 31 abdominal pain, elevated CA 125 salpingo-oophorectomy, ovarian biopsy, partial omentectomy protruding from the fimbria 6 cm, solid polipoid mass with cauliflower-like surface 6 yr, WED
4. (2) borderline serous tumor NS IF during an elective operation NS fimbriated end 1.7 cm, cystic with papillations 2.4 yr, well
5. (4) serous tumor of low malignant potential 31 IF on routine gynecologic examination salpingectomy entire tube 13 cm, cystic with papillations not available
6. (6) serous tumor of low malignant potential 28 lower quadrant pain salpingectomy NS 5 cm, cystic with papillations recently found
7. (current case) serous tumor of borderline malignancy 34 IF on routine gynecologic examination salpingectomy ampulary region 4.4 cm, cystic with papillations 4.6 yr, WED
NS, not specified; IF, incidental finding; WED, without evidence of the disease.
==== Refs
Gatto V Selim MA Lankerani M Primary carcinoma of the fallopian tube in an adolescent J Surg Oncol 1986 33 212 214 3773539
Alvarado-Cabrero I Navani SS Young RH Scully RE Tumors of the fimbriated end of the fallopian tube: a clinicopathologic analysis of 20 cases, including nine carcinomas Int J Gynecol Pathol 1997 16 189 196 9421082
Casasola SV Mindan JP Cystadenofibroma of the fallopian tube Appl Pathol 1989 7 256 259 2803787
Kayaalp E Heller DS Majmudar B Serous tumor of low malignant potential of the fallopian tube Int J Gynecol Pathol 2000 19 398 400 11109174 10.1097/00004347-200010000-00019
Zheng W Wolf S Kramer EE Cox KA Hoda SA Borderline papillary serous tumor of the fallopian tube Am J Surg Pathol 1996 20 30 35 8540606 10.1097/00000478-199601000-00003
Haratz-Rubinstein N Fromberg E Lederman S Sonographic diagnosis of serous tumor of low malignant potential of the fallopian tube J Ultrasound Med 2004 23 869 872 15244314
Seidman JD Kurman RJ Ovarian serous borderline tumors: a critical rewiev of the literature with emphasis on prognostic indicarors Hum Pathol 2000 31 539 557 10836293 10.1053/hp.2000.8048
Kaern J Trope CG Kristensen GB Abeler VM Pettersen EO DNA ploidy; the most important prognostic factor in patients with borderline tumors of the ovary Int J Gynecol Cancer 1993 3 349 358 11578368 10.1046/j.1525-1438.1993.03060349.x
Demirel D Laucirica R Fishman A Ovarian tumors of low malignant potential: correlation of DNA index and S-phase fraction with histopathologic grade and clinical outcome Cancer 1996 77 1494 1500 8608534 10.1002/(SICI)1097-0142(19960415)77:8<1494::AID-CNCR11>3.0.CO;2-V
Gilks CB Alkushi A Yue JJS Lanvin D Ehlen TG Miller DM Advanced-stage serous borderline tumors of the ovary: a clinicopathological study of 49 cases Int J Gynecol Pathol 2003 22 29 36 12496695 10.1097/00004347-200301000-00008
McCarthy JH Aga R A fallopian tube lesion of borderline malignancy associated with pseudo-myxoma peritonei Histopathology 1988 13 223 225 3169690
Friedmann W Minguillon C Wessel J Lichetenegger W Pickel H (Pseudomyxoma peritonei caused by proliferating mucinous adenoma of the mucosa of the fimbriae) Geburtshilfe Frauenheilkd 1990 50 579 580 2167862
Seidman JD Mucinous lesions of the fallopian tube Am J Surg Pathol 1994 18 1205 1212 7977943
|
16212662
|
PMC1282573
|
CC BY
|
2021-01-04 16:03:08
|
no
|
BMC Cancer. 2005 Oct 7; 5:129
|
utf-8
|
BMC Cancer
| 2,005 |
10.1186/1471-2407-5-129
|
oa_comm
|
==== Front
BMC CancerBMC Cancer1471-2407BioMed Central London 1471-2407-5-1291621266210.1186/1471-2407-5-129Case ReportSerous borderline tumor of the fallopian tube presented as hematosalpinx: a case report Krasevic Maja [email protected] Teodora [email protected] Oleg [email protected] Neda [email protected] Department of Pathology, School of Medicine, University of Rijeka, Br. Branchetta 20, Rijeka, Croatia2 Department of Citology, Clinical Hospital Centar, Rijeka, Cambierieva 17/5, Croatia3 Department of Obstetrics and Gynecology, Clinical Hospital Centar, School of Medicine, University of Rijeka, Cambierieva 17/5, Rijeka, Croatia2005 7 10 2005 5 129 129 14 6 2005 7 10 2005 Copyright © 2005 Krasevic et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
Compared with their ovarian counterparts, serous borderline tumors of the fallopian tube are uncommon, with limited experience about their clinical behaviour. We present a case of serous borderline tumor of the fallopian tube with unusual presentation and summarise all the published cases to date.
Case presentation
A case of serous borderline tumor of the fallopian tube in a 34-year old patient is presented, incidentally found during routine gynecologic examination. At laparoscopy the tumor was unusualy presented as hematosalpinx and was treated by salpingectomy. Cell-cycle analysis of the tumor tissue revealed a diploid DNA content and a low S-phase fraction. There was no evidence of the disease during the follow-up period of 4.6 years.
Conclusion
The current case and review of the literature suggest salpingectomy as the optimal treatment for patients with serous borderline tumor of the fallopian tube.
==== Body
Background
Compared with their ovarian counterparts, serous borderline tumors (SBTs) of the fallopian tube are uncommon, with limited experience about their clinical behaviour.
Herein we report a case of SBT of the fallopian tube in a young woman that was unusualy presented as hematosalpinx during the laparoscopic operation.
Case presentation
A 34-year old multigravida was found to have right adnexal mass on her routine gynecologic examination. Her previous medical history was uneventfull and Pap smear was normal. Transvaginal ultrasonography identified a cystic mass adjacent to the right ovary. Serum CA 125 was 5.1 U/ml (reference range: < 35 U/ml). At laparoscopy a dilated fallopian tube with bluish discoloration was found. The contralateral fallopian tube, ovaries and uterus were unremarkable. Exploration of the abdomino-pelvic cavity revealed smooth and shiny peritoneal surphace. Obtained peritoneal and pelvic washing were negative. Fine needle aspiration of dilated part of the fallopian tube revealed a 4 ml of bloody content. Cytological findings were consistent with hematosalpinx. Right salpingectomy was performed without using endoscopic bag. The patient was followed up by means of ultrasonography and serum CA 125 for 4.6 years. During this period she had no evidence of the disease.
Grossly, a 7.0 cm long fallopian tube was irregularly dilated up to 4.4 cm in diameter at the ampulary region. The fimbriae were intact. The serosal surphace was smooth. Sections of the dilated part of the fallopian tube revealed a cystic tumor with focally yellow to tan, soft papillary excrescences protruding into the lumen (Fig. 1A) and foci of intracystic hemorrhage. On microscopic examination the papillae were covered by serous type of epithelium, displaying stratification and budding with focal nuclear atypia (Fig. 1B). Three types of cells were recognised; ciliated cells, hob-nail cells and mesothelium-like cells. Small foci of tumor tissue necroses and hemorrhage were noted. There was no invasion of the supportive stroma of the papillae or into the fallopian tube wall. Focus of endosalpingiosis within the adjacent mesosalpinx was found. DNA analysis determined by flow cytometry paraffin technique revealed DNA diploid tumor with low S-phase fraction of 6.5 %.
Conclusion
To the best of our knowledge, only 6 cases of SBTs of the fallopian tube have been previously reported in the literature. Clinico-pathologic features of these tumors, including the current case are summarized in Table 1. The first case [1] that was originally reported as adenocarcinoma, Alvaredo-Cabrero et al [2] considered it to be a tumor of borderline malignancy. Age of the patients ranged from 19 to 34 (average 29.2) years. Four tumors were found incidentaly [2-4], three were presented with abdominal pain [1,5,6]. Serum CA 125 was elevated in one case [5]. All the tumors were unilateral, measuring from 1.7 – 13 cm. Three of them were located at the fimbriated end, two at the ampulary region, and in one case the whole tube was enlarged. Grossly, all the tumors were cystic with internal papillation, except one that was solid [5]. Histologicaly the epithelium showed the morphology of borderline serous tumors, according to the criteria applied in the ovary.
SBTs were treated exclusively by surgery, most of them by unilateral salpingectomy. Complete staging/restaging was performed only in one case [5], where except salpingo-oophorectomy, contralateral ovarian biopsy, partial omentectomy and appendectomy were performed. Peritoneal and pelvic washing were obtained and multiple pelvic lymph nodes were sampled. Ten months later, contralateral ovary due to adenofibroma was removed and peritoneal and pelvic washing were obtained. The patient was without evidence of the disease during the follow-up period of 6 years.
Seidman and Kurman [7] reviewed 97 reports that included 4,129 patients with ovarian SBTs. They concluded that surgical-pathological stage and type of extraovarian disease (invasive versus noninvasive implants) were the most powerful prognostic indicators for these tumors. Survival rate for stage I SBTs was almost 100 % (99.5 %). Survival for advanced stage SBTs with noninvasive and invasive implants were 95.3 % and 66 % respectively.
In order to predict the behaviour of the SBT in current case, similary to the ovarian tumors, DNA ploidy was determined. In our case the tumor was diploid with low S-phase fraction of 6.3 %. DNA ploidy was found to be important prognostic factor in 321 patients with ovarian borderline tumors (BTs) [8]. Aneuploidy was associated with older age, more advanced disease and non-serous histologic type as well. However, although aneuploidy correlated with higher stage in a study of 42 ovarian BTs, ploidy/DNA index could not predict which borderline tumors would behave in a more aggressive fashion [9]. Analysis of 49 advanced stage ovarian SBTs revealed no correlation between DNA content and frequency of tumor recurrence [10]. No death occurred in a subset of patients with aneuploid tumors. The patients with aneuploid tumors were more likely to have received adjuvant chemotherapy, that could affect the results.
Four additional cases of BTs of the fallopian tube were reported. Three had mucinous type of epithelium [11-13] and one was endometrioid borderline tumour [2]. The mean age of patients with mucinous borderline tumors (MBT) was 54 years. Presented symptoms included constipation [11] and pelvic mass [11,13]. MBT was bilateral in one case [13]. Two cases were associated with pseudomyxoma peritonei [11,12]. Follow-up informations were not available in these cases.
The unique case of borderline endometrioid tumor was reported by Alvaredo-Cabrero et al in a serie of 20 cases of tumors of the fimbriated end of the fallopian tube [2]. The tumor was borderline adenofibroma and contained large number of endometrioid-type glands with focal villoglandular pattern separated by a fibromatous stroma. No follow-up information was available.
In conclusion, confinement to the tube as well as normal DNA content of the tumor may indicate a benign clinical behaviour of the SBTs of fallopian tube, suggesting unilateral salpingectomy as the optimal treatment for these young patients. However, long term follow-up and more cases are needed to make definitive conclusions.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
TS carried out the cytologic examination and participated in the design of the study.
OP and NSS have performed the operation and participated in drafting the manuscript.
MK carried out the pathohistological examination and participated in writing the manuscript.
All authors read and approved the final manuscript.
Pre-publication history
The pre-publication history for this paper can be accessed here:
Figures and Tables
Figure 1 Composite figure of serous borderline tumor of the fallopian tube. A) Papillary tumor protruding into the lumen of the fallopian tube (low magnification). There is no invasion into the wall of the tube (top). B) Stratification and budding of the epithelium with focal nuclear atypia (high magnification).
Table 1 Serous borderline tumors of the fallopian tube: clinical and pathological features of 7 cases.
Case no. (reference no.) Original diagnosis Age (yr) Symptoms Procedure Localization Gross finding Follow-up
1. (1) papillary cystadenocarcinoma 19 lower-guadrant pain, coul-de-sac mass partial salpingectomy fimbriated end 4 cm, cystic with papillations 18 mo, WED
2. (3) borderline cystadenofibroma 32 IF during an early pregnancy salpingectomy ampulary region 2.5 cm, cystic with papillations WED after unspecified time
3. (5) borderline papillary serous tumor 31 abdominal pain, elevated CA 125 salpingo-oophorectomy, ovarian biopsy, partial omentectomy protruding from the fimbria 6 cm, solid polipoid mass with cauliflower-like surface 6 yr, WED
4. (2) borderline serous tumor NS IF during an elective operation NS fimbriated end 1.7 cm, cystic with papillations 2.4 yr, well
5. (4) serous tumor of low malignant potential 31 IF on routine gynecologic examination salpingectomy entire tube 13 cm, cystic with papillations not available
6. (6) serous tumor of low malignant potential 28 lower quadrant pain salpingectomy NS 5 cm, cystic with papillations recently found
7. (current case) serous tumor of borderline malignancy 34 IF on routine gynecologic examination salpingectomy ampulary region 4.4 cm, cystic with papillations 4.6 yr, WED
NS, not specified; IF, incidental finding; WED, without evidence of the disease.
==== Refs
Gatto V Selim MA Lankerani M Primary carcinoma of the fallopian tube in an adolescent J Surg Oncol 1986 33 212 214 3773539
Alvarado-Cabrero I Navani SS Young RH Scully RE Tumors of the fimbriated end of the fallopian tube: a clinicopathologic analysis of 20 cases, including nine carcinomas Int J Gynecol Pathol 1997 16 189 196 9421082
Casasola SV Mindan JP Cystadenofibroma of the fallopian tube Appl Pathol 1989 7 256 259 2803787
Kayaalp E Heller DS Majmudar B Serous tumor of low malignant potential of the fallopian tube Int J Gynecol Pathol 2000 19 398 400 11109174 10.1097/00004347-200010000-00019
Zheng W Wolf S Kramer EE Cox KA Hoda SA Borderline papillary serous tumor of the fallopian tube Am J Surg Pathol 1996 20 30 35 8540606 10.1097/00000478-199601000-00003
Haratz-Rubinstein N Fromberg E Lederman S Sonographic diagnosis of serous tumor of low malignant potential of the fallopian tube J Ultrasound Med 2004 23 869 872 15244314
Seidman JD Kurman RJ Ovarian serous borderline tumors: a critical rewiev of the literature with emphasis on prognostic indicarors Hum Pathol 2000 31 539 557 10836293 10.1053/hp.2000.8048
Kaern J Trope CG Kristensen GB Abeler VM Pettersen EO DNA ploidy; the most important prognostic factor in patients with borderline tumors of the ovary Int J Gynecol Cancer 1993 3 349 358 11578368 10.1046/j.1525-1438.1993.03060349.x
Demirel D Laucirica R Fishman A Ovarian tumors of low malignant potential: correlation of DNA index and S-phase fraction with histopathologic grade and clinical outcome Cancer 1996 77 1494 1500 8608534 10.1002/(SICI)1097-0142(19960415)77:8<1494::AID-CNCR11>3.0.CO;2-V
Gilks CB Alkushi A Yue JJS Lanvin D Ehlen TG Miller DM Advanced-stage serous borderline tumors of the ovary: a clinicopathological study of 49 cases Int J Gynecol Pathol 2003 22 29 36 12496695 10.1097/00004347-200301000-00008
McCarthy JH Aga R A fallopian tube lesion of borderline malignancy associated with pseudo-myxoma peritonei Histopathology 1988 13 223 225 3169690
Friedmann W Minguillon C Wessel J Lichetenegger W Pickel H (Pseudomyxoma peritonei caused by proliferating mucinous adenoma of the mucosa of the fimbriae) Geburtshilfe Frauenheilkd 1990 50 579 580 2167862
Seidman JD Mucinous lesions of the fallopian tube Am J Surg Pathol 1994 18 1205 1212 7977943
|
16262904
|
PMC1282574
|
CC BY
|
2021-01-04 16:30:07
|
no
|
BMC Cardiovasc Disord. 2005 Nov 1; 5:32
|
latin-1
|
BMC Cardiovasc Disord
| 2,005 |
10.1186/1471-2261-5-32
|
oa_comm
|
==== Front
BMC Evol BiolBMC Evolutionary Biology1471-2148BioMed Central London 1471-2148-5-581625578310.1186/1471-2148-5-58Methodology ArticleThe efficiency of different search strategies in estimating parsimony jackknife, bootstrap, and Bremer support Müller Kai F [email protected] Nees-Institut für Biodiversität der Pflanzen, Rheinische Friedrich-Wilhelms-Universität Bonn, Meckenheimer Allee 170, Bonn, D-53115, Germany2005 29 10 2005 5 58 58 9 5 2005 29 10 2005 Copyright © 2005 Müller; licensee BioMed Central Ltd.2005Müller; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
For parsimony analyses, the most common way to estimate confidence is by resampling plans (nonparametric bootstrap, jackknife), and Bremer support (Decay indices). The recent literature reveals that parameter settings that are quite commonly employed are not those that are recommended by theoretical considerations and by previous empirical studies. The optimal search strategy to be applied during resampling was previously addressed solely via standard search strategies available in PAUP*. The question of a compromise between search extensiveness and improved support accuracy for Bremer support received even less attention. A set of experiments was conducted on different datasets to find an empirical cut-off point at which increased search extensiveness does not significantly change Bremer support and jackknife or bootstrap proportions any more.
Results
For the number of replicates needed for accurate estimates of support in resampling plans, a diagram is provided that helps to address the question whether apparently different support values really differ significantly. It is shown that the use of random addition cycles and parsimony ratchet iterations during bootstrapping does not translate into higher support, nor does any extension of the search extensiveness beyond the rather moderate effort of TBR (tree bisection and reconnection branch swapping) plus saving one tree per replicate. Instead, in case of very large matrices, saving more than one shortest tree per iteration and using a strict consensus tree of these yields decreased support compared to saving only one tree. This can be interpreted as a small risk of overestimating support but should be more than compensated by other factors that counteract an enhanced type I error. With regard to Bremer support, a rule of thumb can be derived stating that not much is gained relative to the surplus computational effort when searches are extended beyond 20 ratchet iterations per constrained node, at least not for datasets that fall within the size range found in the current literature.
Conclusion
In view of these results, calculating bootstrap or jackknife proportions with narrow confidence intervals even for very large datasets can be achieved with less expense than often thought. In particular, iterated bootstrap methods that aim at reducing statistical bias inherent to these proportions are more feasible when the individual bootstrap searches require less time.
==== Body
Background
"Without some assessment of reliability, a phylogeny has limited value" (Sanderson 1995: 299) – an examination of the recent phylogenetic literature shows there is a general agreement on this fact, and only few molecular phylogenetic studies of the past 15 years exist in which no estimate of confidence is provided. In the context of cladistic (parsimony) analyses, two basic types are most common: resampling plans (bootstrap, jackknife), and those based on the length difference of trees (Bremer support).
Bootstrap and jackknife in parsimony analyses
Bootstrap and jackknife are computer intensive statistical methods for error estimation [1,2]. With regard to their applicability in phylogenetics [3,4], elaborate discussions exist in the literature [5-9]. The fact that fundamental statistical assumptions may not be met in the phylogenetic context (such as independent, identically distributed variables) could not prevent the bootstrap from becoming the most popular method for reliability assessment, in particular since many researchers consider bootstrap and jackknife merely as indications of relative support, not in a hypothesis-testing framework. Felsenstein [10] provides an easy-to-read but detailed description of resampling plans in phylogenetics and addresses solutions to circumvent some of these more fundamental problems.
Aside from that, the majority of applied phylogenetic studies hitherto do not provide a justification for using a certain number of replicates or a particular search strategy during each bootstrap (or jackknife) replicate. Unfortunately, often the parameter settings that are employed are not those that appear recommendable in view of the existing theoretical and empirical work that provides a guideline for the number of replicates [e.g., [11]] or search strategies [12-14] to be used. Phylogenetic trees are tools for understanding biological processes and gain more and more importance far outside the field of pure phylogenetics. Ideally, biological conclusions based on a given node in the tree should take into account the level of confidence one can have in the existence of the node. Therefore, it appears necessary to more efficiently spread the existing knowledge on the performance and interpretability of the bootstrap and jackknife under different circumstances, but also to address those questions that still remain open.
The number of replicates
Hedges argued that at least 1825 replicates are needed if one wants to attain ±1% accuracy for bootstrap proportions of 95% or higher [11]. The underlying considerations are based on the binomial distribution, which has the favorable characteristic of a variance σ2 that equals np(1-p). In the context of resampling plans, n is the number of replicates and p is the bootstrap or jackknife support value (bootstrap percentage or bootstrap p value) expressed as a fraction of 1, i.e., the proportion of replicates that yielded a tree containing a particular phylogenetic group. Therefore, the n needed to attain ±a accuracy (as 95% confidence interval spanning ±1.96 σ) at a support level p can readily be calculated as
n = p (1-p)(1.95996.../ a)2. (1)
Hedges (1992) based the determination of the confidence interval upon an approximation of the binomial distribution by the normal distribution. However, when np ≤ 5 and n(1-p) ≤ 5 (roughly), much precision is gained when standard errors are based on the binomial distribution. This condition is easily fulfilled at high probabilities. Here, the confidence intervals become asymmetric.
Let Y be the number of successes (replicates that yield a given node) out of n trials (replicates), and p the success probability of each trial. The lower endpoint of the (1 - α)100%-confidence interval around the estimated p (derived from observing m = pn successes) is given by the pl such that
Similarly, the upper endpoint is defined by finding the pu such that
Rather than odd numbers such as "1825", researchers currently use "500", "1000", etc. replicates, and the question arises, what the confidence interval for a given p may be when these number of replicates are used. This is easily derived from Eq. (1) or (more precisely) from Eqs. (2)-(3), and Fig. 1 graphs these intervals for support levels of ≥50 and common replicate numbers. With help of this graph, the frequent issue whether two apparently different support values really differ can relatively easily be addressed visually. The decrease of BS/JS (bootstrap or jackknife support) standard deviations with an increase of the number of replicates (Fig. 1), which theoretically follows from the above equations, was confirmed in detailed empirical studies on the topic using real datasets of Saxifragaceae [12] and Orchidaceae [15].
Figure 1 95% confidence intervals at jackknife and bootstrap frequencies between 50 and 100. 95% confidence intervals at jackknife and bootstrap frequencies between 50 and 100. For a detailed explanation see text.
Search extensiveness
From a theoretical point of view, the question of the optimal number of bootstrap replicates is easier to solve than that of the optimal heuristic search strategy to be applied during bootstrap replicates of cladistic analyses. Farris et al. [4] initially argued that thorough swapping during replicates is unnecessary. The whole point of using resampling is "to avoid drawing poorly-supported conclusions" [4: 117] and identifying those groups that are strongly supported by the data. The most strongly supported groups are certainly easily identified via bootstrapping or jackknifing when no swapping is performed, but at the expense of risking that some nodes are ignored that could gain significant support (whatever one regards as significant). 1 - s is conventionally interpreted as type I error (α) in hypothesis tests where a group of taxa will be considered monophyletic if BS > s (H0: group is not monophyletic, H1: group is monophyletic; α: probability of rejecting H0 while it is true; [16]). Thus, BS/JS that is systematically too low at least does not entail an increased danger of mistakenly inferring monophyly. It does, however, lead to rejecting monophyly incorrectly too often, which is not desirable, either. Nonetheless, one of Farris & al.'s main points was to contrast the speed of their jackknifing approach with the slower neighbor-joining bootstrap and extensive heuristic parsimony searches to identify MP trees (most parsimonious trees); branch swapping during each replicate would have strongly lowered the performance contrast. Therefore, the first version of Farris' jackknifing application, 'JAC', did not perform branch swapping at all.
A number of studies provided practical evidence from real datasets that the non-branch swapping approaches (as in JAC or the "fast" option in PAUP*) yield significantly lower support estimates than analyses performing some kind of branch swapping [12-14,17,18]. Accordingly, branch swapping was later added in the upgrade 'XAC'. In line with Farris & al.'s basic assumptions, however, it was concluded that swapping on more than 1 to 2 trees per iteration does not change support significantly [15]. This implied that the increased computational effort connected with more extensive searches per replicate does not necessarily translate into more accurate estimates. Note that these examinations used the random addition search strategy available in PAUP* [19], which is known to relatively soon fail to find shortest trees as the number of terminal sampled increases (Nixon, 1999). This is why cladistic analyses of datasets approaching or exceeding 100 to 150 taxa (although strongly depending on the dataset) usually make use of the parsimony ratchet [20] or other fast cladistic algorithms, available through a number of software tools [21-24].
These results raise the question in how far bootstrap percentages are affected by trees found in each bootstrap replicate being far from most parsimonious. The above-cited increase in search exhaustiveness, namely moving from non-branch swapping via NNI and SPR to TBR, yielded a considerable increase in support [12,18]. Swapping on more trees (via RAS) apparently did not increase, but sometimes even lessened the average support [15]. These investigations, however, were conducted on datasets for which it was quite likely that shortest trees could be encountered without much search effort per iteration because size and homoplasy still allowed that shortest trees could be found with intermediate effort without algorithms designed for particularly large datasets.
If the standard error of the BS/JS value itself is neglected (e.g., assuming bootstrap/jackknife searches with an infinite number of replicates and a confidence interval for each BS/JS frequency approaching zero width), a dataset-dependent graph can be imagined in which the bootstrap is a function of the exhaustiveness of the search. If we further ignore for a moment the effect of how many trees per replicate are actually used in the majority rule consensus [14], and whether tree weighting is employed for these, this graph will asymptotically approach a "BS/JS level of saturation" at which further increase in exhaustiveness does not increase BS/JS. As a rough guideline we expect that the more taxa, the later this level of saturation will be reached. Certainly other factors such as homoplasy and phylogenetic signal inherent to characters play a significant role, but the mere fact that the number of possible tree topologies soon reaches astronomical dimensions [25] provides the most severe limitations to search algorithms. Somehow a point has now to be chosen at which one decides that the increase in support is not worth the additional search effort. A statistic could be chosen that describes the support level for each discrete level of exhaustiveness and subjected to hierarchical significance tests.
Without the unrealistic assumption of an infinitely narrow confidence interval around each support level, things become more complicated. The size of the (let's say 95%) confidence interval depends on the support level one looks at, less dramatically so when the number of replicates becomes very large (Fig. 1). Thus, to a considerable extend, drawing conclusions on the relative merits of certain search strategies has to take into account the support levels of interest. Commonly, these will fall in the interval [80;100], but less likely only in the standard interval [95;100] due to the frequently cited conservativeness of the nonparametric bootstrap [14,16,26,27].
The relationship between the number of trees per replicate fed into the consensus calculation and the BS/JS is still less straightforward. The more conservative approach of using strict consensus trees of each replicate for the final consensus tree [e.g., [28]], referred to as "strict-consensus approach" (SC) by Davis et al. [29], can be expected to always result in equal or lower support than the standard approach in PAUP*, for which the term "frequency-within-replicates approach" (FWR) has been coined [29,30]. The latter employs tree weights that maintain information on nodes that other trees found in the same replicate lack. This theoretical expectation was empirically corroborated very recently using a 218-terminal dataset [29]. The SC approach [28] may more closely reflect the use of the bootstrap outside the field of phylogenetics, but is only rarely pursued in published analyses [e.g., [31-33]]. Restricting the discussion to the tree-weighting approach in PAUP*, it is hard to predict whether additional trees saved per replicate will decrease resolution of the final majority rule consensus, because these additional trees are usually also swapped upon and thus enhance the probability of finding trees closer to the optimal score of the current replicate. The latter effect is counter to the first, and which effect will be stronger depends on a whole array of parameters, probably above all on the sampling size and thoroughness of the search. Using a 173 taxa data set with 1180 parsimony informative characters, Freudenstein et al. found that beyond 2 trees, saving more trees considerably decreases overall support, if these trees are from the same RAS (random addition search) iteration and not obtained via additional RAS replications [15: 151]. The authors, however, did not test the effect of increasing search extensiveness beyond 2 RAS replicates per jackknife replicate.
Jackknifing versus bootstrapping
Principally, all considerations below apply equally to the bootstrap and jackknife. Farris et al. argue that, in order to directly compare jackknife frequencies with bootstrap frequencies, the probability that a character appears in the resampled matrix has to be set to 1-1/e [4] (along with other requirements). This was recently emphasized again by Freudenstein et al. [15]. Felsenstein disagrees with that view, demonstrating that at least sometimes a 50%-deletion-jackknife more closely reflects bootstrap proportions [10]. In any case, when comparing the behavior of bootstrap and jackknife, differences in the support levels are also a function of the probability that a character appears in the resampled matrix and, thus, the amount of data used for tree inference in the pseudoreplicates.
The Bremer support
The Bremer support [BrS, [34-36]], a synonym of "decay index" [37], "length difference" [38], or "support index" [SI, [39,40]], is a completely different measure of branch support and has been addressed in detail recently [41,42]. For intermediate to large datasets, the calculation of BrS can be problematic, because support values can turn out to be severe overestimations of support [35] unless relatively thorough search strategies are invoked to assess support of each branch. It has been demonstrated that using the parsimony ratchet during Bremer support analysis is highly advantageous in such cases [43]. However, the data-dependent optimal compromise between ratchet search time and improved support accuracy has not been addressed so far and therefore will be briefly dealt with below.
Results and discussion
Bootstrapping and jackknifing
Empirical studies on four different molecular datasets (see Methods) of 86, 89, 385, and 567 taxa, respectively, yielded the following results. Using 10 or 20 parsimony ratchet cycles per jackknife replicate instead of one simple addition search with one saved tree yielded no enhanced support for the 86 and 89 taxa datasets (sign test: p < 0.05). This confirms conclusions from RAS searches on data sets of similar dimension [100 taxa, ref. [13]] or twice as big [173 taxa, ref. [15]]. For the considerably larger 385 and 567 taxa datasets, 10 ratchet cycles could not enhance support significantly, either (sign test: p < 0.05). Moreover, the effect of saving more than one shortest tree per jackknife iteration and using a strict consensus tree of these for the final jackknife consensus tree became obvious. This is detailed in Table 1 for the 567 taxa tree: computing a strict consensus of the n shortest trees found during 10 ratchet iterations (0 <n < 11) provides less support than saving only one out of these shortest trees (p << 0.05). The same effect was previously observed in a 173 taxa dataset [15] when 20 trees were saved (in one RAS cycle per jackknife replicate, applying tree weights rather than using consensus trees): support decreased compared to the outcome of the same analysis saving only one tree (32 clades received at least 4% lower support; the others remained the same. This amounts to a highly significant effect given the number of replicates used by the authors). Consequently, for larger data matrices, reduced search effort (but not as much reduced as using no or less efficient branch swapping) yields rather slightly overestimated support compared to the support found with higher effort. While it has been shown that not applying TBR can severely underestimate support [12,15], using still more thorough search approaches (e.g., several RAS or ratchet cycles, saving several trees) does not significantly raise or even lowers support.
Table 1 Contrasting jackknife support at nodes of the 3-gene jackknife tree for three different search approaches. Contrasting jackknife support (from 500 replicates) at 429 resolved nodes of the 3-gene jackknife-50%-majority-rule-consensus tree (567 terminals) for three different search approaches per jackknife replicate. (1) one heuristic search saving one tree and using simple addition ("simple"); (2) 10 parsimony ratchet iterations, starting from a tree found with simple addition, using the first shortest tree only found within the 10 iterations for consensus tree calculation ("10sv1"); (3) as before, using a strict consensus of all shortest trees found within the 10 ratchet iterations per jackknife replication ("10svAll").
Compared search strategy Sign test, % first < second Sign test, p Wilcoxon test, T Wilcoxon test, p Higher support in...
Simple vs. 10sv1 54.9 0.173 10685.5 0.311 ---
Simple vs. 10svAll 42.3 0.038 7442.5 0.005 Simple
10sv1 vs. 10svAll 6.9 0.000 1107 0.000 10sv1
In all, extending the search extensiveness beyond the rather moderate effort of TBR and saving one tree per replicate does not translate into significantly increased support. Compared to Bremer support, theoretical considerations show that there is a lower risk of significantly overestimating support by using less thorough searches (enhanced type I error of accepting a clade that in fact is not there). The lack of inflated estimates caused by less extensive searches was corroborated using real datasets [12] and simulation studies [13]. These studies, however, compared searches without branch swapping or less effective swapping with searches that include TBR, while not addressing the factor of the number of trees saved and used in the bootstrap/jackknife consensus. The investigation of Freudenstein et al. [15] with 20 saved trees already indicated that support might well drop with more conflicting topologies taken into account per jackknife replicate. In their experiment, however, the option of saving (and swapping on) 20 trees also enhanced the likelihood of finding still shorter trees per replicate. In the present study, the pure effect of ignoring all but one of the topologies with the best score per jackknife replicate becomes evident: support is significantly higher than in a jackknife consensus tree based on all shortest trees found. Thus, for very large trees, there appears to be a small risk of overestimating support. However, this risk probably cannot be judged problematic, in particular not in view of the general conservativeness of nonparametric jackknife and bootstrap estimates in phylogenies [14,16,26,27], which in general should more than counterbalance this slight effect; at least it does so for the dataset analyzed here. As a response to this general conservativeness, α-levels have been raised far above the common 1% – 10% in empirical phylogenetic studies, leading to the acceptance of the presence of nodes with <<90% BS, or at least to referring to such nodes as "highly supported" [e.g., [44]]. One could easily recommend reducing this small risk of overestimating support by representing each bootstrap replicate by a consensus tree derived from multiple searches, but this probably suffers from a too high cost-benefit ratio to be practical in most analyses: 10 ratchet iterations require roughly a 20-fold search time compared to a simple search.
On the other hand, even for large trees, there appears to be no severe risk to underestimate support, as long as one simple-addition tree is swapped using TBR. In contrast, using random addition or random trees as starting trees during each replicate leads to highly significantly underestimated support (tested for the 385 taxa dataset; sign test: p << 0.05).
Note that the above considerations aim at contrasting search strategies and do not extend to fundamental statistical bias existing in bootstrap and jackknife proportions [5,6,45,46] that may deteriorate with increased taxon sampling [14] but improves with increased character sampling. For datasets with many taxa, computational limitations of search strategies become confounded with this bias. Therefore, Sanderson and Wojciechowski [14] could not preclude that part of the decline in BS they observed when increasing sampling size (≤140 taxa) was due to the failure of the simple addition search (saving one tree) to find shortest trees. In view of the results presented here, this effect was probably negligible compared to the statistical bias from random homoplasy distributed among taxa [explanation 3 in 14]. To reduce this bias and achieve more accurate confidence limits, much more computer-intensive, iterated bootstrap methods have to be taken [e.g., [5,45-47]], frequently thought to be too time-consuming to be practical for large amounts of data. Sanderson and Wojciechowski argue that relying on search algorithms with only little branch swapping may circumvent this computational limitation, allowing multiple rounds of bootstrapping by saving time during each individual bootstrap [14: 684]. The outcome, they say, may still be somewhat too conservative because of the failure of these algorithms to find MP trees, but would still be more indicative of true support than conventional BS/JS. In light of the performance of varyingly extensive search strategies on datasets even larger than that in ref. [14], it appears that iterated bootstrap methods are not as impractical as previously thought and should more frequently be considered.
Finally, as shown in Fig. 1, the higher the number of replicates, the lower the error margin for the BS/JS support. To evaluate differences in support for a particular clade, confidence intervals have to be kept in mind to arrive at a statement on the significance of differences. The number of replicates needed to narrow down the confidence intervals to a desired level is a function of the BS/JS. If, for example, one restricts conclusions from a tree topology to nodes >90 and is happy with knowing (at a 5% risk) that a "91" cannot equally likely be a "89" (±1%), more replicates than 3458 are not needed.
Bremer support
Fig. 2 shows how the Bremer support develops at the 25 randomly selected nodes with increasing search extensiveness. Values obtained by a simple search with TBR branch swapping (saving 1 tree) are compared with those obtained with 1 – 50 parsimony ratchet iterations. Vertical lines mark the average number of iterations at which 90% of the final support difference (observed after 50 iterations) are exceeded first. In the smallest dataset (86 taxa) this is the case at the 12th ratchet iteration. For the slightly larger 89 taxa dataset, less iterations are needed until support values become comparably saturated. This dataset, however, displays much less homoplasy than the first and, thus, is easier searchable. Consequently, less effort is needed here to arrive at a similar result. The 385 taxa dataset is comparable to the 89 taxa set in terms of homoplasy but is much larger. Unsurprisingly, the Bremer support settles later here. The same reasoning applies to the largest, 567 taxa dataset, for which the 90%-level is reached only after 17 iterations. In view of the range of different taxon and character sampling covered by these four analyses, saturation appears to happen at quite similar times. Obviously, the 90%-level is an arbitrary measure, and one might want to extend analyses until an average 95% of the final Bremer support differences are reached. This, however, would not change much the relative times at which that is achieved. As a rule of thumb it appears that not much is gained relative to the additional effort when searches are extended beyond 20 iterations, at least not for datasets that fall within the size range observable in current publications. For datasets with a still smaller taxon sampling and still less homoplasy, for which using the ratchet principally does not save time, a single simple addition search (saving a limited number of trees) may already yield Bremer support values close enough to the "true" values (found, e.g, by branch-and-bound searches). As soon as the size and structure of a dataset entails that shortest trees are principally only encountered when at least a random addition search is performed, using the ratchet and the 90%-levels found here as a guideline seems a reasonable strategy.
Figure 2 Differences in Bremer support for 25 randomly selected nodes. Differences in Bremer support for 25 randomly selected nodes (x-axis), comparing values obtained by a simple search (saving 1 tree) with those obtained with subsequent parsimony ratchet iterations (y-axis). Vertical lines mark the average number of iterations at which 90% of the final support difference (observed after 50 iterations) are exceeded first.
Conclusion
One consequence of the above findings is that calculating bootstrap- or jackknife proportions with narrow confidence intervals can be achieved with less expense than often thought, even for very large datasets. In turn, this means that iterated bootstrap methods that aim at reducing statistical bias inherent to bootstrap proportions are more feasible, since the individual bootstrap searches may be performed using less time-intensive heuristic searches during each replicate.
As a further consequence of these reflections, finding bootstrap or jackknife proportions with reasonable confidence can be achieved with much less costs than trying to find the correct Bremer support for large datasets. Bootstrap or jackknife values saturate immediately after one simple addition search, while finding satisfyingly accurate Bremer support may require 20 or more iterations, much depending on what one subjectively accepts as "satisfyingly accurate" (90% of the maximum difference that could be achieved with exhaustive searches ?). If we take the 385 taxa dataset as an example, bootstrap support with 1% accurateness at ≥95% could be gained by 1825 simple searches (with TBR), while Bremer support would require N*(1+13*2) such searches, where N is the number of nodes to test (typically far less than the theoretically possible number of internal nodes). If we approximate that TBR swapping on one tree takes roughly equally long irrespective of the particular resampled matrix or the particular constraints in effect at a given node, calculating Bremer support for the 339 nodes resolved in the strict consensus [48] takes five times longer than bootstrapping or jackknifing (6.5 times if 95% of the final support difference are to be achieved, which on average happens at the 17th iteration for that dataset).
The relative speed of both methods strongly varies with the accurateness aimed at, but even calculating 19592 jackknife iterations needed for 0.5% accurateness at nodes with ≥85% (meaning that the size of the confidence interval matches the precision at which BS/JS proportions usually are reported) is faster than obtaining Bremer support with the precision outlined above, which may only be the case beyond an equivalent of 339*(1+(50*2)) = 34239 such jackknife iterations. The relatively higher performance of jackknifing and bootstrapping further increases with the taxon sampling size. Strictly speaking, a comparison of the speed at which both support types are computable makes not much sense, due to the fundamental differences of both. On the other hand, along with considerations on the interpretability of Bremer support compared to the bootstrap and jackknife [41], such practical considerations may assist in choosing which support type to report when the time available for analyses is limited.
Methods
Four different datasets were chosen that differed not only in the number of taxa sampled but also in the kind of DNA sequence data used (e.g., absolute number of informative characters, overall sequence divergence, overall support levels and levels of homoplasy). Two are comparatively large: a three-gene dataset [44], containing DNA sequences of 18S rDNA, rbcL, and atpB (4592 characters, 2153 parsimony informative) for 567 angiosperm taxa, and a matK dataset [48] for 385 angiosperm taxa (1749 chars, 1075 parsimony informative). Two are of intermediate size: one, using the trnK intron [49], provides high overall support (89 taxa, 3538 characters, 1441 parsimony informative), the other [50] less so (trnL-F and rps4, 86 taxa, 1204 characters, 286 parsimony informative).
Bootstrap and jackknife
The fast search approach (no branch swapping) has been frequently shown to yield considerably lower support values [12-15]. The same applies to less thorough branch swapping algorithms such as NNI (as opposed to TBR; [12,15]. Also, RAS has no beneficial effects on BS/JS as compared to the simple addition sequence – a conclusion based on a 173 taxa set [15]. For still larger datasets, RAS has been frequently shown to be inferior to other strategies such as the parsimony ratchet. In consequence, an extensive testing of the effect of RAS on the datasets analyzed here appeared not warranted. Therefore, the effect of using a limited number of ratchet iterations per jackknife resampling replicate was assessed with help of the author's short C++ program PRAT [24] in conjunction with PAUP* [19].
Since too few replications would hamper contrasting the strategies due to rather wide confidence intervals around the JS values, and since a series of repeated ratchet analyses soon becomes quite time consuming, 500 replicates served as a compromise between variance and computation times. Since the sign test used (see below) ignores the magnitude of differences and since stochastic deviations from the expected BS/JS equally likely are positive or negative, the accurateness achieved with 500 replications should be sufficient in view of the high number of nodes contrasted. Note, however, that knowing the BS or JS of a particular clade with relatively high confidence would require more iterations for values ≥90% (Fig. 1). For the two largest datasets, the ratchet found much shorter trees within the first 10–20 ratchet iterations than simple searches or RAS in PAUP* (in line with common observations; see also Table 1 in [43]). Thus, in terms of the proximity to the putatively minimal tree score, the contrast between TBR on one shortest tree and 20 ratchet iterations is higher than that between 20 ratchet iterations and any number of additional iterations. Therefore, it was first tested whether searches of 10 and 20 ratchet iterations per BS/JS replicate make a difference prior to using more iterations.
To compare jackknife values on two trees, the JS values of all nodes resolved in one tree were compared to the corresponding JS values of the other tree using (a) a sign test and (b) a Wilcoxon test. The first excludes the magnitude of the difference between a pair and only takes pairs with differing values into consideration. The second was used to crosscheck the outcome of the sign test by incorporating information on the magnitude of divergent pairs and retaining information on how many pairs actually consist of equal values. Both tests fully agreed on rejecting or accepting an overall equality of JS (see Table 1).
Bremer support
For each of the four datasets, 25 nodes of the strict consensus tree were randomly chosen (using JAVA's Random class) and subjected to a simple heuristic search in PAUP*, saving only one tree, followed by a parsimony ratchet analysis of 50 iterations (25% characters with double weight, saving one tree). Thus, unlike in the jackknife analyses, not all resolved nodes were monitored due to the considerable search time needed per node. All analyses were performed with PRAP [43] in combination with PAUP*. Monitoring and evaluating the change of support over time was achieved with additional small JAVA classes written for this study.
Acknowledgements
I am indebted to Joseph Felsenstein for very helpful, critical comments on this manuscript. Moreover, John V. Freudenstein, Mark P. Simmons, Thomas Borsch, and anonymous reviewers provided helpful comments that helped to improve the text.
==== Refs
Efron B Gong G A leisurely look at the Bootstrap, the Jackknife, and cross-validation The American Statistician 1983 37 36/48
Efron B The Jackknife, Bootstrap, and other resampling plans CBMS-NSF Regional conference series in applied mathematics 1982 Philadelphia, Society for industrial and applied mathematics
Felsenstein J Confidence limits on phylogenies: An approach using the bootstrap Evolution 1985 39 783 791
Farris JS Albert VA Källersjö M Lipscomb D Kluge AG Parsimony Jackknifing outperforms Neighbor-Joining Cladistics 1996 12 99 124 10.1111/j.1096-0031.1996.tb00196.x
Efron B Halloran E Holmes S Bootstrap confidence levels for phylogenetic trees Proceedings of the National Academy of Sciences of the United States of America 1996 93 13429 13434 8917608 10.1073/pnas.93.23.13429
Holmes S Bootstrapping phylogenetic trees: Theory and methods Statistical Science 2003 18 241 255 10.1214/ss/1063994979
Brown JKM Bootstrap Hypothesis Tests for Evolutionary Trees and Other Dendrograms Proceedings of the National Academy of Sciences of the United States of America 1994 91 12293 12297 7991621
Sanderson MJ Confidence limits on phylogenies: the bootstrap revisited Cladistics 1989 5 113 129
Sanderson MJ Objections to bootstrapping phylogenies: a critique Systematic Biology 1995 44 299 320
Felsenstein J Inferring Phylogenies 2004 Sunderland, Sinauer
Hedges SB The number of replications needed for accurate estimation of the bootstrap p value in phylogenetic studies Molecular Biology and Evolution 1992 9 366 369 1560769
Mort ME Soltis PS Soltis DE Mabry MJ Comparison of three methods for estimating internal support on phylogenetic trees Systematic Biology 2000 49 160 171 12116478 10.1080/10635150050207456
DeBry RW Olmstead RG A simulation study of reduced tree-search effort in bootstrap resampling analysis Systematic Biology 2000 49 171 179 12116479 10.1080/10635150050207465
Sanderson MJ Wojciechowski ME Improved bootstrap confidence limits in large-scale phylogenies, with an example from Neo-Astragalus (Leguminosae) Systematic Biology 2000 49 671 685 12116433 10.1080/106351500750049761
Freudenstein JV van den Berg C Goldman DH Kores PJ Molvray M Chase M An expanded plastid DNA phylogeny of Orchidaceae and analysis of jackknife branch support strategy American Journal of Botany 2004 91 149 157
Felsenstein J Kishino H Is there something wrong with the bootstrap on phylogenies? A reply to Hillis and Bull Systematic Biology 1993 42 193 200
Lipscomb D Farris JS Källersjö M Tehler A Support, ribosomal sequences and the phylogeny of the eukaryotes Cladistics 1998 14 303 338 10.1111/j.1096-0031.1998.tb00341.x
Salamin N Chase MW Hodkinson TR Savolainen V Assessing internal support with large phylogenetic DNA matrices Molecular Phylogenetics and Evolution 2003 27 528 539 12742757 10.1016/S1055-7903(03)00011-3
Swofford DL PAUP*. Phylogenetic Analysis Using Parsimony (*and other Methods) 1998 Sunderland, Sinauer Associates 12064242
Nixon KC The Parsimony Ratchet, a new method for rapid parsimony analysis Cladistics 1999 15 407 414 10.1111/j.1096-0031.1999.tb00277.x
Nixon KC WINCLADA 1996 New York, Program distributed by the author
Goloboff PA NONA: A tree searching program. Program and documentation 1994
Goloboff PA Farris JS Nixon KC TNT: Tree Analysis Using New Technogy 1999
Müller K PRAT: Computer program for phylogenetic analysis of large data sets with PAUP* 2001 Bonn, Program distributed by the author, Nees-Institute, University of Bonn
Felsenstein J The number of evolutionary trees Systematic Zoology 1978 27 401 410
Hillis DM Bull JJ An empirical test of bootstrapping as a method for assessing confidence in phylogenetic analysis Systematic Biology 1993 42 182 192
Zharkikh A Li WH Statistical properties of bootstrap estimation of phylogenetic variability from nucleotide sequences. I. Four taxa with a Molecular Clock Molecular Biology and Evolution 1992 9 1119 1147 1435238
Sanderson MJ Wojciechowski MF Diversification rates in a temperate legume clade: are there 'so many species' of Astragalus (Fabaceae)? American Journal of Botany 1996 83 1488 1502
Davis JI Stevenson DW Petersen G Seberg O Campbell LM Freudenstein JV Goldman DH Hardy CR Michelangi FA Simmons MP Specht CD Vergara-Silva F Gandolfo M A phylogeny of the Monocots, as inferred from rbcL and atpA sequence variation, and a comparison of methods for calculating Jackknife and Bootstrap values Systematic Botany 2004 29 467 510 10.1600/0363644041744365
Soreng RJ Davis JI Phylogtenetics and character evolution in the grass family (Poaceae): simultaneous analysis of morphological and DNA restriction site character sets Botanical Review 1998 64 1 85
Michelangi FA Penneys DS Giza J Soltis DE Hils MH Dan Skean J A preliminary phylogeny of the tribe Miconieae (Melastomataceae) based on nrITS sequence data and its implications on inflorescence position Taxon 2004 53 279 290
Davis JI Simmons MP Stevenson DW Wendel JF Data decisiveness, data quality, and incongruence in phylogenetic analysis: An example from the monocotyledons using mitochondrial atpA sequences. Systematic Biology 1998 47 282 310 12064229 10.1080/106351598260923
Bolmgren K Oxelman B Generic limits in Rhamnus L. s.l. (Rhamnaceae) inferred from nuclear and chloroplast DNA sequence phylogenies Taxon 2004 53 383 390
Bremer K The limits of amino acid sequence data in angiosperm phylogenetic reconstructions Evolution 1988 42 795 803
Bremer K Branch support and tree stability Cladistics 1994 10 295 304 10.1111/j.1096-0031.1994.tb00179.x
Källersjö M Farris JS Kluge AG Bull C Skewness and permutation Cladistics 1992 8 275 287 10.1111/j.1096-0031.1992.tb00071.x
Olmstead RG Bremer B Scott KM Palmer JD A parsimony analysis of the Asteridae s.l. based on rbcL sequences Annals of the Missouri Botanical Garden 1993 80 700 722
Faith DP Cladistic permutation tests for monophyly and nonmonophyly. Syst Zool 1991 40 366 375
Kluge AG Farris JS Quantitative phyletics and the evolution of anurans Systematic Zoology 1969 18 1 32
Davis JI Character removal as a means for assessing stability of clades Cladistics 1993 9 201 210 10.1111/j.1096-0031.1993.tb00218.x
DeBry RW Improving interpretation of the decay index for DNA sequence data Syst Biol 2001 50 742 752 12116944 10.1080/106351501753328866
Oxelman B Backlund M Bremer B Relationships of the Buddlejaceae s.l. investigated using parsimony jackknife and branch support analysis of chloroplast ndhF and rbcL sequence data Systematic Botany 1999 24 164 182
Müller K PRAP - computation of Bremer support for large data sets Molecular Phylogenetics and Evolution 2004 31 780 782 15062810 10.1016/j.ympev.2003.12.006
Soltis DE Soltis PS Chase MW Mort ME Albach DC Zanis M Savolainen V Hahn WH Hoot SB Fay MF Axtell M Swensen SM Prince LM Kress WJ Nixon KC Farris JS Angiosperm phylogeny inferred from 18S rDNA, rbcL, and atpB sequences Botanical Journal of the Linnean Society 2000 133 381 461 10.1006/bojl.2000.0380
Zharkikh A Li WH Estimation of confidence in phylogeny: the complete-and-partial bootstrap technique Molecular Phylogenetics and Evolution 1995 4 44 63 7620635 10.1006/mpev.1995.1005
Rodrigo AG Calibrating the bootstrap test of monophyly. International Journal of Parasitology 1993 23 507 514 8354601 10.1016/0020-7519(93)90040-6
Shimodaira H An approximately unbiased test of phylogenetic tree selection Systematic Biology 2002 51 492 508 12079646 10.1080/10635150290069913
Hilu KW Borsch T Müller K Soltis DE Soltis PS Savolainen V Chase M Powell M Alice LA Evans R Sauquet H Neinhuis C Slotta TA Rohwer JG Campbell CS Chatrou L Angiosperm phylogeny based on matK sequence information American Journal of Botany 2003 90 1758 1776
Müller K Borsch T Legendre L Porembski S Theisen I Barthlott W Evolution of carnivory in Lentibulariaceae and the Lamiales Plant Biology 2004 6 477 490 15248131 10.1055/s-2004-817909
Buck WR Goffinet B Shaw AJ Testing morphological concepts of orders of pleurocarpous mosses (Bryophyta) using phylogenetic reconstructions based on trnL-trnF and rps4 sequences Molecular Phylogenetics and Evolution 2000 16 180 198 10942606 10.1006/mpev.2000.0805
|
16255783
|
PMC1282575
|
CC BY
|
2021-01-04 16:29:17
|
no
|
BMC Evol Biol. 2005 Oct 29; 5:58
|
utf-8
|
BMC Evol Biol
| 2,005 |
10.1186/1471-2148-5-58
|
oa_comm
|
==== Front
BMC Infect DisBMC Infectious Diseases1471-2334BioMed Central London 1471-2334-5-921624889610.1186/1471-2334-5-92Case ReportPneumomediastinum as initial presentation of paralytic rabies: A case report Kietdumrongwong Pongtorn [email protected] Thiravat [email protected] Department of Emergency Medicine, King Chulalongkorn Hospital, Bangkok, Thailand2 Department of Medicine, Chulalongkorn University Hospital, Bangkok, Thailand2005 25 10 2005 5 92 92 7 6 2005 25 10 2005 Copyright © 2005 Kietdumrongwong and Hemachudha; licensee BioMed Central Ltd.2005Kietdumrongwong and Hemachudha; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
Rabies is readily diagnosed when it presents as the classic furious form. Paralytic and atypical forms can pose significant problems in diagnosis. Catastrophic incidents included 7 organ transplant recipients who died of rabies recently in United States and Germany. Although rabies remains top in the lists of differential diagnosis of encephalitis in rabies endemic area, its complication may divert physicians from making a relevant management. We encountered an unusual case of paralytic rabies who presented with spontaneous pneumomediastinum.
Case Presentation
A young male presented with fever and dysphagia. There was a history of fluctuating consciousness and aerophobia but they were absent or could not be demonstrated at the time of admission. He exhibited subcutaneous chest wall emphysema and was found to have pneumomediastinum which resulted in surgical intervention. He developed paralysis followed by seizures during postoperative period. Diagnosis was confirmed by demonstration of rabies RNA in saliva during the preterminal phase and by the autopsy. Over 200 hospital staff subsequently received rabies postexposure prophylaxis.
Conclusion
Spontaneous pneumomediastinum can be a rare complication of rabies. It may lead clinicians to perform inappropriate treatment, particularly when phobic spasms are not present and agitation is not prominent. High level of awareness of rabies in any patient with confusion albeit subtle or with any obscure neurological presentations such as difficulty swallowing with no identifiable causes must be borne in mind.
==== Body
Background
Rabies is an acute viral encephalomyelitis which is virtually 100% fatal. The disease is prevalent in developing countries where it is underreported. Although rabies incidence in humans significantly declined in Thailand from almost 200 to 20 cases annually during the past two decade, more than 400,000 persons required rabies postexposure prophylaxis (PEP) in 2003. This is more than 4 times as many as in 1991 and may explain the lower prevalence in human disease [1]. Furthermore, the percentage of animal brain samples that were confirmed infected with rabies during the 10 year period remained unchanged, within the range of 23–30%. This documents that the main natural vector for rabies in Thailand (the dog) is a remaining threat. Clinical presentations in humans can be categorized as classic (furious and paralytic) and non-classic rabies [2,3]. The latter is almost always associated with bat and some dog rabies variants. Paralytic and non-classic forms are extremely difficult to diagnose. Missing the diagnosis was tragically documented in the United States and Germany where 7 organ transplant recipients died from this disease [4] [ 02/17/2005]. One donor was reportedly bitten by a bat, while the other had a recent travel history to India. Also, several Thai patients who died of rabies had undergone plasma exchanges because of a misdiagnosis of Guillain Barre' syndrome (GBS) [2,5]. Failure to correctly diagnose and manage rabies can be due to unusual clinical manifestations and by distraction of the clinicians by unusual complications [3]. We described a paralytic rabies patient who had dysphagia and pneumomediastinum (ie free air or gas contained within the thoracic cavity from the escape of air into the mediastinal tissues, usually from rupture of interstitial emphysema or pulmonary bleb and also can be found in association with severe asthma, vomiting, excessive coughing or shouting) as initial presentations.
Case presentation
An 18-year old male from Myanmar was brought to the emergency department because of difficulty in swallowing and alteration of consciousness. He had been in good health and had arrived Bangkok 6 days earlier. Three days prior to admission, he developed fever and difficulty in swallowing. He also had pruritus of his right leg and buttock which resulted in extensive excoriation. He had became intermittently confused and refused to eat or drink. There were episodes of agitation and he was reported to have complained of shortness of breath incited by fanning.
One day prior to entry, he was unable to swallow his saliva. His mental status alternated between normal periods and confusion. He gave a history of dog bite 10 years previously but no post exposure vaccination was administered and he denied any recent animal bites or contact with bats.
On admission, he was conscious and well co-operative. He complained of a severely sore throat. He insisted on sitting and continuously spitting out saliva. His body temperature was 39.1°C, blood pressure 130/70 mmHg, heart rate 120 and respiration 30 per minute. There was crepitus on the left side of his neck without signs of inflammation. There were no focal neurological signs. His palatopharyngeal muscle functions were intact but the gag reflex was hyperactive. Aero- and hydrophobic spasms could not be induced. Indirect laryngoscopic examination showed no abnormality. Laboratory studies revealed a leucocytosis of 28,400 with 89.6% of neutrophils. His serum amylase level was slightly elevated (443 U/L, normal range 28–100 U/L). The computerized tomography (CT) scan of the brain was unremarkable. Lumbar puncture revealed an acellular cerebrospinal fluid (CSF) and normal protein and sugar levels. Lateral neck radiographs showed air in retropharyngeal space (Figure 1) and a chest x-ray performed at the same time showed evidence of pneumomediastinum (Figure 2). A liquid barium esophagogram failed to show any point of leakage. Surgical consultation was obtained for exploration. Between admission and operation, his mental state was clear and he showed no signs of aggression and the differential diagnosis of possible rabies was disregarded.
Figure 1 Lateral neck X-Ray showed air in retropharyngeal space.
Figure 2 Chest X-Ray showed pneumomediastinum.
Operative findings were unremarkable. There was no evidence of perforation in upper esophagus and oropharynx. Retropharyngeal area was normal without any accumulation of fluid or purulent discharge.
He required assisted ventilation postoperatively but was able to move limbs voluntarily. He remained rational and cooperative. His hypernatremia (Na 157 mOsm/L) was rapidly corrected to140 mOsm/L within the following 11 hours. On the second postoperative day, he developed brief generalized tonic- clonic seizures. Intravenous phenytoin was initiated. Serial seizures developed on day 6. Repeated lumbar puncture and CT scan were unremarkable. He was conscious but failed to communicate. Paralysis of all limbs with absence of deep tendon reflexes was noted on the same day. He became comatose on day 9. Paralytic rabies was then suspected and rabies virus RNA was demonstrable in saliva but not in the urine by nucleic acid sequence based amplification technique (3). The patient died after 12 days of hospitalization (15 days after first clinical onset). Autopsy findings showed wide spread Negri bodies through out the whole neural axis.
Discussion
This case demonstrates how rabies can escape diagnosis or it is relegated to a secondary position in the differential diagnosis even in a rabies experienced medical center. Most textbooks emphasize the manifestations of furious rabies in humans. Phobic spasms, in the form of aero- and hydrophobia, if present, are pathognomonic of rabies. However, these signs may be absent or not present at all time [2,3]. Difficulty in swallowing usually accompanies phobic spasms. Physical examination only yields a hyperactive gag reflex without any sensory changes at the palate or posterior pharyngeal wall. Other cardinal manifestations, such as fluctuating consciousness and autonomic stimulation signs, also appear periodically or may be absent. They are seen in only half of paralytic rabies patients and are usually not prominent. Once coma develops, phobic spasms are replaced by spontaneously occurred inspiratory spasms which may appear only once or twice during five minutes of observation. They can be difficult to detect in cases of paralytic rabies due to severe weakness of neck and accessory respiratory muscles and the diaphragm.
The patient was admitted to a tertiary care hospital in a canine rabies endemic region. Yet, despite a clear history of fluctuating consciousness and fear of wind or air stream exposure as well as difficulty swallowing without an obvious cause, specialists (such as, internists, otolaryngologists, neurologists and surgeons) ignored such history and were impressed by the dramatic clinical signs of subcutaneous emphysema and pneumomediastinum. Lacking a history of rabies exposure is not uncommon in both furious and paralytic rabies patients. Approximately 10% of rabies patients in rabies endemic areas did not report any exposure, although genetic analysis of rabies virus showed an association with dog rabies variant [6,7]. This may be explained by common minor exposures which are being ignored. This is even more common where bat rabies variants are involved [8,9]. Most deaths occurred because individuals were unaware due to the trivial nature of the wound inflicted by bats and, therefore, they did not seek effective treatment[10]. Some bat rabies variants possess an unique cellular tropism that they can replicate more effectively in the dermis or skin epithelial cells than within muscle cells [11].
Local neuropathic pain, in the form of burning, itching or pruritus, can be found in 30% of dog associated cases and 80% of bat associated cases [12]. This local prodrome was present in this case as pruritus and pain at his right leg and buttock. This was severe to cause extensive excoriation. Dorsal root ganglioneuronitis had been shown to be responsible for such reactions [5].
The findings of air in retropharyngeal space and pneumomediastinum distracted attention from the other history and other symptoms that were suggestive of rabies. We found only one previous report of spontaneous pneumomediastinum in a rabies patient [13], but there was another report of post-mortem finding of ruptured esophagus [14]. The mechanism of spontaneous pneumomediastinum and ruptured esophagus is unknown but may well be due to the violent spasms. Hypernatremia and seizures, well recognized complications are less common in dog associated cases particularly the paralytic type [15,16]. They appeared during the postoperative period in this case. Seizures were suspected to be the consequence of rapid correction of hypernatremia. Percussion myoedema, which can be elicited by tapping on the deltoid region with a tendon hammer and results in swelling of the tissue for a few seconds, was not examined for in this case [17]. Lumbar puncture was performed twice on initial admission and during the time he was comatose. Normal results are typical following rabies infection and may impact on clinical judgment. This was also true in CT scan. [18]. Unlike Japanese encephalitis or its counterparts in a family of Flaviviridae, CT scans are often normal in rabies.
Result of nucleic acid sequence based amplification and the autopsy findings confirmed rabies. Positive result on the saliva but not in the urine was not surprising [19-21]. Rabies virus is excreted intermittently and not simultaneously in all types of secretions (for example, saliva, CSF, urine). Sequencing analysis of the nucleoprotein gene showed that the virus could be categorized in a certain clade of dog rabies variant circulating in Thailand (data not shown).
Over 200 hospital staff received rabies PEP. Later analysis made it doubtful that they all had a potential rabies exposure according to the WHO recommendations [22, 23, 24]. Such anxiety reactions have been observed commonly after hospital deaths of rabies patients. We use an alternative accelerated pre-exposure schedule using an intradermal (ID) regimen for nursing and other staff, which might be exposed. It consists of two ID injections at 2 sites (0.1 ml tissue culture vaccine per site) at both deltoids on days 0, 3 and 7. However, this regimen is not currently recommended by WHO.
In summary, this is a case of classic paralytic rabies. Absence of physical signs and symptoms does not exclude a rabies diagnosis. Greater effort to make an early diagnosis, in this particular case, would have prevented additional suffering for the patient who should have been given palliative care instead of unnecessary surgical intervention. Detailed history and analysis of the whole clinical scenario would avoid such disastrous event and consequence. Pre-exposure vaccination for health workers at risk of exposure should be recommended.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
Dr. Pongtorn Kietdumrongwong took care of the patient and helped coordinating with physicians from various departments and wrote the paper. Dr. Thiravat Hemachudha was consulted during the pre-terminal phase in the intensive care unit and analyzed clinical specimens for rabies RNA detection and wrote the paper.
Pre-publication history
The pre-publication history for this paper can be accessed here:
Acknowledgements
We thank Dr. Henry Wilde for editing manuscript. Drs Thiravat Hemachudha and Henry Wilde are recipients of National Science and Technology Development Agency, Thailand. Dr. Wilde has no conflict of interest.
==== Refs
Hemachudha T Rabies and dog population control in Thailand: success or failure? J Med Assoc Thai 2005 88 120 123 15960231
Hemachudha T Laothamatas J Rupprecht CE Human rabies: a disease of complex neuropathogenetic mechanisms and diagnostic challenges Lancet Neurol 2002 1 101 109 12849514 10.1016/S1474-4422(02)00041-8
Hemachudha T Rupprecht CE Roos K Rabies Principle of Neurological Infectious Diseases 2004 New York , McGraw Hill 151 174
Srinivasan A Burton EC Kuehnert MJ Rupprecht C Sutker WL Ksiazek TG Paddock CD Guarner J Shieh WJ Goldsmith C Hanlon CA Zoretic J Fischbach B Niezgoda M El-Feky WH Orciari L Sanchez EQ Likos A Klintmalm GB Cardo D LeDuc J Chamberland ME Jernigan DB Zaki SR Transmission of rabies virus from an organ donor to four transplant recipients N Engl J Med 2005 352 1103 1111 15784663 10.1056/NEJMoa043018
Hemachudha T Wacharapluesadee S Mitrabhakdi E Wilde H Morimoto K Lewis RA Pathophysiology of human paralytic rabies J Neurovirol 2005 11 93 100 15804967 10.1080/13550280590900409
Hemachudha T Wacharapluesadee S Lumlertdaecha B Orciari LA Rupprecht CE La-Ongpant M Juntrakul S Denduangboripant J Sequence analysis of rabies virus in humans exhibiting encephalitic or paralytic rabies J Infect Dis 2003 188 960 966 14513414 10.1086/378415
Messenger SL Smith JS Rupprecht CE Emerging epidemiology of bat-associated cryptic cases of rabies in humans in the United States Clin Infect Dis 2002 35 738 747 12203172 10.1086/342387
Messenger SL Smith JS Orciari LA Yager PA Rupprecht CE Emerging pattern of rabies deaths and increased viral infectivity Emerg Infect Dis 2003 9 151 154 12603983
Jackson AC Fenton MB Human rabies and bat bites Lancet 2001 357 1714 11428374 10.1016/S0140-6736(00)04852-2
Morimoto K Patel M Corisdeo S Hooper DC Fu ZF Rupprecht CE Koprowski H Dietzschold B Characterization of a unique variant of bat rabies virus responsible for newly emerging human cases in North America Proc Natl Acad Sci U S A 1996 93 5653 5658 8643632 10.1073/pnas.93.11.5653
Omezzine A Hmouda H Jemni L Spontaneous pneumomediastinum: an exceptional complication of rabies Clin Infect Dis 1994 18 663 664 8038338
Oon CT Boerhaave's syndrome (ruptured oesophagus) in a case of rabies Singapore Med J 2000 41 83 85 11063210
Hemachudha T Tirawatnpong S Phanthumchinda K Seizures as the initial manifestation of paralytic rabies J Neurol Neurosurg Psychiatry 1989 52 808 810 2501458
Jackson AC Jackson ACWWH Human disease Rabies 2002 Amsterdam , Academic Press 219 244
Hemachudha T Phanthumchinda K Phanuphak P Manutsathit S Myoedema as a clinical sign in paralytic rabies Lancet 1987 1 1210 2883526 10.1016/S0140-6736(87)92186-6
Laothamatas J Hemachudha T Mitrabhakdi E Wannakrairot P Tulayadaechanont S MR imaging in human rabies AJNR Am J Neuroradiol 2003 24 1102 1109 12812933
Wacharapluesadee S Hemachudha T Urine samples for rabies RNA detection in the diagnosis of rabies in humans Clin Infect Dis 2002 34 874 875 11850870 10.1086/338872
Wacharapluesadee S Hemachudha T Nucleic-acid sequence based amplification in the rapid diagnosis of rabies Lancet 2001 358 892 893 11567709 10.1016/S0140-6736(01)06041-X
Hemachudha T Wacharapluesadee S Antemortem diagnosis of human rabies Clin Infect Dis 2004 39 1085 1086 15472871 10.1086/423813
Wilde H Khawplod P Khamoltham T Hemachudha T Tepsumethanon V Lumlerdacha B Mitmoonpitak C Sitprija V Rabies control in South and Southeast Asia Vaccine 2005 23 2284 2289 15755612 10.1016/j.vaccine.2005.01.030
Rupprecht CE Hanlon CA Hemachudha T Rabies re-examined Lancet Infect Dis 2002 2 327 343 12144896 10.1016/S1473-3099(02)00287-6
|
16248896
|
PMC1282576
|
CC BY
|
2021-01-04 16:28:14
|
no
|
BMC Infect Dis. 2005 Oct 25; 5:92
|
utf-8
|
BMC Infect Dis
| 2,005 |
10.1186/1471-2334-5-92
|
oa_comm
|
==== Front
BMC Public HealthBMC Public Health1471-2458BioMed Central London 1471-2458-5-1071621909510.1186/1471-2458-5-107Research ArticlePrevalence and associations of symptoms of upper extremities, repetitive strain injuries (RSI) and 'RSI-like condition'. A cross sectional study of bank workers in Northeast Brazil Lacerda Eliana M [email protected]ácul Luis C [email protected] S Augusto Lia G [email protected] Maria Teresa A [email protected] Dyhanne C [email protected] Danielle C [email protected] London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UK.2 Centro de Pesquisa Aggeu Magalhães – CPqAM/Fiocruz, Av. Moraes Rego s/n, Recife 50.670-420, Brazil.3 Programa de Pós-Graduação em Saúde Coletiva, Universidade do Vale do Rio dos Sinos, Av Unisinos 950, São Leopoldo 93022-000, Brazil.4 Faculdade de Ciências Médicas de Pernambuco, Universidade de Pernambuco, Rua Arnóbio Marques 310, Recife 50100-130, Brazil.2005 11 10 2005 5 107 107 12 4 2005 11 10 2005 Copyright © 2005 Lacerda et al; licensee BioMed Central Ltd.2005Lacerda et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
The repetitive strain injury syndrome (RSI) is a worldwide occupational health problem affecting all types of economic activities. We investigated the prevalence and some risk factors for RSI and related conditions, namely 'symptoms of upper limbs' and 'RSI-like condition'.
Methods
We conducted a cross-sectional study with 395 bank workers in Recife, Northeast Brazil. Symptoms of upper limbs and 'RSI-like condition' were assessed by a simple questionnaire, which was used to screen probable cases of RSI. The diagnosis of RSI was confirmed by clinical examination. The associations of potential risk factors and the outcomes were assessed by multiple logistic regression analysis.
Results
We found prevalence rates of 56% for symptoms of the upper limbs and 30% for 'RSI-like condition'. The estimated prevalence of clinically confirmed cases of RSI was 22%. Female sex and occupation (as cashier or clerk) increased the risk of all conditions, but the associations were stronger for cases of RSI than for less specific diagnoses of 'RSI-like condition' and symptoms of upper limbs. Age was inversely related to the risk of symptoms of upper limbs but not to 'RSI-like' or RSI.
Conclusion
The variation in the magnitude of risk according to the outcome assessed suggests that previous studies using different definitions may not be immediately comparable. We propose the use of a simple instrument to screen cases of RSI in population based studies, which still needs to be validated in other populations. The high prevalence of RSI and related conditions in this population suggests the need for urgent interventions to tackle the problem, which could be directed to individuals at higher risk and to changes in the work organization and environment of the general population.
==== Body
Background
Repetitive strain injury (RSI) is an occupational disease that considerably impacts workers' lives and has significant socio-economic repercussions. The burden of RSI to people affected and the society are undeniably large. One third of workers' compensation costs in the US private industry are due to RSI[1], and the direct costs with compensation exceeds US$ 20 billion in the Washington State alone[2]. In the Netherlands, 8% of the whole working population take time off work because of RSI symptoms [3]. The Health & Safety Executive, a British institution responsible for the regulation of occupational risks to health, estimated self-reported work related musculoskeletal disorders to affect 448,000 people in 2003/04, corresponding to 1% of the population who has ever worked in Great Britain [4]. In the years 2001/02, it was estimated that 4.1 million full working days were lost due to work related musculoskeletal disorders in the United Kingdom. The estimated costs for employers associated with the condition was between £208 million and £221 million per year (1995/96 prices) [4].
Clinical, epidemiological and social aspects of RSI remain largely controversial in the medical literature [5-9]. Nevertheless, RSI has been widely shown to affect a considerable proportion of the adult population and workers in all levels of economic activities. Certain occupational groups have an increased risk of developing RSI. Among these are bank workers, particularly cashiers, whose activities with repetitive movements increase their risk of upper limb symptoms and RSI [10-12]. However, only a few studies have investigated the prevalence of upper limb symptoms in this high risk group of workers [10,12-14]. We have not located any study in the literature comparing risk factors for clinically diagnosed RSI with the less specific diagnosis of upper limb symptoms.
We studied bank workers of a governmental bank institution in Recife, Northeast Brazil. The aims of the study were: to develop and validate a screening questionnaire for diagnosing RSI; to estimate the prevalence of symptoms of the upper limbs, including those that are potential cases of RSI (which we refer to as having a 'RSI-like condition'), and of clinically confirmed RSI; and to compare some risk factors for RSI with those of 'RSI-like condition' and symptoms of upper limbs.
The comparative description and quantification of upper limb symptoms and RSI among this group of workers at particular risk for the condition adds to the understanding of their epidemiology. It also provides important information for the planning of bank work activities and the design of prevention and control measures for RSI. Of note is that the study tested a simple screening questionnaire for diagnosing cases of RSI, which could be used in similar populations and, subject to further validation, in other populations.
Methods
We conducted a cross-sectional study between March and August 2000. The reference population consisted of all 579 workers employed by the bank and working at one of the 23 branches located in Recife, all of whom were invited to participate in the study. The type of jobs included managers, cashiers, and administrative clerks. For all these job types, we identified the following ergonomic risk factors: intense use of microcomputers (particularly by the cashiers); continuous work without regular periods of rest; extended working hours; poor posture (elevation of shoulders and elbows, forced rotation of low back, typing while gripping the phone between the head and shoulder); high levels of cognitive demand and constant tension and psychological demands related to expected levels of achievement according to set targets [15].
Managers were responsible for planning and defining targets for local branches; monitoring achievements; decision making and attending special clients. These tasks were carried out using personal computers and telephone, in daily 8 hour shifts. The managers worked seated most of the time.
Cashiers' tasks included dealing with deposits and withdrawals, receiving a wide range of payments and selling branch products for clients. These tasks were performed with the worker standing up for most of their 8 hour shifts, and involved intensive use of personal computers for typing alpha numeric data, and the stamping of many documents (using heavy wooden stamps).
The tasks performed by the clerks varied according to the branches' sectors to which they were allocated. They included liaising with personal and business clients in person and by telephone, a range of administrative activities such as preparing and monitoring contracts for loans and concessions, and checking and typing information onto microcomputers. Their job involved nearly continuous use of personal computers and telephones often simultaneously, although their daily routine varied according to the clients' demands. Compared to the cashiers, they used computers less extensively.
Personnel in other jobs, who were not directly contracted by the bank, were excluded from the study.
Study procedures
Development and validation of the screening questionnaire
A short questionnaire was developed and tested with workers in 2 of the 23 bank branches (n = 41; sub-sample 1). They were self-completed and included four questions on upper limb symptoms aiming to screen individuals with symptoms suggestive of RSI (here referred to as having a 'RSI-like condition'), in addition to addressing personal characteristics of the respondents. A consent form summarizing the study objectives and procedures was appended to the questionnaires and all participants were given a chance to clarify any doubts with the research team. Those who wished to give consent signed the form and returned them with the completed questionnaires.
The specific questions on symptoms of upper limbs covered the clinical definition of RSI (Table 1). A worker with a positive answer to the first question (on report of upper limb symptoms described as sensation of weight, discomfort, weakness or pain) was considered as 'symptomatic'. A worker with a positive answer to all four questions was considered as having a 'RSI-like condition', while symptomatic individuals who did not have a 'RSI-like condition' were considered to have 'symptoms of upper limbs unlikely to be related to RSI'.
Table 1 Criteria for diagnosis of RSI included in the screening questionnaire
1. Report of presence of any of the following symptoms in one or both upper limbs: 'sensation of weight', discomfort, weakness or pain in fingers, arms, forearms, elbows or neck
2. Presence of symptom(s) for over one month
3. Presence of symptoms on a daily or nearly daily basis (continuous/recurrent)
4. Relation of symptom(s) with work activities, irrespective of the occurrence of symptoms outside work
The validity of the screening was ascertained by comparing the results of the questionnaire with the clinical diagnosis – used here as the gold standard for the diagnosis of RSI. For that purpose, a full clinical examination was performed in all subjects working in two of the bank branches. Sensitivity, specificity and positive and negative predictive values were derived in the usual way[16].
Diagnosis of RSI
The clinical diagnosis was based on the definition of RSI published on the Brazilian Technical Norms for the diagnosis of RSI [17]. Cases of RSI were considered as those meeting the following criteria, based on occupational history and clinical examination:
- Report of regular (daily or nearly daily) symptoms of pain and/or 'paresthesia' in one or both upper limbs for at least one month,
- Relation of work activities with the appearance, intensification or progression of symptoms,
- Absence of other clinical condition that could justify the symptoms described (e.g. diabetes, Hansen's disease, AIDS),
- Presence, in the comparative examination of the upper limbs, of clinical signs related to the symptoms described – including abnormalities on strength or sensitivity, or neurological signs related to specific syndromes (e.g. Tinel's or Phalen's signs in the carpal tunnel syndrome; or Finkelstein's sign in the deQuervain tenosynovitis).
In addition, cases in initial phase were diagnosed based on a typical clinical history with symptoms for over one month, even in the absence of specific clinical signs. These cases correspond to the zero phase of the RSI evolution according the Japanese Association of Occupational Health classification [18].
The diagnosis was established by the Principal Investigator, an occupational physician with large experience in the diagnosis and management of RSI. Medical history taking and clinical examination were conducted 'in situ', i.e. in properly adapted rooms at the bank branches. The clinical examination was made according to standard procedures, including inspection, palpation, passive and active mobilization of the upper limb segments, and focused neurological examination. The results were formally recorded by the research team.
Outcomes and exposures
The 'screening questionnaire' was completed by workers in all branches, allowing the ascertainment of the following primary outcomes for the study of risk factors: i) self referred symptoms of upper limbs; ii) self referred symptoms of the upper limbs unlikely to be related to RSI (i.e. symptoms of upper limbs and negative screening for 'RSI-like condition'); iii) 'RSI-like condition'; and iv) clinically confirmed RSI. The latter were identified in a further sub-sample of individuals from all participating branches (n = 64; sub-sample 2) as those with 'RSI-like condition' and a confirmed clinical diagnosis of RSI (as above). We obtained data on potential risk factors, including age, sex, marital status, educational level, occupation and time working at the bank.
Data handling
All questionnaires were manually checked shortly after completion, and the respondents queried if necessary. Data from the questionnaires and the results of clinical protocols were double entered onto an EPI-info 6.0 database, checked for consistency, and compared with the original records if necessary.
Data analysis
Data were analysed in EPIINFO 6.0 and STATA version 7. Point prevalence for symptoms of upper limbs, 'RSI-like condition', and confirmed cases of RSI were calculated in standard ways [19].
Study of risk factors
The association of exposure variables and symptoms of upper limbs, 'RSI-like condition' and confirmed cases of RSI were tested by the chi-squared test or chi-squared test for trend [19] for categorical variables, or the Wilcoxon rank test for continuous variables (bivariate analysis); and by multiple logistic regression [20]. The multivariate models included age group, sex and all other independent variables associated with the outcome in the bivariate analysis with a significance level of p < 0.10. The likelihood ratio test was used to compare models with and without specific independent variables [20], with those which did not contribute to the model (p >= 0.05) not included in the final model. The odds ratio was calculated as a measure of strength of association. We preferred to use this rather than the prevalence rate ratio, as the former is adequate for both the bivariate and the logistic regression analyses. We also specifically tested for the possible interaction between independent variables. A significance level of 0.05 or lower was generally assumed to correspond to a significant result in the bivariate analysis and final multivariate model.
Sample size
We studied 395 bank workers. This sample size was calculated as adequate to estimate the prevalence of RSI with a precision of 2.5%, assuming a prevalence of 16% in the study population. This figure, which is slightly higher than the 10% prevalence estimated for Brazilian bank workers by Ribeiro [21], was obtained from occupational registries and from the author's experience with bank workers in Recife. The sample size chosen was also adequate to detect associations between risk factors and outcomes with the following odds ratios, with 90% power and a 5% level of significance: 2 and over for symptoms of upper limbs and 'RSI-like condition', and 3 and over for confirmed cases of RSI (calculations made in Stactalc/Epi-info).
Ethical aspects
The project was approved by the Centro de Pesquisas Aggeu Magalhães Ethics Committee, by the Bank Board of Management, and by the Union of Bank Workers of Pernambuco, the latter representing the workers. All participants signed an informed consent and received feedback on their clinical examinations and were referred to medical care or occupational services as necessary. Participants' anonymity was preserved at all times and their identification was only used for the specific purpose of this research. The ethical principles set by the Declaration of Helsinki were followed.
Results
The validity of the RSI screening was determined based on the evaluation of all 41 workers from 2 bank branches (sub-sample 1), who completed the questionnaires and were fully assessed clinically (Table 2). Based on a positive clinical diagnosis as the gold-standard for the diagnosis of RSI, we found the screening to have a sensitivity of 90% (9/10) and a specificity of 87% (27/31). The positive and negative predictive values were of 69% and 96% respectively. The observed prevalence of RSI was 24% (10 affected cases out of a total of 41 workers)
Table 2 Validity of screening questionnaire for 'RSI-like' cases, showing the clinical diagnosis as the gold standard
Screening Questionnaire Clinical diagnosis Total
Case of RSI Non-case of RSI
Case of RSI-like (+) 9 4 13
Non-case of RSI-like (-) 1 27 28
Total 10 31 41
Following the validation stage, we invited all bank workers in current activity (n = 579) to participate in the study of prevalence and risk factors. The response rate was 68.9% (n = 399). The analysis refers to 395 subjects (99% of respondents) with complete or near complete information.
The sample consisted of 205 (51.9%) women and 190 (48.1%) men. They were on average 40.4 years old (95% confidence interval (95% CI) = 39.9 – 40.9), and had worked at the bank for a median of 13.1 years. Two hundred and fifty (63.3%) were married or in a stable relationship, 72 (18.2%) were separated, 71 (18%) were single, and 2 (0.5%) widowed. The majority (244 or 82.7%) had a university degree, with 22 of them also having a post-graduate degree. Of those with university degree, 24.3% graduated in business, 16.7% in economy, and 9.9% in law. Most of them worked as either administrative clerks (141 or 36.3%) or cashiers (126 or 32.5%); 76 (19.6%) were managers, and 45 (11.6%) had other jobs.
Prevalence estimates
We found prevalence rates of 27.1% (n = 102) for symptoms of upper extremities unrelated to RSI, and 29.9% (n = 118) for 'RSI-like condition', giving a total of 56.2% (n = 222) for symptoms of upper extremities. Two 'symptomatic' individuals were excluded due to insufficient information to classify them as unrelated to RSI or 'RSI-like condition'. Considering the cases of 'RSI-like condition' (118/395) and the projection of the validity of the screening to the study population, we estimate a prevalence of RSI in this population of 22% (Table 3).
Table 3 Estimated prevalence of RSI in the population based on the validity of the screening instrument
Screening Questionnaire Clinical diagnosis Total Estimated prevalence
+ -
+ 78 40 118
- 9 268 277
Total 87 308 395 87/395 = 0.22
Description of RSI cases
From the clinical exams in the sub-sample 2 (n = 64), we diagnosed 55 workers as true cases of RSI, and 9 as false positive cases. Their main clinical features are shown in Table 4.
Table 4 Frequency of features found in clinically diagnosed RSI cases
Features N %
Specific diagnostics:
• De Quervain's tenosynovitis 5 9
• Carpal tunnel syndrome 8 14
• Tenosynovitis of extensors (hands) 1 2
• Tenosynovitis of flexors (hands) 1 2
• Epicondylitis (medial) 8 14
• Epicondylitis (lateral) 7 13
• Biceps tendonitis 3 5
• Supraspinatus tendonitis 9 16
Unspecific signs:
• Tenderness under palpation 11 20
• Pain under active mobilizing 7 13
• Hypertonus 1 2
• Hyperreflexia 2 4
Diagnosis based on occupational history and symptoms in the absence of clinical signs 15 27
From the 9 false positive cases, 5 reported the symptoms to be no longer present when clinically examined; 2 others reported no symptoms after having changed their jobs within the branch; 1 reported that symptoms disappeared after treatment of a dental chronic infection, and 1 after receiving treatment for tennis elbow and giving up the sport. Latter, we identified two further cases. These cases were originally screened as "negatives" for "RSI-like condition" because they had feared disclosing their condition when answering the 'screening questionnaire'.
Risk factors for RSI
Table 5 shows the results of the bivariate analyses. Workers with each of the outcomes were compared with workers with no symptoms of upper limbs (n = 173). The variables sex and occupation were significantly associated with all outcomes. Increasing age appeared to confer protection. The variable age-group was significantly associated with symptoms of upper limbs, marginally associated with 'RSI-like condition', but not associated with RSI. None of the other variables tested were significantly associated with any of the outcomes.
Table 5 Results of bivariate analyses for RSI and related conditions (4 outcomes), compared to asymptomatic workers (n = 173)
Symptoms other than RSI Symptoms 'RSI-like' RSI
Variables N = 102/275 N (%) OR 95% CI n = 222/395 N (%) OR 95% CI n = 118/291 N (%) OR 95% CI n = 55/228 N (%) OR 95% CI
Sex:
Male 40 (39.2) 1* 87 (45.8) 1* 46 (30.9) 1* 18 (14.9) 1*
Female 62 (60.8) 2.28 (1.38 – 3.76) 135 (65.8) 2.28 (1.52 – 3.43) 72 (50.7) 2.30 (1.43 – 3.72) 37 (3.0) 3.02 (1.60 – 5.74)
Job:
Manager 19 (18.1) 1 32 (42.1) 1† 13 (22.8) 1† 6 (12.0) 1†
Clerk 31 (30.7) 1.20 (0.60 – 2.39) 81 (57.4) 1.86 (1.04 – 3.29) 50 (45.4) 2.85 (1.40 – 6.28) 28 (31.8) 3.69 (1.35 – 10.09)
Cashier 41 (40.6) 2.16 (1.09 – 4.28) 82 (65.1) 2.56 (1.40 – 4.67) 39 (47.0) 2.43 (1.20 – 5.65) 18 (29.0) 2.60 (0.92 – 7.38)
Other 10 (9.1) 1.10 (0.44 – 2.78) 24 (53.3) 1.57 (0.74 – 3.32) 14 (40.0) 1.69 (0.88 – 5.76) 2 (8.7) 0.70 (0.13 – 3.76)
Education:
Up to secondary school 33 (32.4) 1 82 (56.2) 1 48 (42.8) 1 22 (25.5) 1
University degree 69 (67.8) 1.23 (0.73 – 2.06) 140 (56.2) 1.00 (0.66 – 1.51) 70 (39.1) 1.52 (0.84 – 2.78) 33 (23.2) 1.24 (0.59 – 2.62)
Marital status:
Single 24 (23.5) 1 41 (57.8) 1 16 (34.8) 1 6 (16.7) 1
Married 58 (56.9) 0.65 (0.35 – 1.22) 139 (55.6) 0.92 (0.54 – 1.56) 80 (41.9) 1.35 (0.69 – 2.64) 42 (27.4) 1.89 (0.73 – 4.87)
Separated/Divorced 20 (19.6) 0.83 (0.38 – 1.82) 42 (56.8) 1.02 (0.53 – 1.99) 22 (40.7) 1.38 (0.61 – 3.12) 7 (18.0) 1.17 (0.35 – 3.88)
Mean Diff (SE) OR 95%CI Mean Diff (SE) OR 95%CI Mean diff (SE) OR 95%CI Mean diff (SE) OR 95%CI
Age (OR by 5 year age-groups) 1.59 (0.67) 0.75 (0.58 – 0.96)† 1.46 (0.53) 0.77 (0.63 – 0.95)∞ 1.34 (0.62) 0.80 (0.62 – 1.02) 0.96 (0.82) 0.82 (0.60 – 1.12)
Time at bank in years
(OR by year worked in bank) 1.29 (0.65) 0.95 (0.91 – 1.00)∞ 1.04 (0.51) 0.96 (0.92 – 1.0) 0.89 (0.61) 0.97 (0.92 – 1.01) 0.30 (0.81) 0.99 (0.93 – 1.05)
* p < 0.001, † p < 0.02, ∞ p <= 0.05
'Symptoms other than RSI" refers to individuals considered to have 'symptoms of upper limbs unlikely to be related to RSI' (n = 102).
The variable 'symptoms' refers to all subjects with symptoms of upper extremities (n = 222).
To be considered as having 'RSI-like condition', positive responses to all 4 screening questions were required (n = 118).
RSI includes confirmed cases of RSI following full clinical assessment (n = 55).
For the variable symptoms, the total sample size of 375 is used as the denominator.
For the other symptom-related variables, the denominator includes 173 cases with no symptoms of upper limbs added to the respective numerator.
In the logistic regression analysis, sex and occupation remained significantly associated with all outcomes (Table 6). Female sex was directly associated with all outcomes, with odds ratios ranging from 2.2 to 3.1 and a stronger association for confirmed cases of RSI. The highest occupational risk for symptoms of upper limbs was for cashiers, while those for 'RSI-like condition' and RSI were for clerks. While the strength of association for cashiers increased slightly though consistently from the less to the more specific outcomes, the risk of clerks increased more dramatically, from 1.4 for symptoms of upper limbs other than due to RSI, up to 3.7 in confirmed cases of RSI. Age-group, as a continuous variable, was a significant predictor of symptoms other than due to RSI (p = 0.05), marginally significant (p = 0.07) for the outcome 'symptoms of upper limbs' – with a trend for protection with increasing age, but became non-significant for the two other outcomes (Table 6).
Table 6 Results of logistic regression analyses for RSI and related conditions (4 outcomes); final model
VARIABLES Symptoms other than RSI OR 95% CI Symptoms OR 95% CI 'RSI-like' OR 95% CI RSI OR 95% CI
Sex:
Male 1* 1* 1*** 1***
Female 2.31 (1.37 – 3.89) 2.24 (1.47 – 3.53) 2.27 (1.36 – 3.79) 3.14 (1.58 – 6.28)
Job:
Manager 1** 1** 1* 1***
Clerk 1.45 (0.70 – 3.0) 2.02 (1.11 – 3.66) 2.85 (1.40 – 6.28) 3.69 (1.35 – 10.09)
Cashier 2.12 (1.04 – 4.31) 2.40 (1.31 – 4.41) 2.43 (1.20 – 5.65) 2.61 (0.92 – 7.38)
Other 0.97 (0.37 – 2.53) 1.48 (0.69 – 3.20) 1.69 (0.88 – 5.76) 0.65 (0.12 – 3.61)
Age in 5 year periods 0.76 (0.59 – 1.0) 0.82 (0.66 – 1.01) 0.86 (0.66 – 1.12) 0.89 (0.62 – 1.26)
* p < 0.001, ** p < 0.02, *** p < 0.01
'Symptoms other than RSI" refers to individuals considered to have 'symptoms of upper limbs unlikely to be related to RSI' (n = 102).
The variable 'symptoms' refers to all subjects with symptoms of upper extremities (n = 222).
To be considered as having 'RSI-like condition', positive responses to all 4 screening questions were required (n = 118).
RSI includes confirmed cases of RSI following full clinical assessment (n = 55).
Discussion
This is the first study investigating the prevalence of RSI in a population of bank workers in Northeast Brazil. The diagnosis of 'RSI-like condition' was made by a simple screening questionnaire based on the definition of RSI, which is easy to apply and has high sensitivity and specificity. Pending further validation in other settings, this could become a widely used instrument in the study of RSI. Diagnostic confirmation and further categorization of cases should be done using clinically available guidelines [22]. We believe the clinical diagnosis of RSI in the study to be very reliable, as it was made by an experienced clinician based on standard procedures. This included a detailed history and physical examination, with exclusion of other conditions that could justify the symptoms in the upper limbs. Nevertheless, the diagnosis is still subject to misclassification, particularly in early or very mild cases of RSI that lack clinical signs. However, in the absence of a pathognomonic diagnostic test, we believe this to be the best means of diagnosing true cases of RSI and of other upper limb conditions. Other screening tests have been proposed to musculoskeletal symptoms related to occupation [14,22-24]. We used a shorter instrument with only 4 questions focusing on upper limb symptoms, which had good sensitivity (90%) and specificity (87%). The false negatives occurred mainly as a result of fear on the part of subjects to disclose their condition. A similar situation was reported by the British RSI association, who claimed "the majority of people experiencing pain, discomfort and loss of function due to musculoskeletal problems in the workplace make no reference to their condition for fear of losing their job"[25]. False positives were mainly due to other temporary conditions of the upper limbs. Such cases can easily be distinguished from true cases of RSI by full clinical assessment, which can be repeated if necessary after a short interval.
The frequency of specific RSI-related diagnoses was crudely similar across anatomic regions (hands and wrists, elbows and shoulders/neck). Carpal tunnel syndrome was unsurprisingly the most frequent specific diagnosis related to RSI, as previously reported in literature [26-29]. This arises due to the specific ergonomic conditions at work sites, e.g. hyperflexion of wrists while typing and other poor postures at work, which may be more important among cashiers than in managers. However, we found a similarly high frequency of epicondylitis. The full understanding of these findings would require further research, including the use of ergonomic techniques.
The literature has been inconsistent on the outcomes studied; some report on work related symptoms of upper limbs [30,31], others refer to more specific diagnoses such as work related carpal tunnel syndrome [28,29,32-36], or to an occupational syndromic condition such as RSI [3,37-41]. It is not clear whether these outcomes are truly comparable, and it is possible that they have different determinants. In this study, we compared some risk factors for a) upper limb symptoms; b) 'RSI-like condition', defined by suggestive symptoms, and c) clinically diagnosed RSI. Although female sex and work as a cashier or clerk were associated with the 3 outcomes, the strength of these associations increased as more specific diagnoses of 'RSI-like condition' and RSI were used. This suggests that while female sex is a risk factor for symptoms of upper limbs, as reported consistently in the literature [14,42-52], much of this risk is accounted for in true cases of RSI, with other cases of symptoms of upper limbs having a weaker association with female sex.
Workers who develop symptoms of RSI and especially those with clinically confirmed diagnosis tend to change their roles within the organisation. Very often this means changing from cashiers to clerks, with the objective of becoming less exposed to repeated movements. Reverse causality may therefore partially explain the higher risk of clerks for RSI than for other upper limb conditions.
Age was significantly associated with symptoms of upper limbs, but not with 'RSI-like condition' or RSI. This suggests increasing age to be protective of conditions other than RSI that present with upper limb symptoms, but caution is warranted when interpreting these findings. Young age may be linked to other problems affecting upper limbs in relatively healthy individuals, such as trauma and sports injuries. Change to lower risk jobs as workers age, 'healthy worker' effects and uncontrolled confounding may also explain these findings. It is also possible that young workers belong to a cohort of individuals that share common risk factors to symptoms of upper limbs not explored here. If this represents a cohort effect, then this would be consistent with an increasing incidence of upper limb conditions. While this is speculative, taken the cross-sectional nature of this study, this confirms a clinical impression of there being an increasing problem in the study area and indeed in other settings. The weakening of the association with age, as we move from symptoms of upper limbs to RSI, could also be interpreted as being due to less specific symptoms representing early stages of RSI in younger individuals, which develop in some of them into full-blown RSI as they age. This would be supported by the age distribution of the sample, with non-RSI related symptoms affecting the younger individuals (mean age of 39.6 years) and confirmed RSI affecting those on average 41.0 years old, with RSI-like condition having intermediate mean ages of 39.9 years. Nevertheless, in contrast to other studies [2,10,12,14,25,29,35,42,53-58], true cases of RSI were not found to be significantly associated with age. The use of a less specific diagnosis in other studies and variation in populations may explain these differences.
We estimated the prevalence of RSI as 22% and that of symptoms of upper limbs as 56%, for this particularly high risk group. Population based prevalence figures for RSI have been typically lower. In Canada, 10% of the population over 20 years old reported RSI serious enough to limit usual activities at some point in the previous 12 months [45]. In the Netherlands, the population prevalence in individuals over 25 years old has been estimated as around 2% for RSI, 11% for epicondylitis, and 16% for tendonitis or capsulitis [44]. Higher prevalence rates have been reported among specific occupational groups, e.g. with prevalence over 60% of pain [57] or musculoskeletal complains of the upper limbs [31] reported in dentists. In bank workers, some researchers reported prevalence of upper limb symptoms varying according to the affected anatomic region e.g. from 6.6% for symptoms in arms to 31.4% in neck [12], and from 16% in elbow to 50% in shoulder [14]. In Southeast Brazil, the prevalence of RSI among bank workers has been reported as 10% with another 10% presenting upper limb symptoms, but not a diagnosis of RSI [21]. We found relatively high prevalence rates of upper limb symptoms and RSI, indicating perhaps a particularly high risk population in our study. The prevalence found was also higher than that predicted based on the author's previous experience as an occupational physician. A possible reason is that many individuals affected by RSI continue to carry out their work activities in spite of the symptoms and do not see the occupational doctor. However, it should be noted that the prevalence of RSI is particularly high among workers at this bank and a more comprehensive study of risk factors in this population could give further insights on why this is the case. The prevalence of RSI was based on the assumed validity of the screening questionnaire for cases of 'RSI-like condition'. As these estimates are subject to random error, this may have lead to some imprecision in the prevalence estimates.
We were not able to include a considerable proportion of the workers in the study. Workers with RSI who were on sick leave, and who probably represent the most severe cases, were excluded from the study due to logistic and ethical reasons. If a relatively large number of RSI cases are likely to be on sickness related absences, this may have led to an underestimation of the prevalence of RSI, and possibly also an underestimation of the association of independent variables with the outcomes. Similarly, if workers with RSI were less likely to participate so as to avoid disclosing their condition (with fear that this might affect their jobs), this would have similar effects.
Only a small selection of variables was used in the analysis. This excluded, for example, variables related to the work organization and environment, which may confound the associations investigated in the study. A comprehensive report on determinants of RSI and related conditions was not, however, the aim of this paper.
The study was conducted in a specific bank in Northeast Brazil. The results are representative of this specific population, but probably also of bank workers in general, particularly those who are subject to similar working conditions.
Conclusion
In conclusion, our study showed a high prevalence of RSI and related conditions in this population of bank workers, raising serious concerns about the magnitude of potentially disabling conditions in this occupational group, and calling for urgent measures to improve work environments and how they are organised. It also confirms gender and certain specific occupational roles in the risk of RSI and related conditions, with stronger associations found among confirmed cases of RSI. The results also suggest age to be more directly linked to symptoms of upper limbs that are not related to RSI. The variable magnitude of the associations suggests that risk factors differ slightly according to the definition/outcome used, and raises a question on the comparability of previous studies using different diagnostic criteria. Further prospective studies with the inclusion of a larger number of potential risk factors would help clarify the role of these and other variables in the aetiology of RSI. We propose the use of a simple screening questionnaire to identify potential cases of RSI, namely cases of 'RSI-like condition'. Pending further validation, the use of such a questionnaire, complemented by full clinical evaluation is a sensible way to identify cases of RSI for population based epidemiological studies in a consistent way.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
EML participated in the conception and study design, carried out the clinical examinations, participated in the analysis and interpretation of data, and has been involved in the drafting the article. LCN participated in the conception and design of the study, in the analysis and interpretation of data, and has been involved in the drafting the article. LASG participated in the conception and study design, and critically revised the article. MTO participated in the interpretation of data and critically revised the article. DCFR was involved in the acquisition of data and critically revised the article. DCAW has been involved in the acquisition of data and revised critically the article. All authors approved the manuscript.
Pre-publication history
The pre-publication history for this paper can be accessed here:
Acknowledgements
We are most grateful for the support received from the Union of Bank Workers in Recife, from the Administrative Section of the study institution and for the bank workers who participated in the study. We thank Anne Tholen for revising the manuscript.
==== Refs
Barr AE Barbe MF Pathophysiological tissue changes associated with repetitive movement: a review of the evidence Phys Ther 2002 82 173 187 11856068
Silverstein B Viikari-Juntura E Kalat J Use of a prevention index to identify industries at high risk for work-related musculoskeletal disorders of the neck, back, and upper extremity in Washington state, 1990-1998 Am J Ind Med 2002 41 149 169 11920960 10.1002/ajim.10054
Bongers PM de Vet HC Blatter BM Repetitive strain injury (RSI): occurrence, etiology, therapy and prevention Ned Tijdschr Geneeskd 2002 146 1971 1976 12420421
Health &Safety Executive
Silversides A Confusion surrounding repetitive strain injury highlighted at conference Cmaj 1997 156 1459 1460 9164411
Santos Filho SB Barreto SM Methodological issues in epidemiological studies of repetitive strain injuries Cad Saude Publica 1998 14 555 563 9761609
Szabo RM King KJ Repetitive stress injury: diagnosis or self-fulfilling prophecy? J Bone Joint Surg Am 2000 82 1314 1322 11005523
Nathan PA Meadows KD Neuromusculoskeletal conditions of the upper extremity: are they due to repetitive occupational trauma? Occup Med 2000 15 677 93, iii 11013051
Verhagen AP Bierma-Zeinstra SMA Feleus A Karels C Dahaghin S Burdorf L de Vet HCW Koes BW Ergonomic and physiotherapeutic interventions for treating upper extremity work related disorders in adults The Cochrane Database of Systematic Reviews 2003 Issue: 3, Art. No.: CD003471. DOI: 10.1002/14651858.CD003471.pub2.
Ferreira Junior M Conceiçäo GM Saldiva PH Work organization is significantly associated with upper extremities musculoskeletal disorders among employees engaged in interactive computer-telephone tasks of an international bank subsidiary in Sao Paulo, Brazil Am J Ind Med 1997 31 468 473 9093663 10.1002/(SICI)1097-0274(199704)31:4<468::AID-AJIM14>3.0.CO;2-Y
Murofuse NT Marziale MH Changes in the work and life of bank employees with repetitive strain injury: RSI Rev Lat Am Enfermagem 2001 9 19 25 12040769
Yu IT Wong TW Musculoskeletal problems among VDU workers in a Hong Kong bank Occup Med (Lond) 1996 46 275 280 8854704
Takala EP Viikari-Juntura E Moneta GB Saarenmaa K Kaivanto K Seasonal variation in neck and shoulder symptoms Scand J Work Environ Health 1992 18 257 261 1411369
Pinheiro FA Troccoli BT Carvalho CV Validity of the Nordic Musculoskeletal Questionnaire as morbidity measurement tool Rev Saude Publica 2002 36 307 312 12131969
Lacerda EM A relação entre lesões por esforços repetitivos e o trabalho bancário - estudo em uma instituição na cidade do recife Departamento de Medicina Social/Núcleo de Saúde Pública 2001 Recife, Universidade Federal de Pernambuco
Greenland S Kenneth J. Rothman SG Basic Methods for Sensitivity Analysis and External Adjustment Modern Epidemiology 1998 2nd Philadelphia, Lippincott Williams & Wilkins 353 355
Instituto Nacional de Seguridade Social Norma Técnica sobre Lesões por Esforços Repetitivos ou Distúrbios Osteomusculares Relacionados ao Trabalho DOU 1998 Seção I
Assunção AA Rocha LE Buschinelli JT, Rocha LE and Rigotto RM Agora até namorar fica difícil: uma história de lesões por esforços repetitivos Vida, doença e trabalho no Brasil 1993 Rio de Janeiro, Atheneu 175 181
Kirkwood BR Sterne JAC Medical statistics 2003 2nd. Oxford, Blackweel Science
Katz MH Multivariable analysis - a pratical guide for clinicians 2000 2nd Cambridge, Cambridge University Press
Ribeiro HP A violência oculta do trabalho. As Lesões por Esforços Repetitivos 2000 Rio de Janeiro, Editora FIOCRUZ
Sluiter JK Rest KM Frings-Dresen MH Criteria document for evaluating the work-relatedness of upper-extremity musculoskeletal disorders Scand J Work Environ Health 2001 27 Suppl 1 1 102 11401243
Kuorinka I Jonsson B Kilbom A Vinterberg H Biering-Sorensen F Andersson G Jorgensen K Standardised Nordic questionnaires for the analysis of musculoskeletal symptoms Appl Ergon 1987 18 233 237 15676628 10.1016/0003-6870(87)90010-X
Viikari-Juntura E Risk factors for upper limb disorders. Implications for prevention and treatment Clin Orthop 1998 39 43 9646745
RSI Awareness
Reis RJ Pinheiro TM Navarro A Martin MM Profile of occupational disease outpatients and the presence of repetitive strain injury Rev Saude Publica 2000 34 292 298 10920453
Leclerc A Landre MF Chastang JF Niedhammer I Roquelaure Y Upper-limb disorders in repetitive work Scand J Work Environ Health 2001 27 268 278 11560341
Kutluhan S Akhan G Demirci S Duru S Koyuncuoglu HR Ozturk M Cirak B Carpal tunnel syndrome in carpet workers Int Arch Occup Environ Health 2001 74 454 457 11563610 10.1007/s004200100246
Kao SY Carpal tunnel syndrome as an occupational disease J Am Board Fam Pract 2003 16 533 542 14963080
Armstrong TJ Buckle P Fine LJ Hagberg M Jonsson B Kilbom A Kuorinka IA Silverstein BA Sjogaard G Viikari-Juntura ER A conceptual model for work-related neck and upper-limb musculoskeletal disorders Scand J Work Environ Health 1993 19 73 84 8316782
Alexopoulos EC Stathi IC Charizani F Prevalence of musculoskeletal disorders in dentists BMC Musculoskelet Disord 2004 5 16 15189564 10.1186/1471-2474-5-16
Atcheson SG Ward JR Lowe W Concurrent medical disease in work-related carpal tunnel syndrome Arch Intern Med 1998 158 1506 1512 9679791 10.1001/archinte.158.14.1506
Barbieri PG Colombini D Rocco A Custureri F Paderno G Outbreak of carpal tunnel syndrome of the upper limbs in automobile seat assemblers: results of exposure evaluation and clinical investigation Med Lav 1996 87 646 655 9148122
Dias JJ Burke FD Wildin CJ Heras-Palou C Bradley MJ Carpal tunnel syndrome and work J Hand Surg [Br] 2004 29 329 333 15234495 10.1016/j.jhsb.2004.03.002
Diaz JH Carpal tunnel syndrome in female nurse anesthetists versus operating room nurses: prevalence, laterality, and impact of handedness Anesth Analg 2001 93 975 980 11574368 10.1097/00000539-200110000-00035
Szabo RM Carpal tunnel syndrome as a repetitive motion disorder Clin Orthop 1998 78 89 9646750
Barbieri PG Colombini D Occhipinti E Vigasio A Poli R [Epidemics of musculotendinous pathologies of the upper limbs (cumulative trauma disorders) in a group of assembly line workers] Med Lav 1993 84 487 500 8177134
Farnsworth EM Diagnosis and management of repetitive strain injury Adv Nurse Pract 2001 9 32 8; quiz 38-9 12420463
Sato L RSI: object and pretext for constructing the work and health field Cad Saude Publica 2001 17 147 152 11241937
Sorgatz H Repetitive strain injuries. Forearm pain caused by tissue responses to repetitive strain Orthopade 2002 31 1006 1014 12376875 10.1007/s00132-002-0354-5
Willems JH Repetitive strain injury (RSI); a report from the Health Council of the Netherlands Ned Tijdschr Geneeskd 2002 146 1969 1970 12420420
Feuerstein M Miller VL Burrell LM Berger R Occupational upper extremity disorders in the federal workforce. Prevalence, health care expenditures, and patterns of work disability J Occup Environ Med 1998 40 546 555 9636935 10.1097/00043764-199806000-00007
Nordander C Ohlsson K Balogh I Rylander L Palsson B Skerfving S Fish processing work: the impact of two sex dependent exposure profiles on musculoskeletal health Occup Environ Med 1999 56 256 264 10450243
Picavet HS Hazes JM Prevalence of self reported musculoskeletal diseases is high Ann Rheum Dis 2003 62 644 650 12810427 10.1136/ard.62.7.644
Tjepkema M Repetitive strain injury Health Rep 2003 14 11 30 14608793
Leclerc A Chastang JF Niedhammer I Landre MF Roquelaure Y Incidence of shoulder pain in repetitive work Occup Environ Med 2004 61 39 44 14691271
Kaergaard A Andersen JH Musculoskeletal disorders of the neck and shoulders in female sewing machine operators: prevalence, incidence, and prognosis Occup Environ Med 2000 57 528 534 10896959 10.1136/oem.57.8.528
Fransson-Hall C Bystrom S Kilbom A Self-reported physical exposure and musculoskeletal symptoms of the forearm-hand among automobile assembly-line workers J Occup Environ Med 1995 37 1136 1144 8528723
Ohlsson K Hansson GA Balogh I Stromberg U Palsson B Nordander C Rylander L Skerfving S Disorders of the neck and upper limbs in women in the fish processing industry Occup Environ Med 1994 51 826 832 7849867
Mathiassen SE Winkel J Sahlin K Melin E Biochemical indicators of hazardous shoulder-neck loads in light industry J Occup Med 1993 35 404 407 8487119
Treaster DE Burr D Gender differences in prevalence of upper extremity musculoskeletal disorders Ergonomics 2004 47 495 526 15204301 10.1080/00140130310001638171
Ferraz MB Frumkin H Helfenstein M Gianeschini C Atra E Inoue O Seiji K Kudo S Jin C Cai SX SJ L Watanabe T Nakatsuka H Ikeda M Upper-extremity Musculoskeletal Disorders in Keyboard Operators in Brazil: A Cross-sectional Study Int J Occup Environ Health 1995 1 239 244 9990162
Banerjee P Gangopadhyay S A study on the prevalence of upper extremity repetitive strain injuries among the handloom weavers of West Bengal J Hum Ergol (Tokyo) 2003 32 17 22 15176126
Hansson GA Balogh I Ohlsson K Palsson B Rylander L Skerfving S Impact of physical exposure on neck and upper limb disorders in female workers Appl Ergon 2000 31 301 310 10855453 10.1016/S0003-6870(99)00047-2
Pascarelli EF Hsu YP Understanding work-related upper extremity disorders: clinical findings in 485 computer users, musicians, and others J Occup Rehabil 2001 11 1 21 11706773 10.1023/A:1016647923501
Nieradko B Borzecki A Computer use and computer related repetitive strain injuries among students of the Medical University of Lublin Ann Univ Mariae Curie Sklodowska [Med] 2003 58 354 357 15315014
Santos Filho SB Barreto SM Occupational activity and prevalence of osteomuscular pain among dentists in Belo Horizonte, Minas Gerais State, Brazil: a contribution to the debate on work-related musculoskeletal disorders Cad Saude Publica 2001 17 181 193 11241941
Arvidsson I Akesson I Hansson GA Wrist movements among females in a repetitive, non-forceful work Appl Ergon 2003 34 309 316 12880741 10.1016/S0003-6870(03)00042-5
|
16219095
|
PMC1282577
|
CC BY
|
2021-01-04 16:28:55
|
no
|
BMC Public Health. 2005 Oct 11; 5:107
|
utf-8
|
BMC Public Health
| 2,005 |
10.1186/1471-2458-5-107
|
oa_comm
|
==== Front
BMC PsychiatryBMC Psychiatry1471-244XBioMed Central London 1471-244X-5-381624625710.1186/1471-244X-5-38Research ArticleExpression of POU-domain transcription factor, Oct-6, in schizophrenia, bipolar disorder and major depression Ubhi Kirenjeet [email protected] Jack [email protected] Institute of Psychiatry, Kings College London, Denmark Hill, London, UK2005 24 10 2005 5 38 38 15 7 2005 24 10 2005 Copyright © 2005 Ubhi and Price; licensee BioMed Central Ltd.2005Ubhi and Price; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
The POU-domain transcription factor Oct-6 has been reported to be differentially expressed between schizophrenic and control post-mortem brains. In this study, we attempted to replicate this finding and to discover whether Oct-6 was also dysregulated in bipolar disorder and major depression.
Methods
Oct-6 mRNA and protein expression were determined by in-situ hybridization and immunohistochemistry respectively in sections of post-mortem brain.
Results
We did not observe any differences in Oct-6 expression between any of the groups under study. Oct-6 mRNA and protein was identically expressed in the hippocampal and cortical regions of most specimens in all groups, including controls.
Conclusion
Oct-6 is, therefore, unlikely to be a specific marker for any psychological disorder; rather its expression in controls suggests that it is normally expressed in most adult brains.
==== Body
Background
The neurodevelopmental hypothesis of schizophrenia proposes that events occurring during foetal development, which adversely effect the development of the brain, may underlie the occurrence of schizophrenia in later life [1]. One approach to try to investigate this hypothesis has been to examine genes involved in the normal development of the brain and determine whether they are altered in schizophrenia.
The POU-domain homeobox transcription factors are one such family of genes involved in development. We have focused on Oct-6 (SCIP/Tst-1/POU3f1), a member of the POU-III subfamily. Much of the initial work on Oct-6 looked at its role in the peripheral nervous system, in particular its role in Schwann cell development, where it is required for the timely initiation of myelination [2]. Oct-6 has also been shown to be expressed in postmitotic neurons of the developing rodent telencephalon as they migrate from the ventricular to the intermediate zone, and is particularly associated with those destined for cortical layers II/III and V. Oct-6 is expressed in these cortical laminae and in the CA1 region of the adult hippocampus [3,4], although some rodent data has suggested that all Oct-6 expression is eventually lost with age [5].
A previous open study has reported the expression of OCT-6 protein in the frontal and temporal lobes of patients diagnosed with schizophrenia but its absence from matched controls, suggesting it may be a putative biological marker for schizophrenia [6]. Affective disorders such as bipolar disorder and major depression share many of the symptoms seen in schizophrenia (DSM-IV) [7], and it has been suggested that a common pathology may underlie schizophrenia and bipolar disorder [8-10]. The potential importance of POU factors in affective disorders was highlighted by a study by Stopkova et al (2004) who reported that a polymorphism within the PIK3C3 promoter, linked to a subset of schizophrenic and bipolar patients. Stopkova et al (2004) suggest that the -432C→T polymorphism, occurring as it does within an octamer binding site (the DNA motif recognised by POU domain transcription factors), may affect POU protein binding to this region [11]. Expression assays in neural stem cell lines have shown that Oct-6-induced expression of the reporter is indeed decreased by the presence of this polymorphism (personal communication, Dafe Uwanogho).
We aimed to replicate the initial finding of Oct-6 expression in the brains of patients diagnosed with schizophrenia and to extend this finding to bipolar disorder and major depression in order to determine whether Oct-6 expression was specific to schizophrenia or a more general marker of psychosis.
Methods
Tissue acquisition
A total of 60 subjects from the Stanley Foundation Neuropathology Consortium were used in these studies. This set consists of 15 samples from patients with schizophrenia, 15 with major depressive disorder, 15 with bipolar disorder, and 15 matched controls patients. A detailed description of this collection has been published [12], and a summary of subject characteristics is shown in Table 1.
Table 1 Summaries of demographic, clinical and histological information of schizophrenic, bipolar, major depression and control cases.
Demographics Group
Controls
n = 15 Schizophrenic
n = 15 Bipolar disorder
n = 15 Major Depression
n = 15
AGE (years, mean ± SD) 48.1 ± 10.7 44.5 ± 13.1 42.3 ± 11.7 46.5 ± 9.3
GENDER 9M, 6F 9M, 6F 9M, 6F 9M, 6F
POSTMORTEM INTERVAL (hours, mean ± SD) 23.7 ± 9.9 33.7 ± 14.6 32.5 ± 16.1 27.5 ± 10.7
CAUSE OF DEATH
CPD 13 7 11 8
Accident 2 0 0 2
Pneumonia 0 0 1 1
Suicide 0 7 1 1
Other 0 1 2 3
pH (mean ± SD) 6.3 ± 0.2 6.2 ± 0.3 6.2 ± 0.2 6.1 ± 0.3
FIXATION (months, mean ± SD) 4.40 ± 3.9 11.20 ± 8.5 9.67 ± 3.6 8.33 ± 6.6
STORAGE OF FROZEN SECTIONS (days, mean ± S D) 621 ± 172.3 434 ± 290.0 338 ± 234.2 621 ± 233.1
BRAIN HEMISPHERE 7R:8L 6R:9L 8R:7L 6R:9L
DURATION OF ILLNESS (years, mean ± SD) 0 ± 0 21.3 ± 11.4 20.1 ± 9.7 12.7 ± 11.1
LIFETIME ANTIPSYCHOTICa DOSE (mg) (min-median-max) 0-0-0 0- 35 000- 200 000 0- 7 500- 60 000 0-0-0
a = lifetime quantity of fluphenazine or equivalent
CPD = cardiac pulmonary disease
Cryostat and paraffin sectioned slides were provided by the Stanley Foundation. They were stored at -80°C and room temperature, respectively, until use. The sections were coded, the investigator had no knowledge of the disease state of the tissue and the experiments were conducted 'blind'.
Antibodies
Anti-Oct-6 antibody 1 : rabbit polyclonal against the N-terminal region of mouse Oct-6 protein (98% homology in human), hereafter referred to as anti-Oct-6 (N). Antibody characterisation and specificity has been previously described [4].
Anti-Oct-6 antibody 2 : rabbit polyclonal against full-length Oct-6 purified from baculovirus infected Sf9 cells hereafter referred to as Anti-Oct-6(FL). Antibody characterisation and specificity has previously been described [13].
Paraffin sections
Slides were dewaxed using xylene, rehydrated and endogenous peroxidase activity was blocked by washing in 0.3% H2O2. Following PBS (phosphate buffered saline) washes, slides were incubated in normal swine serum (1:10 in PBS; Dakocytomation) for 30 mins then incubated at 4°C overnight in Anti-Oct-6(N) (1:500). Slides were then incubated in anti-rabbit secondary (1:500 in PBS; Dakocytomation) for 2 hours. Slides were washed then incubated in ABComplex (Dakocytomation) for a further 2 hours in the dark. Colour was developed using DAB (3, 3'-diaminobenzidine, Vector Laboratories). The reaction took 5–10 minutes and was considered complete when a brown colour was observed on the slides. When suitable colour had developed, the DAB reaction was stopped by washing slides in dH2O, slides were air dried and mounted using Faramount aqueous mounting medium (Dakocytomation).
Frozen sections
Slides were post fixed for 10 minutes in acetone (-20°C) for 10 minutes, air dried and washed in PBS (4°C) for 5 minutes and 0.3% H2O2 for 10 minutes. Slides were blocked for 30 minutes in a 1:10 (in PBS) solution of Swine serum (Dakocytomation) and incubated overnight at room temperature in anti-Oct-6(FL) (1:3000). Slides were washed then incubated in anti- rabbit secondary antibody (1:500, Dakocytomation) for 2 hours. Signal amplification and visualization using ABC and DAB was as described above.
In Situ Hybridization
In situ hybridization was conducted using a 1.6 kb DIG labelled probe complementary to the 3' untranslated region of the Oct-6 gene.
Forward primer: GTGGTGGTGGTGGTGGTGTGTGACGGG
Reverse primer: ACAGCCCTGGGGTACATGTTTATGTGAGTAATAAAAT
Amplicon: 1646 bp
Slides were placed in 4% fomaldehyde at room temperature for 10 mins then washed in DEPC-PBS then in 0.1 M triethanolamine, pH 8.0/acetic anhydride, 400: 1 (vol: vol), on a stir plate for 10 min. Slides were then incubated in pre-hybridization solution (50% formamide, 5 × SSC, 5 × Denharts, 250 μg/ml bakers yeast RNA and 500 μg/ml Herring sperm DNA) for 2–4 hrs at room temperature in a humidified chamber. Slides were incubated in hybridization solution (pre-hybridization solution with probe added at a typical concentration of 1:500), covered with a coverslip and incubated at 60°C for approximately 18 hours. Sense-strand probes were used as specificity controls for hybridization and revealed no binding.
Coverslips were removed by washing in 5 × SSC at 65°C, then the slides were washed in 2 × SSC at 65°C for 30 mins, 0.2 × SSC at 65°C for 30 mins, then in Buffer 1 (0.1 M Tris pH7.5, 0.15 M NaCl) for 5 mins then incubated in block (10% heat inactivated sheep serum in Buffer 1) at room temperature for 1 hour then finally incubated in alkaline phosphatase-conjugated sheep anti-DIG antibody (Roche) at room temperature for 4 hours. Following washed in Buffer 1 and Buffer 3 (0.1 M Tris pH9.5, 0.1 M NaCl, 50 mM MgCl2, 2 mM levamisole), hybridization was visualized by enzyme catalysed colour reaction using nitro blue tetrazolium salt (NBT) and 5-bromo-4-chloro-3-indolyl phosphate (BCIP) (Roche). Reaction was stopped in PBS and then water, slides were air-dried and mounted using Faramount aqueous mounting medium (Dakocytomation).
Image analysis
All sections were analysed with a Nikon light microscope (Eclipse 600) with Lucia 4.0 image analysis software.
Results
OCT-6 immunoreactivity in paraffin sections
We attempted to detect OCT-6 immunoreactivity in hippocampus and cortex in post-mortem human brains. We performed immunohistochemistry on a series of 22 paraffin sections from the hippocampal formation taken from a total of 20 brains from all four groups under study (schizophrenia (n= 9), major depression (n = 3), bipolar disorder (n = 3), and control (n = 5)), using the antibody (Anti-Oct-6 (N)) and staining procedure as in the previous study [6]. In each case, we found OCT-6 expression in all fields (CA4, CA3 and CA1) of the hippocampal formation (Figure 1). OCT-6 immunoreactivity was restricted to large cells with the morphology of pyramidal neurons and appeared to be localised to the perinuclear region of the cytoplasm. There was no distinguishable difference in intensity of staining between different hippocampal fields.
Figure 1 Oct-6 Immunoreactivity in Paraffin-embedded Tissue from Control, Schizophrenic, Bipolar Disorder and Major Depression. Oct-6 immunoreactivity was observed in hippocampal subfields CA4, CA3 and CA1 in controls (CON A – C), schizophrenics (SCH D – F), bipolar disorder (BP G – I) and major depression (MD J – L). Oct-6 immunoreactivity was observed in groups in a cytoplasmic/perinuclear fashion (arrowheads). Oct-6-positive cells had the morphological appearance of pyramidal neurons. Immunoreactivity is visible in the apical dendrites of some neurons (arrows in (H) and (K)) Scale bars = 50 μm
Immunoreactivity was also noted in the cerebral cortex where it assumed a layer-specific pattern, appearing restricted to pyramidal neurons of layers II/III and V (Figure 2). Cells in layer II/III were more intensely stained than those in layer V. OCT-6 expression in all cortical neurons generally appeared to be dispersed throughout the cell, although a few neurons in layer II/III were noted to have the cytoplasmic/perinuclear localization seen in hippocampal cells (Figure 2A (arrowhead)). This laminar distribution in the neocortex is consistent with the pattern described in adult rodents [3].
Figure 2 Oct-6 Immunoreactivity in the Cerebral Cortex. Oct-6 immunoreactivity in the cerebral cortex of a schizophrenic patient. Staining was observed in a layer-specific pattern restricted to layers II/III and V (B) seen at higher magnification in (A) & (C). Oct-6-positive cells had the morphological characteristics of pyramidal neurones. In general, POU3F1 immunoreactivity was dispersed throughout the cell (asterisks in A), though some did have the cytoplasmic/perinuclear localization noted in the hippocampus (arrowheads in A). Immunoreactivity was also observed in the apical dendrites (arrows). Cells in layer V (C) appeared considerably smaller than those in layer II/III (A). A similar cortical expression was observed in all groups. Scale bars: (B) = 200 μm (A & C) = 50 μm
Thus, in contrast to the previous study [6], we observed OCT-6 immunoreactivity in hippocampal field CA1 and were unable to detect any difference in OCT-6 expression between control and schizophrenic specimens, as all the control specimens we examined (n = 5) displayed Oct-6 immunoreactivity.
OCT-6 immunoreactivity in frozen sections
The immunoreactivity in the paraffin sections was predominantly cytoplasmic. Though this has been reported in earlier studies [6, 14], it is nonetheless noteworthy since Oct-6 being a transcription factor would be expected to be nuclear. It was deemed possible that the fixation may be affecting antibody penetration. In order to investigate this further, immunohistochemistry was conducted on fresh frozen sections. We were unable to observe any immunoreactivity in unfixed specimens using the anti-Oct-6 (N) antibody. We were, however, able to observe robust, reproducible staining using the anti-Oct-6(FL) antibody. Oct-6 immunoreactivity was observed in a subset of cells in hippocampal fields CA4, CA3, CA1 in all groups (Figure 4). As with the paraffin sections, staining appeared restricted to pyramidal neurons. Not all specimens had Oct-6 immunoreactive cells, but in each group the majority of specimens were positive (6 of 7 control specimens, 8 of 8 schizophrenic specimens, 6 of 7 bipolar specimens and 7 of 8 major depression specimens). There was no significant difference in the proportion of positive specimens between groups.
In all positive cases staining was evident throughout the hippocampal formation from CA4 to CA1, but different fields showed different degrees of immunoreactivity with the most intense staining generally appearing in CA1.
Differences in staining localization between the different hippocampal fields were apparent. In some specimens, Oct-6 immunoreactivity in CA4 and CA3 was present throughout the neuron cell body (Figure 3A (asterisks), whilst in others it was predominantly nuclear (Figure 3D, E, J and 3K (arrows). Immunoreactivity was also noted in the apical dendrites of some cells (Figure 3B (arrowheads). In all cases the intense staining seen in the CA1 region was predominantly nuclear (Figure 3C, F, I, L and 3O). These differences in localization did not appear to be particularly associated with any of the four groups under study.
In situ hybridization (ISH)
In order to determine areas of Oct-6 gene expression, ISH was performed using a probe complementary to the 3'UTR of the Oct-6 gene. Oct-6 mRNA expression was observed in the cortex and all hippocampal fields in specimens from all four groups (Figure 4). As with immunoreactivity, the majority of specimens in each group were positive (9 of 14 control specimens, 9 of 15 schizophrenic specimens, 9 of 14 bipolar specimens and 9 of 14 major depression specimens). Thus, the proportion of positive specimens was almost identical in each group.
Although levels of immunoreactivity varied between hippocampal fields, the ISH signal appeared to be at similar levels throughout, being equally intense in the CA4 as the CA1 cells. There were some positive cells in cortical regions, but intensity here was much reduced. In the cortex the expression of Oct-6 mRNA was confined to layers II/III and V, similar to the Oct-6 immunoreactivity described above.
Statistical analysis of confounding variables
Kruskal-Wallis non-parametric tests revealed significant differences between groups with regards to the length of time tissue had spent either in formalin or in frozen storage (df = 3 p = 0.017 and df = 3, p = 0.005 respectively) in each case the control samples had spent shorter amounts of time in each.
Discussion
In this study, we sought to reproduce the previously reported finding that OCT-6 was expressed by pyramidal neurons in the hippocampus of schizophrenic brains, but not in controls [6]. We attempted to extend this finding to include bipolar disorder and major depression. Using immunoreactivity and ISH we observed Oct-6 expression in all three disease states, but unlike the previous study we also observed expression in the control group. So while we can confirm that Oct-6 is indeed expressed by pyramidal neurons in the schizophrenic brain (and in bipolar disorder and major depression), there is no observable difference from controls.
Each of the three approaches use in this study gave subtly different results in terms of the localisation of expression and the proportions of specimens that express Oct-6. These differences most likely reflect differences in the techniques themselves, such as fixation. Note also that the paraffin and frozen sections were from different sides of the brain randomly chosen to limit laterality. Nonetheless, no differences between the disease groups and controls emerge with any technique.
The paraffin immunohistochemistry indicated that Oct-6 was expressed in all specimens, regardless of group. The fresh frozen immunohistochemistry and ISH gave positive results in approximately 90% of cases. Therefore, either there are a small number of Oct-6 negative brains, or the fresh frozen sections gave occasional false negatives. A second difference that emerges is that while ISH detected a signal of similar intensity in all hippocampal fields, the immunohistochemistry appeared to give a stronger signal in CA1, though clear staining was observed in all hippocampal fields. It is possible that this difference in expression could be due OCT-6 stability or translational efficiency in the CA1. This is currently under investigation and we have observed evidence of translational control of Oct-6 in studies in vitro (unpublished observations).
The final difference to emerge from our studies was the difference in localisation of immunoreactivity between paraffin and frozen sections. In paraffin sections, the immunoreactivity was exclusively cytoplasmic, whilst in frozen sections it was predominantly nuclear. The cytoplasmic expression was somewhat unexpected for a transcription factor, but not unprecedented. Some transcription factors are regulated by sequestration in the cytoplasm, NFêB for example [15], and a cytoplasmic localisation for Oct-6 has previously been reported in Schwann cells. It was also possible; however, that the absence of nuclear staining was an artefact of fixation, and the fact that we observe nuclear staining in the fresh frozen sections supported this notion.
It is important to note that, despite the fact that no definite statements regarding the cellular localization of Oct-6 immunoreactivity can be made, Oct-6 immunoreactivity was observed in similar regions all groups including controls, regardless of the fixation method used. The expression of Oct-6 mRNA coincided with protein expression and again was observed in all groups including controls. Taken as a whole, the data from this study strongly argue against Oct-6 being a marker of schizophrenia or any other psychiatric condition.
Our overall observations are consistent with Oct-6 expression seen in the rodent [3,4], with the one difference that in the adult rodent Oct-6 expression becomes restricted to the CA1 field of the hippocampus and layer V of the cerebral cortex with age, whereas we see expression in all fields of the human hippocampal formation. It is possible that the restriction may have yet to occur in the brains we investigated, but in rodents the adult pattern is observed by postnatal day 30 [3] while our specimens had an average age of 45, so this explanation is unlikely.
It is unclear why we have been unable to reproduce the earlier result. The few negative specimens seen with fresh frozen sections suggest that Oct-6-ve specimens may occur and thus it is conceivable that all the controls in the previous study were genuinely Oct-6-ve. Unfortunately, we do not have access to these earlier specimens in order to verify this point. Another possibility is that the specimens used by Ilia et al (2002) [6] were different in terms of some other significant uncontrolled variable. We have compared the two groups of specimens and have been unable to identify such a variable. A comparison of the ages of control specimens, for example, indicated no difference between the two (df = 23, p= 0.0913).
Conclusion
The expression of Oct-6, both mRNA and protein, in the majority of samples including controls would seem to suggest that the expression we observe is a reflection of normal Oct-6 expression in the adult. We conclude therefore that Oct-6, rather than being a specific biological marker of schizophrenia, or any other psychotic state, is likely to be normally expressed in the adult brain.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
All experimental work and analysis was carried out by KU. JP contributed to the preparation of the final manuscript.
Figure 3 Oct-6 Immunoreactivity in Fresh-frozen Tissue from Control, Schizophrenic, Bipolar Disorder and Major Depression. Oct-6 immunoreactivity was observed in all groups (CON A – C and CON (2) D- F), schizophrenics (SCH G – I), bipolar disorder (BP J – L) and major depression (MD M – O) and did not differ between groups or in comparison to controls. Oct-6-positive cells had the morphological appearance of pyramidal neurons. In the CA4 and CA3 regions immunoreactivity generally took on a nuclear localization (arrowheads) thought in some cases a more dispersed pattern of expression was noted (asterisks). Oct-6 immunoreactivity was also noted the apical dendrites of some cells (arrows). Though staining was robust throughout, it appeared stronger in the CA1 region (C, F, I, L & O), in all cases Oct-6 immunoreactivity in the CA1 was nuclear. Scale bars = 50 μm
Figure 4 Oct-6 mRNA expression in Control, Schizophrenic, Bipolar Disorder and Major Depression Patients. Oct-6 mRNA expression (arrowheads) was observed in hippocampal fields CA4 (A, E, H & K), CA3 (B, F, I & L) and CA1 (C, G, J and M) in all groups including controls (controls = CON, schizophrenics = SCH bipolar disorder = BP and major depression = MD). No qualitative differences were noted in hybridization signal between different regions or groups. Oct-6 mRNA was observed in a layer-specific pattern in the cortical layers II/III and V (D). Scale bars: (D) = 500 μm, all others = 50 μm
Pre-publication history
The pre-publication history for this paper can be accessed here:
Acknowledgements
This work was funded by the Stanley Foundation and the Medial Research Council, UK. The authors would like to thank Deis Mejier and Micheal Wegner for the gifts of the antibodies, Dafe Uwanogho for advice on probe preparation and Brenda Williams for help in the preparation of the manuscript. This work was funded by the Stanley Foundation, USA.
==== Refs
Jaegle M Meijer D Role of Oct-6 in Schwann cell differentiation Microsc Res Tech 1998 41 372 378 9672420 10.1002/(SICI)1097-0029(19980601)41:5<372::AID-JEMT4>3.0.CO;2-S
Frantz GD Bohner AP Akers RM McConnell SK Regulation of the POU domain gene SCIP during cerebral cortical development J Neurosci 1994 14 472 485 7905511
Zwart R Broos L Grosveld G Meijer D The restricted expression pattern of the POU factor Oct-6 during early development of the mouse nervous system Mech Dev 1996 54 185 194 8652411 10.1016/0925-4773(95)00472-6
Ilia M Bazigou E Price J Expression of the POU domain transcription factor, Oct-6, is attenuated in the adult mouse telencephalon, but increased by neurotoxic damage Exp Neurol 2003 181 159 169 12781989 10.1016/S0014-4886(03)00047-5
Ilia M Beasley C Meijer D Kerwin R Cotter D Everall I Price J Expression of Oct-6, a POU III domain transcription factor, in schizophrenia Am J Psychiatry 2002 159 1174 1182 12091196 10.1176/appi.ajp.159.7.1174
Physicians AAF Diagnostic and Statistical Manual of Mental Disorders IV 1995 Fourth Washington DC
Berrettini WH Are schizophrenic and bipolar disorders related? A review of family and molecular studies Biol Psychiatry 2000 48 531 538 11018225 10.1016/S0006-3223(00)00883-0
Knable MB Schizophrenia and bipolar disorder: findings from studies of the Stanley Foundation Brain Collection Schizophr Res 1999 39 149 152 10507526 10.1016/S0920-9964(99)00114-0
Moller HJ Bipolar disorder and schizophrenia: distinct illnesses or a continuum? J Clin Psychiatry 2003 64 Suppl 6 23 27 12720477
Torrey EF Webster M Knable M Johnston N Yolken RH The stanley foundation brain collection and neuropathology consortium Schizophr Res 2000 44 151 155 10913747 10.1016/S0920-9964(99)00192-9
Leger H Sock E Renner K Grummt F Wegner M Functional interaction between the POU domain protein Tst-1/Oct-6 and the high-mobility-group protein HMG-I/Y Mol Cell Biol 1995 15 3738 3747 7791781
Karin M Ben-Neriah Y Phosphorylation meets ubiquitination: the control of NF-[kappa]B activity Annu Rev Immunol 2000 18 621 663 10837071 10.1146/annurev.immunol.18.1.621
|
16246257
|
PMC1282578
|
CC BY
|
2021-01-04 16:33:02
|
no
|
BMC Psychiatry. 2005 Oct 24; 5:38
|
utf-8
|
BMC Psychiatry
| 2,005 |
10.1186/1471-244X-5-38
|
oa_comm
|
==== Front
BMC Struct BiolBMC Structural Biology1472-6807BioMed Central London 1472-6807-5-191622568710.1186/1472-6807-5-19Research ArticleNatural history of S-adenosylmethionine-binding proteins Kozbial Piotr Z [email protected] Arcady R [email protected] Stowers Institute for Medical Research, 1000 E. 50th St., Kansas City, MO 64110, USA2 Department of Microbiology, Molecular Genetics, and Immunology, University of Kansas Medical Center, Kansas City, Kansas 66160, USA2005 14 10 2005 5 19 19 21 7 2005 14 10 2005 Copyright © 2005 Kozbial and Mushegian; licensee BioMed Central Ltd.2005Kozbial and Mushegian; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
S-adenosylmethionine is a source of diverse chemical groups used in biosynthesis and modification of virtually every class of biomolecules. The most notable reaction requiring S-adenosylmethionine, transfer of methyl group, is performed by a large class of enzymes, S-adenosylmethionine-dependent methyltransferases, which have been the focus of considerable structure-function studies. Evolutionary trajectories of these enzymes, and especially of other classes of S-adenosylmethionine-binding proteins, nevertheless, remain poorly understood. We addressed this issue by computational comparison of sequences and structures of various S-adenosylmethionine-binding proteins.
Results
Two widespread folds, Rossmann fold and TIM barrel, have been repeatedly used in evolution for diverse types of S-adenosylmethionine conversion. There were also cases of recruitment of other relatively common folds for S-adenosylmethionine binding. Several classes of proteins have unique unrelated folds, specialized for just one type of chemistry and unified by the theme of internal domain duplications. In several cases, functional divergence is evident, when evolutionarily related enzymes have changed the mode of binding and the type of chemical transformation of S-adenosylmethionine. There are also instances of functional convergence, when biochemically similar processes are performed by drastically different classes of S-adenosylmethionine-binding proteins.
Comparison of remote sequence similarities and analysis of phyletic patterns suggests that the last universal common ancestor of cellular life had between 10 and 20 S-adenosylmethionine-binding proteins from at least 5 fold classes, providing for S-adenosylmethionine formation, polyamine biosynthesis, and methylation of several substrates, including nucleic acids and peptide chain release factor.
Conclusion
We have observed several novel relationships between families that were not known to be related before, and defined 15 large superfamilies of SAM-binding proteins, at least 5 of which may have been represented in the last common ancestor.
==== Body
Background
S-adenosylmethionine (SAM or AdoMet) is a conjugate of nucleotide adenosine and amino acid methionine, two ubiquitous biological compounds that almost certainly were present in the common ancestor of living cells and may have been found in the prebiotic environment on Earth, predating the origin of Life itself [1]. SAM is an essential metabolic intermediate in every studied cellular life form, and each cellular organism has several SAM-utilizing enzymes. One relatively well-understood biological role of SAM is to donate methyl groups for covalent modification of different substrates – from as simple as oxidized arsenic, chloride, bromide, and iodine ions [2-4], to as complex as rRNA, tRNA, and essential proteins, whose methylation status can serve as a regulatory signal for maturation and control interactions with other macromolecules ([5-7] and references therein).
Methyl transfer is but one of many biochemical processes requiring SAM. Enzymatic reactions that involve interaction of proteins with SAM or its structurally similar derivatives include transfer or methylene, aminoalkyl, ribosyl, and 5'deoxyadenosyl groups; formation of 5'deoxyadenosyl radical, which can be used as a redox intermediate in many reactions; SAM decarboxylation; and de novo synthesis of SAM from adenosine and methionine. There are also numerous interactions between SAM and non-enzymatic proteins, where SAM serves as a ligand triggering a regulatory change in the effector protein.
Despite the interest in this amazing variety of functions associated with SAM, and the known three-dimensional structures for representatives of almost every class of SAM-dependent enzymes, the structural, functional, and evolutionary relationships between the SAM-binding domains remain not well understood. Do all or some of the SAM-binding proteins share common evolutionary ancestry? How many distinct structural modes of interaction between SAM and protein are there? Is there strong or weak correlation between conservation of sequence and structure, the mode of SAM binding, and the chemical reaction facilitated by the enzyme? Finally, what may have been the repertoire of SAM-binding proteins in the ancestral organisms – in particular, in LUCA, the Last Universal Common Ancestor of the three present-day domains of Life – Bacteria, Archaea, and Eukarya?
We sought to address these questions by comparing sequences and structures of various groups of SAM-binding domains recognized in proteins. We describe several previously unsuspected relationships between some of such groups, predict novel members for many of them, and conclude that LUCA may have had more than a dozen of SAM-binding proteins, belonging to several distinct folds.
Results and discussion
We have adopted the iterative comparison strategy, using the known or suspected SAM-binding protein domains as the queries in increasingly sensitive probabilistic methods of sequence modeling and database searching. In many cases, a SAM-binding part of the protein constitutes only part of the polypeptide chain. For example, methyltransferases typically consist of well-conserved SAM-binding portions and highly variable substrate-binding regions, sometimes further supplemented with portable domains also found in otherwise unrelated proteins, such as chromo domain interacting with methylated histone tails in eukaryotes, or PUA domain that probably interacts with RNA [8]. In this work, we are concerned with the protein moieties that bind SAM, so we neither examine these other domains, not consider methyltransferases that utilize other sources of methyl groups, like folate or methylcobalamin derivatives. We did not describe isoprenylcysteine carboxyl methyltransferase (ICMT), an integral endoplasmic reticulum membrane protein with unknown structure [9,10] (reviewed in ref. [11]).
The phylogenetic relationships inside of several recognized groups of SAM-binding proteins, especially within Rossmann-fold SAM-dependent methyltransferases, have been reviewed recently [12,13]. Although we summarize and extend their observations, our main focus is on the analysis of more distant, previously unexamined, relationships.
Versatile α/β architectures adapted for SAM binding
Rossmanoids: ancient and ubiquitous SAM-dependent transferases
The majority of SAM-dependent methyltransferases belong to a large class of enzymes with the Rossmann-like fold, one of the more common arrangements of protein spatial structure, observed in dozens of diverse families of enzymes [14]. SAM-dependent methyltransferases are a large group of enzymes within the Rossmanoid class, and they account for a substantial fraction of all proteins in completely sequenced genomes; for example, with 1.7% of genes in Helicobacter pylori J99 coding for known or predicted SAM-dependent methyltransferases, this group makes the list of 10 most commonly used sequence and structure families in that species [15].
In the most basic arrangement, the Rossmann-like fold consists of alternating β-stranded and α-helical regions, with all strands forming a central relatively planar β-sheet, and helices filling two layers, one on each side of the plane. As with many other Rossmann-like folds, the N-terminal β-strand of methyltransferases is located in the middle of the sheet, and the strand topology is 3214576, with the 7th strand antiparallel to all other strands (Figure 1a). Yet another typical feature of Rossmanoid enzymes is that the functionally important, conserved residues are often located in the C-termini of the β-strands or in the adjoining loops [14]. Some methyltransferases conform to this plan quite well, with an occasional addition of an extra helix or a β-hairpin [16], or, rarely, deletion of one or both of strands 6 and 7 [17]. Most methyltransferases, however, contain additional domains appended or inserted into the basic Rossmann fold [16].
Figure 1 Fold and topology of SAM binding proteins. Corresponding fragments of cartoon and topology representations of selected structures were rainbow colored from N-terminal (blue) to C-terminal (red) end. Less significant fragments of secondary structure were left white in topology diagrams. Reference to other representative structures are provided (in parentheses) as SCOP sunid numbers (i.e. ref. [139]). (a) 1ej0A, Rossmann-fold methyltransferase (53335 excluding: 102555, 69556, 69557, and 69560); (b) 1l1eA, cyclopropane fatty acid synthase (69560); (c) 1mjfA, spermidine synthase (69557); (d) 1vhvA, porphyrin C-methyltransferase (53789); (e) 1j6rB, Met synthase reactivation domain (methyltransferase; 56506); (f) 1mxiA, SPOUT methyltransferase (89629, 75218); (g) 1r30A, SAM dependent radical enzyme (102114); (h) 1mvhA, SET domain methyltransferase (82199); (i) 1rqpB, 5'-fluoro-5'-deoxyadenosine synthase (102521, 101851); (j) 1m7y, 1-aminocyclopropane-1-carboxylate (ACC) synthase (53441, 64130, and similar but different: 53439); (k) 1vkyA, tRNA-ribosyl transferase-isomerase (111338); (l) 1o9tB, methionine adenosyltransferase (55972); (m) 1cmc, MetJ – methionine repressor (dimer; 100972); (n) 1msvA, SAM decarboxylase (56275); (o) 1pbjA, CBS domain (dimer; 54630 – not all CBS domains bind SAM); (p) 1kzfA, acyl-homoserine lactone synthase (75508).
Notwithstanding the insertions of additional domains and structural elaborations, comparative sequence analysis of the Rossmann-fold methyltransferases identifies the set of five highly conserved regions of the SAM-binding region, each centered on one or more nearly-invariant residues (Figure 2). They correspond to motifs I-V from motifs initially proposed for DNA:m5C MTases by Posfai et al. [18] (reviewed in refs. [12,19]), but some of the conserved residues highlighted in this work have not been pointed out before (see below). Each motif has a clear counterpart at the structural level. Five motifs are arranged in the same linear order in almost all known methyltransferases, with a notable exception of several groups of DNA- and RNA-methyltransferases, where circular permutation of the sequence results in a main chain fission after motif II, while the spatial structure of the domain and mode of SAM binding remain virtually unperturbed (discussed in more detail by Bujnicki [20]).
Figure 2 Multiple sequence alignment of Rossmann-fold methyltransferases and nicotianamine synthase. Sequences are denoted by NCBI gi number, short protein name (when available, otherwise COG/KOG/Pfam number was used), and abbreviated species name (as in UniProt Knowledgebase [140]). Nicotianamine synthase is marked by a blue box. Conserved motifs are labeled above the alignment. Conserved residues are marked by asterisk. Consensus positions of the secondary structure elements are shown above the alignment. Numbers in parentheses indicate number of residues omitted for clarity. Residues are highlighted according to the amino acid properties. Gray shading indicates conservation of single residue. Red font indicates conservation of acidic residues (D, E). Cyan font indicates conservation of Ser/Thr (S, T). Blue gray font with yellow shading indicates conservation of aliphatic residues (I, L, V). Dark blue font indicates conservation of basic residues (H, K, and R). Green font indicates conservation of tiny residues (A, G, and S). Blue font with yellow shading indicates conservation of aromatic residues (F, H, W, and Y). Pink font indicates conservation of charged residues (D, E, H, K, and R). Dark green font indicates conservation of small residues (A, C, D, G, N, P, S, T, and V). Bright blue font indicates conservation of polar residues (C, D, E, H, K, N, Q, R, S, and T). Blue font with pale yellow shading indicates conservation of big residues (E, F, H, I, K, L, M, Q, R, W, and Y). Black font with yellow shading indicates conservation of hydrophobic residues (A, C, F, G, H, I, L, M, T, V, W, and Y).
The first conserved sequence block (Motif I) includes in its C-terminal part the consensus GxGxG, considered the hallmark SAM-binding site of the Rossmann-fold SAM-dependent methyltransferases. None of three glycine residues is universally conserved, but the replacements are typically by the residues with small side chains, or with propensity of bending the main chain. This agrees with the structure data, indicating that the consensus is located in a loop connecting the first β-strand and the α-helix in the Rossmann fold core. The complete β-strand and part of the preceding loop are also part of Motif I. In the middle of β-strand 1, there is an exceptionally well conserved acidic residue (D or E); one or more conserved positively charged residues are found close to the N-terminus of this strand (Figure 2).
Motif II encompasses β-strand 2 and adjoining turn. A partially conserved acidic residue is common at the C terminus of this strand. Motif III corresponds to β-strand 3, located at the edge of the β-sheet in the Rossmann fold. An acidic residue is partially conserved close to the C-terminus of this strand, too. Whenever the substrate (SAM), its analogs, or reaction product (SAH) are co-crystallized, they are found close to the invariant residues in Motifs I-III (Figure 2 and see below).
Motif IV consists of β-strand 4 and the flanking loops. In this motif again, there is a well-preserved D/E/N residue, located at the extreme N-terminus of the strand, i.e. at the side of the fold that is not involved in substrate binding. Motif V corresponds to the helix following the strand with motif IV. In some Rossmann-fold methyltransferases, it serves as a scaffold for large hydrophobic or aromatic side chains that stabilize the adenine moiety of AdoMet, but it many cases it has been shows these residues are not essential for the MTase activity [21]. Finally, Motif VI corresponds to Strand 5 of the β-sheet, and the preceding tight turn with a nearly-invariant glycine residue.
Several residues from Motifs I-V are known to make direct contact with SAM. In particular, one or more residues in "GxGxG" loop are in contact with the carboxypropyl moiety of SAM, while conserved acidic residue in Motif II forms hydrogen bonds with the ribose hydroxyls (Figure 2; ref. [22]). Variable residues at the C-termini of strand 2 (Motif II) and conserved acidic residue in Strand 3 (Motif III) interact with the nitrous base, while variable residues C-terminal to Strand 4 (Motif IV) appear to contact the amino and sulfonium groups of the methionine moiety of SAM [22]. Residues from motif IV, VI, VIII, and sometimes X are associated with the catalytic pocket, where residues from motif V and VII are important mostly for the structural stability [19,23].
The roles of other conserved residues in SAM-dependent methyltransferases are less well understood. Near-omnipresence of the D/E residue in Motif I suggests that it has an important role. It has been noted [24], that in FtsJ RNA methyltransferase this residue coordinates SAM through a water molecule. In fact, in all 3-D structures of methyltransferases where solvent molecules are present (i.e. PDB structure 1EJ0, 1KYW, and 2ADM), the oxygen atoms in the carboxyl group of this D/E residue make direct contacts with two water molecules, one of which is capable of forming a hydrogen bond with the side chain of methionine moiety of SAM. In some ribose 2'-O-MTases, D/E amino acid conserved in motif I is substituted by tyrosine (Figure 2 and ref. [25]), and it has been proposed that this residue could be used to directly (not via the water molecule) coordinate the amino-carboxyl end of SAM (J.M. Bujnicki, personal communication).
These observations are of interest for understanding the mechanism of methyl transfer by Rossmann-fold methyltransferases. Two best-studied groups of transferases that have Rossmann-like fold and use a nucleotide derivative as a cofactor, namely ATPase-like kinases and nucleoside diphospho-sugar transferases, appear to require a divalent metal cation for polarization of water molecule that can then attack a scissile phosphoester bond [26-29]. Methyltransferases, on the other hand, need to work on a C-S+ bond in SAM, but do not seem to have any metal ion bound in the appropriate position (even though divalent cations have been included in some crystallization media). The proposals for reaction mechanisms of different classes of SAM-dependent methyltransferases include nucleophilic catalysis, with the identity of nucleophile ranging from moderately conserved residues scattered across the SAM-binding domain to bound water molecule [30], as well as SN2 reaction, which would require initiation by concerted action of several side chains, or, perhaps, by the amino group of the substrate itself [31]. The highly conserved D/E residue in motif I may, however, provide a unifying theme in the catalysis, by polarizing a water molecule that is close to the methyl group of SAM. The water molecule could either serve as a nucleophile, or aid bond displacement between the sulfonium ion and methyl group in some other way.
Finally, we noticed that the conserved basic residue at the beginning of Motif I and nearly-invariant acidic residue at the beginning of motif IV are typically located within a short distance (3Å or less) of each other, potentially forming a salt bridge that may be important for locking other elements of the Rossmann fold in place (Figure 2).
Rossmann-fold SAM-binding proteins that do not have methyltransferase activity
I. Methylene transferases
Formation of the cyclopropane ring in unsaturated fatty acids by cyclopropane fatty acid synthase [EC: 2.1.1.79] has been studied extensively in bacteria. The reaction involves transfer of a methylene group from SAM to the double bond of an unsaturated acyl chain [13,32].
Crystal structures of mycolic acid cyclopropane synthases CmaA1, CmaA2, PcaA, and MmaA2 have fold similar to Rossmann-fold methyltransferases (Figure 1b), with the conserved position of SAM and very similar pattern of interactions with the cofactor [32]. A hallmark of methylene transferases is the presence of the carbonate ion (CO3 2-) at the active center, which probably enables the formation of carbocation intermediate required for completion of the reaction (Figure 3) and ref. [32]. Conserved residues involved in carbonate ion binding (Cys35/Ser, His167/Gln, and Tyr232/Phe – numbered as in PDB structure 1L1E) appear to distinguish methylene transferases from Rossmann-fold methyltransferases.
Figure 3 Structural alignment of cyclopropane fatty acid synthase and Rossmann-fold methyltransferase. Red color – YecO methyltransferase (PDB structure 1IM8:A) Yellow color – ligand in YecO methyltransferase. Green color – mycolic acid cyclopropane synthase CmaA2 (PDB structure 1KPI:A). Blue color – three ligands in CmaA2 (PDB structure 1KPI:A). Motifs I-VI conserved in Rossmann-fold methyltransferases are labeled above the alignment.
II. Amino alkyl transferases: nicotianamine synthase and spermidine synthase
Nicotianamine synthase (S-adenosyl-L-methionine: S-adenosyl-L-methionine: S-adenosyl-L-methionine 3-amino-3-carboxypropyltransferase, EC: 2.5.1.43) catalyses direct condensation of three molecules of SAM, followed by the formation of an acetidine ring, to yield one molecule of nicotianamine, a chelator of various transition metals ubiquitously present in higher plants. In graminaceous plants, nicotianamine is the precursor of phytosiderophores that are secreted from the roots to solubilize ferric iron in the soil. Reduced levels of endogenous nicotianamine affect the development of plant reproductive organs and seeds' maturation [33].
Protein structure of nicotianamine synthase is not known, but sequence similarity searches indicate a significant similarity between nicotianamine synthase and Rossmann-fold methyltransferases. A PSI-BLAST search, using with Arabidopsis thaliana NAS [GenBank:O80483] as a query, matched Pseudomonas syringae ubiE/COQ5 methyltransferase [GenBank:YP_233497] with E-value = 1e-21 and score = 105 at 4th iteration; PDBBLAST top match is to (N5)-Glutamine Methyltransferase [PDB:1T43]. Similarity is the highest in conserved motifs II-IV and VI (Figure 2), and motif I is also conserved, in a modified form (F-x-G-S-G-P-x-P). Interestingly, related sequences with the same modification of Motif 1 are found in archaea (Methanothermobacter thermautotrophicus, [GenBank:NP_275817], bacteria (Pseudomonas aeruginosa, [GenBank:NP_253523], and fungi (Neurospora crassa, [GenBank:XP_330777]. The replacement of the conserved D/E/N residue in motif I (see above) may partially explain the change in the functional group that is transferred from SAM: unlike the related Rossmann-fold methyltransferases, nicotianamine synthases lack negatively charged residue in Motif I, so the aminoalkyl moiety of SAM is not bridged to the enzyme by a water molecule and is free to leave in the course of the reaction.
Spermidine synthase (EC: 2.5.1.16) catalyzes the transfer of the aminopropyl group from decarboxylated SAM to putrescine to form spermidine. Putrescine, spermidine and spermine, formed from spermidine, are polyamines essential for the regulation of cell proliferation and differentiation in most species, and gram-negative bacteria outer membrane permeability in response to the acid stress [34,35]. Spermidine synthase is an oligomeric enzyme, each monomer consisting of a C-terminal domain with a Rossmann-like fold and an N-terminal tetramerization β-stranded domain [36].
Spermidine synthase has very high sequence similarity (approx. 70% identity) to putrescine N-methyltransferase. It has been shown that aminopropyl moiety of spermidine synthase inhibitor – AdoDATO (a compound containing both substrate and product moieties) binds in a similar orientation to the homologous part of SAM in Rossmann-fold methyltransferases. However, the binding site in spermidine synthase contains invariant residue Asp101 (PDB structure 1JQ3), located in the middle of glycine-rich loop (corresponding to motif I of Rossmann-fold methyltransferases) making binding cavity too small to accommodate the carboxyl group of SAM, that prevents SAM binding and enables specific binding of decarboxylated SAM [36]. The polyamine moiety of AdoDATO is oriented toward putrescine binding cleft. Invariant Asp170 (corresponding to D/N residue located at the end of β-sheet in motif-IV of Rossmann-fold methyltransferases) is most likely candidate to deprotonate putrescine, enabling it to perform a nucleophilic attack on methylene carbon of decarboxylated SAM [36].
III. Acalcynomycin-10-hydroxylase
Aclacinomycin 10-hydroxylase RdmB is a methyltransferase homolog that catalyses a SAM-dependent hydroxylation of the C-10 carbon atom of 15-demethoxy-ε-rhodomycin, a step in the biosynthesis of the polyketide antibiotic β-rhodomycin. In order to activate molecular oxygen, the enzyme uses SAM instead of cofactors usually associated with hydroxylase activity, such as flavins, 2-oxoglutarate, or metal ions. It has been proposed that positive charge of the SAM cofactor plays a role in delocalisation of electrons into the anthraquinone core of the substrate [37]. The C-terminal domain of RdmB has an α/β Rossmann-like fold, and contains the conserved signature DLGGGxG in motif I.
The enzyme lacks methyltransferase activity due to the positioning of SAM in which the methyl group points in a direction unfavorable for a SN2 type methyl transfer to the substrate [37]. The difference in SAM-substrate positioning is not well understood, but might be related to differential interactions between SAM binding C-terminal domain and substrate binding N-terminal domain or differences in the conserved loop (amino acids 292–298) [37].
Non-catalytic Rossmannoids
The lack of detectable SAM binding motifs in several Rossmann-fold methyltransferases suggests that they may be recruited for a new function. For example, the three-dimensional structure of sc-mtTFB (Saccharomyces cerevisiae mitochondrial transcription factor B) bears strong resemblance to ribosomal RNA adenine dimethylases (i.e. KsgA and ErmC'). However, several residues required for interaction with SAM are not conserved in sc-mtTFB; in particular, the glycine-rich loop (motif I) contains bulky Tyr residue, and motif IV is poorly conserved.
Human co-orthologs of sc-mtTFB (h-mtTFB1 and h-mtTFB2) have rRNA N6-adenine methyltransferase activity (in an Escherichia. coli assay), but mutational analysis of h-mtTFB1 indicates that this activity is not required for transcriptional activation [38].
In Gcd10p/Gcd14p complex – tRNA(1-methyladenosine) methyltransferase of S. cerevisiae, the lack of SAM binding was observed in Gcd10p that directs binding of tRNA, where Gcd14p binds the required cofactor S-adenosylmethionine [39,40].
In another case, Kar4p (pheromone induced, karyogamy-specific transcription factor) does not bind SAM, where similar (circularly permuted) Ime4p is SAM-binding methyltransferase [41].
In bacterial rRNA:m2G methyltransferases RsmC and RsmD the inactivated domain and the catalytic domain are fused together in one polypeptide [42].
The de novo methyltransferase-like protein, DNMT3L, is required for methylation of imprinted genes in germ cells. Although enzymatically inactive, human DNMT3L accelerates DNA and SAM binding to de novo DNA methyltransferases [43-45].
Rossmann-like domain of bacterial fluorinating enzyme
Actinomycete Streptomyces cattleya is able to produce C-F bonds using inorganic fluoride. The fluorinating activity requires SAM, and the primary product of the reaction is SAM derivative, 5'-fluoro-5'-deoxyadenosine [46]. The protein, 5'-fluoro-5'-deoxyadenosine synthase FlA, belongs to a conserved sequence family represented in most archaea and in a subset of bacteria [46].
The structure of FlA consists of two domains – a larger N-terminal domain with α/β fold, and a smaller C-terminal β-barrel. Both domains interact with SAM and with reaction products [46]. FlA is a hexamer in solution and trimer in crystal, and three SAM molecules are bound by a trimer, between the N-terminal domain of one subunit and the C-terminal domain of the adjoining subunit. This arrangement, however, appears to be dependent on a long (24 amino acids) loop in the N-terminal domain, which is missing from the closely related sequences in all other species. On the other hand, the linker connecting two domains in a monomer is long enough to allow significant domain motions, and it is plausible that two domains may interact in other oligomeric arrangements and perhaps even within a monomer. Therefore, we speculate that SAM binding by FlA-like proteins from other species may occur in the crevice formed by the N- and C-terminal domains of the same molecule, and the following discussion does not consider the oligomerization state.
The N-terminal domain makes contacts mostly with methionine, ribose, and fluoride ion, and C-terminal domain contacts methionine and adenine ring. The α/β N-terminal domain exhibits several features similar to other enzymatic domains with Rossmann-like topology, namely: three-layer α/β/α architecture; the planar central, mostly parallel β-sheet filling "inside-out" (strand topology 2135467), and concentration of the substrate-binding and catalytic residues in the loops following the C-termini of strands. More specifically, the loop after strand 1 contains Asp16 (numbered as in PDB structure 1RQP) hydrogen-bonded to both hydroxyls of ribose, Leu17 that may be involved in water-mediated interaction with methionine amino group, Asp21 and Ser23, both of which can form hydrogen bonds with the same amino group. Loop following strand 2 contains Trp50 that is able to contact one ribose hydroxyl and perhaps to have Van der Waals interactions with the adenine ring. Loop between strands 3 and 4 contain two ribose interactors, Thr76 and Tyr77. Loop after strand 6 hosts Thr155, which is part of hydrogen-bond network linking two domains via two water molecules and methionine carboxyl group, as well as catalytic Ser158 that is expected to make two polar contacts with deprotonated fluoride ion [46]. Although most of these interactions are provided by residues located in loops at the edge of β-sheet, there is no specific sequence similarity between Rossmann-like domain in fluorination enzyme and in Rossmann-fold methyltransferases. There is also no similarity to other SAM-utilizing enzymes.
Rossmann-like fold in SPOUT methyltransferases?
A distinct superfamily of SAM-dependent methyltransferases, SPOUT, which includes families specified by bacterial SpoU, TrmD, and TrmH, proteins, as well as many uncharacterized proteins in all three domains of Life, have been shown to share a set of conserved sequence elements and an α/β-type fold [47]. Trm10, a recently characterized tRNA m1G9 methyltransferase, is also predicted to have this fold [48,49]. All experimentally characterized members of this large superfamily are DNA or RNA methyltransferases. A unique structural feature of this α/β fold is a trefoil knot of two crossing loops in the C-terminal region [47].
Several hallmarks of Rossmann-like structure are evident in the SPOUT fold. There are three main layers, with a central β-sheet sandwiched between two helical layers; the β-sheet is formed "inside-out", with the first and one of the last strands in the center of the sheet; and the SAM ligand interacts mostly with the conserved residues located in the loops at the C-termini of β-strands [50]. There is, however, no sequence similarity between SPOUT-fold methyltransferase and any other Rossmann-fold SAM-binding protein.
Structural similarity between SPOUT-fold and Rossmann-fold methyltransferases (i.e. DALI Z-score = 3.1 for 88 aa with RMSD = 3.7 and sequence identity = 17% for PDB structures 1QAO and 1MXI) is confined mainly to the N-terminal half of those folds. There is no similarity in the C-terminal part, where strands 4 and 5 are rearranged.
SAM binding inside and outside of β-barrels
SAM-radical enzymes: recruitment of ancient enzymatic TIM barrel
A (β/α)8 fold, also known as triose phosphate isomerase (TIM)-like barrel, is one of the largest classes of protein structures, exceeding even Rossmann-like fold in omnipresence and versatility [51]. Most of TIM-barrel proteins are enzymes, belonging to almost all of the major EC classes [52]. A well-known version of a TIM barrel is a (β/α)6 "semi-barrel," in which the inner layer of slanted β-strands does not form a complete cylinder, but has a lateral opening (PDB structure 1OLT).
Recently, several structures of proteins from a large sequence family of "SAM radical" enzymes (ref. [53]; Figure 1g) have been determined, and it became evident that members of this family have (β/α)6 and (β/α)8 folds. SAM-radical enzymes utilize non-covalently linked Fe-S cluster and a SAM molecule, in a reductive cleavage reaction that produces methionine and 5'-deoxyadenosyl radical, that can be used to generate further glycyl or thiyl radicals on the same protein molecule or on a coupled enzyme [54]. It has been noted that SAM-radical sequence family is very large, diverse, but can be recognized by a hallmark CxxxCxxC signature close to the N-terminus, followed by another conserved "GG" motif [55].
We detected more than 2000 non-redundant sequences from SAM-radical family in the sequence databases. Interestingly, when the region containing the three characteristic cysteines was deleted from the queries, the searches resulted in almost the same collection of sequences as with full-length domain, indicating strong evolutionary signal along the stretch of 200–250 residues to the C-terminal side of the CxxxCxxC signature. Multiple alignment of many representative sequences identified four regions with high sequence similarity and three weaker conserved motifs (Figure 4). Comparison of the alignment with the known structures of biotin synthase (PDB structure 1R30), coproporphyrinogen III oxidase (PDB structure 1OLT) and molybdenum cofactor biosynthesis enzyme MoaA (PDB structure 1TV7) suggests structural and functional correlates for these regions and for the most conserved residues within them. The best-conserved motifs correspond to the β-strands of the inner barrel and their C-terminal loops, while the regions of additional partial conservation correspond to the outer-shell α-helices.
Figure 4 Multiple sequence alignment of SAM-dependent radical enzymes. Sequences are denoted by NCBI gi number, conserved domain name (as in NCBI CDD database [141]), short protein name (if available), and abbreviated species name. Secondary structure elements extracted from PDB structure 1OLT coordinates are shown above the alignment. Residues are highlighted according to the amino acid properties with designations as in Figure 2.
Motif 1 includes the most N-terminal β-strand in the (β/α)6 barrel (strand 1). Three invariant cysteine residues in the adjoining loop (Figure 4) coordinate the [Fe-S]4 cluster, which is present in a similar configuration in all protein structures resolved thus far. One iron atom has no contact with the cysteine side chains, and is instead ligated by the N and O atoms from the amino- and carboxy groups of SAM. Also highly conserved is aromatic or heterocyclic residue (Y, F, or H) preceding the last of the three cysteines; the main chain of this residue seems to form a hydrogen bond with the adenosyl moiety of SAM, but the significance of the side chain conservation is unclear; perhaps it contributes to the non-polar milieu of the bound Fe-S cofactor, preserving it from oxidation.
The second prominent motif does not contain any invariant amino acids, but includes several residues with small side chains, most often two or three glycines in a row (Figure 4). This motif corresponds to the second strand in the barrel and the tight turn after the strand. The main chain of this turn is within a contact distance from the amino group of the methionine part of SAM. The third motif also corresponds to the strand-turn structure. A signature T/S-N-G that follows strand 3 is well conserved; as a rule, residues in this turn form hydrogen bonds with the carboxyl group of methionine in SAM. The fourth motif consists of strand 4 and the loop with a highly conserved acidic or amide residue (D, E, N, or Q). Typically, this residue is within hydrogen-bonding range from both 2'- and 3'hydroxyl groups of the ribose ring of SAM. The strands 5 and 6 followed by loops provide one or more residues that form hydrogen bonds with the amino group of adenosyl; however, sequence conservation in these regions is moderate.
The heterogeneity of the SAM radical protein superfamily is most pronounced in their C-terminal regions, which are responsible for binding of substrates and auxiliary cofactors. On the other hand, the structure and sequence of the N-terminal, SAM-binding region of SAM-radical proteins is well conserved, analogously to the Rossmann-fold methyltransferases. The SAM-binding region is essentially an incomplete (β/α)6 "semi-barrel," which is typically modified by evolutionarily diverse elements (commonly consisting of α-helices, but sometimes also containing β-hairpins or small sheets) that serve substrate-binding and regulatory roles.
Rossmann folds and TIM-barrels in fact have quite similar β/α architectures. This becomes especially evident in the case of incomplete barrels. The primary difference is lack of one α-layer in TIM-barrels, and correlated changes in sheet curvature and strand orientation. The two classes of SAM-binding enzymes both use loops between strands and helices to interact with various moieties of relatively extended SAM molecules, but the details of this interaction are quite different (see below).
TIM barrel-like catalytic domain in QueA?
Queuosine is a hyper-modified nucleoside in bacterial and eukaryotic tRNAs, produced by a multi-step enzymatic pathway that includes a transfer, with simultaneous isomerization, of ribose moiety from SAM to a modified base in tRNA, called 7-(aminomethyl)-7-deazaguanosine, or preQ1. This step is performed by QueA protein, an S-adenosylmethionine:tRNA ribosyltransferase-isomerase. QueA homologs are found in most bacteria, but their sequence is not strongly similar to any other protein family, and high-resolution structure of QueA in complex with SAM is unavailable. We interrogated the fold recognition meta-servers with individual QueA sequences and with a probabilistic model of aligned QueA homologs. The highest 3D-Jury consensus score (69 units, indicating the upper level of the "gray zone" of provocative, if statistically insignificant, sequence similarities [56]) was to pyruvate kinases, a distinct class of proteins with three structural domains. The C-terminal, regulatory domain of pyruvate kinases has no counterpart in QueA. The other two domains are arranged in such a way that a smaller, β-barrel domain is inserted into the larger, α/β TIM-barrel domain but folds independently. Similar arrangement of two domains is predicted for QueA.
When this manuscript was under preparation, the structure of QueA from Thermotoga maritima was resolved (PDB structure 1VKY). In agreement with the fold recognition data, it shows an α/β domain with insertion of independently folding β-barrel (Figure 1k). The structure of the α/β domain shows one β-sheet, with preponderance of α-helices on one side (Figure 1k). This structure resembles a semi-barrel, given a strongly curved β-sheet, relative absence of α-helices on the concave side, and a lid-like irregular arrangement of elements that covers the cavity. There is an unresolved protein segment of 32 residues, which should be located close to the inner β-layer of the semi-barrel, and may in fact extend its wall. An unresolved ligand is placed in the proximity of the C-termini of several β-sheets, and if this is in fact SAM, its binding mode would be similar to what is observed in other SAM-binding proteins (see below).
β-barrels in QueA and fluorination enzyme
Both QueA and fluorination enzyme structures show fusions of a larger α/β domain and a smaller all-β domain with barrel-like topology. The role of all-β domain in QueA is unclear, but it is not very likely to be involved in interaction with SAM. In contrast, the β-barrel domain in FlA (which, in fact, is more similar to "smashed β-can," with one side caved in, producing a double-concave surface) makes many contacts with the ligand. Proceeding from the N- to C-terminus, the Asp210-His211 (as in PDB structure 1RQP) dipeptide in the loop after the first strand bonds with the amino group of methionine; Asn215 bonds with the amino group of adenine; Ser269 and Arg270 after strand 5 can form 4 hydrogen bonds altogether, all with the carboxyl group of methionine; and at the C-terminus of strand 5, Arg277 and Ala279 provide additional interactions with adenine. While the catalytic mechanism of FlA is dependent on correct positioning of the fluoride atom with regards to ribose, which is mediated by Ser158 in the Rossmann-like domain (see above and reference [46]), the β-barrel domain appears to be essential for correct orientation of SAM, which serves as fluoride acceptor.
Between the sheets: double-β SAM-binding folds with a common theme of internal domain duplication
Decarboxylase
S-adenosylmethionine decarboxylase (EC: 4.1.1.50, SAMDC) is a key enzyme in spermidine and spermine biosynthesis. It produces decarboxylated SAM (dcSAM), which then donates aminopropyl group to putrescine or spermidine, two essential intermediates in polyamine biosynthesis. Because polyamines link diverse pathways in cellular metabolic networks, and because chemical inhibitors of SAMDC display potent antitumor and anti-parasite activities [57,58], structure-function relationships of SAMDC are of considerable interest.
SAMDC activities have been purified from all three domains of Life (bacterium E. coli, archaea Methanococcus jannaschii, and several eukaryotes), and certain common features of the enzymes have been noticed. All three enzymes are processed in vivo, forming a small subunit derived from the N-terminus and a large subunit accounting for the rest of the molecule; the N-termini of all large subunits contain a pyruvoyl group, produced from a serine residue by autoprocessing and required for the formation of the Schiff base during catalysis. All studied enzymes form multimers from the heterodimers of large and small subunits. There are also differences among bacterial, archaeal and eukaryotic SAMDC: the length of the precursor proteins in different species varies from 105 to 460 amino acids; mammalian enzymes require putrescine for full activity, E. coli enzyme requires Mg2+ cation, while archaeal and plant enzymes apparently do not require those factors. The bacterial enzyme is a tetramer of heterodimers, while eukaryal and archaeal enzymes are homodimerized heterodimers.
Comparative sequence analysis has revealed statistically significant sequence similarity between archaeal and bacterial SAMDC [59]. Multiple alignment of these two classes of SAMDC spans the complete length of the shorter (ca. 120 aa) archaeal enzymes, and also suggests that there are two types of bacterial enzymes – some are about the same size as archaeal SAMDC, and some are longer and phylogenetically distinct (reference [60], Figure 5a, and unpublished observations). No sequence similarity has been reported between these enzymes and eukaryotic SAMDC.
Figure 5 Multiple sequence alignment of SAM decarboxylases. Sequences are denoted by gene and species name. A: Decarboxylase alignment – the residues underlined on the top line are those involved in SAM-binding, enzyme self-processing, and catalysis. Residues are highlighted according to the amino acid properties with designations as in Figure 2. B: Superimposition of the conserved residues in SAMdc from T. maritima (PDB structure 1TLU), human (PDB structure 1I7B) and potato (PDB structure 1MHM) are shown on cartoon representation of T. maritima structure (gray). Blue – Ser; Red – Ser and Ser converted into pyruvoyl group (or pyruvoyl group with covalently bound S-adenosylmethionine methyl ester); Orange – His; Cyan – Cys.
High-resolution structures of eukaryotic SAMDC from humans and plants in complex with substrate analogs and various inhibitors have been reported. The heterodimer folds as a sandwich of two β-sheets between α-helical regions, where the smaller subunit forms a half of one β-sheet, and the larger subunit completes this sheet and accounts for all the strands in the other sheet. The arrangement is unique among the known protein folds, but visual inspection and superposition of the two α-β halves of the molecule revealed their remarkable similarity and suggested the hypothesis of internal duplication [61,62]. The evolutionary origin and catalytic mechanism of prokaryotic SAMDC remained unclear.
Searches of sequence databases with the PSI-BLAST program and more involved probabilistic models of aligned SAMDC enzymes confirmed statistically significant sequence similarity between archaeal and bacterial enzymes, and also, intriguingly, produced several statistically insignificant local matches to one-half of eukaryotic SAMDC sandwich, in the area corresponding to the β-strand 12 in the three-dimensional structure. This strand is positioned next to the active center of the enzyme, and contains residues important for catalysis and/or binding of the substrate (see below). Because both these residues appeared to be preserved in the BLAST output, we sought better statistical validation of this similarity using Metaserver [56]. When SAMDC homolog from archaea Archaeoglobus fulgidus was used as a query, the highest 3D-Jury consensus score (46–60) was reported to the set of the eukaryotic SAMDC structures; this score is at the top of the zone with borderline significance, where most of the non-trivial similarities are discovered [56]. The first false positive (bacterial luciferase) was associated with the sharp drop in the 3D-Jury scores (14.5).
Almost complete archaeal sequence can be aligned to the half of eukaryotic template, with just one short gap. Conversely, the aligned region of the template corresponds almost precisely to the C-terminal half of the double sandwich. We conclude that the archaeal enzyme may resemble a half of the eukaryotic SAMDC fold and may be directly related to the pre-duplication ancestor of that fold. Multiple sequence alignment of archaeal, bacterial, and eukaryotic enzymes strongly reinforces these observations (Figure 5a). The C-terminal halves of eukaryotic enzymes could be aligned to prokaryotic homologs directly and unequivocally; the structurally similar N-terminal halves had to be superimposed using the knowledge of secondary structure and information about a few conserved residues.
The functional and evolutionary implications of the alignment are provocative. In mammalian enzymes, SAM decarboxylase is active as a dimer in which each protomer contains one large and one small subunit, and each of the two halves of the sandwich contributes several residues to binding the substrate and actually performing the catalysis. In particular, Ser residue in β-strand 4 of the eukaryotic enzymes, which is converted into catalytic pyruvoyl group, appears to be within a short distance of the carboxyl group of SAM forming a Schiff base adduct with it. Before product release, carbon of decarboxylated SAM is protonated by adjacent Cys (Figure 5b). This protonation regenerate the pyruvoyl group [63]. Also close to the active site is the side chain of the histidine residue in strand 12, which is believed to be responsible for abstraction of a proton from the α-carbon of the catalytic serine during proenzyme processing [64].
Two acidic residues contribute to binding of SAM: glutamic acid at the C-terminus of strand 3 contacts the base, and another glutamate, at the C-terminus of strand 12, interacts with both hydroxyl groups of the ribose ring. All these residues are conserved in eukaryotic SAMDCs – some in the N-terminal half of the sandwich, and others in the structurally equivalent C-terminal half. Interestingly, in archaeal and most bacterial enzymes, the pattern of conservation of these residues appears to be the union of conserved elements in the two halves of eukaryotic enzymes (Figure 5a), as if the bacto-archaeal enzyme is a homolog of one half of the eukaryotic enzyme, and the β-sandwich in the holoenzyme are made of two identical molecules.
When this manuscript was in preparation, the structure of ligand-free holoenzyme from bacterium T. maritima was deposited in the database (PDB structure 1TMI). Analysis of this structure confirms this sequence-based prediction and suggests that the bacto-archaeal form is ancestral, and the eukaryotic form has been derived from it by domain duplication/fusion, followed by functional specialization of two halves (most notably, by mutating the C-half of the enzyme so that it no longer undergoes autoproteolysis – Figure 5b).
SET domain
Discovered as conserved domain shared by chromatin remodeling proteins Su(var)3–9, E(Z) (short for Enhancer of Zeste) and Trithorax, SET domains turned out to be a distinct class of SAM-dependent methyltransferases. All studied SET methyltransferases transfer methyl group to lysine within various nuclear proteins involved in chromatin function and regulation of transcription, such as histones, TAF10, tumor suppressor p53, but also in such diverse proteins as Rubisco and cytochrome C [65-68].
In SET-domain methyltransferases amine of the substrate lysine residue access the methyl donor (SAM) through a narrow channel connecting the substrate and SAM binding surfaces [69]. SAM binding site and the catalytic center of all studied SET domains seem to be constructed on the unusual but conserved, all-β, knot-like structure [70]. Adenosyl moiety of SAM interacts directly and indirectly, through water, with conserved histidine (PDB structure: 1O9S-His297; 1P0Y-His243). This histidine may serve as a proton acceptor for the hydroxyl group of invariant Tyr (PDB structure 1O9S-Tyr335). The -OH of this Tyr is within 4 Å of the presumptive location of the substrate Lys Nζ, and may be involved in Lys side chain deprotonation (deprotonated Lys is presumed to make a nucleophilic attack on the SAM methyl group). Positively charged amino nitrogen from SAM hydrogen bonds with the side chain of invariant asparagine (PDB structure 1O9S-Asn296). This interaction may contribute to the compact conformation of the SAM molecule.
Phylogenetic analysis of the SET domain suggests that it is an evolutionary innovation in the eukaryotic lineage (with secondary lateral transfer to bacteria, archaea and viruses) [71]. SET domains (Figure 1h and SCOP superfamily: 82199) have a fold unique for SAM binding proteins – a substrate binding subdomain between two structural repeats, which may have evolved by duplication of 3-stranded unit with a generic ligand binding role [71,72]. Those repeats have a β-clip fold formed by double-stranded ribbons sharply bent in two places; the ribbon ends form incomplete barrel.
Similar duplication of a basic three-stranded unit containing the β-clip structural motif probably occurred also in related SAF and dUTPase superfamilies [72], which, however, tend to bind sugar and sugar derivatives [72]. There have been several other cases of adaptation of a generic ligand binding domain for SAM-binding, both in enzymes and in regulatory proteins without catalytic activity (see below).
SAM synthetase
S-adenosylmethionine synthetase (SAM synthetase, ATP:L-methionine S-adenosyltransferase, or MAT, EC: 2.5.1.6) is the main, or, possibly, the only enzyme of de novo SAM biosynthesis. SAM synthases from bacteria and eukaryotes are closely related at the sequence level and have very similar structures [73]. SAM synthases transiently interact with SAM prior to its release. The mechanism of reaction is thought to rely on conserved His14 (Figure 6), which acts as an acid to cleave the C5'-O5' bond of ATP, while simultaneously a change in the ribose ring conformation from C4'-exo to C3'-endo occurs, and the S of Met makes a nucleophilic attack on the C5' to form SAM [74].
Figure 6 Multiple sequence alignment of SAM synthetases. Essential amino acids for the first (*) and the second substrate binding subunit are numbered as in gi:46015497 (PDB structure 1P7L and 1RG9). Aligned sequences represent protein from Archaea (gi:3334428 – M. jannaschii), Eukaryota (gi:400245 – H. sapiens, gi:6016553 – C. elegans), and Bacteria (gi:46015497 – E. coli, gi:15836994 – X. fastidiosa, gi:22095828 – F. nucleatum, gi:13357974 – U. parvum, gi:2500686 – M. pneumoniae, gi:21646472 – C. tepidum). Residues are highlighted according to the amino acid properties with designations as in Figure 2. Substrate binding is annotated below the alignment as follows: small letters – water mediated interactions; inverted colors – interactions with ligand from the second subunit of the homodimeric protein ; A/a – adenosyl moiety; R – ribosyl moiety, M/m – methionine moiety; (+) – Mg2+; K – K+; P – PPNP/Phosphate moiety. Consensus positions of the secondary structure elements are shown above the alignment. Numbers in parentheses indicate number of residues omitted for clarity.
The fold of bacto-eukaryal SAM synthetase is unique; each protein chain is based on a β-α-β-β-α-β module that folds into a wedge-like shape. A polypeptide chain consists of three such tandemly repeated modules, so that the complete SAM synthetase fold looks like a three-slice cream pie with topping made of β-sheets (Figure 1L). The active form of the enzyme appears to consist of two pies, with β-layers facing each other. Two SAM molecules are bound between the sheets of this dimer. In the E. coli enzyme, both subunits contribute many residues to SAM binding (Komoto et al. [74], Figure 6, and PDB structure: 1P7L and 1RG9) In particular, adenosine binds to Asp163, Arg229, Phe230 on one subunit and to Ser99 on another subunit, and interacts with many additional amino acids on both subunits via water-mediated hydrogen bond network. Methionine binds to Gln98 and Asp238 on one subunit, to Glu55 on another, and likewise makes many additional water molecule-mediated contacts. Bacto-eukaryal SAM synthetase is an evolutionary unique sequence and structural family. Even SAM decarboxylase, which is superficially similar in that it also sandwiches SAM between two β-sheets, has no detectable sequence or structure similarity to SAM synthetase.
SAM synthetases from Archaea have been isolated on the basis of their biochemical activity [75]. We performed sensitive searches of the conserved domain database, and found clear evidence for common ancestry of all SAM synthetases (Table 2). Multiple sequence alignment and secondary structure prediction indicate that archaeal enzymes share the same three-dimensional structure as their eukaryotic and bacterial homologs. All known SAM synthetases have conserved GHPD signature containing the main catalytic residue (His14 in E. coli gi:46015497) [75]. Despite high sequence divergence between archaeal and bacto-eukaryal enzymes, the complement of substrate binding residues is well-preserved (Figure 6). The apparent common origin of this unique enzyme in all major divisions of Life is of great interest for reconstruction of the repertoire of SAM-binding protein in the ancestral life forms (see below).
Table 1 Classes of SAM binding proteins
Reaction summary Trivial name, EC number Sequence families and spatial folds (summary) Evolutionary roots and status in LUCA
Methyl transfer SAM-dependent methyltransferase, EC:2.1.1.- Five classes: I, Rossmann fold; II, reactivation domain of methionine synthase; III, "corrinoid-like" MTases; IV, SPOUT domain; V, SET domain. Classes I-IV are α/β folds, class V is a β-clip. The Rossmann-fold MTases are the largest class of SAM-dependent enzymes. The folds of Classes II and III are unique. Trm10 and TrmH families of RNA MTases appear to be the modified versions, of, respectively, class IV fold [47], [49] and class I fold fused to PP-superfamily ATPase (this study). GTP MTase of Sindbis-like viruses may belong to α/β class, but specific fold prediction is unavailable Several distinct Rossmann-fold methyl transferases in LUCA.
Methylene transfer Cyclopropane fatty acid synthase, EC:2.1.1.79 Rossmann-fold methyltransferase family Derived from an ancient enzyme; not in LUCA
Aminoalkyl transfer 1 Nicotianamine synthase, EC:2.5.1.43 Rossmann-fold methyltransferase family with permuted order of sequence motifs Derived from an ancient enzyme; not in LUCA
ACC synthase, EC:4.4.1.14 PLP-dependent aminotransferase fold; the SAM-binding domain is derived from generic substrate-binding cleft Derived from an ancient enzyme; not in LUCA
Acyl-homoserine lactone synthase, EC:6.1.- GNAT-type acetyltransferase fold; the SAM-binding domain is derived from generic substrate-binding cleft Derived from an ancient enzyme; not in LUCA
Aminopropyl transfer Spermidine synthase, EC:2.5.1.16 Rossmann-fold methyltransferase family, but the substrate is decarboxy-SAM Probably in LUCA
Ribosyl transfer tRNA-ribosyl transferase-isomerase, EC:5.- QueA family; smaller β-barrel N-terminal domain and a larger C-terminal domain with α/β fold, distantly related to a TIM-barrel Bacterial invention; not in LUCA
5'deoxyadenosyl transfer 5'-fluoro-5'-deoxy-adenosine synthase, EC:2.5.1.63 Two-domains; larger N-terminal domain has distant similarity to Rossmann-fold methyltransferases, smaller C-terminal domain is a β-barrel Bacterial invention; not in LUCA
5'deoxyadenosyl radical synthesis SAM radical enzymes TIM-like α/β barrel with additional inserted elements. May have distant sequence similarity to TIM barrel of corrinoid methyltransferase (see text) Probably in LUCA
SAM decarboxylation SAM decarboxylase, EC:4.1.1.50 α/β/β/α sandwich in eukaryotes, apparently produced by duplication of a half-unit; stand-alone half-units exist in many bacteria and archaea Probably in LUCA
De novo SAM synthesis Methionine adenosyl transferase, EC:2.5.1.6 Unique fold: repeat of 3 β-α-β-β-α-β units Probably in LUCA
Regulatory binding of SAM Methionine repressor All-α SAM-binding domain is derived from generic small molecule-binding domain Bacterial innovation
CBS domain Mostly-β SAM-binding domain is derived from generic small molecule-binding domain Not in LUCA
Transcription factor mtTFB Rossmann-fold methyltransferase family member that has lost catalytic activity
1 An additional reaction in this class is synthesis of acp3U, a modified base in some tRNAs and rRNAs. This amino alkyl transfer requires SAM, but responsible protein has not been identified.
Table 2 Repertoire of SAM-binding proteins in the Last Universal Common Ancestor
COG Phyletic pattern parsimony1 Tree topology2 Protein product name and function Fold class
SAM-binding proteins in LUCA – best-supported COGs
COG0030 1/0 + KsgA, Dimethyladenosine transferase (rRNA methylation) Rossmann fold
COG0144 1/0 + Sun, tRNA and rRNA cytosine-C5-methylases Rossmann fold
COG0192 0/1 3 + 3 MetK, S-adenosylmethionine synthetase (releases SAM as product) SAM synthase
COG0421 1/0 + SpeE, Spermidine synthase Rossmann fold
COG0500 1/0 + SmtA, SAM-dependent methyltransferases Rossmann fold
COG0621 1/0 + MiaB, 2-methylthioadenine synthetase Rossmann fold
COG1180 0/1 + PflA, Pyruvate-formate lyase-activating enzyme SAM radical (TIM barrel)
COG1586 0/1 3 + 3 SpeD, S-adenosylmethionine decarboxylase SAM decarboxylase
COG2890 1/0 + PrmC (HemK), Methylase of polypeptide chain release factors Rossmann fold
SAM-binding proteins in LUCA – tentatively supported COGs
COG0007 1/0 ? CysG, Uroporphyrinogen-III methylase SAM radical (TIM barrel)
COG0220 1/0 ? Predicted S-adenosylmethionine-dependent methyltransferase Rossmann fold
COG0293 0/1 ? FtsJ, 23S rRNA methylase Rossmann fold
COG0320 0/1 ? LipA, Lipoate synthase SAM radical (TIM barrel)
COG0502 0/1 ? BioB, Biotin synthase and related enzymes SAM radical (TIM barrel)
COG0565 0/1 ? LasT, rRNA methylases SPOUT
COG0685 1/0 ? MetF, 5,10-methylenetetrahydrofolate reductase (binds SAM as inhibitor) TIM barrel
COG1189 0/1 ? Predicted rRNA methylase Rossmann fold
COG2226 1/0 ? UbiE, Methylase involved in ubiquinone/menaquinone biosynthesis Rossmann fold
1 Placement of COGs in LUCA based on weighted parsimony analysis of phyletic patterns [138]. 1/0 indicates that this gene has been tracked back to LUCA genome under all tested conditions; 0/1 indicates tentative placement of gene in LUCA under some but not all conditions.
2 Phylogenetic analysis of each COG. Plus sign indicates tree topology that is fully consistent with gene presence in LUCA, and the question mark indicates a tentative support for the LUCA hypothesis (see Materials and Methods)
3 The common ancestry and phylogeny of archaeal, bacterial, and eukaryotic homologs was established in this work (see text for details).
SAM-binding modules derived from generic ligand-binding domains
ACC synthase
ACC synthase (S-adenosyl-L-methionine methylthioadenosine lyase, EC: 4.4.1.14, KOG0256) catalyses the rate-limiting step in biosynthesis of plant hormone ethylene by the α,γ-elimination of methylthioadenosine from SAM to produce 1-aminocyclopropane-1-carboxylate (ACC) [76]. ACC synthases require pyridoxal phosphate (PLP) for activity, and are related in sequence and structure to a large, diverse group of PLP-dependent transferases. The shared catalytic domain of this fold is of α/β/α type, with mixed central β-sheet of 7 strands (order 3245671), where strand 7 is antiparallel to the rest (SCOP fold: 53382; Figure 1j). Several residues are essential for the substrate binding (reviewed by Jakubowicz; numbered as in PDB structure 1B8G): Glu47 is responsible for putative ionic interaction with SAM; Ala46 and Arg407 interact with carboxypropyl moiety of SAM; Arg150 interacts with ribose moiety; and Ser18, Tyr19, Phe20, and Pro146 form hydrophobic pocket for the adenine ring of SAM [77]. Invariant Tyr85 is involved in the substrate recognition, and interacts with active-site Lys273 from the adjacent subunit (Lys273 forms a covalent Schiff base with PLP cofactor) [78,79].
Structure of ACC synthase is similar to other PLP-dependent transferases, such as transaminating aminotransferases, β-eliminating lyases, and cystathionine synthase. The only other enzyme in this group that binds SAM is 7,8-Diaminopelargonic acid (DAPA) synthase, which utilizes SAM in a different way than ACC synthase, namely as amino group donor in aminotransferase reaction. Both ACC synthase and DAPA synthase have similar active site residues involved in PLP cofactor binding. The interactions between SAM and DAPA synthase have not been studied in detail.
It is likely that ACC synthase and DAPA synthase evolved from other aminotransferases with different, perhaps broad, specificity, by accumulating changes in the ligand-binding region that increased its specificity towards SAM [80]. The evolutionary heritage of ACC synthase is manifest in the retained ability of the enzyme to catalyze slow transamination of substrates such as alanine [81]. Structural similarity between ACC-, DAPA-synthase and some other PLP-dependent enzymes (i.e. cystine C-S lyase – PDB structure 1ELQ; cystathionine β-lyase – PDB structure 1CL2) indicate that SAM binding in this case may have originated from ancestor with PLP-dependent binding of amino group of various sulfur containing amino acids or amino acid derivatives.
AHL synthase
Synthesis and detection of acyl-homoserine lactones (AHLs) enables many gram-negative bacteria to engage in quorum/diffusion sensing, an intercellular signaling mechanism that activates differentiation towards virulence and biofilm formation [82-84].
The AHL synthases (COG3916) catalyze acylation and lactonization of SAM, where the acyl group is provided by acylated acyl carrier protein (acyl-ACP) [85,86]. AHL synthases (SCOP family: 75508; Figure 1p) has acyl-CoA N-acyltransferase ("GNAT-like") fold: α/β/α sandwich with highly twisted β-sheet (SCOP fold: 55728). The conserved N-terminal residues: Arg23, Phe27, and Trp33 (numbered as in PDB structure 1RO5) form putative SAM binding pocket and undergo a dramatic conformational rearrangement upon acyl-ACP binding. This conformational change brings conserved residues of the putative SAM binding site in close proximity to the catalytic site [87]. Position of conserved β-bulge formed by Ser103 and Arg104 in β4 (numbered as in PDB structure 1RO5) distinguishes SAM-binding from other proteins with acyl-CoA-N-acyltransferase fold [86,87]. There is no detectable sequence or structural similarity between AHL synthases and other known SAM binding proteins, indicating independent origin of SAM binding in this fold. As with ACC synthase and DAPA synthase, the most likely mechanism of adaptation was by selecting relatively small changes in a generic ligand-binding region that increased relative affinity to SAM.
Met repressor
The E. coli MetJ repressor (Figure 1m; SCOP family: 100972; COG3060) uses SAM as a co-repressor to regulate the production of methionine. MetJ is a homodimeric, DNA-binding protein with ribbon-helix-helix fold. Co-repressor (SAM) binds to each monomer of the protein dimer at sites that lie on the opposite side of the protein from the DNA-binding motif. Binding of co-repressor affects DNA affinity, but apparently not specificity of MetJ [88-91]. Affinity of MetJ DNA binding is affected primarily by the positive charge associated with the ternary sulfur atom in co-repressor (SAM), which creates a region of positive electrostatic potential on the DNA binding surface overlapping the adjacent phosphodiester backbone in the region of the operator [92-94]. The SAM's adenine ring inserts itself deeply inside a hydrophobic pocket, consisted of side chains from both monomers. The positively charged sulfur of the SAM is greatly attracted by the net negatively charged C-terminal end of the β-helix, hence docking the SAM molecule in place. Electrostatic properties of SAM and its ability to serve as a regulatory feedback molecule in the common metabolic pathway of methionine synthesis probably played an important role in the emergence of this unique mode of SAM binding by MetJ.
MetJ is the only known SAM-binding representative of evolutionarily ancient ribbon-helix-helix (RHH) class of DNA-binding proteins [92,95]. Evolution of SAM cofactor binding in this protein was feasible because SAM adenosyl moiety fit into the cleft formed by both monomers, and its sulfonium center conformation was able to adapt to non-catalytic electrostatic interactions with MetJ repressor.
CBS
CBS-domains (COG0517) are widely distributed in all divisions of Life, in the form of fusions with various unrelated proteins, where they usually form tandem pairs. Binding of the adenosyl-containing molecules, such as ATP, AMP, and SAM by CBS-domains is important for their function as energy or redox status-sensing modules [96-98]. Some CBS-domains also binds single stranded nucleic acids [99].
In general, a tandem of CBS domains (encoded by ~120 aa) folds into one domain with a β-sandwich and 4 α-helices extending from one edge (Figure 1o). CBS domains within each pair are asymmetric. CBS-domain is common in multidomain proteins (i.e.: 15 in Bacteria and 9 in Archaea [100]) and is probably derived from generic small molecule-binding domain. The mode of SAM binding to CBS domains remains unknown.
Rare folds
Porphyrin C-methyltransferase
Porphyrin C-methyltransferases are a family of proteins involved in biosynthesis of tetrapyrroles, which are used in chelation of metal ion. Representative structures of Cbif and CysG with bound reaction product (SAH) are known. They have the same fold, which consists of two dissimilar α/β domains. Domain 1 has parallel sheet of 5 strands (order 32415) sandwiched between 3 α-helices; domain 2 has mixed sheet of 5 strands (also sandwiched between 3 α-helices), order 12534; strands 4 & 5 are antiparallel to the rest; (Figure 1d) [101,102]. The reaction product is bound in a large pocket between the N and C-terminal domains. Binding pocket contains conserved glycine-rich loop GAGPGD, similarly to Rossmann-fold methyltransferases, but in crystal structures of porphyrin C-methyltransferases (i.e. PDB structure 1CBF, 1PJQ, and 1S4D) glycine-rich loop does not contact methionine moiety of SAH, at least in the absence of the precorrin substrate. Instead, Pro30 from the glycine-rich loop (numbered as in PDB structure 1CBF) forms hydrogen bond with adenosyl moiety. Other conserved residues participate in (SAH) binding: adenosyl moiety hydrogen-bond with conserved Ala213, ribosyl moiety hydrogen bound with conserved hydrophobic Leu184 and Ala241; carboxyl group bind Asp103 from conserved Gly-Asp-Pro motif and also Tyr131; the amino group hydrogen-bond with conserved hydrophobic Met106. Near the entrance of SAM binding cleft there is conserved aromatic residue (Tyr107) positioned next to the sulfur of SAH.
Despite structural similarity to Rossmann-fold methyltransferases and local sequence similarity in the glycine-rich loop, porphyrin C-methyltransferases have distinct mode of SAM binding. The evolutionary relationships between porphyrin methyltransferases and other SAM-binding (or indeed any other) proteins remain unclear.
Met synthase activation domain
B12-dependent methionine synthases are large multidomain proteins, which catalyze a multistep reaction of the transfer of a methyl group from 5-methyltetrahydrofolate to homocysteine. The C-terminal domain of Met synthase is involved in reactivation of spontaneously oxidized coenzyme B12, and therefore is required for the catalysis. This activation domain (pfam02965 and C-terminal part of COG1410) has an unusual fold: β-α(2)-β(2)-α(2)-β-α-β; antiparallel β-sheet; order 12354 (Figure 1e; SCOP fold: 56506). Large conformational changes are required for reactivation reaction to occur [103-105].
This domain binds SAM in a shallow groove along the edges of the β-strands. There are several conserved residues involved in SAM binding (numbered as in PDB structure 1MSK): Asp946 and Glu1101 bind carboxypropyl moiety; Arg1134 and Tyr1130 (later via water mediated interactions) bind ribosyl moiety; Tyr1139 and Tyr1189 participate in stacking interactions with adenosyl moiety; Tyr1190 main chain (or Phe in other known sequences) hydrogen bond with adenosyl moiety of SAM. This unusual domain is found in bacterial and metazoan proteins. Interestingly, Met synthases of thermophilic bacteria lack the SAM-binding/activation domain altogether. Presumably, the lack of the transferase domain activity is compensated for by the methyltransferase protein also involved in the same reaction [106]. This particular adaptation in thermophiles seems to be partly due to requirement for greater thermal stability of the enzyme – making large conformational changes required for reactivation less favorable.
Divergence and convergence in evolution of SAM binding
The known SAM-binding proteins belong to 15 distinct superfamilies/folds, which comprise at least 20 more sequence families (Figure 1 and Table 4), that is: 1 – Rossmann fold (methyltransferase, methylene transferase, nicotianamine synthase, spermidine synthase, spermine synthase, and acalcynomycin-10-hydroxylase), 2 – Met synthase activating domain, 3 – porphyrin C-methyltransferase, 4 – SPOUT methyltransferase, 5 – SET domain methyltransferase, 6 – isoprenylcysteine carboxyl methyltransferase, 7 – fluorinating enzyme, 8 – tRNA ribosyltransferase-isomerase (QueA), 9 – SAM decarboxylase, 10 – SAM synthetase, 11 – ACC synthase and DAPA synthase, 12 – N-acyl-homoserine lactone synthase, 13 – Met repressor, 14 – CBS domain, 15 – SAM-dependent radical.
These folds belong to all four large structural classes [107], though there is a distinct excess of α-β and especially α/β folds.
Comparison of fold classes and molecular functions reveals a broad picture of complex interplay between sequence divergence and functional convergence in the evolution of SAM-binding proteins and SAM-dependent molecular functions. On the whole, the assemblage of SAM-binding proteins is extremely heterogeneous. There are huge, apparently monophyletic, superfamilies, which in turn belong to even larger superfolds (Class I methyltransferases from a Rossmann-like superfold is an example), and there are small families with unique folds. There are molecular functions confined to just one superfamily, as in the case of SAM radical formation, which thus far is known to be performed only by enzymes belonging to large and diverse, yet apparently monophyletic, TIM-barrel-like SAM-radical fold. On the other hand, there is SAM-dependent methylation, performed by at least five classes of enzymes, which represent two completely different fold classes and may have been "invented" three or four times (ref. [12] and this study).
Duplication of protein domains is relatively common in both prokaryotes and eukaryotes, for example at least 58% of the domains in Mycoplasma [108] and 98% of the domains in humans [109-111] are duplicates. Several SAM-binding proteins appear to have evolved by ancient domain duplications. Examples include SET-domain methyltransferase [71], SAM synthase, where triplication of a basic wedge-shaped module is likely, and eukaryal SAM decarboxylase (this study). Here again, there is no strict correlation between that mechanism of protein emergence and its further evolution trajectory: SAM synthase and SAM decarboxylase have most likely persisted in evolution as single- or low-copy, vertically transmitted genes since the Last Universal Common Ancestor, whereas SET methyltransferases appear to be an eukaryotic innovation that experienced lineage-specific expansions, significant diversification of substrate specificity, and occasional horizontal transfer to prokaryotes.
Examination of the largest, most diverse set of SAM-dependent enzymes unified by common biochemical function, namely, SAM-dependent methyltransferases, shows more of the same trend. All methyltransferases belong to the same EC class (EC: 2.1.1.-), but they comprise five structural families (reviewed in ref. [12]), which appear to lack clearly discernible common ancestor (with the possible exception of Class I and SPOUT methyltransferase, see above). Three classes, I, IV, and V, are large, found across broad groups of genomes, and, at least in the cases of classes I and V, in multiple genomic contexts, including various protein fusions. In contrast, class II is restricted to just one specialized enzymatic system, methionine synthase, with very specific domain composition. Class I displays a huge variety of substrate specificities, whereas the substrates of other classes are much more narrowly defined (tetrapyrroles for class III, mostly rRNAs and tRNAs for class IV, and short list of proteins for class V). Interestingly, each of these classes of substrates are also targeted by class I enzymes. For example, although most of precorrin methyltransferases are class III enzymes, precorrin-C6 methyltransferase CbiT is a typical Class-I Rossmannoid [112]. Similarly, though most histone lysine N-methyltransferases are class V (SET-domain proteins), the Dot1 histone H3-Lys79 N-methyltransferase belongs to Class I [113]. In at least one case, exactly the same base in tRNA is methylated by Class I methyltransferase in bacteria and by class IV enzyme in archaea [114], suggesting either parallel evolution of different molecular solutions for the same task, or functional takeover by an unrelated gene.
Phyletic distribution and phylogeny of SAM-binding proteins suggests multiple roles for SAM in the Last Universal Common Ancestor
SAM-binding proteins perform an unprecedented variety of chemical reactions, and belong to about 22 distinct sequence and structural families of proteins. It has been noted that some of these groups of proteins (most notably, Rossmann-fold methyltransferases) are extremely ancient and were more than likely represented by multiple paralogs in the life forms predating the divergence of bacteria, archaea, and eukarya [5]. On the other hand, some of the SAM-binding proteins appear to have been invented later in evolution. Such scenarios have been documented, for example, for SET-domain methyltransferase family, which is thought to have emerged in early eukaryotes and then passed to a few bacteria by lateral gene transfer, while greatly expanding in size in multicellular eukaryotes [71], and are also likely for Met repressor, which is a member of the large ligand-binding family essentially confined to bacteria [115]. A complex picture of gains, losses, and lineage-specific expansions of genes coding for SAM-binding proteins is illustrated in Table 3.
Table 3 Counts of each family in selected genomes
Protein family name and number of genes (proteins) per genome Additional information about counts of each family in selected genomes.
HUMAN ARATH YEAST PSEAE MYCTU SULSO HELPY
1. Rossmann fold MTase
49 (64) 70 (81) 30 55 37 31 26 Rossmann fold methyltransferase (MTase) makes the list of 10 most commonly used sequence and structure families [15]. In analyzed genomes, 4.6% of Rossmann fold MTase does not have motif-I (Gly-rich) and may lack MTase activity. Fold: Rossmann-fold.
2. Met synthase activating domain
1 0 0 1 1 0 0 Present in many bacteria and metazoan. Fold: Met synthase activation domain-like.
3. MTase class III
1 3 (4) 2 8 6 7 1 Human genome encodes protein similar to diphthine synthase but not other Class-III MTases (i.e. from tetrapyrrole biosynthesis pathway) identified in analyzed genomes. Fold: Tetrapyrrole-methyltransferase.
4. SPOUT MTase
3 5 2 4 6 1 2 Every analyzed here genome has at least one putative SPOUT rRNA MTase, in archaea the same base of tRNA is methylated by Rossmann-fold instead of SPOUT MTase. Fold: α/β knot.
5. SET domain MTase
25 (34) 34 (41) 5 0 0 0 0 Present in all Eukaryota with sporadic lateral transfer to bacteria and archaea (but not identified in prokaryotic genomes analyzed here). In S. cerevisiae all selected sequences similar to SET-domain MTases are conserved within smart00317 domain (in contrast to SET-domain proteins other then MTases). Fold: β-clip.
6. methylene transferase
0 4 0 1 10 0 1 This enzyme performs cyclopropane fatty acid synthesis and is important for Mycobacterium survival. Fold: Rossmann-fold.
7. Nicotianamine synthase
0 4 0 0 0 0 0 This is plant enzyme has similarity to proteins with unknown function infrequently distributed in bacteria, archaea and fungi (discussed earlier). Fold: Rossmann-fold (predicted).
8. spermidine synthase &9. spermine synthase
1 3 2 3 1 1 1 The mammalian enzyme is highly specific but the bacterial enzyme can use other acceptors then SAM and can synthesize spermine. Spermidine synthase but not spermine synthase is essential for survival of Arabidopsis and S. cerevisiae [142-144]. Fold: Rossmann-fold.
10. aclacynomycin-10-hydroxylase
0 0 0 0 0 0 0 Was found in Streptomyces purpurascens only. Fold: Rossmann-fold.
11. isoprenylcysteine O-MTase
2 2 1 0 3 0 0 ICMT enzymes are present in all eukaryotic organisms [145]. Fold: unknown (predicted: all-α).
12. fluorinating enzyme
0 0 0 0 0 0 0 Gene encoding fluorinating enzyme was identified in Streptomyces cattleya but not yet in plants synthesizing fluorinated metabolites. Fold: Rossmann-like and β-barrel.
13. QueA
0 0 0 1 0 0 1 QueA (tRNA ribosyltransferase – isomerase) homologs are found only in bacteria. Fold: QueA-like (TIM-barrel and β-barrel).
14. SAM decarboxylase
1 4 1 2 ? 2 ? SAM decarboxylase activity has been purified from all three domains of Life. However M. tuberculosis and H. pylori lack clearly identifiable homologue of this enzyme. Fold: SAM-decarboxylase.
15. SAM synthetase
2 4 (5) 2 1 1 1 1 SAM synthases from bacteria and eukaryotes are closely related at the sequence level and have very similar structures [73]. SAM synthase is involved in development and abiotic stress tolerance in plants and have complex expression pattern [146, 147]. Fold: SAM-synthetase.
16. ACC synthase
2 12 0 0 0 0 0 ACC is the precursor of important plant hormone – ethylene. Human homologs have different function. Fold: PLP-dependent transferases.
17. N-acyl-homoserine lactone synthase
0 0 0 2 0 0 0 Found in some bacteria only. Fold: GNAT.
18. Met repressor
0 0 0 0 0 0 0 Found in some enterobacteria and gamma proteobacteria only. Fold: ribbon-helix-helix.
19. CBS domain
5 10 5 7 4 9 3 Numbers of SAM-binding CBS domains presented here are approximate because of strong similarity to CBS domains binding other adenosine derivatives. Fold: CBS-domain.
20. DAPA synthase
0 1 1 6 2 0 1 Biotin biosynthesis is unique to plants, some fungi and most bacteria. Fold: PLP-dependent transferases.
21. SAM-dependent radical
9 (13) 13 (14) 5 18 11 24 11 Human lacks SAM-dependent radical enzyme from biotin and thiamine biosynthetic pathways. Those enzymes generate highly oxidizing 5'-deoxyadenosyl radical in an anaerobic reducing environment, and utilize this radical as catalytic and stoichiometric oxidant in many different enzymatic reactions [148]. Those enzymes are essential for anaerobic growth. Fold:TIM-barrel.
Number of genes and proteins encoded by selected genomes were counted for each SAM-binding protein family. SULSO – Sulfolobus solfataricus (Archaea), PSEAE – Pseudomonas aeruginosa (Gamma proteobacteria), HELPY – Helicobacter pylori J99 (Proteobacteria), MYCTU – Mycobacterium tuberculosis (Actinobacteria), YEAST – Saccharomyces cerevisiae (Eukaryota), ARATH – Arabidopsis thaliana (Eukaryota), HUMAN – Homo sapiens (Eukaryota).
We performed a more detailed examination of evolutionary trajectories of SAM-binding proteins, using information from the NCBI COG database. Each COG is a set of orthologous genes in completely sequenced genomes, along with lineage-specific paralogs. A COG is characterized by phyletic pattern, which is the set of genomes that has at least one member of this COG, and by sequence-based phylogenetic tree of COG members [116]. This information can be used, in conjunction with the consensus phylogenetic tree of the completely sequence genomes, to infer the presence or absence of a COG in an ancestral life form. We used a relatively conservative estimate, allowing for occasional horizontal gene transfer and demanding a complete agreement between species' tree, gene family tree, and phyletic pattern (see Methods for more detail). Under these conditions, the SAM-binding complement of LUCA proteins consists of 9 ancestral COGs (Table 2).
Although 6 of 9 COGs in LUCA represent proteins with Rossmann-like fold, the remaining three folds are all different. It should be noted that the set of 9 COGs is almost certainly an underestimation of the SAM-binding proteome in LUCA. Evolutionary model allowing more frequent horizontal gene transfer and/or non-orthologous gene displacements, as well as slight disagreement between different lines of evidence, will increase the list of COGs that can be placed in LUCA genome. The specificity of these additional enzymes varies, although the modification of translation apparatus continues to figure prominently in the increased set; the diversity of folds, however, appears to increase only slightly, mostly due to addition of the SPOUT fold.
An already-diverse group of Rossmann-like SAM-dependent transferases, a variation of TIM-barrel, and two unique αββα architectures thus appear to represent the best-supported ancestral set of SAM-binding proteins. Although less diverse than the present-day variety of SAM-binding proteins, this set is far from simple. Interestingly, it consists of proteins with α/β architecture and is depleted of all-α and all-β proteins, as seems to be the case for other categories of ancestral enzymes [117] and perhaps non-enzymatic proteins too [118].
The substrates and molecular functions of many of these enzymes are hard to ascertain. Ancient enzymes may have had broader specificities than their present-day descendants [119], but several pathways nevertheless emerge from the analysis of the SAM-binding proteome of the LUCA (Table 2). The common ancestor of bacteria, archaea and eukarya appears to have been able to synthesize SAM de novo, from ATP and methionine; to use it for methylation of RNA bases and, probably, proteins such as translation factors; to decarboxylate SAM; and to synthesize polyamines with the aid of dcSAM. In addition, LUCA had the capacity for generating SAM radicals.
Slightly less restrictive evolutionary model enlarges the set of SAM-binding proteins in LUCA, mostly by increasing the number of paralogs in Class I MT and SAM-radical families, but also by supplementing the set with SPOUT methyltransferases. Further sequence and structure comparison may provide for more detailed understanding of these ancestors, perhaps even to the point of reconstructing the ancestral sequences and studying the ancient SAM-binding proteins in the laboratory.
Conclusion
There are 15 distinct superfamilies of SAM-binding proteins, at least 5 of which may have been represented in the last common ancestor.
Methods
Analysis of multiple sequences in the batch mode was handled using the SEALS package [120].
Iterative database searches with position-specific weight matrices (PSSMs) were performed using the PSI-BLAST program, with the expectation value for inclusion into the PSSM (-h parameter) set at 0.01, unless otherwise indicated [121]. Additional profile searches were carried out using hidden Markov models generated from alignments of protein domains using the hmmsearch program of the HMMER2 package [122].
Multiple alignments of protein sequences were constructed in an iterative fashion, alternating between sequence and structure alignments [123]. Structural alignments of representative structures from the SCOP families [107] were produced using CE-MC [124] and DALI [125]. The muscle program [126] was used to refine all alignments.
The 3-dimensional structures of proteins were manipulated using the Rasmol program and ribbon diagrams were drawn using the PyMOL program [127].
The Metaserver approach [56] was used to interrogate the network of programs that perform secondary structure prediction and tertiary fold-recognition of proteins by a variety of probabilistic matching and energy calculations (here we used predictions from: PSIPRED [128], PROFsec [129], SAM-T02 [130], FFAS03 [131], 3DPSSM [132], BIOINBGU [133], ORFeus [134], FUGUE [135], and Pcons server [136]).
Topology diagrams were created by using TOPS [137].
The inference of the ancestral presences/absences of the COGs has been done by Mirkin et al. [138]; their model was used in this study, with modifications described in Mushegian [118].
List of abbreviations used
SAM: S-adenosylmethionine
PDB: Protein Data Bank
SCOP: Structural Classification of Proteins
Authors' contributions
ARM: initiated and supervised the study, wrote parts of the manuscript and performed the analysis of phyletic distribution and phylogeny. PZK: participated in the design of the study, wrote parts of the manuscript and performed the large scale sequence and structure comparison. Both authors contributed to the writing of the manuscript, interpreted all data, and approved the final version.
Acknowledgements
We thank G. Glazko for help with phyletic patterns analysis, M. Coleman and D. Thomasset for programming help. Additionally, the authors thank J.M. Bujnicki who as a non-anonymous reviewer provided insightful comments and suggestions. This work was supported by Stowers Institute for Medical Research.
==== Refs
Waddell TG Eilders LL Patel BP Sims M Prebiotic methylation and the evolution of methyl transfer reactions in living cells. Orig Life Evol Biosph 2000 30 539 548 11196574 10.1023/A:1026523222285
Thomas DJ Waters SB Styblo M Elucidating the pathway for arsenic methylation. Toxicol Appl Pharmacol 2004 198 319 326 15276411 10.1016/j.taap.2003.10.020
Wuosmaa AM Hager LP Methyl chloride transferase: a carbocation route for biosynthesis of halometabolites. Science 1990 249 160 162 2371563
Saxena D Aouad S Attieh J Saini HS Biochemical characterization of chloromethane emission from the wood-rotting fungus Phellinus pomaceus Appl Environ Microbiol 1998 64 2831 2835 9687437
Anantharaman V Koonin EV Aravind L Comparative genomics and evolution of proteins involved in RNA metabolism. Nucleic Acids Res 2002 30 1427 1464 11917006 10.1093/nar/30.7.1427
Hopper AK Phizicky EM tRNA transfers to the limelight. Genes Dev 2003 17 162 180 12533506 10.1101/gad.1049103
Kouzarides T Histone methylation in transcriptional control. Curr Opin Genet Dev 2002 12 198 209 11893494 10.1016/S0959-437X(02)00287-3
Aravind L Koonin EV Novel predicted RNA-binding domains associated with the translation machinery. J Mol Evol 1999 48 291 302 10093218
Romano JD Michaelis S Topological and mutational analysis of Saccharomyces cerevisiae Ste14p, founding member of the isoprenylcysteine carboxyl methyltransferase family Mol Biol Cell 2001 12 1957 1971 11451995
Anderson JL Frase H Michaelis S Hrycyna CA Purification, functional reconstitution, and characterization of the Saccharomyces cerevisiae isoprenylcysteine carboxylmethyltransferase Ste14p J Biol Chem 2005 280 7336 7345 15611058 10.1074/jbc.M410292200
Peterson YK Winter-Vann AM Casey PJ Icmt AfCS-Nature Molecule Pages 2005 doi:10.1038/mp.a001154.01
Schubert HL Blumenthal RM Cheng X Many paths to methyltransfer: a chronicle of convergence. Trends Biochem Sci 2003 28 329 335 12826405 10.1016/S0968-0004(03)00090-2
Fontecave M Atta M Mulliez E S-adenosylmethionine: nothing goes to waste. Trends Biochem Sci 2004 29 243 249 15130560 10.1016/j.tibs.2004.03.007
Lesk AM NAD-binding domains of dehydrogenases. Curr Opin Struct Biol 1995 5 775 783 8749365 10.1016/0959-440X(95)80010-7
Wolf YI Brenner SE Bash PA Koonin EV Distribution of protein folds in the three superkingdoms of life. Genome Res 1999 9 17 26 9927481
Martin JL McMillan FM SAM (dependent) I AM: the S-adenosylmethionine-dependent methyltransferase fold. Curr Opin Struct Biol 2002 12 783 793 12504684 10.1016/S0959-440X(02)00391-3
Bujnicki JM Comparison of protein structures reveals monophyletic origin of the AdoMet-dependent methyltransferase family and mechanistic convergence rather than recent differentiation of N4-cytosine and N6-adenine DNA methylation. In Silico Biol 1999 1 175 182 11479932
Posfai J Bhagwat AS Posfai G Roberts RJ Predictive motifs derived from cytosine methyltransferases. Nucleic Acids Res 1989 17 2421 2435 2717398
Fauman EB Blumenthal RM Cheng X Cheng X and Blumenthal RM Structure and evolution of AdoMet-dependent MTases. In S-Adenosylmethionine-dependent methyltransferases: structures and functions 1999 Singapore, World Scientific Inc. 1 38
Bujnicki JM Sequence permutations in the molecular evolution of DNA methyltransferases. BMC Evol Biol 2002 2 3 11914127 10.1186/1471-2148-2-3
Sankpal UT Rao DN Mutational analysis of conserved residues in HhaI DNA methyltransferase Nucleic Acids Res 2002 30 2628 2638 12060679 10.1093/nar/gkf380
Cheng X Structure and function of DNA methyltransferases. Annu Rev Biophys Biomol Struct 1995 24 293 318 7663118 10.1146/annurev.bb.24.060195.001453
Kumar S Cheng X Klimasauskas S Mi S Posfai J Roberts RJ Wilson GG The DNA (cytosine-5) methyltransferases Nucleic Acids Res 1994 22 1 10 8127644
Bugl H Fauman EB Staker BL Zheng F Kushner SR Saper MA Bardwell JC Jakob U RNA methylation under heat shock control. Mol Cell 2000 6 349 360 10983982 10.1016/S1097-2765(00)00035-6
Feder M Pas J Wyrwicz LS Bujnicki JM Molecular phylogenetics of the RrmJ/fibrillarin superfamily of ribose 2'-O-methyltransferases Gene 2003 302 129 138 12527203 10.1016/S0378-1119(02)01097-1
Liu J Mushegian A Three monophyletic superfamilies account for the majority of the known glycosyltransferases. Protein Sci 2003 12 1418 1431 12824488 10.1110/ps.0302103
Cheek S Zhang H Grishin NV Sequence and structure classification of kinases. J Mol Biol 2002 320 855 881 12095261 10.1016/S0022-2836(02)00538-7
Unligil UM Zhou S Yuwaraj S Sarkar M Schachter H Rini JM X-ray crystal structure of rabbit N-acetylglucosaminyltransferase I: catalytic mechanism and a new protein superfamily. EMBO J 2000 19 5269 5280 11032794 10.1093/emboj/19.20.5269
Qasba PK Ramakrishnan B Boeggeman E Substrate-induced conformational changes in glycosyltransferases Trends Biochem Sci 2005 30 53 62 15653326 10.1016/j.tibs.2004.11.005
Nes WD Marshall JA Jia Z Jaradat TT Song Z Jayasimha P Active site mapping and substrate channeling in the sterol methyltransferase pathway. J Biol Chem 2002 277 42549 42556 12196515 10.1074/jbc.M204223200
Zubieta C He XZ Dixon RA Noel JP Structures of two natural product methyltransferases reveal the basis for substrate specificity in plant O-methyltransferases. Nat Struct Biol 2001 8 271 279 11224575 10.1038/85029
Huang CC Smith CV Glickman MS Jacobs JWR Sacchettini JC Crystal structures of mycolic acid cyclopropane synthases from Mycobacterium tuberculosis. J Biol Chem 2002 277 11559 11569 11756461 10.1074/jbc.M111698200
Takahashi M Terada Y Nakai I Nakanishi H Yoshimura E Mori S Nishizawa NK Role of nicotianamine in the intracellular delivery of metals and plant reproductive development. Plant Cell 2003 15 1263 1280 12782722 10.1105/tpc.010256
Dela Vega AL Delcour AH Polyamines decrease Escherichia coli outer membrane permeability J Bacteriol 1996 178 3715 3721 8682771
Nikaido H Molecular basis of bacterial outer membrane permeability revisited Microbiol Mol Biol Rev 2003 67 593 656 14665678 10.1128/MMBR.67.4.593-656.2003
Korolev S Ikeguchi Y Skarina T Beasley S Arrowsmith C Edwards A Joachimiak A Pegg AE Savchenko A The crystal structure of spermidine synthase with a multisubstrate adduct inhibitor. Nat Struct Biol 2002 9 27 31 11731804 10.1038/nsb737
Jansson A Koskiniemi H Erola A Wang J Mantsala P Schneider G Niemi J Aclacinomycin 10-hydroxylase is a novel substrate-assisted hydroxylase requiring S-adenosyl-L-methionine as cofactor J Biol Chem 2005 280 3636 3644 15548527 10.1074/jbc.M412095200
McCulloch V Shadel GS Human mitochondrial transcription factor B1 interacts with the C-terminal activation region of h-mtTFA and stimulates transcription independently of its RNA methyltransferase activity. Mol Cell Biol 2003 23 5816 5824 12897151 10.1128/MCB.23.16.5816-5824.2003
Bujnicki JM In silico analysis of the tRNA:m1A58 methyltransferase family: homology-based fold prediction and identification of new members from Eubacteria and Archaea FEBS Lett 2001 507 123 127 11684083 10.1016/S0014-5793(01)02962-3
Anderson J Phan L Hinnebusch AG The Gcd10p/Gcd14p complex is the essential two-subunit tRNA(1-methyladenosine) methyltransferase of Saccharomyces cerevisiae Proc Natl Acad Sci U S A 2000 97 5173 5178 10779558 10.1073/pnas.090102597
Bujnicki JM Feder M Radlinska M Blumenthal RM Structure prediction and phylogenetic analysis of a functionally diverse family of proteins homologous to the MT-A70 subunit of the human mRNA:m(6)A methyltransferase J Mol Evol 2002 55 431 444 12355263 10.1007/s00239-002-2339-8
Bujnicki JM Rychlewski L RNA:(guanine-N2) methyltransferases RsmC/RsmD and their homologs revisited-bioinformatic analysis and prediction of the active site based on the uncharacterized Mj0882 protein structure. BMC Bioinformatics 2002 3 10 11929612 10.1186/1471-2105-3-10
Chen ZX Mann JR Hsieh CL Riggs AD Chedin F Physical and functional interactions between the human DNMT3L protein and members of the de novo methyltransferase family J Cell Biochem 2005 95 902 917 15861382 10.1002/jcb.20447
Gowher H Liebert K Hermann A Xu G Jeltsch A Mechanism of stimulation of catalytic activity of Dnmt3A and Dnmt3B DNA-(cytosine-C5)-methyltransferases by Dnmt3L J Biol Chem 2005 280 13341 13348 15671018 10.1074/jbc.M413412200
Bourc'his D Bestor TH Meiotic catastrophe and retrotransposon reactivation in male germ cells lacking Dnmt3L Nature 2004 431 96 99 15318244 10.1038/nature02886
Dong C Huang F Deng H Schaffrath C Spencer JB O'Hagan D Naismith JH Crystal structure and mechanism of a bacterial fluorinating enzyme. Nature 2004 427 561 565 14765200 10.1038/nature02280
Anantharaman V Koonin EV Aravind L SPOUT: a class of methyltransferases that includes spoU and trmD RNA methylase superfamilies, and novel superfamilies of predicted prokaryotic RNA methylases. J Mol Microbiol Biotechnol 2002 4 71 75 11763972
Jackman JE Montange RK Malik HS Phizicky EM Identification of the yeast gene encoding the tRNA m1G methyltransferase responsible for modification at position 9 RNA 2003 9 574 585 12702816 10.1261/rna.5070303
Ginalski K von Grotthuss M Grishin NV Rychlewski L Detecting distant homology with Meta-BASIC Nucleic Acids Res 2004 32 W576 81 15215454
Ahn HJ Kim HW Yoon HJ Lee BI Suh SW Yang JK Crystal structure of tRNA(m1G37)methyltransferase: insights into tRNA recognition. EMBO J 2003 22 2593 2603 12773376 10.1093/emboj/cdg269
Gerstein M A structural census of genomes: comparing bacterial, eukaryotic, and archaeal genomes in terms of protein structure J Mol Biol 1997 274 562 576 9417935 10.1006/jmbi.1997.1412
Nagano N Orengo CA Thornton JM One fold with many functions: the evolutionary relationships between TIM barrel families based on their sequences, structures and functions. J Mol Biol 2002 321 741 765 12206759 10.1016/S0022-2836(02)00649-6
Sofia HJ Chen G Hetzler BG Reyes-Spindola JF Miller NE Radical SAM, a novel protein superfamily linking unresolved steps in familiar biosynthetic pathways with radical mechanisms: functional characterization using new analysis and information visualization methods. Nucleic Acids Res 2001 29 1097 1106 11222759 10.1093/nar/29.5.1097
Ollagnier-de Choudens S Sanakis Y Hewitson KS Roach P Munck E Fontecave M Reductive cleavage of S-adenosylmethionine by biotin synthase from Escherichia coli J Biol Chem 2002 277 13449 13454 11834738 10.1074/jbc.M111324200
Nicolet Y Drennan CL AdoMet radical proteins-from structure to evolution-alignment of divergent protein sequences reveals strong secondary structure element conservation Nucleic Acids Res 2004 32 4015 4025 15289575 10.1093/nar/gkh728
Ginalski K Elofsson A Fischer D Rychlewski L 3D-Jury: a simple approach to improve protein structure predictions Bioinformatics 2003 19 1015 1018 12761065 10.1093/bioinformatics/btg124
Marton LJ Pegg AE Polyamines as targets for therapeutic intervention Annu Rev Pharmacol Toxicol 1995 35 55 91 7598507 10.1146/annurev.pa.35.040195.000415
Gerner EW Meyskens FLJ Polyamines and cancer: old molecules, new understanding Nat Rev Cancer 2004 4 781 792 15510159 10.1038/nrc1454
Kim AD Graham DE Seeholzer SH Markham GD S-Adenosylmethionine decarboxylase from the archaeon Methanococcus jannaschii: identification of a novel family of pyruvoyl enzymes J Bacteriol 2000 182 6667 6672 11073910 10.1128/JB.182.23.6667-6672.2000
Sekowska A Coppee JY Le Caer JP Martin-Verstraete I Danchin A S-adenosylmethionine decarboxylase of Bacillus subtilis is closely related to archaebacterial counterparts Mol Microbiol 2000 36 1135 1147 10844697 10.1046/j.1365-2958.2000.01930.x
Ekstrom JL Mathews II Stanley BA Pegg AE Ealick SE The crystal structure of human S-adenosylmethionine decarboxylase at 2.25 A resolution reveals a novel fold Structure Fold Des 1999 7 583 595 10378277 10.1016/S0969-2126(99)80074-4
Toms AV Kinsland C McCloskey DE Pegg AE Ealick SE Evolutionary links as revealed by the structure of Thermotoga maritima S-adenosylmethionine decarboxylase J Biol Chem 2004 279 33837 33846 15150268 10.1074/jbc.M403369200
Lu ZJ Markham GD Catalytic properties of the archaeal S-adenosylmethionine decarboxylase from Methanococcus jannaschii J Biol Chem 2004 279 265 273 14573607 10.1074/jbc.M308793200
Bennett EM Ekstrom JL Pegg AE Ealick SE Monomeric S-adenosylmethionine decarboxylase from plants provides an alternative to putrescine stimulation Biochemistry 2002 41 14509 14517 12463749 10.1021/bi026710u
Trievel RC Beach BM Dirk LM Houtz RL Hurley JH Structure and catalytic mechanism of a SET domain protein methyltransferase. Cell 2002 111 91 103 12372303 10.1016/S0092-8674(02)01000-0
Yokoyama A Wang Z Wysocka J Sanyal M Aufiero DJ Kitabayashi I Herr W Cleary ML Leukemia proto-oncoprotein MLL forms a SET1-like histone methyltransferase complex with menin to regulate Hox gene expression Mol Cell Biol 2004 24 5639 5649 15199122 10.1128/MCB.24.13.5639-5649.2004
Kouskouti A Scheer E Staub A Tora L Talianidis I Gene-specific modulation of TAF10 function by SET9-mediated methylation Mol Cell 2004 14 175 182 15099517 10.1016/S1097-2765(04)00182-0
Chuikov S Kurash JK Wilson JR Xiao B Justin N Ivanov GS McKinney K Tempst P Prives C Gamblin SJ Barlev NA Reinberg D Regulation of p53 activity through lysine methylation. Nature 2004 432 353 360 15525938 10.1038/nature03117
Xiao B Jing C Wilson JR Walker PA Vasisht N Kelly G Howell S Taylor IA Blackburn GM Gamblin SJ Structure and catalytic mechanism of the human histone methyltransferase SET7/9. Nature 2003 421 652 656 12540855 10.1038/nature01378
Manzur KL Farooq A Zeng L Plotnikova O Koch AW Sachchidanand Zhou MM A dimeric viral SET domain methyltransferase specific to Lys27 of histone H3. Nat Struct Biol 2003 10 187 196 12567185 10.1038/nsb898
Aravind L Iyer LM Provenance of SET-domain histone methyltransferases through duplication of a simple structural unit Cell Cycle 2003 2 369 376 12851491
Iyer LM Aravind L The emergence of catalytic and structural diversity within the beta-clip fold Proteins 2004 55 977 991 15146494 10.1002/prot.20076
Sanchez-Perez GF Bautista JM Pajares MA Methionine adenosyltransferase as a useful molecular systematics tool revealed by phylogenetic and structural analyses J Mol Biol 2004 335 693 706 14687567 10.1016/j.jmb.2003.11.022
Komoto J Yamada T Takata Y Markham GD Takusagawa F Crystal structure of the S-adenosylmethionine synthetase ternary complex: a novel catalytic mechanism of S-adenosylmethionine synthesis from ATP and Met. Biochemistry 2004 43 1821 1831 14967023 10.1021/bi035611t
Graham DE Bock CL Schalk-Hihi C Lu ZJ Markham GD Identification of a highly diverged class of S-adenosylmethionine synthetases in the archaea J Biol Chem 2000 275 4055 4059 10660563 10.1074/jbc.275.6.4055
Huai Q Xia Y Chen Y Callahan B Li N Ke H Crystal structures of 1-aminocyclopropane-1-carboxylate (ACC) synthase in complex with aminoethoxyvinylglycine and pyridoxal-5'-phosphate provide new insight into catalytic mechanisms J Biol Chem 2001 276 38210 38216 11431475 10.1074/jbc.M008127200
Jakubowicz M Structure, catalytic activity and evolutionary relationships of 1-aminocyclopropane-1-carboxylate synthase, the key enzyme of ethylene synthesis in higher plants Acta Biochim Pol 2002 49 757 774 12422245
Capitani G Eliot AC Gut H Khomutov RM Kirsch JF Grutter MG Structure of 1-aminocyclopropane-1-carboxylate synthase in complex with an amino-oxy analogue of the substrate: implications for substrate binding Biochim Biophys Acta 2003 1647 55 60 12686108
Tsuchisaka A Theologis A Heterodimeric interactions among the 1-amino-cyclopropane-1-carboxylate synthase polypeptides encoded by the Arabidopsis gene family Proc Natl Acad Sci U S A 2004 101 2275 2280 14983000 10.1073/pnas.0308515101
Eliot AC Kirsch JF Avoiding the road less traveled: how the topology of enzyme-substrate complexes can dictate product selection Acc Chem Res 2003 36 757 765 14567709 10.1021/ar0202767
Feng L Geck MK Eliot AC Kirsch JF Aminotransferase activity and bioinformatic analysis of 1-aminocyclopropane-1-carboxylate synthase Biochemistry 2000 39 15242 15249 11106504 10.1021/bi002092a
Fuqua C Parsek MR Greenberg EP Regulation of gene expression by cell-to-cell communication: acyl-homoserine lactone quorum sensing Annu Rev Genet 2001 35 439 468 11700290 10.1146/annurev.genet.35.102401.090913
Redfield RJ Is quorum sensing a side effect of diffusion sensing? Trends Microbiol 2002 10 365 370 12160634 10.1016/S0966-842X(02)02400-9
Lerat E Moran NA The evolutionary history of quorum-sensing systems in bacteria Mol Biol Evol 2004 21 903 913 15014168 10.1093/molbev/msh097
Val DL Cronan JEJ In vivo evidence that S-adenosylmethionine and fatty acid synthesis intermediates are the substrates for the LuxI family of autoinducer synthases J Bacteriol 1998 180 2644 2651 9573148
Watson WT Minogue TD Val DL von Bodman SB Churchill ME Structural basis and specificity of acyl-homoserine lactone signal production in bacterial quorum sensing Mol Cell 2002 9 685 694 11931774 10.1016/S1097-2765(02)00480-X
Gould TA Schweizer HP Churchill ME Structure of the Pseudomonas aeruginosa acyl-homoserinelactone synthase LasI Mol Microbiol 2004 53 1135 1146 15306017 10.1111/j.1365-2958.2004.04211.x
He YY Stockley PG Gold L In vitro evolution of the DNA binding sites of Escherichia coli methionine repressor, MetJ J Mol Biol 1996 255 55 66 8568875 10.1006/jmbi.1996.0006
Saint-Girons I Belfaiza J Guillou Y Perrin D Guiso N Barzu O Cohen GN Interactions of the Escherichia coli methionine repressor with the metF operator and with its corepressor, S-adenosylmethionine J Biol Chem 1986 261 10936 10940 3090041
Cooper A McAlpine A Stockley PG Calorimetric studies of the energetics of protein-DNA interactions in the E. coli methionine repressor (MetJ) system FEBS Lett 1994 348 41 45 8026581 10.1016/0014-5793(94)00579-6
Rafferty JB Somers WS Saint-Girons I Phillips SE Three-dimensional crystal structures of Escherichia coli met repressor with and without corepressor Nature 1989 341 705 710 2677753 10.1038/341705a0
Somers WS Phillips SE Crystal structure of the met repressor-operator complex at 2.8 A resolution reveals DNA recognition by beta-strands Nature 1992 359 387 393 1406951 10.1038/359387a0
Phillips K Phillips SE Electrostatic activation of Escherichia coli methionine repressor Structure 1994 2 309 316 8087557 10.1016/S0969-2126(00)00032-0
Parsons ID Persson B Mekhalfia A Blackburn GM Stockley PG Probing the molecular mechanism of action of co-repressor in the E. coli methionine repressor-operator complex using surface plasmon resonance (SPR) Nucleic Acids Res 1995 23 211 216 7862523
Aravind L Koonin EV DNA-binding proteins and evolution of transcription regulation in the archaea Nucleic Acids Res 1999 27 4658 4670 10556324 10.1093/nar/27.23.4658
Scott JW Hawley SA Green KA Anis M Stewart G Scullion GA Norman DG Hardie DG CBS domains form energy-sensing modules whose binding of adenosine ligands is disrupted by disease mutations. J Clin Invest 2004 113 274 284 14722619 10.1172/JCI200419874
Kemp BE Bateman domains and adenosine derivatives form a binding contract. J Clin Invest 2004 113 182 184 14722609 10.1172/JCI200420846
Banerjee R Zou CG Redox regulation and reaction mechanism of human cystathionine-beta-synthase: a PLP-dependent hemesensor protein Arch Biochem Biophys 2005 433 144 156 15581573 10.1016/j.abb.2004.08.037
McLean JE Hamaguchi N Belenky P Mortimer SE Stanton M Hedstrom L Inosine 5'-monophosphate dehydrogenase binds nucleic acids in vitro and in vivo. Biochem J 2004 379 243 251 14766016 10.1042/BJ20031585
Koonin EV Wolf YI Karev GP The structure of the protein universe and genome evolution Nature 2002 420 218 223 12432406 10.1038/nature01256
Stroupe ME Leech HK Daniels DS Warren MJ Getzoff ED CysG structure reveals tetrapyrrole-binding features and novel regulation of siroheme biosynthesis Nat Struct Biol 2003 10 1064 1073 14595395 10.1038/nsb1007
Vevodova J Graham RM Raux E Schubert HL Roper DI Brindley AA Ian Scott A Roessner CA Stamford NP Stroupe ME Getzoff ED Warren MJ Wilson KS Structure/function studies on a S-adenosyl-L-methionine-dependent uroporphyrinogen III C methyltransferase (SUMT), a key regulatory enzyme of tetrapyrrole biosynthesis J Mol Biol 2004 344 419 433 15522295 10.1016/j.jmb.2004.09.020
Dixon MM Huang S Matthews RG Ludwig M The structure of the C-terminal domain of methionine synthase: presenting S-adenosylmethionine for reductive methylation of B12 Structure 1996 4 1263 1275 8939751 10.1016/S0969-2126(96)00135-9
Bandarian V Pattridge KA Lennon BW Huddler DP Matthews RG Ludwig ML Domain alternation switches B(12)-dependent methionine synthase to the activation conformation Nat Struct Biol 2002 9 53 56 11731805 10.1038/nsb738
Jarrett JT Huang S Matthews RG Methionine synthase exists in two distinct conformations that differ in reactivity toward methyltetrahydrofolate, adenosylmethionine, and flavodoxin Biochemistry 1998 37 5372 5382 9548919 10.1021/bi9730893
England JL Shakhnovich BE Shakhnovich EI Natural selection of more designable folds: a mechanism for thermophilic adaptation Proc Natl Acad Sci U S A 2003 100 8727 8731 12843403 10.1073/pnas.1530713100
Murzin AG Brenner SE Hubbard T Chothia C SCOP: a structural classification of proteins database for the investigation of sequences and structures J Mol Biol 1995 247 536 540 7723011 10.1006/jmbi.1995.0159
Teichmann SA Park J Chothia C Structural assignments to the Mycoplasma genitalium proteins show extensive gene duplications and domain rearrangements Proc Natl Acad Sci U S A 1998 95 14658 14663 9843945 10.1073/pnas.95.25.14658
Gough J Karplus K Hughey R Chothia C Assignment of homology to genome sequences using a library of hidden Markov models that represent all proteins of known structure J Mol Biol 2001 313 903 919 11697912 10.1006/jmbi.2001.5080
Madera M Vogel C Kummerfeld SK Chothia C Gough J The SUPERFAMILY database in 2004: additions and improvements Nucleic Acids Res 2004 32 D235 9 14681402 10.1093/nar/gkh117
Muller A MacCallum RM Sternberg MJ Structural characterization of the human proteome Genome Res 2002 12 1625 1641 12421749 10.1101/gr.221202
Keller JP Smith PM Benach J Christendat D deTitta GT Hunt JF The crystal structure of MT0146/CbiT suggests that the putative precorrin-8w decarboxylase is a methyltransferase Structure (Camb) 2002 10 1475 1487 12429089 10.1016/S0969-2126(02)00876-6
Min J Feng Q Li Z Zhang Y Xu RM Structure of the catalytic domain of human DOT1L, a non-SET domain nucleosomal histone methyltransferase. Cell 2003 112 711 723 12628190 10.1016/S0092-8674(03)00114-4
Christian T Evilia C Williams S Hou YM Distinct origins of tRNA(m1G37) methyltransferase J Mol Biol 2004 339 707 719 15165845 10.1016/j.jmb.2004.04.025
Perez-Rueda E Collado-Vides J Segovia L Phylogenetic distribution of DNA-binding transcription factors in bacteria and archaea Comput Biol Chem 2004 28 341 350 15556475 10.1016/j.compbiolchem.2004.09.004
Tatusov RL Fedorova ND Jackson JD Jacobs AR Kiryutin B Koonin EV Krylov DM Mazumder R Mekhedov SL Nikolskaya AN Rao BS Smirnov S Sverdlov AV Vasudevan S Wolf YI Yin JJ Natale DA The COG database: an updated version includes eukaryotes BMC Bioinformatics 2003 4 41 12969510 10.1186/1471-2105-4-41
Anantharaman V Aravind L Koonin EV Emergence of diverse biochemical activities in evolutionarily conserved structural scaffolds of proteins Curr Opin Chem Biol 2003 7 12 20 12547421 10.1016/S1367-5931(02)00018-2
Mushegian A Protein content of minimal and ancestral ribosome RNA 2005 11 1400 1406 16043494 10.1261/rna.2180205
Copley RR Bork P Homology among (betaalpha)(8) barrels: implications for the evolution of metabolic pathways J Mol Biol 2000 303 627 641 11054297 10.1006/jmbi.2000.4152
Walker DR Koonin EV SEALS: a system for easy analysis of lots of sequences. Proc Int Conf Intell Syst Mol Biol 1997 5 333 339 9322058
Altschul SF Madden TL Schaffer AA Zhang J Zhang Z Miller W Lipman DJ Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 1997 25 3389 3402 9254694 10.1093/nar/25.17.3389
Eddy SR Profile hidden Markov models. Bioinformatics 1998 14 755 763 9918945 10.1093/bioinformatics/14.9.755
Sadreyev R Grishin N COMPASS: a tool for comparison of multiple protein alignments with assessment of statistical significance. J Mol Biol 2003 326 317 336 12547212 10.1016/S0022-2836(02)01371-2
Guda C Lu S Scheeff ED Bourne PE Shindyalov IN CE-MC: a multiple protein structure alignment server. Nucleic Acids Res 2004 32 W100 3 15215359 10.1093/nar/gnh098
Holm L Sander C Searching protein structure databases has come of age. Proteins 1994 19 165 173 7937731 10.1002/prot.340190302
Edgar RC MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res 2004 32 1792 1797 15034147 10.1093/nar/gkh340
DeLano WL The PyMOL Molecular Graphics System 2002 http://www.pymol.org
McGuffin LJ Bryson K Jones DT The PSIPRED protein structure prediction server Bioinformatics 2000 16 404 405 10869041 10.1093/bioinformatics/16.4.404
Rost B Yachdav G Liu J The PredictProtein server Nucleic Acids Res 2004 32 W321 6 15215403
Karplus K Karchin R Draper J Casper J Mandel-Gutfreund Y Diekhans M Hughey R Combining local-structure, fold-recognition, and new fold methods for protein structure prediction Proteins 2003 53 Suppl 6 491 496 14579338 10.1002/prot.10540
Rychlewski L Jaroszewski L Li W Godzik A Comparison of sequence profiles. Strategies for structural predictions using sequence information Protein Sci 2000 9 232 241 10716175
Kelley LA MacCallum RM Sternberg MJ Enhanced genome annotation using structural profiles in the program 3D-PSSM J Mol Biol 2000 299 499 520 10860755 10.1006/jmbi.2000.3741
Fischer D Hybrid fold recognition: combining sequence derived properties with evolutionary information Pac Symp Biocomput 2000 119 130 10902162
Ginalski K Pas J Wyrwicz LS von Grotthuss M Bujnicki JM Rychlewski L ORFeus: Detection of distant homology using sequence profiles and predicted secondary structure Nucleic Acids Res 2003 31 3804 3807 12824423 10.1093/nar/gkg504
Shi J Blundell TL Mizuguchi K FUGUE: sequence-structure homology recognition using environment-specific substitution tables and structure-dependent gap penalties J Mol Biol 2001 310 243 257 11419950 10.1006/jmbi.2001.4762
Lundstrom J Rychlewski L Bujnicki J Elofsson A Pcons: a neural-network-based consensus predictor that improves fold recognition Protein Sci 2001 10 2354 2362 11604541 10.1110/ps.08501
Michalopoulos I Torrance GM Gilbert DR Westhead DR TOPS: an enhanced database of protein structural topology Nucleic Acids Res 2004 32 D251 4 14681405 10.1093/nar/gkh060
Mirkin BG Fenner TI Galperin MY Koonin EV Algorithms for computing parsimonious evolutionary scenarios for genome evolution, the last universal common ancestor and dominance of horizontal gene transfer in the evolution of prokaryotes BMC Evol Biol 2003 3 2 12515582 10.1186/1471-2148-3-2
SCOP Superfamily: S-adenosyl-L-methionine-dependent methyltransferases [http://scop.mrc-lmb.cam.ac.uk/scop/searchcgi?sunid=53335]
UniProt Knowledgebase: list of organism identification codes [http://www.expasy.org/cgi-bin/speclist]
Marchler-Bauer A Anderson JB DeWeese-Scott C Fedorova ND Geer LY He S Hurwitz DI Jackson JD Jacobs AR Lanczycki CJ Liebert CA Liu C Madej T Marchler GH Mazumder R Nikolskaya AN Panchenko AR Rao BS Shoemaker BA Simonyan V Song JS Thiessen PA Vasudevan S Wang Y Yamashita RA Yin JJ Bryant SH CDD: a curated Entrez database of conserved domain alignments. Nucleic Acids Res 2003 31 383 387 12520028 10.1093/nar/gkg087
Imai A Matsuyama T Hanzawa Y Akiyama T Tamaoki M Saji H Shirano Y Kato T Hayashi H Shibata D Tabata S Komeda Y Takahashi T Spermidine synthase genes are essential for survival of Arabidopsis Plant Physiol 2004 135 1565 1573 15247389 10.1104/pp.104.041699
Imai A Akiyama T Kato T Sato S Tabata S Yamamoto KT Takahashi T Spermine is not essential for survival of Arabidopsis FEBS Lett 2004 556 148 152 14706842 10.1016/S0014-5793(03)01395-4
Chattopadhyay MK Tabor CW Tabor H Spermidine but not spermine is essential for hypusine biosynthesis and growth in Saccharomyces cerevisiae: spermine is converted to spermidine in vivo by the FMS1-amine oxidase Proc Natl Acad Sci U S A 2003 100 13869 13874 14617780 10.1073/pnas.1835918100
Sapperstein S Berkower C Michaelis S Nucleotide sequence of the yeast STE14 gene, which encodes farnesylcysteine carboxyl methyltransferase, and demonstration of its essential role in a-factor export Mol Cell Biol 1994 14 1438 1449 8289819
Gomez-Gomez L Carrasco P Differential expression of the S-adenosyl-L-methionine synthase genes during pea development Plant Physiol 1998 117 397 405 9625692 10.1104/pp.117.2.397
Sanchez-Aguayo I Rodriguez-Galan JM Garcia R Torreblanca J Pardo JM Salt stress enhances xylem development and expression of S-adenosyl-L-methionine synthase in lignifying tissues of tomato plants Planta 2004 220 278 285 15322882 10.1007/s00425-004-1350-2
Jarrett JT The generation of 5'-deoxyadenosyl radicals by adenosylmethionine-dependent radical enzymes Curr Opin Chem Biol 2003 7 174 182 12714049 10.1016/S1367-5931(03)00022-X
|
16225687
|
PMC1282579
|
CC BY
|
2021-01-04 16:37:47
|
no
|
BMC Struct Biol. 2005 Oct 14; 5:19
|
utf-8
|
BMC Struct Biol
| 2,005 |
10.1186/1472-6807-5-19
|
oa_comm
|
==== Front
Biomed Eng OnlineBioMedical Engineering OnLine1475-925XBioMed Central London 1475-925X-4-631626643110.1186/1475-925X-4-63Book ReviewReview of "Biomedical Engineering Principles", by A.B. Ritter, S. Reisman and B.B. Michniak Gefen Amit [email protected] Department of Biomedical Engineering, Faculty of Engineering, Tel Aviv University, Tel Aviv 69978, Israel2005 2 11 2005 4 63 63 Biomedical Engineering Principles .
Ritter AB , Reisman S , and Michniak BB . Boca Raton, FL: CRC Press . 2005 . ISBN 0-8247-9616-0, xii+665 pages. US$100 (Hardcover). 29 10 2005 2 11 2005 Copyright © 2005 Gefen; licensee BioMed Central Ltd.2005Gefen; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
==== Body
Being a young discipline, Biomedical Engineering (BME) offers too few good textbooks for undergraduates. Moreover, in some of the available undergraduate BME textbooks, the levels of mathematics and engineering science were compromised in favor of a more qualitative approach. I was therefore happy to discover that in the recent title by Ritter et al., there is not only in-breadth overview on current BME fields, but also, in-depth analysis of specific selected topics that are supported by appropriate mathematical models, equations, quantitative data charts and tables. This indicates, however, that this book is intended for BME undergraduates at the sophomore, junior or senior years (but not for freshmen). Specifically, to work with this textbook, a student should have already learned calculus, differential and integral equations, and basic engineering courses such as statics, dynamics, solid and fluid mechanics and analog electrical circuits. It should be acknowledged that the authors provide overviews on some of these topics where relevant (e.g. they develop the general equations of continuity and motion before discussing hemodynamic flows in Chapter 3), however, such overviews are too brief to actually be able to introduce students to the theoretical basis. Instead, it appears that they were aimed at refreshing prior knowledge. Hence, instructors may use this book in a BME program to support advanced-year courses in biomechanics (cardiovascular or musculoskeletal) or biomedical signal processing, but not in a first-year Introduction to BME course. Considering the broad scope of this book, however, it is not recommended as a sole textbook in a biomechanics or a biomedical signal processing class. Contrarily, in a non-BME engineering program (e.g. which includes some elective BME advanced-year courses), this title can certainly serve as the major textbook to be used in class.
The book contains 12 chapters that are organized in 3 parts. Part 1 (Chapters 1–4) concerns transport processes, cell physiology and the cardiovascular system. Specifically, Chapter 1 includes an overview of engineering analysis of physiological systems with the aid of models, with a thorough discussion on model validation and parameter estimation. Chapter 2 focuses on the cellular level, and discusses cellular processes such as diffusion, transport, generation of membrane potential and propagation of action potentials. Particular attention has been given to muscle contraction at the microscopic scale, with distinction between processes in cardiac and smooth muscles. Chapter 3 is back at the organ scale, and concerns hemodynamic flows as analyzed by engineering tools, including electrical analogs, models of flows in tubes (steady, pulsatile, and turbulent), and characterization of heart sounds. Chapter 4 is a natural continuation, which extends the basic hemodynamic principles described in the previous chapter to flow, pressure and volume relations in the cardiovascular system. It also contains a comprehensive description of the electrocardiogram (ECG), its method of measurement, and its interpretation (e.g. how to identify arrhythmias, myocardial infraction, arterial and ventricular diseases, etc.). A useful feature in this chapter is mathematical models for the left ventricle and the arterial system with their embodiment in Simulink (MathWorks Co., Natick MA, USA) [1]. A detailed description of heart failure conditions closes this chapter.
The second part of the book is a deep and thorough introduction to biomedical signal processing. It is linked to the previous part on modeling by the concept that ultimately, bioengineering models are compared with and validated against experimental data. Acquisition, processing and reduction of physiological signals are therefore essential bioengineering tools that should be thought at the undergraduate BME level. Accordingly, Chapter 5 presents biomedical signals such as the ECG, the electromyogram (EMG), and the electroencephalogram (EEG). It then describes the Fourier transform and its utility in processing these signals. Practical aspects, such as 60 Hz component rejection are provided in this context. The following Chapter 6 concerns acquisition, sampling, and signal windowing, and discusses typical problems such as selection of a sampling frequency and aliasing. Chapter 7 introduces the reader to common techniques for physiological signal processing such as autoregressive modeling, time-frequency and wavelet analyses. Examples of physiological signal processing follow in Chapter 8. These include heart rate variability analysis, blood pressure variability, body temperature oscillations, EMG and EEG analyses in the frequency domains and particularly, median frequency of the EMG as a measure of muscle fatigue. The second part ends with two chapters on biomechanics: Chapter 9 presents some principles of biomechanics, mostly in whole-body musculoskeletal biomechanics but also in tissue mechanics and thermal regulation, and Chapter 10 details some applications. Applications are given in postural stability, biomechanics of swimming, design of ankle prostheses, human power augmentation devices and exoskeleton prototypes, functional electrical stimulation and movement control, decompression models for divers, prediction of pressure ulcer formation on the skin and analysis of lung sounds in normals and emphysematic patients. This is a broad collection of examples, however, as mentioned earlier, all examples are thoroughly analyzed by means of quantitative engineering tools, including biostatistical analyses and models that are based on empirical data.
The third and last part (which is also the shortest part) of this book is dedicated to tissue engineering. Chapter 11 begins with a historical perspective on the field of tissue engineering, which is followed by the paradigms of tissue engineering, a review of materials used as scaffolds in tissue engineering with particular emphasis on degradable polymers but also some reference to ceramics and composites, and a discussion on biological interactions. The chapter ends with an overview on tissue engineering applications including skin substitutes, cardiovascular components, bone regeneration, tissue-engineered muscles and nerve regeneration. The book closes with Chapter 12 on future trends in BME, which contains the authors' perspective on the future of cardiac care, minimally invasive robotic surgery, the next generation of computed tomography imaging and nanotechnology in BME.
In addition to the thorough quantitative handling of the topics in this book, there are review problems at the end of each chapter that allow a student to rehearse the material covered, however, no solutions are included, and this may be frustrating for some students. One reservation that should be noted is that the quality of artwork in the book is not uniform, and some figures are actually blurred, e.g. in Chapter 6 on signal acquisition and processing. Nevertheless, this is a minor issue and the authors should be commended for their achievement – a good, thorough BME textbook for undergraduates that does not move away from using advanced mathematics or engineering theories when needed. I will certainly use examples from the relevant chapters in my Musculoskeletal Biomechanics and Tissue Biomechanics courses, and will add this book to the recommended literature in these courses.
==== Refs
Dabney JB Harman TL Mastering Simulink 2004 Prentice Hall, Upper Saddle River, New Jersey
|
0
|
PMC1282580
|
CC BY
|
2021-01-04 16:37:36
|
no
|
Biomed Eng Online. 2005 Nov 2; 4:63
|
utf-8
|
Biomed Eng Online
| 2,005 |
10.1186/1475-925X-4-63
|
oa_comm
|
==== Front
J CarcinogJournal of Carcinogenesis1477-3163BioMed Central London 1477-3163-4-191623616210.1186/1477-3163-4-19ResearchThe role of c-kit and imatinib mesylate in uveal melanoma Pereira Patricia Rusa [email protected] Alexandre Nakao [email protected] Jean Claude [email protected] Zelia Maria [email protected] Rubens [email protected] Miguel N [email protected] Department of Ophthalmology, Federal University of São Paulo, São Paulo, Brazil2 Henry C. Witelson Ocular Pathology Laboratory, Department of Ophthalmology, McGill University, Montreal – Canada2005 19 10 2005 4 19 19 4 7 2005 19 10 2005 Copyright © 2005 Pereira et al; licensee BioMed Central Ltd.2005Pereira et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
Uveal melanoma (UM) is the most common primary intraocular tumor in adults, leading to metastasis in 40% of the cases and ultimately to death in 10 years, despite local and/or systemic treatment. The c-kit protein (CD117) is a membrane-bound tyrosine kinase receptor and its overexpression has been observed in several neoplasms. Imatinib mesylate is a FDA approved compound that inhibits tyrosine quinase receptors, as well as c-kit. Imatinib mesylate controls tumor growth in up to 85% of advanced gastrointestinal stromal tumors, a neoplasia resistant to conventional therapy.
Methods
Fifty-five specimens of primary UM selected from the archives of the Ocular Pathology Laboratory, McGill University, Montreal, Canada, were immunostained for c-kit. All cells displaying distinct immunoreactivity were considered positive. Four human UM cell lines and 1 human uveal transformed melanocyte cell line were tested for in vitro proliferation Assays (TOX-6) and invasion assay with imatinib mesylate (concentration of 10 μM).
Results
The c-kit expression was positive in 78.2% of the UM. There was a statistical significant decrease in the proliferation and invasion rates of all 5 cell lines.
Conclusion
The majority of UM expressed c-kit, and imatinib mesylate does decrease the proliferation and invasion rates of human UM cell lines. These results justify the need for a clinical trial to investigate in vivo the response of UM to imatinib mesylate.
C-kitCD117melanomauvealocularGleevec®Imatinib mesylate
==== Body
Background
Uveal melanoma (UM) is the most common primary intra-ocular tumour in adults [1], with an incidence of five to six individuals per million people [2]. Forty percent of UM patients will progress from local to systemic disease developing metastasis that will ultimately lead to death after 10 years of diagnosis, despite the treatment options such as local radiotherapy, enucleation and systemic chemotherapy [3].
C-kit (Kit, CD117, stem cell factor receptor) is a 145 kDa transmembrane tyrosine kinase protein that acts as a type-III receptor. The c-kit proto-oncogene, located on chromosome 4q11-21, encodes the c-kit, which ligand is the stem cell factor (SCF, steel factor, kit ligand, mast cell growth factor) [4,5]. Tyrosine phosphorylation by protein tyrosine kinases is of particular importance in cellular signalling and can mediate signals for major cellular processes, such as proliferation, differentiation, apoptosis, attachment, and migration. The role of c-kit expression has been studied in hematologic and solid tumours, such as acute leukemias [6], and gastrointestinal stromal tumors (GIST) [7].
The clinical importance of c-kit expression in malignant tumors relies on the existence of a compound (imatinib mesylate, STI571, Gleevec®, Novartis Pharma AG Basel, Switzerland) that specifically inhibits tyrosine kinase receptors [8]. Moreover, a clinically relevant breakthrough has been the finding of remarkable anti-tumor effects of this compound in GIST, a group of tumors regarded as being generally resistant to conventional chemotherapy [9]. Imatinib mesylate has been approved by the United States-FDA to treat c-kit positive GIST and Philadelphia-chromosome-positive chronic myelogenous leukemia [10]. The purpose of this article is to study the immunoexpression of c-kit in UM, as well as the in vitro effects of imatinib mesylate on UM cell lines.
Materials and methods
Paraffin blocks
Formalin-fixed, paraffin-embedded blocks from enucleation of primary choroidal melanoma were collected from the archives of the Henry C. Witelson Ocular Pathology Laboratory, McGill University, Montreal, Canada from the years 1980–2004. All the cases have sufficient tumor material for analysis. The tumors with irradiation were excluded.
Immunohistochemistry
Immunohistochemistry was performed using the polyclonal anti-CD117 antibody A4502 (Dako, Mississauga, Ontario, Canada). The antibody was applied at a dilution of 1:300 for 18 h at 4°C, after 15 minutes in 10 nmol/l citrate buffer (pH 6.0) for antigen retrieval. Endogenous peroxidase was blocked using 0.3% hydrogen peroxidase diluted in methanol for 30 minutes. A standard avidin-biotin complex (ABC) technique using diaminobenzidine was used for visualization, with a red colouring stain to avoid misinterpretation in pigmented tumors. A case of KIT-positive Gastrointestinal Stromal Tumor (GIST) was used as control. Negative control sections were incubated with normal rabbit serum instead of the primary antibody.
After tissue processing, all cells displaying distinct immunoreactivity were considered positive, irrespective of staining intensity. We assigned the results of c-kit staining as negative when no staining was present, low when less than 50%, and high when more than 50% of melanoma cells were positive. The immunoreactivity was categorized as cytoplasmatic or membranous expression using the grade system described above. In order to better characterize the c-kit expression in uveal melanomas, the expression in the different cell types (spindle and epithelioid – modified Callender's classification) was analysed, in the mixed cell, predominant epithelioid and predominant spindle tumors.
Cell Culture
Four human UM cell lines (92.1, SP6.5, MKT-BR, OCM-1) and one human uveal transformed melanocyte cell line (UW-1) were incubated at 37°C in a humidified 5% CO2-enriched atmosphere. The cells were cultured in RPMI-1640 medium (Invitrogen, Burlington, Ontario, Canada), supplemented with 5% heat inactivated fetal bovine serum (FBS), 1% fungizone, and 1% penicillin-streptomycin purchased from Invitrogen (Burlington, Ontario, Canada). Cells were cultured as a monolayer in 25 cm2 flasks (Fisher, Whitby, Ontario, Canada) and observed twice weekly, at every media change, for normal growth by phase contrast microscopy. The cultures were grown to confluence and passage by treatment with 0.05% trypsin in EDTA (Fisher) at 37°C and washed in 7 ml RPMI-1640 media before being centrifuged at 120 g for 10 minutes to form a pellet. Cells were then suspended in 1 ml of medium and counted using the Trypan Blue dye exclusion test.
The UM cell lines 92.1, SP6.5, and MKT-BR were established by Dr. Jager (University Hospital Leiden, The Netherlands), Dr. Pelletier (Laval University, Quebec, Canada), Dr. Belkhou (CJF INSERM, France), respectively. Dr. Albert (University of Wisconsin-Madison, USA) established the OCM-1 and UW-1 cell lines [11,12].
In Vitro Invasion Assay
A modified Boyden chamber consisting of a polyethelene teraphthalate membrane (PET) with 8-um diameter pores, precoated with Matrigel, an artificial basement membrane, (Beckton Dickenson Labware, Bedford, MA) was used as previously described [13], to assay for invasive ability. PET membrane without Matrigel was used as a control.
Briefly, 1.25 × 105 cells were added to the upper chamber in RPMI-1640 medium with 0.1% FBS. RPMI-1640 medium with 10% FBS was added to the lower chamber as a chemoattractant to obtain the baseline invasive ability of the cell lines. The effect of imatinib mesylate on invasion was assayed by adding 10 μM of imatinib mesylate to the RPMI-1640 medium supplemented with 10% FBS in the upper chamber. The concentrations of imatinib mesylate was chosen based on the blood levels reported in previous studies with the maximum tolerated dose of imatinib mesylate on clinical trials [14]. The chambers were then incubated at 37°C in a 5% CO2-enriched atmosphere for 48 hours to allow for cellular invasion through the Matrigel.
Non-invading cells were removed from the upper chamber by gently wiping the surface of the membrane with a moist cotton swab. Membranes were removed and then stained using a Diff-Quick staining set, which stains cell nuclei purple and cytoplasm pink. Stained cells were counted microscopically in 20 high-powered fields, randomly. Only cells whose nuclei had completely invaded through the membrane were counted. Each experimental condition, including control, was performed in triplicate and the average number of invading cells was then calculated for all experimental conditions.
Percent invasion was determined for each cell line under each experimental condition using the following formula: % invasion = (mean number of cells invading through the matrigel/mean number of cells migrating through control PET membrane) multiplied by a hundred. The cell lines were then ranked according to their invasive ability.
In Vitro Proliferation Assay
The Sulforhodamine-B based assay kit (TOX-6, Sigma-Aldrich) was performed as per the National Cancer Institute protocol [15]. Briefly, five human UM cell lines were seeded into wells at a concentration of 2.5 × 103 cells per well, in a minimum of six wells per cell line. A row of 8 wells exposed to only RPMI-1640 medium was used as a control. Twenty-four hours following seeding, imatinib mesylate was added to the experimental wells. The concentration of imatinib mesylate was 10 μM [14]. Cells were allowed to incubate for 48 hours following cell seeding. Following this 48 hour period, cells were fixed to the bottom of the wells using a solution of 50% Trichloroacetic acid (TCA) for 1 hour at 4°C. Plates were then rinsed with distilled water, to remove TCA and medium, and air dried. The Sulforhodamine-B dye was added to each well and allowed to stain for 25 minutes. The Sulforhodamine-B dye was subsequently removed by washing with a 10% acetic acid solution and once more allowed to air dry. The dye that was incorporated into the fixed cells at the bottom of the wells was solubilized in a 10 mM solution of Tris. The absorbance of the solute was measured using a microplate reader at a wavelength of 510 nm. This gave a comparison of control cell proliferation rate over 48 hours compared to proliferation rate of cells exposed to imatinib mesylate during the same time period at a dose of 10 μM.
Statistical Analysis
The differences in invasion rates under three experimental conditions for each uveal melanoma cell line were determined using the ANOVA test. A p value of less than 0.05 was considered statistically significant. Calculations were computer-based (SPSS 11.5, SPSS Inc., Chicago, Illinois, USA).
Results
Fifty-five cases of UM were studied. Eight seven percent of the tumors (n = 48) were classified as mixed cell type (spindle and epithelioid cells), 9% (n = 5) had predominance of epithelioid cells, and 3.6% (n = 2) of spindle cells.
Seventy-eight percent of cases (n = 43) were found to be c-kit positive (Figure 1A). Among the positive cases, 46.5% (n = 20) presented with what was considered as high expression. All lesions with high immunoreactivity (n = 20) had cytoplasmic and cell membrane expression. Meanwhile, among lesions with low immunoreactivity (n = 23), 100% presented a cytoplasmatic reaction and just 30.4% (n = 7) presented with a cell membrane stain-pattern (Figure 1B). (Table 1)
Figure 1 A. Choroidal UM displaying high expression of c-kit. The above retina is somewhat detached. Figure 1B. UM composed of epithelioid cells showing a cell membrane pattern of immunostain.
Table 1 High, low and negative expression of c-kit in different cell types and different patterns of stain.
High expression Low expression Negative
Spindle cells 36.8% 57.9% 5.3%
Epithelioid cells 41.5% 53.6% 4.9%
Cell membrane 100% 30.4% ------
Cytoplasmatic 100% 100% ------
The percent invasion of cell lines according to the baseline invasion without imatinib mesylate was: MKT-BR (38.4%) > OCM-1 (21.7%) > 92.1 (14.4%) > UW-1 (12%) > SP6.5 (3%). The addition of imatinib mesylate decreased the invasion in all cell lines: MKT-BR (1.03%); OCM-1 (0.1%); 92.1 (0.2%); UW-1 (0%); SP6.5 (0%). The results are shown in Figure 3.
Figure 3 The graph shows the invasion assay results for each cell line with experimental conditions. The blue bar for each represents the control proliferation without Imatinib Mesylate, and the red bar with the compound. A statistical difference was seen for the five cell lines after 10 μM exposure.
No visible changes in cytomorphology were seen in reaction to the presence of imatinib mesylate (figure 2).
Statistically significant differences between the invasion rates for the control group and imatinib mesylate group were found in all cell lines (T test p value < 0.05).
Figure 4 depicts the results from the Sulforhodamine-B assay. The mean and standard deviation for each cell line per condition is shown in Table 2. Cells that were directly exposed imatinib mesylate showed a decrease in proliferation for all five human cell lines (92.1, MKT-BR, OCM-1, SP6.5, UW-1) as compared to control (p value of 0.001354991, 0.012655861, 9.47698 × 10-7, 0.002754018 and 5.79576 × 10-6 respectively).
Figure 4 The graph shows the proliferation assay results comparing a control and the Imatinib mesylate exposure cells. A statistical difference was seen for the five cell lines after 10 μM exposure.
Table 2 Number of cells counted per control membrane, and percent invasion (mean ± standard deviation) of cells in response to experimental conditions.
Cell Line Average Number of Cells in Control PET Percent Invasion of Cells in Response to 10% FBS Percent Invasion of Cells in Response to Imatinib mesylate
MKT-BR 247 38.43% ± 8.5 1.03% ± 0.2
OCM-1 246.67 21.7% ± 1.9 0.1% ± 0.2
92.1 223 14.4% ± 2.7 0.2% ± 0.2
UW-1 66.33 12% ± 4.8 0% ± 0
SP6.5 266 3% ± 1.5 0% ± 0
Discussion
It is known that protein tyrosine kinases (PTK) have an important role in cellular mechanisms, such as differentiation, proliferation and regulatory mechanisms, as well as in signal transduction. C-kit is one of these PTK, which is expressed in a wide variety of human malignancies [16] including chronic and acute myelogenous leukemia [6], GIST [7], mastocytosis [17], small cell lung carcinoma [18], chromophobe renal cell carcinoma [19], cutaneous [20] and UM [16]. As c-kit is expressed in normal interstitial cells of Cajal, the progenitor cell of GIST [7], the present article studies the expression of c-kit in uveal melanomas, as normal choroidal melanocytes do express this marker [21].
We demonstrated that more than 75% of UM from our series are positive for c-kit. This finding, per se, supports the idea of a clinical trial of imatinib mesylate for UM, especially in metastatic cases. Once metastatic disease is detected, no effective method of systemic therapy has been identified [3]. Moreover, not 100% of GIST is positive for c-kit. In fact, 6% of GIST are c-kit negative [22]. Before the imatinib mesylate "era", metastatic GIST had a median survival times ranging between 10–20 months [23]. Nowadays, imatinib mesylate controls tumor growth in up to 85% of advanced GIST [24], with 90% of acceptable toxicity [25].
In cutaneous melanoma, c-kit is strongly expressed in radial growth phase, and weak or no expression is seen in vertical growth phase and metastatic disease [26]. Therefore, in cutaneous melanomas c-kit expression appears to be related with stage of the disease. To further investigate a similar expression of the c-kit in UM, we observed the cell type (spindle and epithelioid) in which the c-kit was expressed, as it is well known that spindle cell type is less aggressive than epithelioid type [1]. None of the previous studies recorded the cell type in which the expression was occurring. Mouriaux et al [21] compared the c-kit expression with cell type tumor (Callendar's classification) as a correlation to prognostic factor, but did not study which cells were in fact expressing the receptor. In overall cases, including epithelioid, spindle and mixed tumors, 90.7% of the epithelioid cells was positive for c-kit, whereas 83.7% of the spindle cells expressed c-kit. This finding suggests a different role of c-kit expression in cutaneous and UMs. Therefore, c-kit can be used to differentiate primary UM from metastatic cutaneous melanomas to the uveal tract [27]. As in UM, GIST can be categorized into spindle cell, epitheloid cell and mixed cell type, although the prognostic relevance of cell type in GIST seems to be limited [7]. Like in UM showed by the present study, the expression of c-kit seems to be equally distributed among epithelioid and spindle cell morphology GISTs [28].
Pache et al [29] and Mouriaux et al [21] found a membrane pattern of immunoreactivity in all their positive cases (n = 72, 87% of all UM, and n = 43, 75% of all UM, respectively). The last suggested that cytoplasmatic immunoreactivity would be due to a non-mature c-kit rather than an internalized or truncated form, and the membrane staining could correspond to the c-kit active form. All-Ericsson et al [16] considered positive all tumors expressing c-kit, regardless of the staining location, meaning cell membrane or cytoplasmatic (n = 84, 63% of all UM). The present study demonstrated that all cases with high c-kit expression showed both cytoplasmatic and cell-membrane staining. Regarding the cases with low expression, 100% had cytoplasmatic expression, but only 30.4% had cell-membrane pattern of stain. In GISTs, the c-kit expression is diffuse strong cytoplasmatic and up to 50% of the cases show cytoplasmatic dot-like (so-called "golgi pattern") staining [7]. Moreover, there are no studies concerning important differences between cytoplasmatic and membranous staining in GIST. Therefore, we hypothesize that all cases expressing c-kit, cytoplasmatic or membranous, should be considered c-kit positive.
GIST is a sarcoma arising from the interstitial cells of Cajal, harbouring mutation of c-kit. Mutations are detected in approximately 71% of tumors, the majority (over 60%) involving exon 11, and less exons 9 and 13 [28]. Choroidal melanoma does not have alterations of exons 2, 8, 9, 11, 13 and 17 [29]. However, imatinib mesylate selectively inhibits not only c-kit, but also other tyrosine kinases such as Bcr-Abl and platelet-derived growth factor (PDGF) receptor [30]. We could demonstrate that the decrease of proliferation of UM cells with imatinib mesylate was very significant, in the 4 UM cell lines tested and in the human uveal transformed melanocyte cell line, compared to the control group. Other studies also support that imatinib mesylate can decrease UM cells proliferation rates [8]. The mechanism that c-kit would interfere in UM proliferation could be other than c-kit mutation, but further studies are necessary to investigate this hypothesis.
The concentration of imatinib mesylate used for the in vitro studies was 10 μM. This concentration is equivalent to the highest drug concentration achieved in the blood of patients receiving 1000 mg/day of imatinib mesylate, the maximum tolerated dose reported by clinical trials [14]. At that dose, the blood concentration of imatinib mesylate ranged from 6 to 10 μM [14]. According to clinical trials, the current treatment for GIST is 800 mg/day of imatinib mesylate [25]. All-Ericsson et al [16] demonstrated that concentrations of 10 μM of imatinib mesylate could inhibit the proliferation of 5 UM cell lines in 50% (2 of them, OCM-1 and 92.1, were also studied in this article). The different concentrations of imatinib mesylate tested had different responses according to the cell line. Pache et al [29] also had the same conclusion. Moreover, the last demonstrated that imatinib mesylate does not influence the proliferation of normal uveal melanocytes. The human uveal transformed melanocyte cell line UW-1 studied in the present article demonstrated a significant decrease in proliferation and invasion rates when treated with imatinib mesylate. UW-1 was originally derived from uveal melanocytes, and transformed into malignant melanoma cells throughout culture. We hypothesize that imatinib mesylate could act in a more general pathway than c-kit receptor, as it does inhibit UW-1 proliferation and invasion rates.
The effect on invasion of UM cells in response to imatinib mesylate has never been published before. Tumor cells must possess invasive abilities in order for metastasis to occur. Due to the lack of lymphatics in the eye, uveal melanoma cells must leave the primary tumor via hematogenous dissemination, with metastasis almost exclusively occurring in the liver [2]. Our study demonstrated that imatinib mesylate markedly reduced the invasiveness of all cell lines tested. The invasion assay is important to show the ability of cells to invade a basement membrane, simulating the escape of cells from the primary tumor, as well as the implantation of cells at the site of the metastasis. The use of artificial basement membrane can study the invasive response of cells to drugs by counting the amount of cells that invade the matrigel layer. A drug that can inhibit or reduce the invasiveness ability of the UM cells could be beneficial, as most of the UM are nowadays treated conservatively [31]. Decreasing the invasiveness of the tumoral cells, the drug would also decrease the ability of implantation of cells at the site of metastasis. Therefore, imatinib mesylate would be beneficial not only for UM patients that already developed metastasis, but also for patients without any sign of metastatic disease.
Conclusion
We could demonstrate that primary choroidal melanomas express c-kit and imatinib mesylate decrease the proliferation rate and invasiveness of uveal melanoma cells in vitro. Therefore, our data supports a clinical trial for studying imatinib mesylate in uveal melanoma.
Figure 2 A. Photomicrograph of UM cells without Imatinib mesylate (control) invading through Matrigel (200×). Figure 2B. Photomicrograph of UM cells treated with Imatinib mesylate invading through Matrigel (200×).
Acknowledgements
We'd like to acknowledge Novartis Pharmaceutics to kindly supply imatinib mesylate for this experiment.
Dr. Alexandre Nakao Odashiro is supported by CAPES foundation.
==== Refs
McLean IW Armed Forces Institute of Pathology (U.S.), Universities Associated for Research and Education in Pathology Tumors of the eye and ocular adnexa 1994 Washington, D.C. Bethesda, Md.: Armed Forces Institute of Pathology; under the auspices of Universities Associated for Research and Education in Pathology Inc
Egan KM Seddon JM Glynn RJ Gragoudas ES Albert DM Epidemiologic aspects of uveal melanoma Surv Ophthalmol 1988 32 239 251 3279559 10.1016/0039-6257(88)90173-7
Assessment of metastatic disease status at death in 435 patients with large choroidal melanoma in the Collaborative Ocular Melanoma Study (COMS): COMS report no. 15 Arch Ophthalmol 2001 119 670 676 11346394
Morstyn G Brown S Gordon M Crawford J Demetri G Rich W McGuire B Foote M McNiece I Stem cell factor is a potent synergistic factor in hematopoiesis Oncology 1994 51 205 214 7515171
Yarden Y Kuang WJ Yang-Feng T Coussens L Munemitsu S Dull TJ Chen E Schlessinger J Francke U Ullrich A Human proto-oncogene c-kit: a new cell surface receptor tyrosine kinase for an unidentified ligand Embo J 1987 6 3341 3351 2448137
Cortes J Giles F O'Brien S Thomas D Albitar M Rios MB Talpaz M Garcia-Manero G Faderl S Letvak L Results of imatinib mesylate therapy in patients with refractory or recurrent acute myeloid leukemia, high-risk myelodysplastic syndrome, and myeloproliferative disorders Cancer 2003 97 2760 2766 12767088 10.1002/cncr.11416
Fletcher CD Berman JJ Corless C Gorstein F Lasota J Longley BJ Miettinen M O'Leary TJ Remotti H Rubin BP Diagnosis of gastrointestinal stromal tumors: A consensus approach Hum Pathol 2002 33 459 465 12094370 10.1053/hupa.2002.123545
Lefevre G Glotin AL Calipel A Mouriaux F Tran T Kherrouche Z Maurage CA Auclair C Mascarelli F Roles of stem cell factor/c-Kit and effects of Glivec/STI571 in human uveal melanoma cell tumorigenesis J Biol Chem 2004 279 31769 31779 15145934 10.1074/jbc.M403907200
de Silva CM Reid R Gastrointestinal stromal tumors (GIST): C-kit mutations, CD117 expression, differential diagnosis and targeted cancer therapy with Imatinib Pathol Oncol Res 2003 9 13 19 12704441
Cohen MH Johnson JR Pazdur R U.S. Food and Drug Administration drug approval summary: conversion of imatinib mesylate (ST1571; Gleevec) tablets from accelerated approval to full approval. Clin Cancer Res 2005 11 12 19 15671523
Diebold Y Blanco G Saornil MA Fernandez N Lazaro MC Morphologic and immunocytochemical characterization of four human uveal cell lines (melanoma- and melanocytes-derived) Curr Eye Res 1997 16 487 495 9154388 10.1076/ceyr.16.5.487.7047
De Waard-Siebinga I Blom DJ Griffioen M Schrier PI Hoogendoorn E Beverstock G Danen EH Jager MJ Establishment and characterization of an uveal-melanoma cell line Int J Cancer 1995 62 155 161 7622289
Woodward JK Elshaw SR Murray AK Nichols CE Cross N Laws D Rennie IG Sisley K Stimulation and inhibition of uveal melanoma invasion by HGF, GRO, IL-1alpha and TGF-beta Invest Ophthalmol Vis Sci 2002 43 3144 3152 12356817
Druker BJ Taking aim at Ewing's sarcoma: is KIT a target and will imatinib work? J Natl Cancer Inst 2002 94 1660 1661 12441314
Skehan P Storeng R Scudiero D Monks A McMahon J Vistica D Warren JT Bokesch H Kenney S Boyd MR New colorimetric cytotoxicity assay for anticancer-drug screening J Natl Cancer Inst 1990 82 1107 1112 2359136
All-Ericsson C Girnita L Muller-Brunotte A Brodin B Seregard S Ostman A Larsson O c-Kit-dependent growth of uveal melanoma cells: a potential therapeutic target? Invest Ophthalmol Vis Sci 2004 45 2075 2082 15223779 10.1167/iovs.03-1196
Akin C Fumo G Yavuz AS Lipsky PE Neckers L Metcalfe DD A novel form of mastocytosis associated with a transmembrane c-kit mutation and response to imatinib Blood 2004 103 3222 3225 15070706 10.1182/blood-2003-11-3816
Krug LM Crapanzano JP Azzoli CG Miller VA Rizvi N Gomez J Kris MG Pizzo B Tyson L Dunne M Imatinib mesylate lacks activity in small cell lung carcinoma expressing c-kit protein Cancer 2005 103 2128 2131 15812822 10.1002/cncr.21000
Petit A Castillo M Santos M Mellado B Alcover JB Mallofre C KIT expression in chromophobe renal cell carcinoma: comparative immunohistochemical analysis of KIT expression in different renal cell neoplasms Am J Surg Pathol 2004 28 676 678 15105658
Potti A Hille RC Koch M Immunohistochemical determination of HER-2/neu overexpression in malignant melanoma reveals no prognostic value, while c-Kit (CD117) overexpression exhibits potential therapeutic implications J Carcinog 2003 2 8 14617373 10.1186/1477-3163-2-8
Mouriaux F Kherrouche Z Maurage CA Demailly FX Labalette P Saule S Expression of the c-kit receptor in choroidal melanomas Melanoma Res 2003 13 161 166 12690299 10.1097/00008390-200304000-00008
Tzen CY Mau BL Analysis of CD117-negative gastrointestinal stromal tumors World J Gastroenterol 2005 11 1052 1055 15742414
Bauer S Hartmann JT de Wit M Lang H Grabellus F Antoch G Niebel W Erhard J Ebeling P Zeth M Resection of residual disease in patients with metastatic gastrointestinal stromal tumors responding to treatment with imatinib Int J Cancer 2005
Blay JY Bonvalot S Casali P Choi H Debiec-Richter M Dei Tos AP Emile JF Gronchi A Hogendoorn PC Joensuu H Consensus meeting for the management of gastrointestinal stromal tumors. Report of the GIST Consensus Conference of 20–21 March 2004, under the auspices of ESMO Ann Oncol 2005 16 566 578 15781488 10.1093/annonc/mdi127
Candelaria M de la Garza J Duenas-Gonzalez A A Clinical and Biological Overview of Gastrointestinal Stromal Tumors Med Oncol 2005 22 1 10 15750190 10.1385/MO:22:1:001
Stefanou D Batistatou A Zioga A Arkoumani E Papachristou DJ Agnantis NJ Immunohistochemical expression of vascular endothelial growth factor (VEGF) and C-KIT in cutaneous melanocytic lesions Int J Surg Pathol 2004 12 133 138 15173918
Guerriere-Kovach PM Hunt EL Patterson JW Glembocki DJ English JC 3rdWick MR Primary melanoma of the skin and cutaneous melanomatous metastases: comparative histologic features and immunophenotypes Am J Clin Pathol 2004 122 70 77 15272532 10.1309/R8N1-CN04-KKFV-6M7T
Romagnoli S Graziani D Bramerio M Gambacorta M Colombo P Roncalli M Coggi G Bosari S Immunohistochemical profile and c-kit mutations in gastrointestinal stromal tumors Pathol Res Pract 2005 201 71 81 15901127 10.1016/j.prp.2005.01.005
Pache M Glatz K Bosch D Dirnhofer S Mirlacher M Simon R Schraml P Rufle A Flammer J Sauter G Sequence analysis and high-throughput immunohistochemical profiling of KIT (CD 117) expression in uveal melanoma using tissue microarrays Virchows Arch 2003 443 741 744 14513377 10.1007/s00428-003-0883-2
Buchdunger E Cioffi CL Law N Stover D Ohno-Jones S Druker BJ Lydon NB Abl protein-tyrosine kinase inhibitor STI571 inhibits in vitro signal transduction mediated by c-kit and platelet-derived growth factor receptors J Pharmacol Exp Ther 2000 295 139 145 10991971
Damato B Developments in the management of uveal melanoma Clin Experiment Ophthalmol 2004 32 639 647 15575836 10.1111/j.1442-9071.2004.00917.x
|
16236162
|
PMC1282581
|
CC BY
|
2021-01-04 16:39:19
|
no
|
J Carcinog. 2005 Oct 19; 4:19
|
utf-8
|
J Carcinog
| 2,005 |
10.1186/1477-3163-4-19
|
oa_comm
|
==== Front
Chiropr OsteopatChiropractic & Osteopathy1746-1340BioMed Central London 1746-1340-13-221624203510.1186/1746-1340-13-22ResearchAn online survey of chiropractors' opinions of continuing education Stuber Kent J [email protected] Jaroslaw P [email protected] Dean L [email protected] Paul [email protected] Private practice KS: Calgary, AB, Canada, DS: Eaton, OH, USA, PP: Rocky Hill, CT, USA2 Post-graduate faculty, University of Bridgeport Chiropractic College, Bridgeport, CT, USA3 Division of Continuing Education, Canadian Memorial Chiropractic College, Toronto, ON, Canada4 Department of Physical Education, Health and Sports Studies, Miami University, Oxford, OH, USA2005 21 10 2005 13 22 22 9 9 2005 21 10 2005 Copyright © 2005 Stuber et al; licensee BioMed Central Ltd.2005Stuber et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
Continuing Education (CE) for chiropractors is mandatory for licensure in most North American jurisdictions. Numerous chiropractic colleges have begun collaborating with universities to offer master's degree programs. Distance education master's degree programs may be desirable to allow full-time practicing doctors to further their post-graduate education. The present survey sought to answer three questions. First, what is the level of satisfaction of chiropractors with their continuing education? Second, what is the level of interest of chiropractors in online master's degree programs? Lastly, what is the response rate of chiropractors to an online survey?
Methods
An online survey consisting of 22 multiple choice questions was e-mailed to 1000 chiropractors randomly selected from the mailing list of an online chiropractic newsletter. Upon completion of the questionnaire, participants' answers were saved on a secure site. Data analysis included evaluation of the demographic characteristics of the respondents, their opinions of and patterns of taking CE including online education, preferred learning formats, and their interest in proposed online master's degree programs. A survey response rate was determined.
Results
Nearly 86% of respondents felt their previously completed CE courses were either somewhat or extremely satisfactory. Over ninety percent of respondents who had completed online or distance CE coursesfound them to be somewhat or extremelysatisfactory. Almost half the respondents indicated that they most preferred online distance learning, while 34.08% most preferred face-to-face interaction. Fifty-three percent of respondents indicated an interest in starting a master's degree program; however 70.46% of respondents were interested in an online master's degree program that would offer CE credit. A response rate of 35.8% was obtained.
Conclusion
Satisfaction among chiropractors with CE programs is high. The notion of completing a part-time online master's degree (or online combined with face-to-face interaction) appears to be popular among respondents, with a M.Sc. in Chiropractic Sciences being the most popular of those mentioned. Online surveys are a viable method of obtaining opinion in a cost and time efficient manner; there are some sources of bias involved in this type of research, and numerous steps need to be taken to obtain a suitable response rate.
==== Body
Background
The objective of this study was to conduct an online survey of chiropractors to determine their satisfaction with current chiropractic continuing education programs and to ascertain their level of interest in a variety of online master's degree programs particularly online continuing education programs. This study also sought to assess the response rate of chiropractors to online surveys.
There is a paucity of information on opinions of chiropractors through survey studies regarding online education. The recent NBCE Job Analysis of Chiropractic [1] provides some minimal recent data. The input of practicing chiropractors as well as educators and regulators is important in developing and sculpting the future programs for chiropractors. Without appropriate surveys and environmental scans there is the danger of creating programs that are of no interest or practicality to field practitioners.
As with most health care professions, chiropractors in North America have mandatory continuing education requirements in most jurisdictions [2]. These continuing education requirements can be obtained through online continuing education courses, live seminars, workshops, and courses, and an assortment of other distance learning formats (telephone conferencing, text-based courses). In many provinces and states, continuing education courses are approved if offered or sponsored by a chiropractic college accredited by the Council on Chiropractic Education (CCE) or approved by the state/provincial regulatory board.
Many chiropractors choose to augment their professional education (Doctor of Chiropractic) with courses that can certify them as proficient in certain techniques or aspects of practice (such as the Certified Chiropractic Sports Physician certificate program). Others seek specialist level training, and this training can be obtained as either part of a 2–3 year residency program or in a part-time weekend course format over the course of three years (several of these programs now have distance learning components available as well). Organizations that offer these programs aim to provide more detailed or relevant information on topics such as: orthopedics, nutrition, neurology, pediatrics, etc. than an undergraduate chiropractic program. There is a problem with this state of affairs: most health care professionals (including chiropractors) do not know the value or significance of these "letters." It is common in the chiropractic literature to see numerous designations listed by an individual author – What does this mean to the reader? Is this person an expert? Is this person a specialist?
This trend has been called "credential inflation" [3] – many journals now only publish terminal degrees (MD, DC, PhD etc.), or none at all. Objectively assessing the expertise of doctors holding these certifications requires knowledge regarding the process of attaining the certification. The most common areas of certification include: orthopaedics, neurology, paediatrics, sports and rehabilitation. Some are related (e.g. CCSP leading to DABCSP) and some are proving difficult to track down.
It is fundamental to define what "chiropractic certification/specialty program" means. These are programs offered to chiropractors or chiropractic students who are enrolled in, or have graduated from, any accredited chiropractic college/university anywhere in the world. Upon completion of the program, the right to use the acronym of the particular specialty board, governing body, or program is granted to the doctor. There are numerous organizations within the chiropractic profession that offer these "specialty" certifications. Some are offered through chiropractic college or university settings, while others are offered through private organizations.
Chiropractic specialties are often compared to, and promoted as, the equivalent of medical specialties. This idea deserves further discussion, as certain standards should be met in order to legitimately achieve specialist standing. It is logical to assume that a specialist is someone who practices a specialty after receiving advanced clinical training. That is, someone who allots a significant portion of his or her practice to the specialty, and has significantly advanced knowledge and training in a particular area compared to an average chiropractor. This may only apply to a small number of chiropractic "specialties". These, along with other attributes of defining oneself as chiropractic "specialist", have been suggested by Nelson and Lawrence [3], and are listed in Table 1.
Table 1 Suggested Criteria to Define a Chiropractic Specialist
1) The training of a specialist should be substantially greater than that of the average chiropractor.
2) The training should consist primarily of actual patient contact, not simply repeated lectures (i.e. lectures they have already had in their undergraduate education).
3) A substantial part of the chiropractor's practice should be devoted to the specialty.
4) It should be possible to fail the certification process.
5) There should be some scholarly effort in the area of specialization: research, publication, or some other recognized contribution to the field.
Adapted from: Nelson C & awrence DJ. Degree and certification proliferation and the JMPT. Journal of Manipulative and Physiological Therapeutics 1995; 18(2): 55–56.
Table 2 Chiropractic Colleges, Partner Universities, Degrees and Study Format Offered.
Chiropractic College Partner University Degree Offered Study Format Offered
New York Chiropractic College None Master of Science in Diagnostic Imaging Full-time, on-site, residential program
None Master of Science in Acupuncture and Oriental Medicine Full-time, on-site, residential program
None Master of Science in Acupuncture Full-time, on-site, residential program
National University of Health Sciences A.T. Still University Master of Public Health (MPH) On-line distance, part-time
Palmer College of Chiropractic None Master of Science in Anatomy Full-time, on-site, residential program
University of Iowa Master of Science in Clinical Research Full-time, on-site, residential program
Anglo-European Chiropractic College University of Portsmouth MSc in Advanced Professional Development in Chiropractic Tutorials, workshops, seminars, practical skills classes, small group work, and lectures. Part-time.
MSc in Advanced Professional Development in Chiropractic Pediatrics Tutorials, workshops, seminars, practical skills classes, small group work, and lectures. Part-time.
Southern California University of Health Sciences Royal Melbourne Institute of Technology University Master of Applied Science in Musculoskeletal Management Text-based distance, part-time. Distance education programs are designed as self-contained learning packages. Group sessions are scheduled periodically and communications are transmitted through e-mail and the internet.
Having said that, a master's degree offered by an accredited institution can potentially provide its graduates credentialed recognition (in areas such as health policy development, health administration, or research) and opportunities not available to those who complete a chiropractic specialty program. Chiropractors taking graduate-level courses from a non-chiropractic college or university are at times unable to obtain continuing education credit for their coursework. To counter this problem, numerous chiropractic colleges have begun either on their own or in collaboration with universities to offer master's degree programs. This collaboration allows chiropractors in these programs to obtain recognized graduate level education, training, and qualification while still obtaining continuing education credit (as the courses are offered by a chiropractic college). It is noteworthy to point out that master's degrees can be obtained in both chiropractic and non-chiropractic fields (see Table 2). Table 2 lists the chiropractic schools that are members of the Association of Chiropractic Colleges that currently offer a part-time master's degree program, their university partners, and the program and format offered at the time of publication.
Several of the programs listed in Table 2 require classroom interaction and instruction, and may be difficult for the practicing clinician to complete without taking extended time away from their practice. Predominantly distance learning master's degree programs may be desirable to allow full-time practicing doctors to obtain graduate level education. Numerous studies and reviews have found that online continuing education methods are at least as effective in increasing participant knowledge as traditional continuing education methods [4,5]. In fact, one recent randomized controlled trial found that evidence-based online continuing medical educationcan produce objectively measured changes in behavior as well as sustained gains in knowledge that are comparable or superior to those realized from effective live activities [6].
Methods
An online survey consisting of 22 multiple choice questions was e-mailed to 1000 chiropractors (out of a pool of over 40,000) randomly selected by computer from the mailing list of ChiroWire. This is an online newsletter sent out by a continuing education provider for chiropractors in the United States and Canada. The survey was uniquely generated. Although the reliability of this instrument was not assessed, face and content validity was accepted by a panel of 3 chiropractors and a psychologist.
We surveyed 1000 chiropractors with the goal of achieving a sample size of 384. The target sample size was determined by using an alpha level of .05, with a conservative estimate of proportion (proportion of the population answering a particular way) equal to .50. We used the method proposed by Pittenger to calculate our survey sample size [7]. Hawk et al used a comparable method and determined that 380 chiropractors would be a desirable sample size for a similar survey [8]. Since we planned on using our results to describe chiropractors' opinions of CE as opposed to implementing a specific change in behavior with respect to CE, we were willing to tolerate a larger margin of sampling error [9].
At the time of conducting the study, it was not known what proportion of doctors on the mailing list had actually taken online CE programs in the past. This convenience sample of doctors was asked for their participation in the study by way of an e-mail message with the subject line of "Chiropractic continuing education study – please participate". The e-mail message consisted of an introduction, explaining the purpose of the survey, along with a link to an electronic informed consent form. After reading the introductory letter, the recipient could click on the link to the informed consent form if they wanted to participate in the study. The subject was then asked to read the informed consent form. If the subject consented to participation they clicked on an "I Consent" button at the end of the informed consent form that immediately linked them to a secure website consisting of the survey for completion (the secure page was hosted by ). If the subject did not consent to participation, they were asked to erase the message.
The email containing the link to the survey was sent to the database only once. No follow-up emails or other contact was attempted by the researchers. Doctors who responded within twelve days from the email were included in the survey results. Those who agreed to participate completed the 22 multiple choice questions (by simply 'pointing and clicking' on their answer) and then clicked on a "Submit" button upon completion of the survey. Results were stored by until data analysis took place. Survey responses were anonymous and subjects were given the option of contacting the authors before or after completion of the study to have any questions answered. The online survey software has cookie technology and respondent session keys that prevented multiple responses from the same user. The questions in the survey consisted of demographic information, questions about preferred learning styles for continuing education, and which master's degree format and program content they would be most interested in. Ethics approval for this study was obtained from the University of Bridgeport Institutional Review Board.
Data analysis consisted of determining the demographic characteristics of the respondents, their opinions of and patterns of taking continuing education programs including online programs, and their interest in proposed master's degree programs and formats. A survey response rate was also determined.
Results
Demographics
Some respondents did not answer all of the questions; the number of respondents to several of the questions is indicated in the respective figures. Palmer College of Chiropractic alumni were the most common respondents (19.55%), followed by Life University (14.8%), National University of Health Sciences (10.34%), Logan College of Chiropractic (7.54%), Cleveland Chiropractic College (6.42%), Western States Chiropractic College (5.87%), and New York Chiropractic College (5.03%) (the preceding colleges had alumni who represented greater than five percent of respondents to this survey). The highest percentage of respondents had been in practice for 6–10 years (20.28%), followed closely by those in practice 16–20 years (19.44%), 21–25 years (18.03%), 11–15 years (17.46%), 0–5 years (16.34%), and finally by those in practice for 26 years or greater (8.45%).
Over two thirds (67.6%) of the respondents to this survey had a bachelor's degree, while 10.56% had master's degrees and 1.76% had a Ph.D. Nearly thirty percent (27.89%) of respondents have completed a diplomate program or equivalent indicating a practice specialty, and 10.42% were in the process of completing one.
Continuing education opinions and practices
Nearly ninety-five percent (94.65%) of respondents indicated that they had mandatory continuing education requirements for re-licensure in their state or province, compared with 1.69% who indicated that continuing education was recommended, and 3.66% who indicated that their jurisdiction had no continuing education requirements. When asked how many continuing education hours they felt that chiropractors should have to complete annually, the most common response was eleven to fifteen hours by 37.61% of the respondents as seen in figure 1.
Figure 1 Opinion of number of hours of continuing education that chiropractors should be required to complete annually (n = 351) and number of hours actually completed by respondents (n = 354).
With respect to the importance of chiropractic continuing education, greater than ninety five percent of responding chiropractors (95.24%) felt that CE was important, including 72.83% who felt that it was very important (figure 2). When asked which areas of study were most important in continuing education the doctors indicated that internal disorders & prevention (9.87%), neurology (9.66%), and nutrition (9.62%) were the three most important areas (figure 3). Additional topics of study were asked about, but we chose to report only those that were selected by greater than five percent of respondents (respondents could pick as many responses as they desired).
Figure 2 Opinion of importance of continuing education for chiropractors (n = 357).
Figure 3 Areas of study that are most important for chiropractic continuing education (more than one answer could be selected) (only answers with greater than 5% response shown) (n = 2463).
Figures one and four depict the number of continuing education hours completed in the previous year by the respondents, with greater than 26 hours being the most common response, by over one third of the doctors (33.9%). When asked about the number of online or distance continuing education hours completed in the previous year (figure 4), close to thirty percent (29.38%) of respondents did not do any online or distance continuing education, while approximately one in four (25.14%) completed one to five hours of online or distance CE.
Figure 4 Number of hours of chiropractic continuing education and online or distance continuing education completed in the previous year (n = 354).
When asked for their opinion of continuing education courses that they had taken, over eighty percent (85.64%) of respondents replied that they were satisfactory (figure 5). Just over six percent (6.2%) of respondents felt that the courses they had completed were unsatisfactory. With respect to the doctors' opinions of online or distance continuing education courses that they had completed (figure 5) over ninety percent (90.36%) of those who had done some online or distance CE course responded that they were satisfactory. Slightly more than three percent (3.21%) felt that they were unsatisfactory. Over one in five respondents (20.45%) from the entire sample indicated that they hadn't taken any online or distance continuing education courses and could not comment.
Figure 5 Opinion of completed continuing education programs, overall (n = 355) and online or distance (n = 280).
When asked for their preference among different formats to fulfil their continuing education hours, close to fifty percent (46.48%) of the respondents indicated that they most preferred online distance learning, while over one third (34.08%) most preferred face-to-face interaction in the form of seminars, classes, and workshops. Nearly eighteen percent (17.75%) of the respondents indicated no preference among the different continuing education formats, while the remaining respondents preferred either telephone conferencing (0.28%) or text-based distance learning (1.41%).
The doctors were also asked for their opinions on the following statement, "One of the things that frustrates me with online or distance chiropractic continuing education programs is that I cannot get anything out of it to put on my resume, like a certificate, diplomate, or master's degree" (figure 6). Over one-third (36.52%) of respondents were in agreement with that statement. Nearly one quarter (23.87%) of respondents disagreed with the statement. Eleven percent of respondents indicated that they had not completed any online or distance continuing education programs and could not comment.
Figure 6 Response to the following statement, "One of the things that frustrates me with online or distance chiropractic continuing education programs is that I cannot get anything out of it to put on my resume like a diploma, etc." (n = 356).
Master's degree program interest
Over half (52.8%) of the respondents indicated an interest in starting a master's degree program; while over twenty percent (20.11%) of the respondents indicated that they were uninterested in a master's degree program. The remaining respondents either had no opinion one way or the other (21.51%), or were already in or had completed graduated studies (master's or PhD level) (5.59%). Nearly all of the respondents (96.91%) indicated being most interested in a part-time master's degree program in one form or another. This includes almost two thirds (63.16%) who indicated being most interested in a part-time online distance program, and nearly one third (31.27%) who indicated being most interest in a combination of online distance learning and face-to-face interaction in the form of seminars, classes, etc. Less than five percent (3.10%) of the respondents indicated that they were most interested in a full-time master's degree program.
Over forty percent (43.81%) of the respondents indicated being most interested in a Master of Science degree in Chiropractic Sciences, compared with 20.85% each for a Master of Science degree in Pain Management and a Master of Science in Rehabilitation. Close to fifteen percent (14.50%) of respondents indicated a preference in "Other" master's degree programs. We are unable to determine specifically what that may mean, other than those doctors were likely not interested in any of the three programs listed. Nearly eighty-five percent (84.59%) of the respondents indicated that if they were to take a master's degree program that they would rather take an elective course than complete a thesis or dissertation.
When asked if they would be interested in starting an online master's degree program with which they could obtain continuing education credit, over seventy percent (70.46%) of the respondents indicated an interest. One in eight (12.5%) respondents indicated that they were uninterested in such a program. The remaining 17.04% were neither interested nor disinterested (13.92%), or else they had already completed or were working on a master's degree or PhD (3.12%).
Survey response rate
The response rate was calculated according to the following formula:
Calculated response rate = number of respondents/number of people who received the survey.
Calculated response rate = 358/1000 = 0.358 = 35.8%
Our final sample size of 358 was close to our target size of 384. Since we planned on using our results to describe chiropractors' opinions of CE as opposed to implementing a specific change in behavior with respect to CE, we were willing to tolerate a larger margin of sampling error [9]. Therefore, our final sample size was deemed satisfactory.
Discussion
As stated earlier, the opinions of chiropractors surveyed assist in defining needs and developing the appropriate programs for improved clinical education. A patient-centred approach is both sought and preferred.
In the current NBCE survey (Job Analysis Survey NBCE 2005), 42.3% of chiropractors had practiced five to fifteen years [1]. The present study results show 37.74% of respondents who have been in practice from six to fifteen years, indicating that the results are similar. In 2003 the average chiropractor was in practice 15.6 years [1].
Nearly 60% (59.6%) of chiropractors have bachelor's degrees [1], the finding in the present study show a slightly higher number at 67.6%. The portion of chiropractors with bachelor's and master's degrees has increased [1]. Over six percent (6.3%) of chiropractors have master's degrees and 1.6% have doctorates in non-chiropractic fields [1]. Our results show that 10.56% of respondents have a master's degree and 1.76% have a Ph.D. The current results suggest that more chiropractors today have graduate level training than in the past. The two surveys (NBCE survey and our online survey) share similar results. The few minimal discrepancies between the two studies are most likely due to number of respondents and the resulting power of the studies. The NBCE Job Analysis surveyed 10,189 chiropractors out of a pool of 68,799 [1]. 3,370 (33.3%) responses were received [1]. In our study we received 358 (35.8%) responses from a pool of 1,000.
Specialization
In the NBCE Job Analysis Survey, 64% of respondents had no diplomate status or equivalent specialty certification. The present survey obtained a considerably higher percentage of respondents who indicated diplomate status at 27.89%, compared with 14.4% reported previously [1]. Reasons for the difference between the present survey and the NBCE results are unknown.
Continuing education
The 2005 NBCE Job Analysis Survey provided data regarding CE and other professional education activities. Most relevant to the present survey was that 97.1% of chiropractic practitioners continue their professional educational by attending conferences and seminars, 30% of chiropractic practitioners attend diplomate courses to continue their education, and 12.7% respondents indicated that they had participated on online credit courses [1]. The results from the present study show that 70.62% of respondents had completed some online or distance CE in the previous year, a much higher result than the NBCE survey.
The 2003 NBCE data indicates that chiropractors took more units of continuing education than the previous survey in 1998 [1]. One-fifth of respondents (19.4%) had completed 11–15 hours of CE, compared with 23.73% in the present study. In the NBCE survey, nearly one quarter (22.9%) of respondents had completed 16 to 20 hours, slightly higher than the 16.95% in our survey. One-fifth (21.6%) of NBCE respondents had completed 21 to 25 hours, comparable with our result of 19.49%. Almost thirty percent (29.2%) of NBCE respondents had completed 26 or more hours, a result similar to our 33.9%.
The results of this survey indicate that chiropractors overwhelmingly feel that continuing education is important. This agrees with Bolton's findings that chiropractors have a positive attitude towards continuing professional education along with continuing professional development and are aware of the need to keep current on new knowledge and technology [10]
The most popular choice among the respondents in this survey for the number of annual continuing education hours that chiropractors should complete was eleven to fifteen hours. Interestingly, the most frequently indicated choice for the number of continuing education hours actually completed in the previous year was twenty-six hours or greater. These results are difficult to compare with Bolton's findings that 40.1% of surveyed chiropractors regularly attended continuing professional education/development (CPE/D) courses and 52.6% occasionally attended CPE/D courses, while 7.3% never attended CPE/D [10]. Certain areas of continuing education study appear to be considered more important by the respondents, particularly internal disorders and prevention, neurology, nutrition, rehabilitation, clinical sciences, and orthopedics.
Among those indicating a preference, online distance was favored most by those surveyed, followed by face-to-face seminars and classes. These results differ from Cobb's review of online continuing health professional education and Stancic et al's study where both studies found that in-person continuing education was the most popular format of continuing education learning over online learning, although it was noted that the popularity of online formats was increasing [4,11]. In contrast, our sample would prefer to take online CE compared to other formats such as live, face-to-face interaction. These results indicate that chiropractic colleges and continuing education providers certainly should consider offering online distance education programs, but traditional face-to-face seminars, conventions, and classes should still be offered as well.
Nearly 30% of those surveyed indicated that they had not completed any online continuing education in the previous year, while approximately one in four indicated that they completed one to five hours of online continuing education in the past year (meaning that close to 55% of respondents completed five hours or less of online continuing education in the previous year). This result is lower than the results from Casebeer et al's survey of physicians that found that 70% of those surveyed rarely or never accessed CME online, perhaps indicating that chiropractors access online continuing education courses more than physicians [12].
While online continuing education appears to be popular among chiropractors, most of the doctors are not completing tremendous numbers of online continuing education hours. This is demonstrated by the fewer than 15% of the doctors who indicated that they completed sixteen or more hours of online continuing education in the previous year. Stancic et al found that there were four factors that helped a physician to determine whether they take a continuing education program: cost, personal control over content, personal development, and convenience of access [11]. Casebeer et al found that physicians desire online professional development courses that are easy-to-use, relevant, valid in content, credible, and available for immediate access [12]. Online continuing education providers need to design courses with these factors and wishes in mind, as these qualities are likely important to all groups of health care professionals.
A very strong majority (>85%) of the chiropractors surveyed appear to be satisfied with the continuing education courses that they have completed thus far. Another strong majority (>90%) of the chiropractors indicated satisfaction with online or distance continuing education courses they had completed. This is in agreement with what Cobb found in her literature review of online continuing health education, in that health care professionals were satisfied with computer-based continuing education programs that they had participated in [4].
Of those who had completed online continuing education nearly 37% agreed with a statement indicating frustration with not obtaining certificates, diplomate status, or a master's degree from online or distance continuing education programs. This would seem to indicate that those offering online CE programs should consider the possibility of offering diplomate, master's degree, or certifications.
The results of this survey indicate that the chiropractors who responded generally appear to have an interest in completing a master's degree program (greater than 50%). This number increases to over 70% if the proposed program would be online and count towards continuing education credit, including greater than one in three respondents who were very interested such a program. There is also much greater interest from respondents in part-time master's degree programs than full-time programs, particularly if those part-time programs are either entirely or partially online. This is not an unexpected result as it is likely that most of the chiropractors completing our survey are in full-time practice (although we did not ask this question of respondents) and leaving practice to pursue a master's degree full-time may not be feasible. A part-time completely online master's degree program was of greatest interest for respondents, being preferred at a rate of over 2 to 1 to a program that would combine online learning with face-to-face interaction in the form of seminars or classes.
An M.Sc in Chiropractic Sciences was preferred at a rate of greater than 2 to 1 when compared to either an M.Sc in Pain Management or an M.Sc in Rehabilitation, for whom doctors had equal preference. The respondents in this survey showed a greater preference towards completing elective courses in a master's degree program when compared to completing a thesis or dissertation. This is not surprising as it is most likely that field practitioners may not have the interest or time necessary for conducting the research needed to be done to complete a thesis or dissertation. A question that a follow-up study could consider is which areas would be of most interest for an elective course.
Online survey response
The calculated response rate for this study was 35.8%. This is within the range reported by Russell et al (7.0% to 91.4%) for surveys of chiropractors [13]. However, this is not an extremely high response rate and thus a potential source of bias due to non-response. There are numerous ways that the response rate in this study could have been improved. First, the survey was only sent to the sample population once. Sending the survey numerous times or sending out reminders out to those sampled may have improved the response rate, as mentioned by Russell et al [13]. Use of advance notice is also known to lead to higher response rates. We did not employ this technique.
An additional issue in this study is that the initial e-mail inviting each doctor to participate was sent to 1000 random chiropractors out of a pool of 40,000. A blinded randomizing program was used to select those who received the e-mail containing the survey and thus it could not actually be ascertained as to who received e-mails and who did not. This would make it extremely difficult to re-send the survey or send a reminder. It is possible to create a blinded program that would randomly select those to be surveyed and send them an initial e-mail as advanced notice of the survey, followed by an e-mail containing the survey; followed by periodic e-mails containing reminders and/or additional copies of the survey. However, this was not done in this study. Finally, by only allowing 12 days for subjects to respond to the survey before tallying the data, this study may not have yielded the optimal number of responses. A longer period for responses combined with advanced notice, reminder e-mails and/or additional copies of the survey may allow for higher response rates.
Online surveys are a potentially attractive option for opinion based survey research in the future. They have the potential to be cost-effective, as the costs of postage (and return postage) and copying is not a factor. However, in the case of this survey, a third-party had to be contracted to construct the survey (after they were provided with the questions) and to receive and tally the results from the subjects. It is possible that this could be done from a single e-mail address or web page set up by the authors, but having a service perform this function worked in this case and the cost involved was reasonable, certainly less than the above-mentioned costs for postage and copying. Online surveys can also be sent out and responses received quickly. As mentioned previously, this study obtained 358 responses in 12 days; this would take far longer using conventional postal surveys.
One potential drawback or source of bias in online surveys is that they can only be taken by internet capable individuals. It is possible that older doctors or other doctors who do not frequently use the internet or do not have internet access would not be able to be surveyed in this manner. Another potential source of bias in this study is that the sample came from a group of chiropractors who signed themselves up with an online continuing education company. It is reasonable to assume that these individuals may be more interested in online continuing education than other doctors who have not signed up with such a service.
A final source of bias is that the online continuing education company whose subscribers were surveyed is an American company, and this could explain the small numbers of Canadian and international doctors responding (4.23% and 3.10%, respectively). As mentioned previously, some respondents did not answer all of the questions, and so the complete results from those surveyed were not obtained. It is unknown as to why some doctors chose not to answer some questions.
Future studies could delve into whether there is a difference in continuing education patterns and opinions between Canadian and American doctors, and graduates of different schools.
Conclusion
The satisfaction rate among chiropractors taking CE programs (including online programs) is high. The notion of completing a part-time online master's degree (or online combined with in class sessions) appears to be popular among respondents, with a M.Sc. in Chiropractic Sciences being most popular among respondents. Finally, online surveys are a viable method of obtaining opinion in a cost and time efficient manner. There are some sources of bias (e.g. sampling bias) involved in this type of research, and numerous steps need to be taken to obtain a suitable response rate.
Chiropractors feel that continuing education is important. Among chiropractors surveyed, online CE is the most preferred method of learning over live lectures or workshops, and other methods of distance learning. Certain areas of chiropractic CE appear to be more important to those surveyed, particularly internal disorders and prevention, neurology, nutrition, rehabilitation, clinical sciences, and orthopedics.
Competing interests
KJS: post-graduate online instructor for chirocredit.com, an online continuing education provider sponsored by the University of Bridgeport Chiropractic College, author of an online continuing education course for Canadian Memorial Chiropractic College's Division of Continuing Education.
JPG: Director of Canadian Memorial Chiropractic College's Division of Continuing Education
DS: post-graduate online instructor for chirocredit.com, an online continuing education provider sponsored by the University of Bridgeport Chiropractic College
PP: post-graduate online instructor for chirocredit.com, an online continuing education providersponsored bythe University of Bridgeport Chiropractic College; CEO of OnlineContinuingEd, LLC, anonline continuing education provider.
Authors' contributions
KJS was involved in study conception, study design, data analysis and interpretation, and manuscript preparation and revision. JPG carried out data analysis and interpretation, and was involved with manuscript preparation and revision. DLS took part in study conception, study design, and manuscript preparation and revision. PP was involved with study conception, study design, data acquisition, and manuscript revision. All authors read and approved the final manuscript.
Acknowledgements
The authors wish to acknowledge and thank the Canadian Memorial Chiropractic College Division of Continuing Education and ChiroCredit.com, a division of OnlineContinuingCE, LLC, for funding this study.
==== Refs
Christensen M Kollasch M Job Analysis of Chiropractic 2005 Greeley, CO: National Board of Chiropractic Examiners 78 89
Pallister S Continuing education for chiropractors in Canada J Can Chiro Assoc 1989 33 121 122
Nelson C Lawrence DJ Degree and certification proliferation and the JMPT Journal of Manipulative and Physiological Therapeutics 1995 18 55 56
Cobb SC Internet continuing education for health care professionals: an integrative review J Contin Educ Health Prof 2004 24 171 180 15490549 10.1002/chp.1340240308
Wutoh R Boren SA Balas EA E-learning: a review of internet based continuing medical education J Contin Educ Health Prof 2004 24 20 30 15069909 10.1002/chp.1340240105
Fordis M King JE Ballantyne CM Jones PH Schneider KH Spann SJ Greenberg SB Greisinger AJ Comparison of the instructional efficacy of internet-based CME with live interactive CME workshops: a randomized controlled trial J Am Med Assoc 2005 294 1043 1051 10.1001/jama.294.9.1043
Pittinger DJ Behavioral Research Design and Analysis 2003 New York, NY: McGraw-Hill 196 197
Hawk C Dusio ME A survey of 492 US chiropractors on primary care and prevention-related issues Journal of Manipulative and Physiological Therapeutics 1995 18 57 64 7790784
Bordens KS Abbott BB Research Design and Methods: A Process Approach 2002 New York, NY: McGraw-Hill 247 248
Bolton JE Chiropractors' attitudes to, and perceptions of, the impact of continuing professional education on clinical practice Med Educ 2002 36 317 324 11940171 10.1046/j.1365-2923.2002.01164.x
Stancic N Mullen PD Prokhorov AV Frankowski RF McAlister AL Continuing medical education: what delivery format do physicians prefer? J Contin Educ Health Prof 2003 23 162 167 14528787 10.1002/chp.1340230307
Casebeer L Bennett N Kristofco R Carillo A Centor R Physician internet medical information seeking and on-line continuing education use patterns J Contin Educ Health Prof 2002 22 33 42 12004639 10.1002/chp.1340220105
Russell ML Vehoef MJ Injeyan HA McMorland DG Response rates for surveys of chiropractors J Manipulative Physiol Ther 2004 27 43 48 14739873 10.1016/j.jmpt.2003.11.005
|
16242035
|
PMC1282582
|
CC BY
|
2021-01-04 16:38:23
|
no
|
Chiropr Osteopat. 2005 Oct 21; 13:22
|
utf-8
|
Chiropr Osteopat
| 2,005 |
10.1186/1746-1340-13-22
|
oa_comm
|
==== Front
Clin Mol AllergyClinical and molecular allergy : CMA1476-7961BioMed Central London 1476-7961-3-141625314010.1186/1476-7961-3-14Case ReportSevere allergic reactions to guinea pig Zacharisen Michael C [email protected] Michael B [email protected] Jeffrey L [email protected] Viswanath P [email protected] Section of Allergy/Immunology, Department of Pediatrics, Medical College of Wisconsin, Milwaukee, Wisconsin, USA2 University Pediatrics and Family Allergy, Huntington, West Virginia, USA3 Zablocki VA Medical Research Center, Milwaukee, Wisconsin, USA2005 27 10 2005 3 14 14 15 6 2005 27 10 2005 Copyright © 2005 Zacharisen et al; licensee BioMed Central Ltd.2005Zacharisen et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
Allergic sensitization and reactions to guinea pig (Cavia porcellus) have been well documented in laboratory animal handlers, primarily manifesting as rhinitis, conjunctivitis, and asthma. Severe allergic reactions, however, are rare.
Methods
We report two patients with severe allergic reactions following non-occupational exposure to guinea pigs. The first patient, an 11-year-old female, developed ocular, nasal, skin and laryngeal edema symptoms immediately after handling a guinea pig. The second patient, a 24-year-old female, developed symptoms of isolated laryngeal edema after cleaning a guinea pig cage. Percutaneous skin testing, RAST, ELISA and ELISA inhibition testing with guinea pig extract were performed.
Results
Both patients had IgE-mediated allergy to guinea pig confirmed by ELISA and either RAST or skin testing. ELISA inhibition studies confirmed the specificity of the IgE reactivity to guinea pig.
Conclusion
Severe IgE-mediated reactions can occur following non-occupational guinea pig exposure. Physicians should be aware of this possibility.
guinea pigallergy
==== Body
Introduction
Guinea pigs are popular household pets and also used in laboratory research. Allergic symptoms including rhinitis, conjunctivitis, and asthma have been documented in laboratory animal workers exposed to guinea pigs [1-5]. An extensive review of the literature revealed no reports of severe allergic reactions resulting from guinea pig exposure. We report two patients with severe allergic reactions following direct exposure to guinea pigs in domestic settings.
Case Reports
Case One
An 11-year-old female with a history of migraine headaches and exercise-induced asthma (EIA) was evaluated in the Allergy clinic two months after experiencing symptoms while holding a guinea pig at her hairdresser's home. This was the only episode of symptoms associated with guinea pig exposure; she had handled the pet previously without exhibiting symptoms. Within minutes of holding the guinea pig, she developed ocular itching, lacrimation, and periorbital angioedema. Symptoms rapidly progressed to facial urticaria and angioedema, rhinorrhea, throat tightness, and dyspnea. She had difficulty speaking, repeatedly attempted to clear her throat, and expressed feelings of impending doom. There was no coughing or audible wheezing.
Treatment within 20 minutes included diphenhydramine 25 mg, followed by nebulized albuterol at an urgent care clinic. Although her vital signs were not available, her mother denied she had low blood pressure. The symptoms resolved within two hours without recurrence. Epinephrine and corticosteroids were not administered.
She has since avoided guinea pig exposure and has had no further symptoms other than those related to EIA. She reported previous casual exposure to guinea pigs without adverse reaction, but had not kept any rodents as pets. She denied symptoms on previous exposures to dogs, cats, and other caged rodents.
Her past medical history was significant for episodic bronchitis, croup, sinusitis, and migraine headaches. She had no history of perennial or seasonal rhinitis. Her only medication was propranolol for migraine prophylaxis. Prior and current spirometry was normal. An exercise challenge within the past year was consistent with EIA. Physical examination, at the time of the evaluation, was significant for allergic shiners and pale, swollen inferior nasal turbinates. Her lungs were clear, and the remainder of her examination was normal.
Case Two
A 24-year-old female smoker with allergic rhinitis, EIA, and known cat-induced rhinitis was evaluated for a several year history of perennial rhinitis and conjunctivitis. She also described an episode of severe allergic symptoms resulting from guinea pig exposure. Within minutes of cleaning her pet guinea pig's cage, she developed throat tightness, severe dyspnea, and anxious feelings. She denied coughing, wheezing, and urticaria. A feeling of "impending doom" was not specifically stated. Her symptoms resolved spontaneously one hour after departing outdoors. She did not take medication or seek medical attention.
Her past medical history was significant for irritable bowel syndrome and gastroesophageal reflux disease. Daily medication included sertraline, nasal fluticasone, and oral contraceptives. Physical examination was significant for bilateral serous otitis media and edematous nasal turbinates. Her lungs were clear, and the remainder of her examination was normal. Spirometry was equivocal due to submaximal effort.
Methods
Case One
Commercial radioallergosorbent testing (RAST) [Quest Diagnostics, San Juan Capistrano, CA] was performed to a variety of animal and environmental antigens. Animal antigens included dog, cat, cow, gerbil, goat, hamster, horse, mouse, rabbit, rat, sheep, swine, and guinea pig. Environmental allergens included common pollens and molds, as well as dust mites. Percutaneous skin testing was not performed due to concurrent usage of beta-blocker medication.
Case Two
Percutaneous skin testing with a variety of environmental allergen extracts (Greer Labs, Inc., Lenoir, NC), including cat, dog, and guinea pig antigens, was performed using DermaPIK (Greer Labs, Inc., Lenoir, NC). Histamine and albumin-saline controls were included. Commercial RAST testing was not performed.
Both Cases
Sera from both patients and three non-atopic, adult controls were assayed for specific IgE to a freshly prepared extract from the fur of a guinea pig by enzyme-linked immunosorbent assay (ELISA) as previously described [6]. The guinea pig antigen was prepared by extracting 46 mg of fur with 7.5 ml of sterile phosphate buffered saline (PBS) incubated overnight at 4°C. The extract demonstrated 40 μg/ml of protein by bis-cinchoninic acid (BCA) protein assay. ELISA inhibition testing was performed as follows. 100 μl of sera (1:20) in PBS and 5% milk was incubated with 50 μl of several concentrations of guinea pig extract for one hour at room temperature and overnight at 4°C. Immulon II polystyrene microtiter plates (Fisher Scientific, Itasca, IL) were coated with the antigen by incubating for 2 hours at room temperature and overnight at 4°C with a 5 μg/ml dilution of the extract. The plates were then washed and blocked for one hour with PBS and 0.3 % Tween 20. Sera inhibition was continued for another hour at room temperature, then 50 μl of PBS and 0.3% Tween 20 was added to the sera to prepare for ELISA. Following washing of the plates, the sera were added to the wells and incubated for 3 hours at room temperature. Plates were washed and a 1:500 dilution of biotinylated-goat anti-human IgE was added for 1 hour. Following a 1-hour incubation with 1:1000 streptavidin-labeled peroxidase, o-phenylenediamine was added for 30 minutes, and the reaction stopped with 6N sulfuric acid. The optical density (OD) was read by spectrophotometry at 490 nm.
Results
Case One
RAST to guinea pig was strongly positive (>17.5 kU/L). All other antigens tested were negative (<35 kU/L). Complete blood count was normal. Serum IgE was 64 kU/L (<114 kU/L).
Case Two
Percutaneous skin testing was positive (equivalent to histamine control with negative saline control) to guinea pig epithelium extract. Skin reactivity was also detected to cat dander as well as ragweed, grass, and tree pollens.
Both Cases
ELISA demonstrated elevated levels of serum-specific IgE to crude extracts of guinea pig fur in both patients with net optical density of 1.08 and 1.29 for case 1 and 2, respectively. There was no serum-specific IgE identified in the control sera. ELISA inhibition with guinea pig allergen resulted in complete absorption of specific IgE antibody (Figure 1). The results indicate that there is no cross reactivity between hamster and guinea pig and that the antibody detected in the two cases are antigen-specific and therefore relevant. There also was minimal inhibition with hamster extract.
Figure 1 ELISA inhibition assay with guinea pig allergen resulted in complete absorption of specific IgE antibody in both patients.
Discussion
Guinea pigs are popular household pets because of their small size and the minimal time and expense involved in their care. Two major guinea pig allergens, Cav p I and Cav p II, have been identified [7,8]. Guinea pig dust, dander, fur, urine and saliva have been found to be the more potent extracts when compared to whole pelt, feces, and serum [9]. Inhalant allergens may be derived from material shed from the guinea pig coat after contamination with saliva and urine [8]. The size of airborne particles derived from guinea pig urine and dander resulting in the most allergenic activity have been shown to be of a diameter either greater than 5 microns or less than 0.8 microns, thus small enough to penetrate the lower respiratory tract when inhaled [10]. Therefore, it is not surprising that asthma can occur when sensitized individuals are exposed to guinea pigs.
In contrast to domestic settings, laboratory animal allergy (LAA) is well documented [1-5]. Approximately one third of laboratory animal workers have occupational allergy to animal dander [3]. While rats and mice are primarily used in the laboratory setting, guinea pig use is also common [4]. In a large epidemiologic study of LAA utilizing a questionnaire, the subjects handling guinea pigs reported the highest prevalence of symptoms suggestive of LAA (31%) [5].
We have described two cases of severe allergic reactions following direct contact with guinea pigs in a domestic setting. These cases are unique because they occurred with non-occupational exposure. Both sensitization and subsequent exposures were limited to domestic environments. Both patients experienced dyspnea and symptoms consistent with laryngeal edema immediately after direct contact with a guinea pig. One patient had feelings of impending doom. ELISA confirmed elevated levels of serum-specific IgE to guinea pig. Both patients demonstrated IgE inhibition with guinea pig fur extract and minimally with hamster fur extract, confirming the specificity.
Patients experiencing severe allergic reactions should be treated initially with epinephrine followed by antihistamines and corticosteroids. Although epinephrine was not administered in these cases, fortunately both patients recovered uneventfully. Primary preventative treatment should include avoidance of the offending allergen, although strong emotional attachments to pets may make adherence to this recommendation difficult. Both patients are actively avoiding contact. In addition, patients should be prescribed self-injectable epinephrine and oral antihistamines in case of accidental exposure, and be supplied with information for obtaining a medical information bracelet. Beta-blocker medications should also be avoided if possible. The patient taking propranolol for migraine headache prophylaxis was prescribed an alternative medication.
While exposure to guinea pigs rarely causes severe allergic reactions, their presence in homes, schools, and laboratories underscores the need for physicians to be aware of this possibility.
Table 1 Test Patient 1 Patient 2
CBC Normal ND
Total serum IgE 64 lU/L ND
Guinea pig RAST* > 17.5 kU/L ND
Environmental allergen RAST* Negative ND
Guinea pig ELISA¶ Positive at 1.08 Positive at 1.29
Percutaneous skin test
Guinea pig ND Positive
Environmental allergens ND Positive#
* Commercial RAST
¶ In-house ELISA
ND Not done
Acknowledgements
The authors express their appreciation to Nancy Elms for her diligent efforts in the laboratory on behalf of this project and to Karen Lindberg for her efforts in manuscript preparation.
==== Refs
Lutsky I Neuman I Laboratory animal dander allergy: I. An occupational disease Ann Allergy 1975 35 201 205 1238037
Beeson M Dewdney J Edwards R Prevalence and diagnosis of laboratory animal allergy Clin Allergy 1983 13 433 442 6627621
Bush R Wood R Eggleston P Laboratory animal allergy J Allergy Clin Immunol 1998 102 99 112 9679853 10.1016/S0091-6749(98)70060-0
Fisher R Saunders W Murray S Stave G Prevention of laboratory animal allergy J Occup Environ Med 1998 40 609 13 9675719 10.1097/00043764-199807000-00005
Aoyama K Ueda A Manda F Allergy to laboratory animals: an epidemiological study Br J Ind Med 1992 49 41 47 1733454
Prince E Zacharisen MC Kurup VP Anaphylaxis to rabbit: a case report Ann Allergy Asthma Immunol 1998 81 272 273 9759807
Schou C Defining allergens of mammalian origin Clin Exp Allergy 1993 23 7 14 8439823
Walls A Taylor A Longbottom J Allergy to guinea pigs: II Identification of specific allergens in guinea pig dust by crossed radio-immunoelectrophoresis and investigation of the possible origin Clin Allergy 1985 15 535 546 2416489
Walls A Taylor A Longbottom J Allergy to guinea pigs: I Allergenic activities of extracts derived from the pelt, saliva, urine and other sources Clin Allergy 1985 15 241 251 4006174
Swanson M Agarwal M Yunginger J Reed C Guinea-pig-derived allergens: Clinicoimmunologic studies, characterization, airborne quantitation, and size distribution Am Rev Respir Dis 1984 129 844 849 6721283
|
16253140
|
PMC1282583
|
CC BY
|
2021-01-04 16:36:25
|
no
|
Clin Mol Allergy. 2005 Oct 27; 3:14
|
utf-8
|
Clin Mol Allergy
| 2,005 |
10.1186/1476-7961-3-14
|
oa_comm
|
==== Front
Curr Control Trials Cardiovasc MedCurrent Controlled Trials in Cardiovascular Medicine1468-67081468-6694BioMed Central 1468-6708-6-151621910010.1186/1468-6708-6-15Study ProtocolThe clopidogrel after surgery for coronary artery disease (CASCADE) randomized controlled trial: clopidogrel and aspirin versus aspirin alone after coronary bypass surgery [NCT00228423] Kulik Alexander [email protected] May Michel [email protected] George A [email protected] Thierry G [email protected] Marc [email protected] Division of Cardiac Surgery, University of Ottawa Heart Institute, Ottawa, Canada2 Division of Cardiology, University of Ottawa Heart Institute, Ottawa, Canada3 Department of Epidemiology and Community Medicine, University of Ottawa, Ottawa, Canada4 Division of Cardiac Surgery, University of Ottawa Heart Institute, Ottawa, Canada5 Division of Cardiac Surgery, University of Ottawa Heart Institute, and Department of Epidemiology and Community Medicine, University of Ottawa, Ottawa, Canada2005 11 10 2005 6 1 15 15 3 10 2005 11 10 2005 Copyright © 2005 Kulik et al; licensee BioMed Central Ltd.2005Kulik et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
Saphenous vein graft disease remains a major limitation of coronary artery bypass graft surgery. The process of saphenous vein intimal hyperplasia begins just days after surgical revascularization, setting the stage for graft atherosclerotic disease and its sequalae. Clopidogrel improves outcomes in patients with atherosclerotic disease, and is effective at reducing intimal hyperplasia in animal models of thrombosis. Therefore, the goal of this study will be to evaluate the efficacy of clopidogrel and aspirin therapy versus aspirin alone in the prevention of saphenous vein graft intimal hyperplasia following coronary artery bypass surgery.
Methods
Patients undergoing multi-vessel coronary artery bypass grafting and in whom at least two saphenous vein grafts will be used are eligible for the study. Patients will be randomized to receive daily clopidogrel 75 mg or placebo, in addition to daily aspirin 162 mg, for a one year duration starting on the day of surgery (as soon as postoperative bleeding has been excluded). At the end of one year, all patients will undergo coronary angiography and intravascular ultrasound assessment of one saphenous vein graft as selected by randomization. The trial will be powered to test the hypothesis that clopidogrel and aspirin will reduce vein graft intimal hyperplasia by 20% compared to aspirin alone at one year following bypass surgery.
Discussion
This trial is the first prospective human study that will address the question of whether clopidogrel therapy improves outcomes and reduces saphenous vein graft intimal hyperplasia following cardiac surgery. Should the combination of clopidogrel and aspirin reduce the process of vein graft intimal hyperplasia, the results of this study will help redefine modern antiplatelet management of coronary artery bypass patients.
coronary artery bypass graft surgeryantiplatelet therapysaphenous vein
==== Body
Background
Coronary artery bypass graft surgery (CABG) is the most durable approach for the treatment of ischemic heart disease [1], with >400,000 procedures performed annually in the United States alone [2]. Despite the increasing application of arterial conduits during CABG, the saphenous vein remains the most common conduit, employed for more than 70% of grafts [3]. However, saphenous vein graft (SVG) disease presents an important clinical problem. Even with aggressive medical therapy [4-11], up to 15% of vein grafts occlude in the first year after bypass surgery. Between 1 and 6 years, the graft attrition rate is 1% to 2% per year, and between 6 and 10 years it rises to 4% per year. By 10 years after surgery, only 60% of grafts are patent and only 50% of patent vein grafts are free of significant stenosis. In addition, native coronary artery disease progresses in 5% of patients annually [12-15]. Reflecting the graft and native vessel attrition, this population is at high risk for subsequent ischemic events, including death, myocardial infarction (MI) and stroke [14]. Further revascularization, either reoperation or percutaneous coronary intervention, is required in 4% of patients by 5 years, 19% of patients by 10 years and 31% of patients by 12 years after the initial bypass surgery [3,16].
The process of SVG disease is composed of three mechanistically interlinked stages: thrombosis, intimal hyperplasia, and atherosclerosis [13,14,16-18]. Early graft thrombosis can occur at the time of surgery secondary to focal endothelial disruption [19,20]. Grafts that survive this early period develop a progressive thickening of the media that begins within days after implantation. This process, termed intimal hyperplasia, is a consequence of smooth muscle cell proliferation and extracellular matrix protein synthesis [21,22]. Platelets play a fundamental role in the process of smooth muscle cell proliferation and intimal hyperplasia [23,24]. Intimal hyperplasia is present in all grafts 1 month after implantation [25] and forms a template for the development of superimposed atherosclerotic changes [17,18]. With the passage of a sufficient period of time, the thrombotic occlusion of vein grafts is almost inevitable due to progressive atherosclerosis [17].
Despite its established benefit in patients with coronary artery disease, aspirin therapy has numerous limitations. It is a relatively weak antiplatelet agent and has no effect on thrombin, which is believed to play a major role in acute coronary syndrome [26]. Even with aspirin therapy for secondary prevention, a large number of recurrent events occur [27]. A significant proportion of patients undergoing CABG may be aspirin resistant, defined as undetectable platelet inhibition after one week of therapy [28,29]. Depending on the population studied and the specific definition of aspirin resistance, anywhere from 10–40% of patients appear to have an inadequate antiplatelet response to aspirin [28,30]. Such patients appear to be at increased risk for the development of vascular events. In theory, these aspirin-resistant patients may derive particular benefit from additional antiplatelet therapy [31].
Clopidogrel is a thienopyridine antiplatelet agent that inhibits ADP-dependent platelet activation and aggregation [32]. Sevenfold more potent than ticlopidine, clopidogrel is free of its adverse side effects such as neutropenia, diarrhea and rash [33]. Unlike aspirin [24,34], clopidogrel has been shown to inhibit the process of platelet-mediated intimal proliferation and smooth muscle hyperplasia in laboratory experiments. In a cell culture model, clopidogrel significantly inhibited platelet adhesion to immobilized fibrinogen and also inhibited platelet-dependent mitogenic signaling and DNA synthesis in cultured coronary artery smooth muscle cells [35]. Similarly, in animal thrombosis models, clopidogrel but not aspirin significantly inhibited platelet-mediated intimal proliferation and smooth muscle hyperplasia [33,36]. Furthermore, the combination of clopidogrel with aspirin led to potent synergistic antithrombotic effects and a decrease in myointimal proliferation compared to either therapy alone [24,36,37].
Several large clinical trials have demonstrated that clopidogrel reduces ischemic events and mortality in patients with coronary and vascular disease [38-41]. In the CAPRIE (Clopidogrel versus Aspirin in Patients with Ischemic Events) trial, clopidogrel (75 mg/day) was demonstrated to be significantly more effective than aspirin (325 mg/day) in preventing vascular thrombotic events (ischemic stroke, MI or vascular death) in patients with clinical evidence of atherosclerotic disease (clopidogrel 9.78% vs. aspirin 10.64%, relative risk reduction [RRR] 8.7%, p = 0.045) [38]. In patients presenting with acute coronary syndromes, the CURE (Clopidogrel in Unstable angina to prevent Recurrent ischemic Events) study demonstrated that the combination of clopidogrel and aspirin was more effective at reducing the primary outcome (cardiovascular death, nonfatal MI or stroke) compared to aspirin alone (clopidogrel and aspirin 9.3% vs. aspirin alone 11.4%, RRR 20%, p < 0.001) [40]. Subgroup analysis from these trials suggested that patients that underwent surgical revascularization also benefited from clopidogrel [39,42,43]. However, no trial to date has prospectively evaluated the combined effects of clopidogrel plus aspirin on saphenous vein graft disease after CABG.
There currently exists a clinical equipoise regarding the optimal antiplatelet therapy for patients who have undergone coronary artery bypass surgery. While some clinicians believe in the beneficial effects of clopidogrel, the increased risk of bleeding and the lack of data in CABG patients make it impossible to establish definitive recommendations. We therefore propose the Clopidogrel after Surgery for Coronary Artery Disease (CASCADE) study, a randomized, placebo-controlled trial comparing clopidogrel plus aspirin versus aspirin alone in CABG patients revascularized with saphenous vein. The primary aim of this study will be to evaluate the effect of combined antiplatelet therapy on the reduction of SVG intimal hyperplasia one year after CABG, through the assessment of intimal area by intravascular ultrasound (IVUS). Secondary aims will evaluate the safety of clopidogrel administration following CABG with regards to bleeding complications. We hypothesize that the combination of clopidogrel with aspirin will reduce the SVG intimal hyperplasia (intimal area) by 20% one year post-CABG compared to the usual antiplatelet therapy of aspirin alone.
Methods
Study Population and Recruitment Procedure
The study population will include all patients undergoing multi-vessel elective or urgent CABG using at least two saphenous vein grafts at the University of Ottawa Heart Institute (OHI) over the study period (see Table 1 for inclusion and exclusion criteria). Patients undergoing off-pump CABG (OPCAB) will also be eligible for this study, as long as at least two saphenous vein grafts are used. OPCAB is performed in order to avoid the hazards associated with standard CABG, such as cardiopulmonary bypass and aortic cross-clamping, and is carried out at the discretion of the surgeon in patients deemed to be at higher risk of thromboembolic or renal complications during surgery. All CABG patients at the OHI will be triaged pre-operatively, and study eligible patients will be selected and approached by the study nurse to explain the trial and obtain consent.
Description of Intervention and Control
A prospective randomized double-blinded placebo-controlled study will be conducted from November 2005 to November 2007 in order to achieve the study objectives. Patients will be recruited over the first 12 months of the study, and graft evaluation for each patient will occur over the following 12 months (one year after surgery for each patient). Patients will be randomized into an experimental group (receiving clopidogrel) or a control group (placebo). The placebo and clopidogrel medications will be prepared by the Bristol Myers-squibb Sanofi Canada Partnership and appear identical. Medication administration and data collection will be performed in a double-blind manner, such that neither the patient nor the healthcare personnel will be aware of the medication assignment. Recruitment and written consent will be performed prior to surgery. However, patients will not be randomized until after surgery has been completed and clinical stability ensured. Patients that are bleeding excessively after surgery (chest tube output > 200 cc/hr) or those requiring high levels of hemodynamic support (more than 2 inotropes and/or intra-aortic balloon pump) will not be randomized into the study.
After surgery, the study medication will be administered via nasogastric tube when the chest tube drainage has decreased to ≤50 cc/hr for 2 hours. Each patient will receive either clopidogrel 75 mg or placebo, in addition to enteric coated aspirin 162 mg (Figure 1). The study drug and aspirin 162 mg will be repeated orally in the same dose once daily for the duration of one year. Because post-CABG patients are relatively aspirin resistant following surgery, the aspirin dose of 162 mg will ensure adequate platelet inhibition in those patients randomized to aspirin alone. Furthermore, 162 mg falls within the safety window of combining aspirin with clopidogrel [30,40,44].
Figure 1 text
Allocation Procedure
A stratified random design allocation will be utilized to account for the presence or absence of diabetes, as well as the use or nonuse of cardiopulmonary bypass (standard CABG versus OPCAB). A block randomization technique will ensure an equal distribution of diabetic patients in both arms of the trial and an equal distribution of OPCAB patients. The randomization schedule will be generated using SAS 9.1 software (SAS, Cary, NC). All patients and study personnel will be blinded to the treatment assignment, which will be performed by the hospital pharmacy.
Concomitant Medication and Treatments
Patients will receive concomitant therapies in both groups as recommended by the current American College of Cardiology / American Heart Association guidelines. This will include smoking cessation counseling and the administration of aspirin, beta blockers, angiotensin converting enzyme inhibitors, and lipid lowering medications. Target LDL values will be those recommended as per current guidelines [45-47] and will be assessed during the study follow-up period. Routine peptic ulcer prophylaxis will not be administered in order to fully evaluate gastrointestinal side effects.
Diabetic patients will be eligible for enrollment in this study, regardless of their preoperative need for insulin therapy, and will be allocated equally into both groups through stratified randomization. Diabetic patients will have aggressive perioperative glycemic control, including an intravenous insulin infusion both in the operating room and in the intensive care unit, and a subcutaneous insulin sliding scale while recovering on the surgical ward. Once drinking well, diabetic patients will be restarted on their original preoperative diabetic regimens (oral agents and/or insulin therapy). The treatment of diabetes during this study will be closely monitored in collaboration with an endocrinologist specializing in the management of diabetes.
Primary Outcomes
The primary endpoint of this study will be to assess whether the addition of clopidogrel to aspirin reduces intimal hyperplasia in saphenous vein grafts 12 months after bypass surgery, as assessed by IVUS. Patients will undergo IVUS imaging 12 months post-CABG, and the average intimal area in the proximal 40 mm of one vein graft per patient will be assessed.
Secondary Outcomes
At the time of intravascular ultrasound, coronary and graft angiography will also be performed to assess vein graft patency and areas of stenosis. Although this trial will not be sufficiently powered for such a purpose, this data will be obtained whilst gathering information pertaining to the primary outcome.
Endpoints related to safety will also be documented, both at the time of surgery as well as during the one year of study drug administration. After surgery, data will be recorded regarding chest tube blood loss, blood product transfusions, bleeding requiring tube thoracostomy or sternal re-opening, perioperative MI, and gastrointestinal complications. Complete blood counts (CBC) will assess the hemoglobin level in the immediate postoperative period and during the one year follow-up. In addition, the incidence of major adverse cardiovascular events following CABG (mortality, MI, cerebrovascular accident, hospitalization for coronary ischemia, need for coronary intervention) will be recorded.
IVUS Procedure
Intravascular ultrasound will be used to assess the area of intimal hyperplasia present in saphenous vein grafts at 12 months after surgery. IVUS differs from angiography by providing cross-sectional images of both the vessel wall and lumen with high resolution. The process of intimal hyperplasia is easily detectable and quantifiable by IVUS, but may completely escape visualization by angiography. At one year following CABG, vein graft intimal hyperplasia is universally present.
Efforts will be made to schedule all patients at 52 ± 2 weeks from the day of randomization (Figure 1). The IVUS procedure will first start with angiography of the native coronary arteries and the coronary bypass grafts. This will allow assessment of the progression of native coronary artery atherosclerosis (both in grafted and non-grafted vessels). Patency of all bypass grafts will be scored using 1) the Fitzgibbon method [14] and 2) the TIMI classification [48]. Patency of the coronary arteries will be assessed by the TIMI method.
IVUS studies will be performed using a 40 MHz imaging catheter (Atlantis® SR Pro, Boston Scientific). This catheter is a monorail system and has 6F guiding catheter compatibility. All IVUS imaging will be done with the administration of unfractionated heparin (70 units per kg, minimum 4000 units) before the introduction of the guidewire into a vein graft. Each patient will have at least two vein grafts implanted at the time of surgery. However, imaging of more than one graft is not clinically advisable for safety and practical reasons. In order to minimize bias, the cardiologist performing the IVUS procedure will be blinded to the treatment allocation, and the selection of the SVG for IVUS imaging will be randomized. For this purpose, a sequence randomization scheme (based on the number of vein grafts) will produce a random sequence of numbers for each patient, such as "2, 1 and 3". These numbers will correspond to the position of the proximal anastomosis of each SVG on the ascending aorta (by increasing number from cranial to caudal, starting from 1). The graft whose random number is produced first will be selected for intubation with the IVUS catheter, unless 1) this particular graft showed more than 50% stenosis on the selective angiography performed immediately prior, or 2) access to this graft is technically difficult (graft tortuosity, difficult graft intubation). If any of the above two instances is encountered, the graft with the next number in the randomization sequence will instead be selected for IVUS, and so on. Graft exclusion for whatever basis (either because >50% stenosis or technically difficult access) will be recorded for all patients in each group.
Once the randomized graft selection has been completed, the actual IVUS procedure will take place. Intracoronary nitroglycerin (200 μg) will be given before advancing the IVUS catheter, and a 6F guiding catheter will be used to engage the vein graft. Then, a 0.014" coronary angioplasty guidewire will be advanced distally through the vein graft and positioned into the native coronary artery. The IVUS catheter will then be advanced into the graft at least 50 mm beyond the aorto-ostial anastomosis. The guiding catheter will then be disengaged to ensure visualization of the aorto-ostial anastomosis on pullback. This is essential because the aorto-ostial anastomosis will be the only landmark available to assure measurements of the proximal portion of each graft are comparable between groups. IVUS imaging will be done using a validated motorized pullback device at 0.5 mm/sec. Each study will be recorded on a separate S-VHS videotape. Only one quality pullback will be needed per graft, but a second pullback may be repeated if the first set of pullback images were judged suboptimal in quality.
IVUS Analysis
The IVUS images will be sent to an independent core laboratory and interpreted by an experienced cardiologist blinded to treatment allocation. The methods for the analysis have been previously reported and validated [49]. Briefly, using the aorto-ostial anastomosis as a landmark, the most proximal 40 mm of vein graft will be analyzed. The video images will be digitized and measurements of lumen, intimal hyperplasia, and external elastic lamina areas will be available for each digitized cross section. Intimal hyperplasia volumes will be computed by multiplying the corresponding areas of each of the cross-sections by the distance between slices and by adding the products. For the purpose of this study, the mean plaque area per patient for the 40 mm-analyzed segment will be used for comparison between treatment groups.
Sample Size
The trial will be powered to test the hypothesis that clopidogrel and aspirin should reduce vein graft intimal hyperplasia by 20% compared to aspirin alone at one year following bypass surgery. According to Hozumi et al., at one year, the mean intimal area of angiographically normal saphenous vein grafts is 5.26 mm2, with a standard deviation of 1.38 mm2 [50]. Intravascular ultrasound imaging will be performed at follow-up in one vein graft per patient. In order to account for potential angiographic refusals and study withdrawals, approximately 100 patients in total will be required to test the null hypothesis with an α value of 0.05 and a power of 0.90.
Data Collection and Safety Monitoring
Intimal area will be recorded at the time of IVUS one year after CABG. Vein graft patency and stenosis will be assessed by angiography. The incidence of major adverse coronary events and bleeding complications will be documented during postoperative clinic visits at one month, six months and twelve months after surgery. Telephone home assessments every three months will also be used to document events. All serious adverse events will be reported to the ethics committee. The development of serious adverse events that might be attributable to clopidogrel will lead to the termination of the study drug.
Ethics
The study will adhere to the highest research ethics standards of the OHI. This protocol follows the CONSORT guidelines [51] and was approved on July 19, 2005 by the University of Ottawa Heart Institute Human Research Ethics Board.
Statistical Analysis
Vein graft intimal area, the primary outcome of the study, will be compared between the two randomization groups using two-sided Student's t tests. Vein graft patency will be compared using a Fisher's exact test. With respect to the secondary outcomes, continuous data will be compared between the two groups using two-sided Student's t tests, two-sample Wilcoxon rank-sum tests, or ANOVA as appropriate, and a Fisher's exact test will be used for categorical data. In order to assess possible interactions between patient characteristics (such as diabetes) and treatment outcomes, an exploratory analysis using multivariate linear regression will be performed. In the improbable event that all vein grafts are occluded and IVUS cannot be performed in a particular patient undergoing the one-year study, a value of five times the mean intimal area (decided a priori) will be assigned for that patient.
Discussion
The CASCADE study is a novel randomized double-blind placebo-controlled trial that will help clarify the controversial issue of antiplatelet therapy following CABG surgery. Specifically, it will answer the questions of whether the addition of clopidogrel to aspirin is safe following cardiac surgery and whether it reduces saphenous vein graft intimal hyperplasia. Although subgroup analyses of previous trials have suggested a benefit of clopidogrel in cardiac surgery patients [39,42,43], no trial to date has specifically focused on the clinical or angiographic outcomes in patients treated with clopidogrel therapy immediately after surgical revascularization. Should the combination of clopidogrel and aspirin reduce the process of vein graft intimal hyperplasia, the CASCADE trial has the potential to redefine modern antiplatelet management of coronary artery bypass patients.
We believe that the strengths of this study are its randomized and blinded design. Patients will be randomized into two treatment groups, and graft selection for IVUS imaging will be randomized within each patient. Blinding will occur for all patients and health care providers, and the IVUS and angiogram images will be interpreted by an external blinded core laboratory. While the process of intimal hyperplasia may be reduced in the trial, the study will not be powered to demonstrate a difference in angiographic patency or freedom from coronary events following CABG. In order to demonstrate a difference in either of these outcomes, a considerably larger sample size and longer follow-up period would be required.
The results of this randomized trial will be generalizable to all patients undergoing standard CABG or OPCAB surgery with saphenous vein. Although the trial enrolment is limited to lower risk patients, it is anticipated that the results will also be applicable to higher risk patients with severe ischemic left ventricular dysfunction or those undergoing redo-CABG. High-risk patients participating in previous clopidogrel studies have been found to derive an enhanced benefit with the combined treatment of clopidogrel and aspirin [42]. Furthermore, there is no reason to anticipate differences in the pathophysiology of vein graft disease between high risk patients and those of the proposed study group.
Conclusion
Saphenous vein graft disease continues to be a major limitation of surgical revascularization for coronary artery disease. The process of saphenous vein intimal hyperplasia begins just days after surgery, setting the stage for graft atherosclerotic disease and its sequalae. Clopidogrel has been demonstrated to improve outcomes in patients with coronary and vascular disease, and it is effective at reducing intimal hyperplasia in animal models of thrombosis. However, no prospective study to date has been conducted in humans regarding the use of clopidogrel after CABG to prevent vein graft intimal hyperplasia. The CASCADE trial is a randomized, placebo-controlled trial comparing clopidogrel plus aspirin versus aspirin alone in CABG patients revascularized with saphenous vein. The effects of clopidogrel on vein graft intimal hyperplasia will be studied with coronary angiography and intravascular ultrasound one year following CABG.
Abbreviations
CABG – coronary artery bypass graft surgery
CASCADE – Clopidogrel after Surgery for Coronary Artery Disease Trial
CBC – complete blood counts
IVUS – intravascular ultrasound
MI – myocardial infarction
OHI – University of Ottawa Heart Institute
OPCAB – off-pump coronary artery bypass graft surgery
RRR – relative risk reduction
SVG – saphenous vein graft
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
All authors read and approved the final manuscript. Specifically, AK conceived of the study, participated in the protocol design and helped draft the manuscript. MLM participated in the protocol design, helped draft the manuscript, and will perform the intravascular ultrasound procedures on study patients. GAW helped draft the manuscript, design the study protocol, and will coordinate the statistical analysis. TGM helped design the study protocol and draft the manuscript. MR is the principle investigator and helped conceive the study, design the protocol and draft the manuscript.
Table 1 Study Inclusion and Exclusion Criteria
Inclusion Criteria:
Reason:
- Patients undergoing primary multi-vessel CABG with at least two saphenous vein grafts, with or without cardiopulmonary bypass - The population of clinical interest
Preoperative Exclusion Criteria:
Reason:
- Emergency surgery - Unable to obtain consent
- Valve surgery - Requirement of postoperative anticoagulation
- Redo CABG - Higher risk of postoperative bleeding and low cardiac output syndrome
- Left ventricle ejection fraction <25% - Higher risk of postoperative low cardiac output syndrome and mortality
- Serum creatinine >130 μmol/L - Contraindication to use of postoperative angiography
- Preoperative use of clopidogrel (with the exception of the current admission) - Contraindication to randomization, confounding of results
- Preoperative use of warfarin - Requirement of postoperative anticoagulation
- Allergy to aspirin or clopidogrel - Contraindication to use of aspirin or clopidogrel
- History of cerebrovascular accident - Concurrent indication for clopidogrel
- History of severe liver disease - Contraindication to use of clopidogrel
- Morbid obesity - Unable to perform postoperative angiography
- Residence outside of Ottawa region - Unable to perform postoperative angiography
- Current malignancy - Higher risk of early postoperative mortality
- Inability to provide informed consent - Ineligible for research study enrollment
Postoperative Exclusion Criteria:
Reason:
- Low cardiac output syndrome with inotropic support for greater than 24 hours - Unstable patient unlikely to benefit from treatment
- Recurrent ventricular arrhythmias - Unstable patient unlikely to benefit from treatment
- Intubation for more than 24 hours - Unstable patient unlikely to benefit from treatment
- Postoperative bleeding or cardiac tamponade requiring surgical exploration - Contraindication to use of clopidogrel
- Postoperative gastrointestinal bleeding - Contraindication to use of clopidogrel
- Postoperative warfarin requirement - Contraindication to use of clopidogrel
Acknowledgements
The CASCADE trial is funded through several sources, including a Physicians' Services Incorporated Foundation resident research grant, as well as grants from Boston Scientific Inc. and financial support from the Bristol-Myers Squibb Sanofi Canada Partnership. Boston Scientific will supply the intravascular ultrasound (IVUS) catheters used during the course of this study. The Bristol-Myers Squibb Sanofi Canada Partnership will supply the study medications (clopidogrel and placebo), and it has agreed to provide direct financial support during the trial. Drs. Laurent Carty and Stanislav Glezer from Sanofi-Aventis and Dr. Olga Tarasova from Bristol-Myers Squibb critically reviewed the protocol design during its development.
==== Refs
Hannan EL Racz MJ Walford G Jones RH Ryan TJ Bennett E Culliford AT Isom OW Gold JP Rose EA Long-term outcomes of coronary-artery bypass grafting versus stent implantation N Engl J Med 2005 352 2174 2183 15917382 10.1056/NEJMoa040316
American Heart Association: Heart Disease and Stroke Statistics - 2004 Update
Weintraub WS Jones EL Craver JM Guyton RA Frequency of repeat coronary bypass or coronary angioplasty after coronary artery bypass surgery using saphenous venous grafts Am J Cardiol 1994 73 103 112 8296729 10.1016/0002-9149(94)90198-8
Collaborative meta-analysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk patients Bmj 2002 324 71 86 11786451 10.1136/bmj.324.7336.S71
Goldman S Copeland J Moritz T Henderson W Zadina K Ovitt T Doherty J Read R Chesler E Sako Y Improvement in early saphenous vein graft patency after coronary artery bypass surgery with antiplatelet therapy: results of a Veterans Administration Cooperative Study Circulation 1988 77 1324 1332 3286040
Goldman S Copeland J Moritz T Henderson W Zadina K Ovitt T Doherty J Read R Chesler E Sako Y Saphenous vein graft patency 1 year after coronary artery bypass surgery and effects of antiplatelet therapy. Results of a Veterans Administration Cooperative Study Circulation 1989 80 1190 1197 2680158
Goldman S Copeland J Moritz T Henderson W Zadina K Ovitt T Kern KB Sethi G Sharma GV Khuri S Long-term graft patency (3 years) after coronary artery surgery. Effects of aspirin: results of a VA Cooperative study Circulation 1994 89 1138 1143 8124800
Blankenhorn DH Nessim SA Johnson RL Sanmarco ME Azen SP Cashin-Hemphill L Beneficial effects of combined colestipol-niacin therapy on coronary atherosclerosis and coronary venous bypass grafts Jama 1987 257 3233 3240 3295315 10.1001/jama.257.23.3233
The effect of aggressive lowering of low-density lipoprotein cholesterol levels and low-dose anticoagulation on obstructive changes in saphenous-vein coronary-artery bypass grafts. The Post Coronary Artery Bypass Graft Trial Investigators N Engl J Med 1997 336 153 162 8992351 10.1056/NEJM199701163360301
Sacks FM Pfeffer MA Moye LA Rouleau JL Rutherford JD Cole TG Brown L Warnica JW Arnold JM Wun CC Davis BR Braunwald E The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. Cholesterol and Recurrent Events Trial investigators N Engl J Med 1996 335 1001 1009 8801446 10.1056/NEJM199610033351401
Sarjeant JM Rabinovitch M Understanding and treating vein graft atherosclerosis Cardiovasc Pathol 2002 11 263 271 12361836 10.1016/S1054-8807(02)00125-4
Campeau L Enjalbert M Lesperance J Bourassa MG Kwiterovich PJ Wacholder S Sniderman A The relation of risk factors to the development of atherosclerosis in saphenous-vein bypass grafts and the progression of disease in the native circulation. A study 10 years after aortocoronary bypass surgery N Engl J Med 1984 311 1329 1332 6333635
Bourassa MG Fate of venous grafts: the past, the present and the future J Am Coll Cardiol 1991 17 1081 1083 2007707
Fitzgibbon GM Kafka HP Leach AJ Keon WJ Hooper GD Burton JR Coronary bypass graft fate and patient outcome: angiographic follow-up of 5,065 grafts related to survival and reoperation in 1,388 patients during 25 years J Am Coll Cardiol 1996 28 616 626 8772748 10.1016/0735-1097(96)00206-9
Alderman EL Corley SD Fisher LD Chaitman BR Faxon DP Foster ED Killip T Sosa JA Bourassa MG Five-year angiographic follow-up of factors associated with progression of coronary artery disease in the Coronary Artery Surgery Study (CASS). CASS Participating Investigators and Staff J Am Coll Cardiol 1993 22 1141 1154 8409054
van Brussel BL Ernst JM Ernst NM Kelder HC Knaepen PJ Plokker HW Vermeulen FE Voors AA Clinical outcome in venous coronary artery bypass grafting: a 15-year follow-up study Int J Cardiol 1997 58 119 126 9049676 10.1016/S0167-5273(96)02858-6
Motwani JG Topol EJ Aortocoronary saphenous vein graft disease: pathogenesis, predisposition, and prevention Circulation 1998 97 916 931 9521341
Bryan AJ Angelini GD The biology of saphenous vein graft occlusion: etiology and strategies for prevention Curr Opin Cardiol 1994 9 641 649 7819622
Roubos N Rosenfeldt FL Richards SM Conyers RA Davis BB Improved preservation of saphenous vein grafts by the use of glyceryl trinitrate-verapamil solution during harvesting Circulation 1995 92 II31 6 7586430
Verrier ED Boyle EMJ Endothelial cell injury in cardiovascular surgery Ann Thorac Surg 1996 62 915 922 8784042 10.1016/S0003-4975(96)00528-0
Fingerle J Johnson R Clowes AW Majesky MW Reidy MA Role of platelets in smooth muscle cell proliferation and migration after vascular injury in rat carotid artery Proc Natl Acad Sci U S A 1989 86 8412 8416 2813399
Ross R The pathogenesis of atherosclerosis--an update N Engl J Med 1986 314 488 500 3511384
Jawien A Bowen-Pope DF Lindner V Schwartz SM Clowes AW Platelet-derived growth factor promotes smooth muscle migration and intimal thickening in a rat model of balloon angioplasty J Clin Invest 1992 89 507 511 1531345
Herbert JM Dol F Bernat A Falotico R Lale A Savi P The antiaggregating and antithrombotic activity of clopidogrel is potentiated by aspirin in several experimental models in the rabbit Thromb Haemost 1998 80 512 518 9759636
Butany JW David TE Ojha M Histological and morphometric analyses of early and late aortocoronary vein grafts and distal anastomoses Can J Cardiol 1998 14 671 677 9627522
Cavusoglu E Cheng J Bhatt R Kunamneni PB Marmur JD Eng C Clopidogrel in the management of ischemic heart disease Heart Dis 2003 5 144 152 12713681 10.1097/01.HDX.0000061697.22030.e5
Bhatt DL Topol EJ Antiplatelet and anticoagulant therapy in the secondary prevention of ischemic heart disease Med Clin North Am 2000 84 163 79, ix 10685133 10.1016/S0025-7125(05)70212-1
Helgason CM Bolin KM Hoff JA Winkler SR Mangat A Tortorice KL Brace LD Development of aspirin resistance in persons with previous ischemic stroke Stroke 1994 25 2331 2336 7974569
Gum PA Kottke-Marchant K Welsh PA White J Topol EJ A prospective, blinded determination of the natural history of aspirin resistance among stable patients with cardiovascular disease J Am Coll Cardiol 2003 41 961 965 12651041 10.1016/S0735-1097(02)03014-0
Buchanan MR Brister SJ Individual variation in the effects of ASA on platelet function: implications for the use of ASA clinically Can J Cardiol 1995 11 221 227 7889440
Chopra V Marmur JD Cavusoglu E The role of clopidogrel in the management of patients with ischemic heart disease Cardiovasc Drugs Ther 2003 17 467 477 15107602 10.1023/B:CARD.0000015862.62649.c8
Hollopeter G Jantzen HM Vincent D Li G England L Ramakrishnan V Yang RB Nurden P Nurden A Julius D Conley PB Identification of the platelet ADP receptor targeted by antithrombotic drugs Nature 2001 409 202 207 11196645 10.1038/35051599
Herbert JM Tissinier A Defreyn G Maffrand JP Inhibitory effect of clopidogrel on platelet adhesion and intimal proliferation after arterial injury in rabbits Arterioscler Thromb 1993 13 1171 1179 8343491
Clowes AW Karnovsky MJ Failure of certain antiplatelet drugs to affect myointimal thickening following arterial endothelial injury in the rat Lab Invest 1977 36 452 464 846193
Hermann A Weber AA Schror K Clopidogrel inhibits platelet adhesion and platelet-dependent mitogenesis in vascular smooth muscle cells Thromb Res 2002 105 173 175 11958809 10.1016/S0049-3848(01)00403-0
Harker LA Marzec UM Kelly AB Chronos NR Sundell IB Hanson SR Herbert JM Clopidogrel inhibition of stent, graft, and vascular thrombogenesis with antithrombotic enhancement by aspirin in nonhuman primates Circulation 1998 98 2461 2469 9832493
Cadroy Y Bossavy JP Thalamas C Sagnard L Sakariassen K Boneu B Early potent antithrombotic effect with combined aspirin and a loading dose of clopidogrel on experimental arterial thrombogenesis in humans Circulation 2000 101 2823 2828 10859288
A randomised, blinded, trial of clopidogrel versus aspirin in patients at risk of ischaemic events (CAPRIE). CAPRIE Steering Committee Lancet 1996 348 1329 1339 8918275 10.1016/S0140-6736(96)09457-3
Bhatt DL Chew DP Hirsch AT Ringleb PA Hacke W Topol EJ Superiority of clopidogrel versus aspirin in patients with prior cardiac surgery Circulation 2001 103 363 368 11157686
Yusuf S Zhao F Mehta SR Chrolavicius S Tognoni G Fox KK Effects of clopidogrel in addition to aspirin in patients with acute coronary syndromes without ST-segment elevation N Engl J Med 2001 345 494 502 11519503 10.1056/NEJMoa010746
Saw J Topol EJ Steinhubl SR Brennan D Berger PB Moliterno DJ Comparison of long-term usefulness of clopidogrel therapy after the first percutaneous coronary intervention or coronary artery bypass grafting versus that after the second or repeat intervention Am J Cardiol 2004 94 623 625 15342293 10.1016/j.amjcard.2004.05.026
Hirsh J Bhatt DL Comparative benefits of clopidogrel and aspirin in high-risk patient populations: lessons from the CAPRIE and CURE studies Arch Intern Med 2004 164 2106 2110 15505123 10.1001/archinte.164.19.2106
Fox KA Mehta SR Peters R Zhao F Lakkis N Gersh BJ Yusuf S Benefits and risks of the combination of clopidogrel and aspirin in patients undergoing surgical revascularization for non-ST-elevation acute coronary syndrome: the Clopidogrel in Unstable angina to prevent Recurrent ischemic Events (CURE) Trial Circulation 2004 110 1202 1208 15313956 10.1161/01.CIR.0000140675.85342.1B
Zimmermann N Wenk A Kim U Kienzle P Weber AA Gams E Schror K Hohlfeld T Functional and biochemical evaluation of platelet aspirin resistance after coronary artery bypass surgery Circulation 2003 108 542 547 12874188 10.1161/01.CIR.0000081770.51929.5A
Pearson T Rapaport E Criqui M Furberg C Fuster V Hiratzka L Little W Ockene I Williams G Optimal risk factor management in the patient after coronary revascularization. A statement for healthcare professionals from an American Heart Association Writing Group Circulation 1994 90 3125 3133 7994868
Eagle KA Guyton RA Davidoff R Edwards FH Ewy GA Gardner TJ Hart JC Herrmann HC Hillis LD Hutter AMJ Lytle BW Marlow RA Nugent WC Orszulak TA ACC/AHA 2004 guideline update for coronary artery bypass graft surgery: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Update the 1999 Guidelines for Coronary Artery Bypass Graft Surgery) Circulation 2004 110 e340 437 15466654 10.1161/01.CIR.0000138790.14877.7D
Smith SCJ Blair SN Bonow RO Brass LM Cerqueira MD Dracup K Fuster V Gotto A Grundy SM Miller NH Jacobs A Jones D Krauss RM Mosca L Ockene I Pasternak RC Pearson T Pfeffer MA Starke RD Taubert KA AHA/ACC Scientific Statement: AHA/ACC guidelines for preventing heart attack and death in patients with atherosclerotic cardiovascular disease: 2001 update: A statement for healthcare professionals from the American Heart Association and the American College of Cardiology Circulation 2001 104 1577 1579 11571256
The Thrombolysis in Myocardial Infarction (TIMI) trial. Phase I findings. TIMI Study Group N Engl J Med 1985 312 932 936 4038784
Tardif JC Gregoire J Lesperance J Lambert J L'Allier PL Rodes J Anderson T Blue JW Imus J Heinonen T Design features of the Avasimibe and Progression of coronary Lesions assessed by intravascular UltraSound (A-PLUS) clinical trial Am Heart J 2002 144 589 596 12360153
Hozumi T Yoshikawa J Yoshida K Akasaka T Takagi T Honda Y Okura H Use of intravascular ultrasound for in vivo assessment of changes in intimal thickness of angiographically normal saphenous vein grafts one year after aortocoronary bypass surgery Heart 1996 76 317 320 8983677
Moher D Schulz KF Altman DG The CONSORT statement: revised recommendations for improving the quality of reports of parallel-group randomised trials Lancet 2001 357 1191 1194 11323066 10.1016/S0140-6736(00)04337-3
|
16219100
|
PMC1282584
|
CC BY
|
2021-01-04 16:47:32
|
no
|
Curr Control Trials Cardiovasc Med. 2005 Oct 11; 6(1):15
|
utf-8
|
Curr Control Trials Cardiovasc Med
| 2,005 |
10.1186/1468-6708-6-15
|
oa_comm
|
==== Front
Curr Control Trials Cardiovasc MedCurrent Controlled Trials in Cardiovascular Medicine1468-67081468-6694BioMed Central 1468-6708-6-151621910010.1186/1468-6708-6-15Study ProtocolThe clopidogrel after surgery for coronary artery disease (CASCADE) randomized controlled trial: clopidogrel and aspirin versus aspirin alone after coronary bypass surgery [NCT00228423] Kulik Alexander [email protected] May Michel [email protected] George A [email protected] Thierry G [email protected] Marc [email protected] Division of Cardiac Surgery, University of Ottawa Heart Institute, Ottawa, Canada2 Division of Cardiology, University of Ottawa Heart Institute, Ottawa, Canada3 Department of Epidemiology and Community Medicine, University of Ottawa, Ottawa, Canada4 Division of Cardiac Surgery, University of Ottawa Heart Institute, Ottawa, Canada5 Division of Cardiac Surgery, University of Ottawa Heart Institute, and Department of Epidemiology and Community Medicine, University of Ottawa, Ottawa, Canada2005 11 10 2005 6 1 15 15 3 10 2005 11 10 2005 Copyright © 2005 Kulik et al; licensee BioMed Central Ltd.2005Kulik et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
Saphenous vein graft disease remains a major limitation of coronary artery bypass graft surgery. The process of saphenous vein intimal hyperplasia begins just days after surgical revascularization, setting the stage for graft atherosclerotic disease and its sequalae. Clopidogrel improves outcomes in patients with atherosclerotic disease, and is effective at reducing intimal hyperplasia in animal models of thrombosis. Therefore, the goal of this study will be to evaluate the efficacy of clopidogrel and aspirin therapy versus aspirin alone in the prevention of saphenous vein graft intimal hyperplasia following coronary artery bypass surgery.
Methods
Patients undergoing multi-vessel coronary artery bypass grafting and in whom at least two saphenous vein grafts will be used are eligible for the study. Patients will be randomized to receive daily clopidogrel 75 mg or placebo, in addition to daily aspirin 162 mg, for a one year duration starting on the day of surgery (as soon as postoperative bleeding has been excluded). At the end of one year, all patients will undergo coronary angiography and intravascular ultrasound assessment of one saphenous vein graft as selected by randomization. The trial will be powered to test the hypothesis that clopidogrel and aspirin will reduce vein graft intimal hyperplasia by 20% compared to aspirin alone at one year following bypass surgery.
Discussion
This trial is the first prospective human study that will address the question of whether clopidogrel therapy improves outcomes and reduces saphenous vein graft intimal hyperplasia following cardiac surgery. Should the combination of clopidogrel and aspirin reduce the process of vein graft intimal hyperplasia, the results of this study will help redefine modern antiplatelet management of coronary artery bypass patients.
coronary artery bypass graft surgeryantiplatelet therapysaphenous vein
==== Body
Background
Coronary artery bypass graft surgery (CABG) is the most durable approach for the treatment of ischemic heart disease [1], with >400,000 procedures performed annually in the United States alone [2]. Despite the increasing application of arterial conduits during CABG, the saphenous vein remains the most common conduit, employed for more than 70% of grafts [3]. However, saphenous vein graft (SVG) disease presents an important clinical problem. Even with aggressive medical therapy [4-11], up to 15% of vein grafts occlude in the first year after bypass surgery. Between 1 and 6 years, the graft attrition rate is 1% to 2% per year, and between 6 and 10 years it rises to 4% per year. By 10 years after surgery, only 60% of grafts are patent and only 50% of patent vein grafts are free of significant stenosis. In addition, native coronary artery disease progresses in 5% of patients annually [12-15]. Reflecting the graft and native vessel attrition, this population is at high risk for subsequent ischemic events, including death, myocardial infarction (MI) and stroke [14]. Further revascularization, either reoperation or percutaneous coronary intervention, is required in 4% of patients by 5 years, 19% of patients by 10 years and 31% of patients by 12 years after the initial bypass surgery [3,16].
The process of SVG disease is composed of three mechanistically interlinked stages: thrombosis, intimal hyperplasia, and atherosclerosis [13,14,16-18]. Early graft thrombosis can occur at the time of surgery secondary to focal endothelial disruption [19,20]. Grafts that survive this early period develop a progressive thickening of the media that begins within days after implantation. This process, termed intimal hyperplasia, is a consequence of smooth muscle cell proliferation and extracellular matrix protein synthesis [21,22]. Platelets play a fundamental role in the process of smooth muscle cell proliferation and intimal hyperplasia [23,24]. Intimal hyperplasia is present in all grafts 1 month after implantation [25] and forms a template for the development of superimposed atherosclerotic changes [17,18]. With the passage of a sufficient period of time, the thrombotic occlusion of vein grafts is almost inevitable due to progressive atherosclerosis [17].
Despite its established benefit in patients with coronary artery disease, aspirin therapy has numerous limitations. It is a relatively weak antiplatelet agent and has no effect on thrombin, which is believed to play a major role in acute coronary syndrome [26]. Even with aspirin therapy for secondary prevention, a large number of recurrent events occur [27]. A significant proportion of patients undergoing CABG may be aspirin resistant, defined as undetectable platelet inhibition after one week of therapy [28,29]. Depending on the population studied and the specific definition of aspirin resistance, anywhere from 10–40% of patients appear to have an inadequate antiplatelet response to aspirin [28,30]. Such patients appear to be at increased risk for the development of vascular events. In theory, these aspirin-resistant patients may derive particular benefit from additional antiplatelet therapy [31].
Clopidogrel is a thienopyridine antiplatelet agent that inhibits ADP-dependent platelet activation and aggregation [32]. Sevenfold more potent than ticlopidine, clopidogrel is free of its adverse side effects such as neutropenia, diarrhea and rash [33]. Unlike aspirin [24,34], clopidogrel has been shown to inhibit the process of platelet-mediated intimal proliferation and smooth muscle hyperplasia in laboratory experiments. In a cell culture model, clopidogrel significantly inhibited platelet adhesion to immobilized fibrinogen and also inhibited platelet-dependent mitogenic signaling and DNA synthesis in cultured coronary artery smooth muscle cells [35]. Similarly, in animal thrombosis models, clopidogrel but not aspirin significantly inhibited platelet-mediated intimal proliferation and smooth muscle hyperplasia [33,36]. Furthermore, the combination of clopidogrel with aspirin led to potent synergistic antithrombotic effects and a decrease in myointimal proliferation compared to either therapy alone [24,36,37].
Several large clinical trials have demonstrated that clopidogrel reduces ischemic events and mortality in patients with coronary and vascular disease [38-41]. In the CAPRIE (Clopidogrel versus Aspirin in Patients with Ischemic Events) trial, clopidogrel (75 mg/day) was demonstrated to be significantly more effective than aspirin (325 mg/day) in preventing vascular thrombotic events (ischemic stroke, MI or vascular death) in patients with clinical evidence of atherosclerotic disease (clopidogrel 9.78% vs. aspirin 10.64%, relative risk reduction [RRR] 8.7%, p = 0.045) [38]. In patients presenting with acute coronary syndromes, the CURE (Clopidogrel in Unstable angina to prevent Recurrent ischemic Events) study demonstrated that the combination of clopidogrel and aspirin was more effective at reducing the primary outcome (cardiovascular death, nonfatal MI or stroke) compared to aspirin alone (clopidogrel and aspirin 9.3% vs. aspirin alone 11.4%, RRR 20%, p < 0.001) [40]. Subgroup analysis from these trials suggested that patients that underwent surgical revascularization also benefited from clopidogrel [39,42,43]. However, no trial to date has prospectively evaluated the combined effects of clopidogrel plus aspirin on saphenous vein graft disease after CABG.
There currently exists a clinical equipoise regarding the optimal antiplatelet therapy for patients who have undergone coronary artery bypass surgery. While some clinicians believe in the beneficial effects of clopidogrel, the increased risk of bleeding and the lack of data in CABG patients make it impossible to establish definitive recommendations. We therefore propose the Clopidogrel after Surgery for Coronary Artery Disease (CASCADE) study, a randomized, placebo-controlled trial comparing clopidogrel plus aspirin versus aspirin alone in CABG patients revascularized with saphenous vein. The primary aim of this study will be to evaluate the effect of combined antiplatelet therapy on the reduction of SVG intimal hyperplasia one year after CABG, through the assessment of intimal area by intravascular ultrasound (IVUS). Secondary aims will evaluate the safety of clopidogrel administration following CABG with regards to bleeding complications. We hypothesize that the combination of clopidogrel with aspirin will reduce the SVG intimal hyperplasia (intimal area) by 20% one year post-CABG compared to the usual antiplatelet therapy of aspirin alone.
Methods
Study Population and Recruitment Procedure
The study population will include all patients undergoing multi-vessel elective or urgent CABG using at least two saphenous vein grafts at the University of Ottawa Heart Institute (OHI) over the study period (see Table 1 for inclusion and exclusion criteria). Patients undergoing off-pump CABG (OPCAB) will also be eligible for this study, as long as at least two saphenous vein grafts are used. OPCAB is performed in order to avoid the hazards associated with standard CABG, such as cardiopulmonary bypass and aortic cross-clamping, and is carried out at the discretion of the surgeon in patients deemed to be at higher risk of thromboembolic or renal complications during surgery. All CABG patients at the OHI will be triaged pre-operatively, and study eligible patients will be selected and approached by the study nurse to explain the trial and obtain consent.
Description of Intervention and Control
A prospective randomized double-blinded placebo-controlled study will be conducted from November 2005 to November 2007 in order to achieve the study objectives. Patients will be recruited over the first 12 months of the study, and graft evaluation for each patient will occur over the following 12 months (one year after surgery for each patient). Patients will be randomized into an experimental group (receiving clopidogrel) or a control group (placebo). The placebo and clopidogrel medications will be prepared by the Bristol Myers-squibb Sanofi Canada Partnership and appear identical. Medication administration and data collection will be performed in a double-blind manner, such that neither the patient nor the healthcare personnel will be aware of the medication assignment. Recruitment and written consent will be performed prior to surgery. However, patients will not be randomized until after surgery has been completed and clinical stability ensured. Patients that are bleeding excessively after surgery (chest tube output > 200 cc/hr) or those requiring high levels of hemodynamic support (more than 2 inotropes and/or intra-aortic balloon pump) will not be randomized into the study.
After surgery, the study medication will be administered via nasogastric tube when the chest tube drainage has decreased to ≤50 cc/hr for 2 hours. Each patient will receive either clopidogrel 75 mg or placebo, in addition to enteric coated aspirin 162 mg (Figure 1). The study drug and aspirin 162 mg will be repeated orally in the same dose once daily for the duration of one year. Because post-CABG patients are relatively aspirin resistant following surgery, the aspirin dose of 162 mg will ensure adequate platelet inhibition in those patients randomized to aspirin alone. Furthermore, 162 mg falls within the safety window of combining aspirin with clopidogrel [30,40,44].
Figure 1 text
Allocation Procedure
A stratified random design allocation will be utilized to account for the presence or absence of diabetes, as well as the use or nonuse of cardiopulmonary bypass (standard CABG versus OPCAB). A block randomization technique will ensure an equal distribution of diabetic patients in both arms of the trial and an equal distribution of OPCAB patients. The randomization schedule will be generated using SAS 9.1 software (SAS, Cary, NC). All patients and study personnel will be blinded to the treatment assignment, which will be performed by the hospital pharmacy.
Concomitant Medication and Treatments
Patients will receive concomitant therapies in both groups as recommended by the current American College of Cardiology / American Heart Association guidelines. This will include smoking cessation counseling and the administration of aspirin, beta blockers, angiotensin converting enzyme inhibitors, and lipid lowering medications. Target LDL values will be those recommended as per current guidelines [45-47] and will be assessed during the study follow-up period. Routine peptic ulcer prophylaxis will not be administered in order to fully evaluate gastrointestinal side effects.
Diabetic patients will be eligible for enrollment in this study, regardless of their preoperative need for insulin therapy, and will be allocated equally into both groups through stratified randomization. Diabetic patients will have aggressive perioperative glycemic control, including an intravenous insulin infusion both in the operating room and in the intensive care unit, and a subcutaneous insulin sliding scale while recovering on the surgical ward. Once drinking well, diabetic patients will be restarted on their original preoperative diabetic regimens (oral agents and/or insulin therapy). The treatment of diabetes during this study will be closely monitored in collaboration with an endocrinologist specializing in the management of diabetes.
Primary Outcomes
The primary endpoint of this study will be to assess whether the addition of clopidogrel to aspirin reduces intimal hyperplasia in saphenous vein grafts 12 months after bypass surgery, as assessed by IVUS. Patients will undergo IVUS imaging 12 months post-CABG, and the average intimal area in the proximal 40 mm of one vein graft per patient will be assessed.
Secondary Outcomes
At the time of intravascular ultrasound, coronary and graft angiography will also be performed to assess vein graft patency and areas of stenosis. Although this trial will not be sufficiently powered for such a purpose, this data will be obtained whilst gathering information pertaining to the primary outcome.
Endpoints related to safety will also be documented, both at the time of surgery as well as during the one year of study drug administration. After surgery, data will be recorded regarding chest tube blood loss, blood product transfusions, bleeding requiring tube thoracostomy or sternal re-opening, perioperative MI, and gastrointestinal complications. Complete blood counts (CBC) will assess the hemoglobin level in the immediate postoperative period and during the one year follow-up. In addition, the incidence of major adverse cardiovascular events following CABG (mortality, MI, cerebrovascular accident, hospitalization for coronary ischemia, need for coronary intervention) will be recorded.
IVUS Procedure
Intravascular ultrasound will be used to assess the area of intimal hyperplasia present in saphenous vein grafts at 12 months after surgery. IVUS differs from angiography by providing cross-sectional images of both the vessel wall and lumen with high resolution. The process of intimal hyperplasia is easily detectable and quantifiable by IVUS, but may completely escape visualization by angiography. At one year following CABG, vein graft intimal hyperplasia is universally present.
Efforts will be made to schedule all patients at 52 ± 2 weeks from the day of randomization (Figure 1). The IVUS procedure will first start with angiography of the native coronary arteries and the coronary bypass grafts. This will allow assessment of the progression of native coronary artery atherosclerosis (both in grafted and non-grafted vessels). Patency of all bypass grafts will be scored using 1) the Fitzgibbon method [14] and 2) the TIMI classification [48]. Patency of the coronary arteries will be assessed by the TIMI method.
IVUS studies will be performed using a 40 MHz imaging catheter (Atlantis® SR Pro, Boston Scientific). This catheter is a monorail system and has 6F guiding catheter compatibility. All IVUS imaging will be done with the administration of unfractionated heparin (70 units per kg, minimum 4000 units) before the introduction of the guidewire into a vein graft. Each patient will have at least two vein grafts implanted at the time of surgery. However, imaging of more than one graft is not clinically advisable for safety and practical reasons. In order to minimize bias, the cardiologist performing the IVUS procedure will be blinded to the treatment allocation, and the selection of the SVG for IVUS imaging will be randomized. For this purpose, a sequence randomization scheme (based on the number of vein grafts) will produce a random sequence of numbers for each patient, such as "2, 1 and 3". These numbers will correspond to the position of the proximal anastomosis of each SVG on the ascending aorta (by increasing number from cranial to caudal, starting from 1). The graft whose random number is produced first will be selected for intubation with the IVUS catheter, unless 1) this particular graft showed more than 50% stenosis on the selective angiography performed immediately prior, or 2) access to this graft is technically difficult (graft tortuosity, difficult graft intubation). If any of the above two instances is encountered, the graft with the next number in the randomization sequence will instead be selected for IVUS, and so on. Graft exclusion for whatever basis (either because >50% stenosis or technically difficult access) will be recorded for all patients in each group.
Once the randomized graft selection has been completed, the actual IVUS procedure will take place. Intracoronary nitroglycerin (200 μg) will be given before advancing the IVUS catheter, and a 6F guiding catheter will be used to engage the vein graft. Then, a 0.014" coronary angioplasty guidewire will be advanced distally through the vein graft and positioned into the native coronary artery. The IVUS catheter will then be advanced into the graft at least 50 mm beyond the aorto-ostial anastomosis. The guiding catheter will then be disengaged to ensure visualization of the aorto-ostial anastomosis on pullback. This is essential because the aorto-ostial anastomosis will be the only landmark available to assure measurements of the proximal portion of each graft are comparable between groups. IVUS imaging will be done using a validated motorized pullback device at 0.5 mm/sec. Each study will be recorded on a separate S-VHS videotape. Only one quality pullback will be needed per graft, but a second pullback may be repeated if the first set of pullback images were judged suboptimal in quality.
IVUS Analysis
The IVUS images will be sent to an independent core laboratory and interpreted by an experienced cardiologist blinded to treatment allocation. The methods for the analysis have been previously reported and validated [49]. Briefly, using the aorto-ostial anastomosis as a landmark, the most proximal 40 mm of vein graft will be analyzed. The video images will be digitized and measurements of lumen, intimal hyperplasia, and external elastic lamina areas will be available for each digitized cross section. Intimal hyperplasia volumes will be computed by multiplying the corresponding areas of each of the cross-sections by the distance between slices and by adding the products. For the purpose of this study, the mean plaque area per patient for the 40 mm-analyzed segment will be used for comparison between treatment groups.
Sample Size
The trial will be powered to test the hypothesis that clopidogrel and aspirin should reduce vein graft intimal hyperplasia by 20% compared to aspirin alone at one year following bypass surgery. According to Hozumi et al., at one year, the mean intimal area of angiographically normal saphenous vein grafts is 5.26 mm2, with a standard deviation of 1.38 mm2 [50]. Intravascular ultrasound imaging will be performed at follow-up in one vein graft per patient. In order to account for potential angiographic refusals and study withdrawals, approximately 100 patients in total will be required to test the null hypothesis with an α value of 0.05 and a power of 0.90.
Data Collection and Safety Monitoring
Intimal area will be recorded at the time of IVUS one year after CABG. Vein graft patency and stenosis will be assessed by angiography. The incidence of major adverse coronary events and bleeding complications will be documented during postoperative clinic visits at one month, six months and twelve months after surgery. Telephone home assessments every three months will also be used to document events. All serious adverse events will be reported to the ethics committee. The development of serious adverse events that might be attributable to clopidogrel will lead to the termination of the study drug.
Ethics
The study will adhere to the highest research ethics standards of the OHI. This protocol follows the CONSORT guidelines [51] and was approved on July 19, 2005 by the University of Ottawa Heart Institute Human Research Ethics Board.
Statistical Analysis
Vein graft intimal area, the primary outcome of the study, will be compared between the two randomization groups using two-sided Student's t tests. Vein graft patency will be compared using a Fisher's exact test. With respect to the secondary outcomes, continuous data will be compared between the two groups using two-sided Student's t tests, two-sample Wilcoxon rank-sum tests, or ANOVA as appropriate, and a Fisher's exact test will be used for categorical data. In order to assess possible interactions between patient characteristics (such as diabetes) and treatment outcomes, an exploratory analysis using multivariate linear regression will be performed. In the improbable event that all vein grafts are occluded and IVUS cannot be performed in a particular patient undergoing the one-year study, a value of five times the mean intimal area (decided a priori) will be assigned for that patient.
Discussion
The CASCADE study is a novel randomized double-blind placebo-controlled trial that will help clarify the controversial issue of antiplatelet therapy following CABG surgery. Specifically, it will answer the questions of whether the addition of clopidogrel to aspirin is safe following cardiac surgery and whether it reduces saphenous vein graft intimal hyperplasia. Although subgroup analyses of previous trials have suggested a benefit of clopidogrel in cardiac surgery patients [39,42,43], no trial to date has specifically focused on the clinical or angiographic outcomes in patients treated with clopidogrel therapy immediately after surgical revascularization. Should the combination of clopidogrel and aspirin reduce the process of vein graft intimal hyperplasia, the CASCADE trial has the potential to redefine modern antiplatelet management of coronary artery bypass patients.
We believe that the strengths of this study are its randomized and blinded design. Patients will be randomized into two treatment groups, and graft selection for IVUS imaging will be randomized within each patient. Blinding will occur for all patients and health care providers, and the IVUS and angiogram images will be interpreted by an external blinded core laboratory. While the process of intimal hyperplasia may be reduced in the trial, the study will not be powered to demonstrate a difference in angiographic patency or freedom from coronary events following CABG. In order to demonstrate a difference in either of these outcomes, a considerably larger sample size and longer follow-up period would be required.
The results of this randomized trial will be generalizable to all patients undergoing standard CABG or OPCAB surgery with saphenous vein. Although the trial enrolment is limited to lower risk patients, it is anticipated that the results will also be applicable to higher risk patients with severe ischemic left ventricular dysfunction or those undergoing redo-CABG. High-risk patients participating in previous clopidogrel studies have been found to derive an enhanced benefit with the combined treatment of clopidogrel and aspirin [42]. Furthermore, there is no reason to anticipate differences in the pathophysiology of vein graft disease between high risk patients and those of the proposed study group.
Conclusion
Saphenous vein graft disease continues to be a major limitation of surgical revascularization for coronary artery disease. The process of saphenous vein intimal hyperplasia begins just days after surgery, setting the stage for graft atherosclerotic disease and its sequalae. Clopidogrel has been demonstrated to improve outcomes in patients with coronary and vascular disease, and it is effective at reducing intimal hyperplasia in animal models of thrombosis. However, no prospective study to date has been conducted in humans regarding the use of clopidogrel after CABG to prevent vein graft intimal hyperplasia. The CASCADE trial is a randomized, placebo-controlled trial comparing clopidogrel plus aspirin versus aspirin alone in CABG patients revascularized with saphenous vein. The effects of clopidogrel on vein graft intimal hyperplasia will be studied with coronary angiography and intravascular ultrasound one year following CABG.
Abbreviations
CABG – coronary artery bypass graft surgery
CASCADE – Clopidogrel after Surgery for Coronary Artery Disease Trial
CBC – complete blood counts
IVUS – intravascular ultrasound
MI – myocardial infarction
OHI – University of Ottawa Heart Institute
OPCAB – off-pump coronary artery bypass graft surgery
RRR – relative risk reduction
SVG – saphenous vein graft
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
All authors read and approved the final manuscript. Specifically, AK conceived of the study, participated in the protocol design and helped draft the manuscript. MLM participated in the protocol design, helped draft the manuscript, and will perform the intravascular ultrasound procedures on study patients. GAW helped draft the manuscript, design the study protocol, and will coordinate the statistical analysis. TGM helped design the study protocol and draft the manuscript. MR is the principle investigator and helped conceive the study, design the protocol and draft the manuscript.
Table 1 Study Inclusion and Exclusion Criteria
Inclusion Criteria:
Reason:
- Patients undergoing primary multi-vessel CABG with at least two saphenous vein grafts, with or without cardiopulmonary bypass - The population of clinical interest
Preoperative Exclusion Criteria:
Reason:
- Emergency surgery - Unable to obtain consent
- Valve surgery - Requirement of postoperative anticoagulation
- Redo CABG - Higher risk of postoperative bleeding and low cardiac output syndrome
- Left ventricle ejection fraction <25% - Higher risk of postoperative low cardiac output syndrome and mortality
- Serum creatinine >130 μmol/L - Contraindication to use of postoperative angiography
- Preoperative use of clopidogrel (with the exception of the current admission) - Contraindication to randomization, confounding of results
- Preoperative use of warfarin - Requirement of postoperative anticoagulation
- Allergy to aspirin or clopidogrel - Contraindication to use of aspirin or clopidogrel
- History of cerebrovascular accident - Concurrent indication for clopidogrel
- History of severe liver disease - Contraindication to use of clopidogrel
- Morbid obesity - Unable to perform postoperative angiography
- Residence outside of Ottawa region - Unable to perform postoperative angiography
- Current malignancy - Higher risk of early postoperative mortality
- Inability to provide informed consent - Ineligible for research study enrollment
Postoperative Exclusion Criteria:
Reason:
- Low cardiac output syndrome with inotropic support for greater than 24 hours - Unstable patient unlikely to benefit from treatment
- Recurrent ventricular arrhythmias - Unstable patient unlikely to benefit from treatment
- Intubation for more than 24 hours - Unstable patient unlikely to benefit from treatment
- Postoperative bleeding or cardiac tamponade requiring surgical exploration - Contraindication to use of clopidogrel
- Postoperative gastrointestinal bleeding - Contraindication to use of clopidogrel
- Postoperative warfarin requirement - Contraindication to use of clopidogrel
Acknowledgements
The CASCADE trial is funded through several sources, including a Physicians' Services Incorporated Foundation resident research grant, as well as grants from Boston Scientific Inc. and financial support from the Bristol-Myers Squibb Sanofi Canada Partnership. Boston Scientific will supply the intravascular ultrasound (IVUS) catheters used during the course of this study. The Bristol-Myers Squibb Sanofi Canada Partnership will supply the study medications (clopidogrel and placebo), and it has agreed to provide direct financial support during the trial. Drs. Laurent Carty and Stanislav Glezer from Sanofi-Aventis and Dr. Olga Tarasova from Bristol-Myers Squibb critically reviewed the protocol design during its development.
==== Refs
Hannan EL Racz MJ Walford G Jones RH Ryan TJ Bennett E Culliford AT Isom OW Gold JP Rose EA Long-term outcomes of coronary-artery bypass grafting versus stent implantation N Engl J Med 2005 352 2174 2183 15917382 10.1056/NEJMoa040316
American Heart Association: Heart Disease and Stroke Statistics - 2004 Update
Weintraub WS Jones EL Craver JM Guyton RA Frequency of repeat coronary bypass or coronary angioplasty after coronary artery bypass surgery using saphenous venous grafts Am J Cardiol 1994 73 103 112 8296729 10.1016/0002-9149(94)90198-8
Collaborative meta-analysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk patients Bmj 2002 324 71 86 11786451 10.1136/bmj.324.7336.S71
Goldman S Copeland J Moritz T Henderson W Zadina K Ovitt T Doherty J Read R Chesler E Sako Y Improvement in early saphenous vein graft patency after coronary artery bypass surgery with antiplatelet therapy: results of a Veterans Administration Cooperative Study Circulation 1988 77 1324 1332 3286040
Goldman S Copeland J Moritz T Henderson W Zadina K Ovitt T Doherty J Read R Chesler E Sako Y Saphenous vein graft patency 1 year after coronary artery bypass surgery and effects of antiplatelet therapy. Results of a Veterans Administration Cooperative Study Circulation 1989 80 1190 1197 2680158
Goldman S Copeland J Moritz T Henderson W Zadina K Ovitt T Kern KB Sethi G Sharma GV Khuri S Long-term graft patency (3 years) after coronary artery surgery. Effects of aspirin: results of a VA Cooperative study Circulation 1994 89 1138 1143 8124800
Blankenhorn DH Nessim SA Johnson RL Sanmarco ME Azen SP Cashin-Hemphill L Beneficial effects of combined colestipol-niacin therapy on coronary atherosclerosis and coronary venous bypass grafts Jama 1987 257 3233 3240 3295315 10.1001/jama.257.23.3233
The effect of aggressive lowering of low-density lipoprotein cholesterol levels and low-dose anticoagulation on obstructive changes in saphenous-vein coronary-artery bypass grafts. The Post Coronary Artery Bypass Graft Trial Investigators N Engl J Med 1997 336 153 162 8992351 10.1056/NEJM199701163360301
Sacks FM Pfeffer MA Moye LA Rouleau JL Rutherford JD Cole TG Brown L Warnica JW Arnold JM Wun CC Davis BR Braunwald E The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. Cholesterol and Recurrent Events Trial investigators N Engl J Med 1996 335 1001 1009 8801446 10.1056/NEJM199610033351401
Sarjeant JM Rabinovitch M Understanding and treating vein graft atherosclerosis Cardiovasc Pathol 2002 11 263 271 12361836 10.1016/S1054-8807(02)00125-4
Campeau L Enjalbert M Lesperance J Bourassa MG Kwiterovich PJ Wacholder S Sniderman A The relation of risk factors to the development of atherosclerosis in saphenous-vein bypass grafts and the progression of disease in the native circulation. A study 10 years after aortocoronary bypass surgery N Engl J Med 1984 311 1329 1332 6333635
Bourassa MG Fate of venous grafts: the past, the present and the future J Am Coll Cardiol 1991 17 1081 1083 2007707
Fitzgibbon GM Kafka HP Leach AJ Keon WJ Hooper GD Burton JR Coronary bypass graft fate and patient outcome: angiographic follow-up of 5,065 grafts related to survival and reoperation in 1,388 patients during 25 years J Am Coll Cardiol 1996 28 616 626 8772748 10.1016/0735-1097(96)00206-9
Alderman EL Corley SD Fisher LD Chaitman BR Faxon DP Foster ED Killip T Sosa JA Bourassa MG Five-year angiographic follow-up of factors associated with progression of coronary artery disease in the Coronary Artery Surgery Study (CASS). CASS Participating Investigators and Staff J Am Coll Cardiol 1993 22 1141 1154 8409054
van Brussel BL Ernst JM Ernst NM Kelder HC Knaepen PJ Plokker HW Vermeulen FE Voors AA Clinical outcome in venous coronary artery bypass grafting: a 15-year follow-up study Int J Cardiol 1997 58 119 126 9049676 10.1016/S0167-5273(96)02858-6
Motwani JG Topol EJ Aortocoronary saphenous vein graft disease: pathogenesis, predisposition, and prevention Circulation 1998 97 916 931 9521341
Bryan AJ Angelini GD The biology of saphenous vein graft occlusion: etiology and strategies for prevention Curr Opin Cardiol 1994 9 641 649 7819622
Roubos N Rosenfeldt FL Richards SM Conyers RA Davis BB Improved preservation of saphenous vein grafts by the use of glyceryl trinitrate-verapamil solution during harvesting Circulation 1995 92 II31 6 7586430
Verrier ED Boyle EMJ Endothelial cell injury in cardiovascular surgery Ann Thorac Surg 1996 62 915 922 8784042 10.1016/S0003-4975(96)00528-0
Fingerle J Johnson R Clowes AW Majesky MW Reidy MA Role of platelets in smooth muscle cell proliferation and migration after vascular injury in rat carotid artery Proc Natl Acad Sci U S A 1989 86 8412 8416 2813399
Ross R The pathogenesis of atherosclerosis--an update N Engl J Med 1986 314 488 500 3511384
Jawien A Bowen-Pope DF Lindner V Schwartz SM Clowes AW Platelet-derived growth factor promotes smooth muscle migration and intimal thickening in a rat model of balloon angioplasty J Clin Invest 1992 89 507 511 1531345
Herbert JM Dol F Bernat A Falotico R Lale A Savi P The antiaggregating and antithrombotic activity of clopidogrel is potentiated by aspirin in several experimental models in the rabbit Thromb Haemost 1998 80 512 518 9759636
Butany JW David TE Ojha M Histological and morphometric analyses of early and late aortocoronary vein grafts and distal anastomoses Can J Cardiol 1998 14 671 677 9627522
Cavusoglu E Cheng J Bhatt R Kunamneni PB Marmur JD Eng C Clopidogrel in the management of ischemic heart disease Heart Dis 2003 5 144 152 12713681 10.1097/01.HDX.0000061697.22030.e5
Bhatt DL Topol EJ Antiplatelet and anticoagulant therapy in the secondary prevention of ischemic heart disease Med Clin North Am 2000 84 163 79, ix 10685133 10.1016/S0025-7125(05)70212-1
Helgason CM Bolin KM Hoff JA Winkler SR Mangat A Tortorice KL Brace LD Development of aspirin resistance in persons with previous ischemic stroke Stroke 1994 25 2331 2336 7974569
Gum PA Kottke-Marchant K Welsh PA White J Topol EJ A prospective, blinded determination of the natural history of aspirin resistance among stable patients with cardiovascular disease J Am Coll Cardiol 2003 41 961 965 12651041 10.1016/S0735-1097(02)03014-0
Buchanan MR Brister SJ Individual variation in the effects of ASA on platelet function: implications for the use of ASA clinically Can J Cardiol 1995 11 221 227 7889440
Chopra V Marmur JD Cavusoglu E The role of clopidogrel in the management of patients with ischemic heart disease Cardiovasc Drugs Ther 2003 17 467 477 15107602 10.1023/B:CARD.0000015862.62649.c8
Hollopeter G Jantzen HM Vincent D Li G England L Ramakrishnan V Yang RB Nurden P Nurden A Julius D Conley PB Identification of the platelet ADP receptor targeted by antithrombotic drugs Nature 2001 409 202 207 11196645 10.1038/35051599
Herbert JM Tissinier A Defreyn G Maffrand JP Inhibitory effect of clopidogrel on platelet adhesion and intimal proliferation after arterial injury in rabbits Arterioscler Thromb 1993 13 1171 1179 8343491
Clowes AW Karnovsky MJ Failure of certain antiplatelet drugs to affect myointimal thickening following arterial endothelial injury in the rat Lab Invest 1977 36 452 464 846193
Hermann A Weber AA Schror K Clopidogrel inhibits platelet adhesion and platelet-dependent mitogenesis in vascular smooth muscle cells Thromb Res 2002 105 173 175 11958809 10.1016/S0049-3848(01)00403-0
Harker LA Marzec UM Kelly AB Chronos NR Sundell IB Hanson SR Herbert JM Clopidogrel inhibition of stent, graft, and vascular thrombogenesis with antithrombotic enhancement by aspirin in nonhuman primates Circulation 1998 98 2461 2469 9832493
Cadroy Y Bossavy JP Thalamas C Sagnard L Sakariassen K Boneu B Early potent antithrombotic effect with combined aspirin and a loading dose of clopidogrel on experimental arterial thrombogenesis in humans Circulation 2000 101 2823 2828 10859288
A randomised, blinded, trial of clopidogrel versus aspirin in patients at risk of ischaemic events (CAPRIE). CAPRIE Steering Committee Lancet 1996 348 1329 1339 8918275 10.1016/S0140-6736(96)09457-3
Bhatt DL Chew DP Hirsch AT Ringleb PA Hacke W Topol EJ Superiority of clopidogrel versus aspirin in patients with prior cardiac surgery Circulation 2001 103 363 368 11157686
Yusuf S Zhao F Mehta SR Chrolavicius S Tognoni G Fox KK Effects of clopidogrel in addition to aspirin in patients with acute coronary syndromes without ST-segment elevation N Engl J Med 2001 345 494 502 11519503 10.1056/NEJMoa010746
Saw J Topol EJ Steinhubl SR Brennan D Berger PB Moliterno DJ Comparison of long-term usefulness of clopidogrel therapy after the first percutaneous coronary intervention or coronary artery bypass grafting versus that after the second or repeat intervention Am J Cardiol 2004 94 623 625 15342293 10.1016/j.amjcard.2004.05.026
Hirsh J Bhatt DL Comparative benefits of clopidogrel and aspirin in high-risk patient populations: lessons from the CAPRIE and CURE studies Arch Intern Med 2004 164 2106 2110 15505123 10.1001/archinte.164.19.2106
Fox KA Mehta SR Peters R Zhao F Lakkis N Gersh BJ Yusuf S Benefits and risks of the combination of clopidogrel and aspirin in patients undergoing surgical revascularization for non-ST-elevation acute coronary syndrome: the Clopidogrel in Unstable angina to prevent Recurrent ischemic Events (CURE) Trial Circulation 2004 110 1202 1208 15313956 10.1161/01.CIR.0000140675.85342.1B
Zimmermann N Wenk A Kim U Kienzle P Weber AA Gams E Schror K Hohlfeld T Functional and biochemical evaluation of platelet aspirin resistance after coronary artery bypass surgery Circulation 2003 108 542 547 12874188 10.1161/01.CIR.0000081770.51929.5A
Pearson T Rapaport E Criqui M Furberg C Fuster V Hiratzka L Little W Ockene I Williams G Optimal risk factor management in the patient after coronary revascularization. A statement for healthcare professionals from an American Heart Association Writing Group Circulation 1994 90 3125 3133 7994868
Eagle KA Guyton RA Davidoff R Edwards FH Ewy GA Gardner TJ Hart JC Herrmann HC Hillis LD Hutter AMJ Lytle BW Marlow RA Nugent WC Orszulak TA ACC/AHA 2004 guideline update for coronary artery bypass graft surgery: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Update the 1999 Guidelines for Coronary Artery Bypass Graft Surgery) Circulation 2004 110 e340 437 15466654 10.1161/01.CIR.0000138790.14877.7D
Smith SCJ Blair SN Bonow RO Brass LM Cerqueira MD Dracup K Fuster V Gotto A Grundy SM Miller NH Jacobs A Jones D Krauss RM Mosca L Ockene I Pasternak RC Pearson T Pfeffer MA Starke RD Taubert KA AHA/ACC Scientific Statement: AHA/ACC guidelines for preventing heart attack and death in patients with atherosclerotic cardiovascular disease: 2001 update: A statement for healthcare professionals from the American Heart Association and the American College of Cardiology Circulation 2001 104 1577 1579 11571256
The Thrombolysis in Myocardial Infarction (TIMI) trial. Phase I findings. TIMI Study Group N Engl J Med 1985 312 932 936 4038784
Tardif JC Gregoire J Lesperance J Lambert J L'Allier PL Rodes J Anderson T Blue JW Imus J Heinonen T Design features of the Avasimibe and Progression of coronary Lesions assessed by intravascular UltraSound (A-PLUS) clinical trial Am Heart J 2002 144 589 596 12360153
Hozumi T Yoshikawa J Yoshida K Akasaka T Takagi T Honda Y Okura H Use of intravascular ultrasound for in vivo assessment of changes in intimal thickness of angiographically normal saphenous vein grafts one year after aortocoronary bypass surgery Heart 1996 76 317 320 8983677
Moher D Schulz KF Altman DG The CONSORT statement: revised recommendations for improving the quality of reports of parallel-group randomised trials Lancet 2001 357 1191 1194 11323066 10.1016/S0140-6736(00)04337-3
|
16255773
|
PMC1282585
|
CC BY
|
2021-01-04 16:37:29
|
no
|
Dyn Med. 2005 Oct 28; 4:9
|
latin-1
|
Dyn Med
| 2,005 |
10.1186/1476-5918-4-9
|
oa_comm
|
==== Front
Curr Control Trials Cardiovasc MedCurrent Controlled Trials in Cardiovascular Medicine1468-67081468-6694BioMed Central 1468-6708-6-151621910010.1186/1468-6708-6-15Study ProtocolThe clopidogrel after surgery for coronary artery disease (CASCADE) randomized controlled trial: clopidogrel and aspirin versus aspirin alone after coronary bypass surgery [NCT00228423] Kulik Alexander [email protected] May Michel [email protected] George A [email protected] Thierry G [email protected] Marc [email protected] Division of Cardiac Surgery, University of Ottawa Heart Institute, Ottawa, Canada2 Division of Cardiology, University of Ottawa Heart Institute, Ottawa, Canada3 Department of Epidemiology and Community Medicine, University of Ottawa, Ottawa, Canada4 Division of Cardiac Surgery, University of Ottawa Heart Institute, Ottawa, Canada5 Division of Cardiac Surgery, University of Ottawa Heart Institute, and Department of Epidemiology and Community Medicine, University of Ottawa, Ottawa, Canada2005 11 10 2005 6 1 15 15 3 10 2005 11 10 2005 Copyright © 2005 Kulik et al; licensee BioMed Central Ltd.2005Kulik et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
Saphenous vein graft disease remains a major limitation of coronary artery bypass graft surgery. The process of saphenous vein intimal hyperplasia begins just days after surgical revascularization, setting the stage for graft atherosclerotic disease and its sequalae. Clopidogrel improves outcomes in patients with atherosclerotic disease, and is effective at reducing intimal hyperplasia in animal models of thrombosis. Therefore, the goal of this study will be to evaluate the efficacy of clopidogrel and aspirin therapy versus aspirin alone in the prevention of saphenous vein graft intimal hyperplasia following coronary artery bypass surgery.
Methods
Patients undergoing multi-vessel coronary artery bypass grafting and in whom at least two saphenous vein grafts will be used are eligible for the study. Patients will be randomized to receive daily clopidogrel 75 mg or placebo, in addition to daily aspirin 162 mg, for a one year duration starting on the day of surgery (as soon as postoperative bleeding has been excluded). At the end of one year, all patients will undergo coronary angiography and intravascular ultrasound assessment of one saphenous vein graft as selected by randomization. The trial will be powered to test the hypothesis that clopidogrel and aspirin will reduce vein graft intimal hyperplasia by 20% compared to aspirin alone at one year following bypass surgery.
Discussion
This trial is the first prospective human study that will address the question of whether clopidogrel therapy improves outcomes and reduces saphenous vein graft intimal hyperplasia following cardiac surgery. Should the combination of clopidogrel and aspirin reduce the process of vein graft intimal hyperplasia, the results of this study will help redefine modern antiplatelet management of coronary artery bypass patients.
coronary artery bypass graft surgeryantiplatelet therapysaphenous vein
==== Body
Background
Coronary artery bypass graft surgery (CABG) is the most durable approach for the treatment of ischemic heart disease [1], with >400,000 procedures performed annually in the United States alone [2]. Despite the increasing application of arterial conduits during CABG, the saphenous vein remains the most common conduit, employed for more than 70% of grafts [3]. However, saphenous vein graft (SVG) disease presents an important clinical problem. Even with aggressive medical therapy [4-11], up to 15% of vein grafts occlude in the first year after bypass surgery. Between 1 and 6 years, the graft attrition rate is 1% to 2% per year, and between 6 and 10 years it rises to 4% per year. By 10 years after surgery, only 60% of grafts are patent and only 50% of patent vein grafts are free of significant stenosis. In addition, native coronary artery disease progresses in 5% of patients annually [12-15]. Reflecting the graft and native vessel attrition, this population is at high risk for subsequent ischemic events, including death, myocardial infarction (MI) and stroke [14]. Further revascularization, either reoperation or percutaneous coronary intervention, is required in 4% of patients by 5 years, 19% of patients by 10 years and 31% of patients by 12 years after the initial bypass surgery [3,16].
The process of SVG disease is composed of three mechanistically interlinked stages: thrombosis, intimal hyperplasia, and atherosclerosis [13,14,16-18]. Early graft thrombosis can occur at the time of surgery secondary to focal endothelial disruption [19,20]. Grafts that survive this early period develop a progressive thickening of the media that begins within days after implantation. This process, termed intimal hyperplasia, is a consequence of smooth muscle cell proliferation and extracellular matrix protein synthesis [21,22]. Platelets play a fundamental role in the process of smooth muscle cell proliferation and intimal hyperplasia [23,24]. Intimal hyperplasia is present in all grafts 1 month after implantation [25] and forms a template for the development of superimposed atherosclerotic changes [17,18]. With the passage of a sufficient period of time, the thrombotic occlusion of vein grafts is almost inevitable due to progressive atherosclerosis [17].
Despite its established benefit in patients with coronary artery disease, aspirin therapy has numerous limitations. It is a relatively weak antiplatelet agent and has no effect on thrombin, which is believed to play a major role in acute coronary syndrome [26]. Even with aspirin therapy for secondary prevention, a large number of recurrent events occur [27]. A significant proportion of patients undergoing CABG may be aspirin resistant, defined as undetectable platelet inhibition after one week of therapy [28,29]. Depending on the population studied and the specific definition of aspirin resistance, anywhere from 10–40% of patients appear to have an inadequate antiplatelet response to aspirin [28,30]. Such patients appear to be at increased risk for the development of vascular events. In theory, these aspirin-resistant patients may derive particular benefit from additional antiplatelet therapy [31].
Clopidogrel is a thienopyridine antiplatelet agent that inhibits ADP-dependent platelet activation and aggregation [32]. Sevenfold more potent than ticlopidine, clopidogrel is free of its adverse side effects such as neutropenia, diarrhea and rash [33]. Unlike aspirin [24,34], clopidogrel has been shown to inhibit the process of platelet-mediated intimal proliferation and smooth muscle hyperplasia in laboratory experiments. In a cell culture model, clopidogrel significantly inhibited platelet adhesion to immobilized fibrinogen and also inhibited platelet-dependent mitogenic signaling and DNA synthesis in cultured coronary artery smooth muscle cells [35]. Similarly, in animal thrombosis models, clopidogrel but not aspirin significantly inhibited platelet-mediated intimal proliferation and smooth muscle hyperplasia [33,36]. Furthermore, the combination of clopidogrel with aspirin led to potent synergistic antithrombotic effects and a decrease in myointimal proliferation compared to either therapy alone [24,36,37].
Several large clinical trials have demonstrated that clopidogrel reduces ischemic events and mortality in patients with coronary and vascular disease [38-41]. In the CAPRIE (Clopidogrel versus Aspirin in Patients with Ischemic Events) trial, clopidogrel (75 mg/day) was demonstrated to be significantly more effective than aspirin (325 mg/day) in preventing vascular thrombotic events (ischemic stroke, MI or vascular death) in patients with clinical evidence of atherosclerotic disease (clopidogrel 9.78% vs. aspirin 10.64%, relative risk reduction [RRR] 8.7%, p = 0.045) [38]. In patients presenting with acute coronary syndromes, the CURE (Clopidogrel in Unstable angina to prevent Recurrent ischemic Events) study demonstrated that the combination of clopidogrel and aspirin was more effective at reducing the primary outcome (cardiovascular death, nonfatal MI or stroke) compared to aspirin alone (clopidogrel and aspirin 9.3% vs. aspirin alone 11.4%, RRR 20%, p < 0.001) [40]. Subgroup analysis from these trials suggested that patients that underwent surgical revascularization also benefited from clopidogrel [39,42,43]. However, no trial to date has prospectively evaluated the combined effects of clopidogrel plus aspirin on saphenous vein graft disease after CABG.
There currently exists a clinical equipoise regarding the optimal antiplatelet therapy for patients who have undergone coronary artery bypass surgery. While some clinicians believe in the beneficial effects of clopidogrel, the increased risk of bleeding and the lack of data in CABG patients make it impossible to establish definitive recommendations. We therefore propose the Clopidogrel after Surgery for Coronary Artery Disease (CASCADE) study, a randomized, placebo-controlled trial comparing clopidogrel plus aspirin versus aspirin alone in CABG patients revascularized with saphenous vein. The primary aim of this study will be to evaluate the effect of combined antiplatelet therapy on the reduction of SVG intimal hyperplasia one year after CABG, through the assessment of intimal area by intravascular ultrasound (IVUS). Secondary aims will evaluate the safety of clopidogrel administration following CABG with regards to bleeding complications. We hypothesize that the combination of clopidogrel with aspirin will reduce the SVG intimal hyperplasia (intimal area) by 20% one year post-CABG compared to the usual antiplatelet therapy of aspirin alone.
Methods
Study Population and Recruitment Procedure
The study population will include all patients undergoing multi-vessel elective or urgent CABG using at least two saphenous vein grafts at the University of Ottawa Heart Institute (OHI) over the study period (see Table 1 for inclusion and exclusion criteria). Patients undergoing off-pump CABG (OPCAB) will also be eligible for this study, as long as at least two saphenous vein grafts are used. OPCAB is performed in order to avoid the hazards associated with standard CABG, such as cardiopulmonary bypass and aortic cross-clamping, and is carried out at the discretion of the surgeon in patients deemed to be at higher risk of thromboembolic or renal complications during surgery. All CABG patients at the OHI will be triaged pre-operatively, and study eligible patients will be selected and approached by the study nurse to explain the trial and obtain consent.
Description of Intervention and Control
A prospective randomized double-blinded placebo-controlled study will be conducted from November 2005 to November 2007 in order to achieve the study objectives. Patients will be recruited over the first 12 months of the study, and graft evaluation for each patient will occur over the following 12 months (one year after surgery for each patient). Patients will be randomized into an experimental group (receiving clopidogrel) or a control group (placebo). The placebo and clopidogrel medications will be prepared by the Bristol Myers-squibb Sanofi Canada Partnership and appear identical. Medication administration and data collection will be performed in a double-blind manner, such that neither the patient nor the healthcare personnel will be aware of the medication assignment. Recruitment and written consent will be performed prior to surgery. However, patients will not be randomized until after surgery has been completed and clinical stability ensured. Patients that are bleeding excessively after surgery (chest tube output > 200 cc/hr) or those requiring high levels of hemodynamic support (more than 2 inotropes and/or intra-aortic balloon pump) will not be randomized into the study.
After surgery, the study medication will be administered via nasogastric tube when the chest tube drainage has decreased to ≤50 cc/hr for 2 hours. Each patient will receive either clopidogrel 75 mg or placebo, in addition to enteric coated aspirin 162 mg (Figure 1). The study drug and aspirin 162 mg will be repeated orally in the same dose once daily for the duration of one year. Because post-CABG patients are relatively aspirin resistant following surgery, the aspirin dose of 162 mg will ensure adequate platelet inhibition in those patients randomized to aspirin alone. Furthermore, 162 mg falls within the safety window of combining aspirin with clopidogrel [30,40,44].
Figure 1 text
Allocation Procedure
A stratified random design allocation will be utilized to account for the presence or absence of diabetes, as well as the use or nonuse of cardiopulmonary bypass (standard CABG versus OPCAB). A block randomization technique will ensure an equal distribution of diabetic patients in both arms of the trial and an equal distribution of OPCAB patients. The randomization schedule will be generated using SAS 9.1 software (SAS, Cary, NC). All patients and study personnel will be blinded to the treatment assignment, which will be performed by the hospital pharmacy.
Concomitant Medication and Treatments
Patients will receive concomitant therapies in both groups as recommended by the current American College of Cardiology / American Heart Association guidelines. This will include smoking cessation counseling and the administration of aspirin, beta blockers, angiotensin converting enzyme inhibitors, and lipid lowering medications. Target LDL values will be those recommended as per current guidelines [45-47] and will be assessed during the study follow-up period. Routine peptic ulcer prophylaxis will not be administered in order to fully evaluate gastrointestinal side effects.
Diabetic patients will be eligible for enrollment in this study, regardless of their preoperative need for insulin therapy, and will be allocated equally into both groups through stratified randomization. Diabetic patients will have aggressive perioperative glycemic control, including an intravenous insulin infusion both in the operating room and in the intensive care unit, and a subcutaneous insulin sliding scale while recovering on the surgical ward. Once drinking well, diabetic patients will be restarted on their original preoperative diabetic regimens (oral agents and/or insulin therapy). The treatment of diabetes during this study will be closely monitored in collaboration with an endocrinologist specializing in the management of diabetes.
Primary Outcomes
The primary endpoint of this study will be to assess whether the addition of clopidogrel to aspirin reduces intimal hyperplasia in saphenous vein grafts 12 months after bypass surgery, as assessed by IVUS. Patients will undergo IVUS imaging 12 months post-CABG, and the average intimal area in the proximal 40 mm of one vein graft per patient will be assessed.
Secondary Outcomes
At the time of intravascular ultrasound, coronary and graft angiography will also be performed to assess vein graft patency and areas of stenosis. Although this trial will not be sufficiently powered for such a purpose, this data will be obtained whilst gathering information pertaining to the primary outcome.
Endpoints related to safety will also be documented, both at the time of surgery as well as during the one year of study drug administration. After surgery, data will be recorded regarding chest tube blood loss, blood product transfusions, bleeding requiring tube thoracostomy or sternal re-opening, perioperative MI, and gastrointestinal complications. Complete blood counts (CBC) will assess the hemoglobin level in the immediate postoperative period and during the one year follow-up. In addition, the incidence of major adverse cardiovascular events following CABG (mortality, MI, cerebrovascular accident, hospitalization for coronary ischemia, need for coronary intervention) will be recorded.
IVUS Procedure
Intravascular ultrasound will be used to assess the area of intimal hyperplasia present in saphenous vein grafts at 12 months after surgery. IVUS differs from angiography by providing cross-sectional images of both the vessel wall and lumen with high resolution. The process of intimal hyperplasia is easily detectable and quantifiable by IVUS, but may completely escape visualization by angiography. At one year following CABG, vein graft intimal hyperplasia is universally present.
Efforts will be made to schedule all patients at 52 ± 2 weeks from the day of randomization (Figure 1). The IVUS procedure will first start with angiography of the native coronary arteries and the coronary bypass grafts. This will allow assessment of the progression of native coronary artery atherosclerosis (both in grafted and non-grafted vessels). Patency of all bypass grafts will be scored using 1) the Fitzgibbon method [14] and 2) the TIMI classification [48]. Patency of the coronary arteries will be assessed by the TIMI method.
IVUS studies will be performed using a 40 MHz imaging catheter (Atlantis® SR Pro, Boston Scientific). This catheter is a monorail system and has 6F guiding catheter compatibility. All IVUS imaging will be done with the administration of unfractionated heparin (70 units per kg, minimum 4000 units) before the introduction of the guidewire into a vein graft. Each patient will have at least two vein grafts implanted at the time of surgery. However, imaging of more than one graft is not clinically advisable for safety and practical reasons. In order to minimize bias, the cardiologist performing the IVUS procedure will be blinded to the treatment allocation, and the selection of the SVG for IVUS imaging will be randomized. For this purpose, a sequence randomization scheme (based on the number of vein grafts) will produce a random sequence of numbers for each patient, such as "2, 1 and 3". These numbers will correspond to the position of the proximal anastomosis of each SVG on the ascending aorta (by increasing number from cranial to caudal, starting from 1). The graft whose random number is produced first will be selected for intubation with the IVUS catheter, unless 1) this particular graft showed more than 50% stenosis on the selective angiography performed immediately prior, or 2) access to this graft is technically difficult (graft tortuosity, difficult graft intubation). If any of the above two instances is encountered, the graft with the next number in the randomization sequence will instead be selected for IVUS, and so on. Graft exclusion for whatever basis (either because >50% stenosis or technically difficult access) will be recorded for all patients in each group.
Once the randomized graft selection has been completed, the actual IVUS procedure will take place. Intracoronary nitroglycerin (200 μg) will be given before advancing the IVUS catheter, and a 6F guiding catheter will be used to engage the vein graft. Then, a 0.014" coronary angioplasty guidewire will be advanced distally through the vein graft and positioned into the native coronary artery. The IVUS catheter will then be advanced into the graft at least 50 mm beyond the aorto-ostial anastomosis. The guiding catheter will then be disengaged to ensure visualization of the aorto-ostial anastomosis on pullback. This is essential because the aorto-ostial anastomosis will be the only landmark available to assure measurements of the proximal portion of each graft are comparable between groups. IVUS imaging will be done using a validated motorized pullback device at 0.5 mm/sec. Each study will be recorded on a separate S-VHS videotape. Only one quality pullback will be needed per graft, but a second pullback may be repeated if the first set of pullback images were judged suboptimal in quality.
IVUS Analysis
The IVUS images will be sent to an independent core laboratory and interpreted by an experienced cardiologist blinded to treatment allocation. The methods for the analysis have been previously reported and validated [49]. Briefly, using the aorto-ostial anastomosis as a landmark, the most proximal 40 mm of vein graft will be analyzed. The video images will be digitized and measurements of lumen, intimal hyperplasia, and external elastic lamina areas will be available for each digitized cross section. Intimal hyperplasia volumes will be computed by multiplying the corresponding areas of each of the cross-sections by the distance between slices and by adding the products. For the purpose of this study, the mean plaque area per patient for the 40 mm-analyzed segment will be used for comparison between treatment groups.
Sample Size
The trial will be powered to test the hypothesis that clopidogrel and aspirin should reduce vein graft intimal hyperplasia by 20% compared to aspirin alone at one year following bypass surgery. According to Hozumi et al., at one year, the mean intimal area of angiographically normal saphenous vein grafts is 5.26 mm2, with a standard deviation of 1.38 mm2 [50]. Intravascular ultrasound imaging will be performed at follow-up in one vein graft per patient. In order to account for potential angiographic refusals and study withdrawals, approximately 100 patients in total will be required to test the null hypothesis with an α value of 0.05 and a power of 0.90.
Data Collection and Safety Monitoring
Intimal area will be recorded at the time of IVUS one year after CABG. Vein graft patency and stenosis will be assessed by angiography. The incidence of major adverse coronary events and bleeding complications will be documented during postoperative clinic visits at one month, six months and twelve months after surgery. Telephone home assessments every three months will also be used to document events. All serious adverse events will be reported to the ethics committee. The development of serious adverse events that might be attributable to clopidogrel will lead to the termination of the study drug.
Ethics
The study will adhere to the highest research ethics standards of the OHI. This protocol follows the CONSORT guidelines [51] and was approved on July 19, 2005 by the University of Ottawa Heart Institute Human Research Ethics Board.
Statistical Analysis
Vein graft intimal area, the primary outcome of the study, will be compared between the two randomization groups using two-sided Student's t tests. Vein graft patency will be compared using a Fisher's exact test. With respect to the secondary outcomes, continuous data will be compared between the two groups using two-sided Student's t tests, two-sample Wilcoxon rank-sum tests, or ANOVA as appropriate, and a Fisher's exact test will be used for categorical data. In order to assess possible interactions between patient characteristics (such as diabetes) and treatment outcomes, an exploratory analysis using multivariate linear regression will be performed. In the improbable event that all vein grafts are occluded and IVUS cannot be performed in a particular patient undergoing the one-year study, a value of five times the mean intimal area (decided a priori) will be assigned for that patient.
Discussion
The CASCADE study is a novel randomized double-blind placebo-controlled trial that will help clarify the controversial issue of antiplatelet therapy following CABG surgery. Specifically, it will answer the questions of whether the addition of clopidogrel to aspirin is safe following cardiac surgery and whether it reduces saphenous vein graft intimal hyperplasia. Although subgroup analyses of previous trials have suggested a benefit of clopidogrel in cardiac surgery patients [39,42,43], no trial to date has specifically focused on the clinical or angiographic outcomes in patients treated with clopidogrel therapy immediately after surgical revascularization. Should the combination of clopidogrel and aspirin reduce the process of vein graft intimal hyperplasia, the CASCADE trial has the potential to redefine modern antiplatelet management of coronary artery bypass patients.
We believe that the strengths of this study are its randomized and blinded design. Patients will be randomized into two treatment groups, and graft selection for IVUS imaging will be randomized within each patient. Blinding will occur for all patients and health care providers, and the IVUS and angiogram images will be interpreted by an external blinded core laboratory. While the process of intimal hyperplasia may be reduced in the trial, the study will not be powered to demonstrate a difference in angiographic patency or freedom from coronary events following CABG. In order to demonstrate a difference in either of these outcomes, a considerably larger sample size and longer follow-up period would be required.
The results of this randomized trial will be generalizable to all patients undergoing standard CABG or OPCAB surgery with saphenous vein. Although the trial enrolment is limited to lower risk patients, it is anticipated that the results will also be applicable to higher risk patients with severe ischemic left ventricular dysfunction or those undergoing redo-CABG. High-risk patients participating in previous clopidogrel studies have been found to derive an enhanced benefit with the combined treatment of clopidogrel and aspirin [42]. Furthermore, there is no reason to anticipate differences in the pathophysiology of vein graft disease between high risk patients and those of the proposed study group.
Conclusion
Saphenous vein graft disease continues to be a major limitation of surgical revascularization for coronary artery disease. The process of saphenous vein intimal hyperplasia begins just days after surgery, setting the stage for graft atherosclerotic disease and its sequalae. Clopidogrel has been demonstrated to improve outcomes in patients with coronary and vascular disease, and it is effective at reducing intimal hyperplasia in animal models of thrombosis. However, no prospective study to date has been conducted in humans regarding the use of clopidogrel after CABG to prevent vein graft intimal hyperplasia. The CASCADE trial is a randomized, placebo-controlled trial comparing clopidogrel plus aspirin versus aspirin alone in CABG patients revascularized with saphenous vein. The effects of clopidogrel on vein graft intimal hyperplasia will be studied with coronary angiography and intravascular ultrasound one year following CABG.
Abbreviations
CABG – coronary artery bypass graft surgery
CASCADE – Clopidogrel after Surgery for Coronary Artery Disease Trial
CBC – complete blood counts
IVUS – intravascular ultrasound
MI – myocardial infarction
OHI – University of Ottawa Heart Institute
OPCAB – off-pump coronary artery bypass graft surgery
RRR – relative risk reduction
SVG – saphenous vein graft
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
All authors read and approved the final manuscript. Specifically, AK conceived of the study, participated in the protocol design and helped draft the manuscript. MLM participated in the protocol design, helped draft the manuscript, and will perform the intravascular ultrasound procedures on study patients. GAW helped draft the manuscript, design the study protocol, and will coordinate the statistical analysis. TGM helped design the study protocol and draft the manuscript. MR is the principle investigator and helped conceive the study, design the protocol and draft the manuscript.
Table 1 Study Inclusion and Exclusion Criteria
Inclusion Criteria:
Reason:
- Patients undergoing primary multi-vessel CABG with at least two saphenous vein grafts, with or without cardiopulmonary bypass - The population of clinical interest
Preoperative Exclusion Criteria:
Reason:
- Emergency surgery - Unable to obtain consent
- Valve surgery - Requirement of postoperative anticoagulation
- Redo CABG - Higher risk of postoperative bleeding and low cardiac output syndrome
- Left ventricle ejection fraction <25% - Higher risk of postoperative low cardiac output syndrome and mortality
- Serum creatinine >130 μmol/L - Contraindication to use of postoperative angiography
- Preoperative use of clopidogrel (with the exception of the current admission) - Contraindication to randomization, confounding of results
- Preoperative use of warfarin - Requirement of postoperative anticoagulation
- Allergy to aspirin or clopidogrel - Contraindication to use of aspirin or clopidogrel
- History of cerebrovascular accident - Concurrent indication for clopidogrel
- History of severe liver disease - Contraindication to use of clopidogrel
- Morbid obesity - Unable to perform postoperative angiography
- Residence outside of Ottawa region - Unable to perform postoperative angiography
- Current malignancy - Higher risk of early postoperative mortality
- Inability to provide informed consent - Ineligible for research study enrollment
Postoperative Exclusion Criteria:
Reason:
- Low cardiac output syndrome with inotropic support for greater than 24 hours - Unstable patient unlikely to benefit from treatment
- Recurrent ventricular arrhythmias - Unstable patient unlikely to benefit from treatment
- Intubation for more than 24 hours - Unstable patient unlikely to benefit from treatment
- Postoperative bleeding or cardiac tamponade requiring surgical exploration - Contraindication to use of clopidogrel
- Postoperative gastrointestinal bleeding - Contraindication to use of clopidogrel
- Postoperative warfarin requirement - Contraindication to use of clopidogrel
Acknowledgements
The CASCADE trial is funded through several sources, including a Physicians' Services Incorporated Foundation resident research grant, as well as grants from Boston Scientific Inc. and financial support from the Bristol-Myers Squibb Sanofi Canada Partnership. Boston Scientific will supply the intravascular ultrasound (IVUS) catheters used during the course of this study. The Bristol-Myers Squibb Sanofi Canada Partnership will supply the study medications (clopidogrel and placebo), and it has agreed to provide direct financial support during the trial. Drs. Laurent Carty and Stanislav Glezer from Sanofi-Aventis and Dr. Olga Tarasova from Bristol-Myers Squibb critically reviewed the protocol design during its development.
==== Refs
Hannan EL Racz MJ Walford G Jones RH Ryan TJ Bennett E Culliford AT Isom OW Gold JP Rose EA Long-term outcomes of coronary-artery bypass grafting versus stent implantation N Engl J Med 2005 352 2174 2183 15917382 10.1056/NEJMoa040316
American Heart Association: Heart Disease and Stroke Statistics - 2004 Update
Weintraub WS Jones EL Craver JM Guyton RA Frequency of repeat coronary bypass or coronary angioplasty after coronary artery bypass surgery using saphenous venous grafts Am J Cardiol 1994 73 103 112 8296729 10.1016/0002-9149(94)90198-8
Collaborative meta-analysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk patients Bmj 2002 324 71 86 11786451 10.1136/bmj.324.7336.S71
Goldman S Copeland J Moritz T Henderson W Zadina K Ovitt T Doherty J Read R Chesler E Sako Y Improvement in early saphenous vein graft patency after coronary artery bypass surgery with antiplatelet therapy: results of a Veterans Administration Cooperative Study Circulation 1988 77 1324 1332 3286040
Goldman S Copeland J Moritz T Henderson W Zadina K Ovitt T Doherty J Read R Chesler E Sako Y Saphenous vein graft patency 1 year after coronary artery bypass surgery and effects of antiplatelet therapy. Results of a Veterans Administration Cooperative Study Circulation 1989 80 1190 1197 2680158
Goldman S Copeland J Moritz T Henderson W Zadina K Ovitt T Kern KB Sethi G Sharma GV Khuri S Long-term graft patency (3 years) after coronary artery surgery. Effects of aspirin: results of a VA Cooperative study Circulation 1994 89 1138 1143 8124800
Blankenhorn DH Nessim SA Johnson RL Sanmarco ME Azen SP Cashin-Hemphill L Beneficial effects of combined colestipol-niacin therapy on coronary atherosclerosis and coronary venous bypass grafts Jama 1987 257 3233 3240 3295315 10.1001/jama.257.23.3233
The effect of aggressive lowering of low-density lipoprotein cholesterol levels and low-dose anticoagulation on obstructive changes in saphenous-vein coronary-artery bypass grafts. The Post Coronary Artery Bypass Graft Trial Investigators N Engl J Med 1997 336 153 162 8992351 10.1056/NEJM199701163360301
Sacks FM Pfeffer MA Moye LA Rouleau JL Rutherford JD Cole TG Brown L Warnica JW Arnold JM Wun CC Davis BR Braunwald E The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. Cholesterol and Recurrent Events Trial investigators N Engl J Med 1996 335 1001 1009 8801446 10.1056/NEJM199610033351401
Sarjeant JM Rabinovitch M Understanding and treating vein graft atherosclerosis Cardiovasc Pathol 2002 11 263 271 12361836 10.1016/S1054-8807(02)00125-4
Campeau L Enjalbert M Lesperance J Bourassa MG Kwiterovich PJ Wacholder S Sniderman A The relation of risk factors to the development of atherosclerosis in saphenous-vein bypass grafts and the progression of disease in the native circulation. A study 10 years after aortocoronary bypass surgery N Engl J Med 1984 311 1329 1332 6333635
Bourassa MG Fate of venous grafts: the past, the present and the future J Am Coll Cardiol 1991 17 1081 1083 2007707
Fitzgibbon GM Kafka HP Leach AJ Keon WJ Hooper GD Burton JR Coronary bypass graft fate and patient outcome: angiographic follow-up of 5,065 grafts related to survival and reoperation in 1,388 patients during 25 years J Am Coll Cardiol 1996 28 616 626 8772748 10.1016/0735-1097(96)00206-9
Alderman EL Corley SD Fisher LD Chaitman BR Faxon DP Foster ED Killip T Sosa JA Bourassa MG Five-year angiographic follow-up of factors associated with progression of coronary artery disease in the Coronary Artery Surgery Study (CASS). CASS Participating Investigators and Staff J Am Coll Cardiol 1993 22 1141 1154 8409054
van Brussel BL Ernst JM Ernst NM Kelder HC Knaepen PJ Plokker HW Vermeulen FE Voors AA Clinical outcome in venous coronary artery bypass grafting: a 15-year follow-up study Int J Cardiol 1997 58 119 126 9049676 10.1016/S0167-5273(96)02858-6
Motwani JG Topol EJ Aortocoronary saphenous vein graft disease: pathogenesis, predisposition, and prevention Circulation 1998 97 916 931 9521341
Bryan AJ Angelini GD The biology of saphenous vein graft occlusion: etiology and strategies for prevention Curr Opin Cardiol 1994 9 641 649 7819622
Roubos N Rosenfeldt FL Richards SM Conyers RA Davis BB Improved preservation of saphenous vein grafts by the use of glyceryl trinitrate-verapamil solution during harvesting Circulation 1995 92 II31 6 7586430
Verrier ED Boyle EMJ Endothelial cell injury in cardiovascular surgery Ann Thorac Surg 1996 62 915 922 8784042 10.1016/S0003-4975(96)00528-0
Fingerle J Johnson R Clowes AW Majesky MW Reidy MA Role of platelets in smooth muscle cell proliferation and migration after vascular injury in rat carotid artery Proc Natl Acad Sci U S A 1989 86 8412 8416 2813399
Ross R The pathogenesis of atherosclerosis--an update N Engl J Med 1986 314 488 500 3511384
Jawien A Bowen-Pope DF Lindner V Schwartz SM Clowes AW Platelet-derived growth factor promotes smooth muscle migration and intimal thickening in a rat model of balloon angioplasty J Clin Invest 1992 89 507 511 1531345
Herbert JM Dol F Bernat A Falotico R Lale A Savi P The antiaggregating and antithrombotic activity of clopidogrel is potentiated by aspirin in several experimental models in the rabbit Thromb Haemost 1998 80 512 518 9759636
Butany JW David TE Ojha M Histological and morphometric analyses of early and late aortocoronary vein grafts and distal anastomoses Can J Cardiol 1998 14 671 677 9627522
Cavusoglu E Cheng J Bhatt R Kunamneni PB Marmur JD Eng C Clopidogrel in the management of ischemic heart disease Heart Dis 2003 5 144 152 12713681 10.1097/01.HDX.0000061697.22030.e5
Bhatt DL Topol EJ Antiplatelet and anticoagulant therapy in the secondary prevention of ischemic heart disease Med Clin North Am 2000 84 163 79, ix 10685133 10.1016/S0025-7125(05)70212-1
Helgason CM Bolin KM Hoff JA Winkler SR Mangat A Tortorice KL Brace LD Development of aspirin resistance in persons with previous ischemic stroke Stroke 1994 25 2331 2336 7974569
Gum PA Kottke-Marchant K Welsh PA White J Topol EJ A prospective, blinded determination of the natural history of aspirin resistance among stable patients with cardiovascular disease J Am Coll Cardiol 2003 41 961 965 12651041 10.1016/S0735-1097(02)03014-0
Buchanan MR Brister SJ Individual variation in the effects of ASA on platelet function: implications for the use of ASA clinically Can J Cardiol 1995 11 221 227 7889440
Chopra V Marmur JD Cavusoglu E The role of clopidogrel in the management of patients with ischemic heart disease Cardiovasc Drugs Ther 2003 17 467 477 15107602 10.1023/B:CARD.0000015862.62649.c8
Hollopeter G Jantzen HM Vincent D Li G England L Ramakrishnan V Yang RB Nurden P Nurden A Julius D Conley PB Identification of the platelet ADP receptor targeted by antithrombotic drugs Nature 2001 409 202 207 11196645 10.1038/35051599
Herbert JM Tissinier A Defreyn G Maffrand JP Inhibitory effect of clopidogrel on platelet adhesion and intimal proliferation after arterial injury in rabbits Arterioscler Thromb 1993 13 1171 1179 8343491
Clowes AW Karnovsky MJ Failure of certain antiplatelet drugs to affect myointimal thickening following arterial endothelial injury in the rat Lab Invest 1977 36 452 464 846193
Hermann A Weber AA Schror K Clopidogrel inhibits platelet adhesion and platelet-dependent mitogenesis in vascular smooth muscle cells Thromb Res 2002 105 173 175 11958809 10.1016/S0049-3848(01)00403-0
Harker LA Marzec UM Kelly AB Chronos NR Sundell IB Hanson SR Herbert JM Clopidogrel inhibition of stent, graft, and vascular thrombogenesis with antithrombotic enhancement by aspirin in nonhuman primates Circulation 1998 98 2461 2469 9832493
Cadroy Y Bossavy JP Thalamas C Sagnard L Sakariassen K Boneu B Early potent antithrombotic effect with combined aspirin and a loading dose of clopidogrel on experimental arterial thrombogenesis in humans Circulation 2000 101 2823 2828 10859288
A randomised, blinded, trial of clopidogrel versus aspirin in patients at risk of ischaemic events (CAPRIE). CAPRIE Steering Committee Lancet 1996 348 1329 1339 8918275 10.1016/S0140-6736(96)09457-3
Bhatt DL Chew DP Hirsch AT Ringleb PA Hacke W Topol EJ Superiority of clopidogrel versus aspirin in patients with prior cardiac surgery Circulation 2001 103 363 368 11157686
Yusuf S Zhao F Mehta SR Chrolavicius S Tognoni G Fox KK Effects of clopidogrel in addition to aspirin in patients with acute coronary syndromes without ST-segment elevation N Engl J Med 2001 345 494 502 11519503 10.1056/NEJMoa010746
Saw J Topol EJ Steinhubl SR Brennan D Berger PB Moliterno DJ Comparison of long-term usefulness of clopidogrel therapy after the first percutaneous coronary intervention or coronary artery bypass grafting versus that after the second or repeat intervention Am J Cardiol 2004 94 623 625 15342293 10.1016/j.amjcard.2004.05.026
Hirsh J Bhatt DL Comparative benefits of clopidogrel and aspirin in high-risk patient populations: lessons from the CAPRIE and CURE studies Arch Intern Med 2004 164 2106 2110 15505123 10.1001/archinte.164.19.2106
Fox KA Mehta SR Peters R Zhao F Lakkis N Gersh BJ Yusuf S Benefits and risks of the combination of clopidogrel and aspirin in patients undergoing surgical revascularization for non-ST-elevation acute coronary syndrome: the Clopidogrel in Unstable angina to prevent Recurrent ischemic Events (CURE) Trial Circulation 2004 110 1202 1208 15313956 10.1161/01.CIR.0000140675.85342.1B
Zimmermann N Wenk A Kim U Kienzle P Weber AA Gams E Schror K Hohlfeld T Functional and biochemical evaluation of platelet aspirin resistance after coronary artery bypass surgery Circulation 2003 108 542 547 12874188 10.1161/01.CIR.0000081770.51929.5A
Pearson T Rapaport E Criqui M Furberg C Fuster V Hiratzka L Little W Ockene I Williams G Optimal risk factor management in the patient after coronary revascularization. A statement for healthcare professionals from an American Heart Association Writing Group Circulation 1994 90 3125 3133 7994868
Eagle KA Guyton RA Davidoff R Edwards FH Ewy GA Gardner TJ Hart JC Herrmann HC Hillis LD Hutter AMJ Lytle BW Marlow RA Nugent WC Orszulak TA ACC/AHA 2004 guideline update for coronary artery bypass graft surgery: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Update the 1999 Guidelines for Coronary Artery Bypass Graft Surgery) Circulation 2004 110 e340 437 15466654 10.1161/01.CIR.0000138790.14877.7D
Smith SCJ Blair SN Bonow RO Brass LM Cerqueira MD Dracup K Fuster V Gotto A Grundy SM Miller NH Jacobs A Jones D Krauss RM Mosca L Ockene I Pasternak RC Pearson T Pfeffer MA Starke RD Taubert KA AHA/ACC Scientific Statement: AHA/ACC guidelines for preventing heart attack and death in patients with atherosclerotic cardiovascular disease: 2001 update: A statement for healthcare professionals from the American Heart Association and the American College of Cardiology Circulation 2001 104 1577 1579 11571256
The Thrombolysis in Myocardial Infarction (TIMI) trial. Phase I findings. TIMI Study Group N Engl J Med 1985 312 932 936 4038784
Tardif JC Gregoire J Lesperance J Lambert J L'Allier PL Rodes J Anderson T Blue JW Imus J Heinonen T Design features of the Avasimibe and Progression of coronary Lesions assessed by intravascular UltraSound (A-PLUS) clinical trial Am Heart J 2002 144 589 596 12360153
Hozumi T Yoshikawa J Yoshida K Akasaka T Takagi T Honda Y Okura H Use of intravascular ultrasound for in vivo assessment of changes in intimal thickness of angiographically normal saphenous vein grafts one year after aortocoronary bypass surgery Heart 1996 76 317 320 8983677
Moher D Schulz KF Altman DG The CONSORT statement: revised recommendations for improving the quality of reports of parallel-group randomised trials Lancet 2001 357 1191 1194 11323066 10.1016/S0140-6736(00)04337-3
|
16266430
|
PMC1282586
|
CC BY
|
2021-01-04 16:38:25
|
no
|
Filaria J. 2005 Nov 2; 4:10
|
latin-1
|
Filaria J
| 2,005 |
10.1186/1475-2883-4-10
|
oa_comm
|
==== Front
Front ZoolFrontiers in Zoology1742-9994BioMed Central London 1742-9994-2-151621611810.1186/1742-9994-2-15ReviewConservation and co-option in developmental programmes: the importance of homology relationships Sanetra Matthias [email protected] Gerrit [email protected] May-Britt [email protected] Axel [email protected] Lehrstuhl für Zoologie und Evolutionsbiologie, Fachbereich Biologie, Universität Konstanz, 78457 Konstanz, Germany2005 10 10 2005 2 15 15 20 6 2005 10 10 2005 Copyright © 2005 Sanetra et al; licensee BioMed Central Ltd.2005Sanetra et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
One of the surprising insights gained from research in evolutionary developmental biology (evo-devo) is that increasing diversity in body plans and morphology in organisms across animal phyla are not reflected in similarly dramatic changes at the level of gene composition of their genomes. For instance, simplicity at the tissue level of organization often contrasts with a high degree of genetic complexity. Also intriguing is the observation that the coding regions of several genes of invertebrates show high sequence similarity to those in humans. This lack of change (conservation) indicates that evolutionary novelties may arise more frequently through combinatorial processes, such as changes in gene regulation and the recruitment of novel genes into existing regulatory gene networks (co-option), and less often through adaptive evolutionary processes in the coding portions of a gene. As a consequence, it is of great interest to examine whether the widespread conservation of the genetic machinery implies the same developmental function in a last common ancestor, or whether homologous genes acquired new developmental roles in structures of independent phylogenetic origin. To distinguish between these two possibilities one must refer to current concepts of phylogeny reconstruction and carefully investigate homology relationships. Particularly problematic in terms of homology decisions is the use of gene expression patterns of a given structure. In the future, research on more organisms other than the typical model systems will be required since these can provide insights that are not easily obtained from comparisons among only a few distantly related model species.
==== Body
Introduction
Evolutionary developmental biology (evo-devo) seeks to unravel the bases of developmental changes in body plan evolution of complex organisms such as animals and plants. The significance of this relatively new discipline is based on the premise that evolution cannot be fully understood without understanding the evolution of developmental programmes [1], and a number of novel conceptual frameworks have emerged from evo-devo research to supplement those of traditional evolutionary biology, such as DEVELOPMENTAL REPROGRAMMING [1-6]. The latter concept describes the process that acts between mutation and selection on the level of the organism, leading from an altered gene product to a new ontogeny and phenotype. Reprogramming has been proposed to constitute an additional evolutionary mechanism because some ontogenetic changes may be promoted by existing developmental mechanisms while other alterations are prevented [1,3,7] (referred to as 'developmental drive' and 'constraint', respectively [8]). It seems therefore likely that evolution can be biased by development, and this may have a powerful impact on the direction of evolutionary change [1,7,8].
During the past two decades it was discovered that most animals, no matter how divergent in form, share specific gene families that regulate major aspects of body patterning, for instance many homeobox-containing genes [9,10], which are even present in the Cnidaria [11]. Recent findings show that morphologically simple organisms often possess genes, such as members of the pax gene family, that are homologous and show a high level of sequence similarity to those of higher vertebrates [12-15]. Despite this astonishing extent of evolutionary conservation in developmental regulatory genes across major taxonomic groups, there are also cases where gene expression patterns differ markedly among closely related taxa, for instance in the molecular mechanisms that determine the spatial axes of the tetrapod limb [16]. In the recent past, one of the goals of evo-devo research was to search for putative phylum-specific genes, which may have given rise to phylum-specific evolutionary novelties. However, the view hat new phyla arose in concert with the advent of novel genes has been increasingly challenged [17,18]. Instead, there is mounting evidence that the evolution of lineage-specific body plans does not primarily depend on the invention of new genes but rather on the deployment of new gene regulatory circuitries. Changes in the transcriptional regulation of genes may thus be more significant than changes in gene number or protein function [18,19]. Moreover, the use of 'old' genes for novel structures has recently been demonstrated in a number of instances [20,21].
Among the main controversies that have emerged from evo-devo research is whether or not the utilization of conserved molecular components in developmental programmes across animal phyla can be taken as evidence for a shared developmental function in their latest common ancestor. The alternative would be the co-option of conserved genes and gene pathways to new functions, most likely operating in non-homologous structures. It is therefore fundamental to employ phylogenetic methods and homology criteria meticulously to resolve such issues, including the idea of retention of genetic programmes or 're-awakening' [22-25]. Recent gene expression studies, for instance, have revealed some common molecular aspects of segment ontogenesis between insects and annelids (shared segmental expression of engrailed and wingless) [26], and between arachnids and chordates (shared role of the notch signalling pathway) [27], which have been used to reinforce the hypothesis that the last common ancestor of all bilaterian animals was segmented [17,28-30]. However, one must be particularly critical about such deep homologies since the probability of gene recruitment to non-homologous roles grows with phylogenetic distance [1,31]. The use of gene expression patterns to establish homologies between morphologically similar features among distantly related organisms is another matter of ongoing debate [1,31,32].
In this review, we provide a brief introduction to evolutionary developmental biology for newcomers to the field who may be overwhelmed by the abundant literature. We outline how recent advances in evo-devo research have changed our understanding of the genesis of species differences and morphological novelties. In particular, we present examples to show that the contribution of phylogenetics to test hypotheses for the interpretation of problems in the evolution of developmental processes [20,33,34] is becoming more and more recognized. We also stress the importance of accurately assessing homology relationships and appropriate phylogenetic sampling of organisms for evo-devo studies.
Morphological versus Genetic Complexity
It would seem plausible to think that an organism with only a handful of tissue types has a much simpler genetic machinery than morphologically more complex creatures, such as vertebrates. But surprisingly little correlation has been found between genetic complexity and the degree of morphological organization so far. Some recent studies have revealed an astoundingly large number of similarities in the genetic make-up between morphologically complex organisms (e.g., vertebrates) and relatively simple forms (e.g., corals, molluscs). The most spectacular examples come from cnidarians, which are among the most basal metazoan animals composed of only two cell layers and yet exhibit a rather advanced suite of genes, such as pax, wnt, and genes involved in organizing the bilaterian head [12,35,36]. About 12 % of the genes found in the EST library of the coral Acropora (Anthozoa) were shared with vertebrates but had no match with Caenorhabditis elegans or Drosophila melanogaster [14]. Until this finding, many of those 'vertebrate genes' were presumed to be lineage-specific, e.g., Churchill and Tumorhead [14], which are functionally associated with a highly differentiated nervous system. In cases where a particular gene sequence was present in all three animals, the coral sequence matched the human counterpart much more strongly than any of the corresponding Caenorhabditis or Drosophila sequences [14]. This suggests higher rates of divergence in these invertebrate model systems, and a recent analysis of the Pufferfish genome likewise indicates that many fish proteins have diverged markedly faster than their mammalian homologues [37,38]. Certainly, the genetic complexity of corals is surprising considering that they possess only a few tissue types and a simple nervous system.
Similarly, lineage-specific gene losses in some model organisms can obscure ancient orthology relationships among genes. For example, a steroid receptor gene has been discovered in the mollusc Aplysia, but no orthologues are known from the fully sequenced genomes of the invertebrates C. elegans and D. melanogaster [13]. Instead, the sequence was most similar to that of the human estrogen receptor. Much in line with another study [39] these results suggest the unexpected evolution of genes for the major receptor types, such as steroid receptors and thyroid hormone receptors, very early in metazoan history at the base of the Bilateria. Nuclear receptor genes would then be far older than previously thought, with an estimated origin for the protostome-deuterostome split of approximately 960 milllion years ago [40]. Under this hypothesis one has to assume that these genes, previously thought to be vertebrate innovations, were lost independently in several bilaterian lineages (Fig. 1). The most extensive gene losses of ancestral gene families have been reported in C. elegans and D. melanogaster, with 31.0% and 25.8%, respectively [14,41]. However, the sparse taxon sampling presently hampers a thorough analysis and more informative interpretation of character changes, as does the fact that some critical nodes in metazoan phylogeny are still unresolved [42-44].
Figure 1 Phylogeny of the Metazoa (adopted and simplified from [112-114]) depicting several new and controversial hypotheses of character evolution with regard to bilateral symmetry, central nervous system, segmentation, and steroid receptors. PDA: last common protostome-deuterostome ancestor; CNS: central nervous system. Filled bars indicate the origin of a character while open bars in the same colour indicate loss of that character. Open grey bars show that data for the assumed loss of steroid receptors are missing.
The above considerations imply that the ancestral metazoan might have contained many more genes than previously thought (see also [45] on opsin genes, and [36] on the wnt gene family in sea anemones). The similarity in the genomic repertoire between some invertebrate and human sequences can be explained by the atypical condition (i.e., rapid divergence) of the commonly used model organisms [14,15]. Another important point is whether genes shared between morphologically simple and complex organisms suggest that their ancestor already had the developmental programmes that are now implemented, for example, in modern vertebrates. In several cases, and especially for the corals, it seems more likely that many genes in simple organisms are confined to a single role, probably the ancestral role, while they acquired additional functions later in evolution, which allowed for greater complexity in the organism. Many nuclear receptor genes did probably not yet have their present function in the organisms in which they first occurred [39].
Evolutionary Genetics of Morphological Novelties
In order to understand how novel characters can arise at the cellular level it is important to study the history of the genes involved and their regulatory interactions. An apparent paradox is that the morphological novelty we observe at the level of different animal phyla is not always reflected in similar changes of gene composition and sequence divergence of the genes controlling development. Instead, the emergence of evolutionary novelties largely appears to be based on the CO-OPTION of already existing genes for new functions. One molecular mechanism acting at the gene level is the acquisition of new regulatory sequences that leads to novel patterns of transcriptional activation [7,22,23,46]. As a consequence, new genes can be recruited into existing regulatory gene networks and result in functional changes to the network. Genes may gain novel expression domains by chance mutations or recombination events in their cis-regulatory elements. Alternatively, changes in the expression of upstream transcription factors themselves can initiate activation of target genes in new domains (Fig. 2). When combined with an increase in gene family size through duplications, the acquisition of novel functions in the duplicates (sub- or neo-functionalization) by co-option [47] is a powerful mechanism to increase the pool of developmental building blocks (modules).
Figure 2 Mechanisms of regulatory evolution. The diagram depicts the two possible modes of regulatory changes that generate novel developmental control mechanisms. New domains of gene activation are achieved by the gain of a novel regulatory element or the gain of an upstream transcription factor (right). If such novel regulatory wiring follows a gene duplication, it may liberate protein evolution and open a developmental trajectory towards co-option (m/M = mesodermal enhancer/transcription factor. e/E = ectodermal enhancer /transcription factor).
The mechanism of co-option is facilitated by the modular character of gene interactions. MODULARITY has become one of the central paradigms of molecular evolutionary biology [48,49]. This concept proposes the use of pre-existing building blocks in novel ways, rather than the origin of completely new elements, as the main source of molecular and regulatory innovations [48,50,51]. In a gene-based, developmental context it suggests that individual genes together perform a given "network function" (e.g., the RTK-Ras or wnt pathways) [52]. Such modules can be visualized as being composed of a set of interacting genes that can associate in novel ways with other modules, forming networks of higher level organization. This gene-set, also called the 'GENETIC TOOLKIT', determines the overall body plan and the number, identity and pattern of body parts [53]. It appears that the evolution of metazoan development and body plans is based on an increase in the complexity of the control circuitry regulating an ancestral toolkit of genes, rather than on the invention of novel developmental genes [17]. Extensive comparisons of gene functions in relation to animal evolutionary history will be needed to uncover the ancestral functions of these toolkit genes. Present-day organisms that have retained a particular gene are good candidates for reconstructing ancestral character states if they exhibit shared functions (or consensus functions) [15]. Another possibility to consider is that the ancestral function might have become lost in the course of evolution.
Proteins required for mineralization were co-opted for vertebrate-specific innovations
One particularly instructive example illustrates this idea. Major transitions in evolution are often accompanied by phylum-specific innovations, such as the occurrence of mineralized tissue in vertebrates [54], which was fundamental to the radiation of modern vertebrates. Its development enabled the evolution of endoskeleton, body armour and teeth, thereby providing adaptive phenotypes for improved locomotion, protection and predation, respectively. At the heart of this tissue type are members of the secretory calcium binding phosphoprotein (SCPP) gene family, which produce the special ionic conditions in the extracellular matrix required for skeletal mineralization. How did this gene family evolve, and what was its likely ancestral function in invertebrates? The SCPP gene family has been traced back to an ancestral gene, SPARC (encoding a non-collagenous bone matrix protein), present in both protostomes and deuterostomes [20]. Tetrapod SCPPs arose from SPARC by several gene duplications followed by co-option for new functions (Fig. 3), and include genes for enamel, dentin/bone, as well as milk caseins and salivary proteins in mammals [55]. Birds have an eggshell SCPP but lack the genes for enamel, milk and salivary-associated proteins, probably reflecting the loss of teeth in birds and the evolution of mammary glands in mammals [56]. SPARC homologues have been identified in a variety of invertebrates where they influence cell behaviour and interactions with the extracellular matrix, rather than being involved in the generation of mineralized tissues. Thus, an evolutionary modification of these proteins seems likely to have occurred through a functional shift from facilitating the stabilization of structural proteins towards enabling more diverse interactions between cells and proteins of the extracellular matrix in the course of vertebrate evolution.
Figure 3 Evolution of the secretory calcium binding phosphoprotein (SCPP) gene family involving gene duplication and co-option. A duplication event of an ancestral SPARC gene (1) in the vertebrate lineage generated two paralogues (2), and one of these (SPARCL1) gave rise to the SCPP gene family through subsequent tandem gene duplications after the divergence of cartilaginous and bony fish (3). In mammals, the formation of tooth and bone tissue is based upon the presence of various members of the SCPP gene family, including genes for enamel SCPPs, dentin/bone SCPPs, as well as milk caseins and salivary proteins. Most of these are closely linked on one chromosome in humans, as shown in the diagram.
Regulatory DNA evolution in invertebrates
Significant morphological transformations in the body plan of invertebrates have been found to correlate with developmental changes of Hox gene expression patterns [57,58]. Interestingly, as two examples in molluscs demonstrate, some of these changes are not related to the characteristic Hox function of establishing pattern along the anteroposterior axis. In the gastropod Haliotis asinina, two Hox genes (Has-Hox1 and Has-Hox4) are expressed in the mantle margin, where they have been co-opted into a new developmental role in shell formation [59]. Since morphological novelties derived from the ancestral molluscan body plan are striking in cephalopods, it seems promising to explore the molecular mechanisms of cephalopod innovations. For example, patterns of Hox expression in the squid Euprymna scolopes strengthen the argument that co-option events are often associated with the origin of new morphological structures [58]. The acquisition of three innovations derived from the ancestral molluscan foot, namely brachial crown, funnel tube and stellate ganglia, could be ascribed to Hox gene recruitment during cephalopod evolution. Given the large diversity of molluscan body plans, it has been suggested that this morphological flexibility may result from a relaxation of regulatory constraints on the recruitment of morphological patterning genes [58].
Not surprisingly, variation at the species level is also frequently based upon changes to gene regulation. Expression of the yellow gene at the wing tips of the fruitfly Drosophila biarmipes, a species closely related to D. melanogaster, results in conspicuous black pigment spots. It was shown that, for the evolution of this pigment pattern, the gene's regulatory sequences had gained additional binding sites for highly conserved transcription factors [60]. When experimentally introduced into D. melanogaster, these regulatory elements are capable of driving reporter gene expression (and thus yellow expression) in the distal-anterior region of the wing. Interestingly, among those newly evolved binding sites is one for the transcription factor engrailed, which perfectly illustrates how regulatory pathways already present in the wing have been co-opted to control wing pigmentation through chance mutations of ancestral enhancer sequences. It certainly is an appealing concept that the combinatorial nature of transcriptional regulation creates a large reservoir for morphological diversity, and may provide more variation for natural selection than changes in the gene product alone. Changes in the cis-regulatory systems of genes may therefore be more significant than changes in gene number or protein function [19].
Developmental innovations through protein sequence evolution
Examples in which evolutionary changes in gene regulation lead to morphological changes are numerous [61], yet there are also a number of well studied cases in which changes in protein sequence have been linked to new adaptations (reviewed by [46]). Since many genes have pleiotropic functions, changes to their protein sequence are potentially deleterious. Thus most cases involve gene duplications, as exemplified above for the SPARC gene family, or alternative splicing, in which one copy retains its function while the other acquires a new one. In the latter case, even mutations resulting from detrimental mechanisms, such as frameshift mutations, have been shown to be retained for up to hundreds of millions of years and have evolved new protein functions [62]. Finally, changes to protein-protein interactions can lead to alterations in developmental mechanisms, by integrating novel regulators into existing pathways, or by eliminating old ones. For example, the transcription factor brachyury is expressed in the circumference of the blastopore of most animal phyla and its orthologues from most Bilateria are capable of inducing mesoderm when assayed in Xenopus animal caps; however, in this assay, brachyury orthologues from Drosophila and tunicates strongly induce formation of both mesoderm and endoderm, and this is strictly correlated with the loss of a short protein-protein interaction motif, N-terminal of the DNA-binding domain [63]. Interestingly, insects and tunicates have not only lost circumferential blastopore brachyury expression independently, but also have a derived mode of gastrulation which is largely independent of brachyury. Messenger et al. [64] have recently identified Smadl as the cofactor that binds to the conserved brachyury N-terminal peptide and inhibits endoderm induction. The possibility therefore exists that this repression module, which is absent in the diploblastic Hydra, evolved in the bilaterians to separate mesoderm from mesendoderm. In Drosophila and tunicates, these tissue types are derived from topologically separate regions, and the derived mode of development may have relaxed the selective pressure required to maintain this motif. It is evident that in the future many more such examples will be found that are associated with the gain or loss of a particular structure or mode of development.
Phylogeny, Homology, and Gene Expression Patterns
Given sufficient periods of evolutionary time a certain gene or gene cascade may be conserved within a lineage, yet might be highly divergent among lineages. Therefore, while examining the different roles of conserved versus co-opted developmental mechanisms, it is important to recognize that conservation (lack of change) is a relative term whose interpretation depends on the selection of the appropriate phylogenetic framework. Knowledge about the phylogenetic relationships among model organisms and their relatives will thus substantially improve the understanding of developmental processes and uncover general evolutionary patterns [32,65-67]. Phylogenies are statements not only of relationships among taxa, but also about the evolution of characters along the tree. Mapping characters onto a robust phylogeny is a good way to determine if those characters may be homologous [34].
The basic concepts
Homology as a historical concept is typically defined as the shared inheritance of a trait from a most recent common ancestor (e.g., eyes of fish and mammals; Fig. 4A), though plain similarity has inadequately been used in a number of instances (for further discussions see [24,25,68,69]). HISTORICAL HOMOLOGY can be applied to different levels of biological organization, including morphological structures, developmental processes, and genes [23]. Abouheif [70] suggested a hierarchical analysis of homology relationships, which is designed to reveal alternative evolutionary hypotheses, for example the recruitment of homologous genes and networks to function in convergent embryonic and morphological structures (e.g., eyes of vertebrates and insects; Fig 4A). Gene family phylogenies can further be used to indicate if genes are orthologous (arose by speciation and thus are homologous in the strict sense) or paralogous (arose by gene duplication). On the other hand, the concept of GENERATIVE HOMOLOGY (Fig. 4A), or syngeny, uses shared developmental pathways to imply that a given character is generated by the same genetic machinery inherited from a common ancestor [71]. Wilkinson [6] has recently expanded this idea to use a combination of shared key genes (one or more) plus a shared biological or developmental function for which those genes are crucial as an indicator of homology. Both these concepts incorporate the view that the continuity of inheritance of the potential to make a particular trait rather than the continuity of its appearance is what matters most in homologous relationships [72]. The phylogenetic re-appearance of characters (reversal) is often observed even in well established phylogenies, therefore the genetic potential to produce those characters is probably retained, a phenomenon which can be referred to as 'LATENT HOMOLOGY' [69,73] or 'REAWAKENING' [66] (Fig. 4B). Whenever using the term homology it is important to state explicitly which type, subset, or level of homology is alluded to [24,69,74], and to work within a well-defined phylogenetic framework.
Figure 4 Homology relationships at different hierarchical levels of biological organization (ge: gene; gn: gene network; gi: gene interactions in the network; e: embryonic origin; m: morphological structure) following the suggestions detailed by Abouheif [65]. (A) Functioning of homologous genes in structures of independent evolutionary origin. In insect and vertebrate eyes, a very similar genetic machinery is used to generate structures that are historically non-homologous and morphologically dissimilar. Note that this generative homology is different from latent homology (B) in that the latter is concerned with morphologically homologous structures. This figure illustrates the difference between two main concepts, historical vs. generative homology. (B) Latent homology or re-awakening. If morphologically homologous characters, such as arthropod compound eyes [104], have multiple independent origins on a phylogeny then this may be due to retained genetic programmes that are not expressed in some of the ancestors, but whose functionality has been recovered after further speciation, giving the incorrect impression of convergent evolution. (C) Partial homology of gene networks. Novel genes (V, W, X, Y) are recruited into an ancestral network consisting of genes A, B, and C in two different lineages of taxa (a+b and e). Thus, the black part of the network is homologous (inherited from a common ancestor) while the coloured segments are not.
The concept of homology is undoubtedly challenged when it comes to tracking down the evolutionary origins of developmental programmes [24,71,74,75], and the use of gene expression patterns to infer homology is a matter of intense debate [5,23,76]. Of particular difficulty is how to distinguish superficial similarity from phylogenetic information content. As an important rule, comparisons of gene expression patterns to assess homology hypotheses should be restricted to orthologous gene copies, since new expression domains evolving among paralogues are likely to be convergent (Fig. 5). These gene relationships may be complex and thus need to be tested based on a phylogenetic tree including the whole gene family by identification of the timing of speciation events relative to gene duplications [23] (Fig. 5). An analysis of the Snail/Slug gene family revealed functional modifications after the original copy had become duplicated during vertebrate evolution. In addition, the study demonstrated unexpected evolutionary changes in Snail/Slug expression domains, which imply a large degree of plasticity and highlight the risk of using expression or function as homology indicators when studying the evolution of gene families [32]. Patterns of gene expression are particularly complicated to interpret in terms of homology because they interconnect different levels of biological organization, in this case genes and morphological structures [23]. Conserved expression patterns do not necessarily indicate conservation of gene function, and we know today that homologous genes can be expressed in structures that have different evolutionary origin (thus are non-homologous), such as distal-less in animal appendages [76] (see also previous sections for more examples). Likewise, similar developmental roles may be the result of convergent evolution [25]. To overcome some of these difficulties and ambiguities in terminology, the term HOMOCRACY was recently proposed to describe organs/structures that are governed by identical patterning genes [76]. It is important to note that a given structure can be both homocratic and homologous, thus the two terms are not mutually exclusive (Fig. 5).
Figure 5 The use of phylogenetic testing in the study of gene expression patterns and to infer ancestral developmental roles. A mixture of orthologous and paralogous genes is included in this phylogenetic tree, therefore reflecting gene relationships rather than relationships of taxa. The phylogeny of taxa can be inferred by analyzing either set of orthologous genes. The two lineages (A and B) of orthologous gene copies are shown in capital letters, and the B lineage is additionally highlighted in grey, while recent paralogues are numbered. Triangles, squares and hexagons indicate gene expression in three tissues/organs, which are grouped by superficial similarity as primary homology hypothesis. Homocratic structures are defined by co-expression of identical patterning genes, but these can be either orthologues or paralogues. In the latter case they do not strengthen a homology hypothesis because convergent evolution is likely (as in the depicted case).
Gene network evolution
Behind the scenes are complex interacting gene networks (pathways) that form the genetic machinery required for the origin and functioning of morphological structures. These networks can be considered a distinct level of biological organization, the homology relationships of which may differ from other such levels [22,70]. However, some authors have dismissed the strictly hierarchical view of biology calling for a combinatorial approach to homology [74,77], because interactive combinatorial processes, such as co-option and modularity, play a significant role in biological systems. One consequence of this would be to accept that homology assessments cannot be reduced to a yes-or-no question even at the molecular level [74,78]. The obvious problem is how much (and which parts) of a given entity, for instance a gene network, must be continuous between lineages. Clearly, two networks are homologous if all genes and their interactions are derived from an identical network in the most recent common ancestor. Quite often networks share certain elements but differ in others, and thus they are neither fully homologous nor independently evolved; rather, they might be considered partially homologous. The recruitment of novel genes into an existing regulatory gene network, for instance, will lead to partial homology (Fig. 4C). As a consequence, gene expression patterns need to be evaluated at the level of gene networks to reveal their true evolutionary relationships [70,79]. Studying closely related species in a phylogenetic context is helpful in this case, because smaller changes are expected, and this knowledge can then be used to connect more ancient character states. For example, the wing patterning network in Drosophila is well studied, and when the expression patterns of its constituent genes in wingless castes of ants were compared, the results provided surprising insights into the evolution of wing polyphenism: several closely related species do not share a common mechanism to interrupt wing development in the wingless castes, which is an unexpected finding given that a common origin of all wingless castes can be assumed [33].
Similar expression patterns do not necessarily indicate homology
We have emphasized along with other authors [5,73,76] that gene expression patterns should not be used to infer morphological homology of structures without employing phylogenetic criteria to test hypotheses about orthology of genes as well as partial homology, or convergence of gene networks. In addition, genes that are expressed during basic cellular processes, such as cell proliferation or epithelial-mesenchymal transitions, are likely to be frequently utilized in non-homologous organs. As they have a high developmental and evolutionary constraint, they are not informative to support homology of structures. Conversely, if homology of structures is to be based upon gene expression, members of the genetic toolkit should be consulted, as they control diverse and general patterning processes. Some recent studies that seem to indicate deep homologies in the body plan of animal phyla, although employing this latter strategy, should nevertheless be interpreted with caution. For example, Lowe et al. [80] proposed that a comparable expression pattern in the nervous system of chordates, hemichordates and Drosophila is an indication of homology (at least 14 out of 22 genes involved in neural patterning), suggesting that the ancestor of deuterostomes (and probably all bilaterians) had a diffuse nervous system that was centralized independently in arthropods and chordates (Fig. 1). This idea disagrees with the prevailing theory of a single origin of the central nervous system with a dorsoventral axis inversion, which is supported by the inversion of TGFβ-signalling in chordates and Drosophila [81]. The example further illustrates the undesirable fact that there is no obvious boundary as to which number of genes need to be co-expressed to proclaim a structure homologous, i.e., would 10 out of 22 genes still justify homology of these nervous systems? As with morphological analyses, similarity measures cannot be used as evidence to indicate common descent, and are thus incongruent with the concept of homology. Such approaches are clearly phenetic and by no means phylogenetic. In another example, the use of homologous genes to achieve bilateral symmetry in larvae of sea anemones was used to infer that bilaterality originated before the split of Cnidaria and Bilateria [82] (Fig. 1). But is bilateral symmetry of sea anemones really homologous to that observed in more derived animals or did it arise by convergent evolution [83]? To answer this question we concur with Holland [31] that if orthologous genes are used to control the development of a similar structure in two otherwise morphologically different animals, this lends some support to the hypothesis of homology but is in itself not sufficient and requires more rigorous tests. Further, early metazoan interrelationships have remained particularly difficult to understand, once more stressing the importance of phylogenetic certainty in relation to the co-option or homology question.
Nevertheless, one adequate method of testing morphological homology hypotheses by using gene expression data has recently been proposed [84]. This approach involves the investigation of changes in developmental systems in a parsimony analyses by mapping gene expression data onto a phylogenetic tree along with other characters. Different homology hypotheses can then be evaluated in terms of the number of evolutionary steps required for each hypothesis to be valid, and the most parsimonious solution (involving the smallest number of steps) can be identified. An example can be seen in Figure 5, depicting a single co-option event leading to shared gene expression in organ 2 in frog and mouse. This supports the homology of organ 2 in these two species. But if instead we were to assume organ 2 in frog as being homologous to organ 3 in mouse, then two independent changes in gene expression would be required (gain of the purple hexagon and loss of the orange square in organ 3 to explain the observed gene expression pattern), which would be less parsimonious and thus would not support this homology relationship. In this way, several competing homology hypotheses can be compared in a combined analysis including many genes and other characters [84]. In general, whenever a co-option event can be indentified as a SYNAPOMORPHY for a set of taxa, it should be informative to support the homology of a structure. Gene expression data could also be useful to distinguish between convergence and re-awakening, as in the former case we expect distinguishing features to reflect different origins of two independently evolved structures while in the latter we do not (see also chapter on the retention of genetic programmes).
What does Shared Genetic Potential suggest: Conserved Developmental Programmes or Repeated Evolution?
Despite growing evidence for a widespread conservation of the genetic toolkit that is used to produce the complex body plan of bilaterian animals, there is reason to believe that the last common ancestor (Urbilateria) did not necessarily employ the same developmental programmes of extant animals. One principal idea is that the function of many homologous developmental genes in a last common ancestor was of the same general kind as now observed, but in a different developmental context [4]. Conserved genes, or entire networks (modules), might have been co-opted repeatedly into new regulatory regions or morphological structures. If the available genetic toolkit is of limited size, then the possibility of co-option for similar functions, i.e., the repeated evolution of analogous developmental processes, does not appear unlikely [76]. For instance, the homeobox gene distal-less (dll) and its vertebrate homologue dlx are expressed across extant animals in various types of appendages (e.g., vertebrate limbs and echinoderm tube feet) that are clearly non-homologous [22,85]. Shared functions of dll genes among animal phyla are few and very general, so that the consensus function is reduced to a general role in regulating cell proliferation [76]. Hence, the bilaterian ancestor probably had no legs but perhaps some inconspicuous body wall outgrowths triggered by dll expression. Co-option of this pre-existing mechanism into more specific building blocks (e.g., for structures that grow out distally but are historically non-homologous), could have subsequently occurred during the evolution of different animal phyla.
It is currently intensely debated whether segmentation in different animal phyla has had a common origin or not [28,29,85-88]. Specifically, the question is whether the last common protostome-deuterostome ancestor was already segmented or whether segmentation arose on three separate occasions in arthropods, annelids and vertebrates (Fig. 1). Current views on this have partly changed towards the 'single origin of segmentation' hypothesis due to the finding that notch and delta genes participate in the segmentation of both spiders and vertebrates [26]. By contrast, segmentation in Drosophila is notch and delta independent. The more basal phylogenetic position of spiders in relation to Drosophila suggests a derived mode of segmentation in the latter [34], and thus allows the evaluation of more distant ancestors within the arthropod clade. Similarly, the arthropod-like expression pattern of engrailed and wingless genes in segment formation of the annelid Platynereis points to a segmented ancestor of all protostomes [27]. However, there are also reasonable objections to the 'single origin of segmentation' hypothesis (e.g., for parsimony reasons [89]). Part of the problem might reside in the definition of segmentation since some authors relax the definition to include other repeated structures, such as paired coeloms in echinoderms, which would render the lack of segmentation an uncommon trait throughout the animal kingdom [29,34]. Furthermore, segmentation is a mesodermal process in annelids and vertebrates, whereas in arthropods it is primarily ectodermal [88]. It is therefore conceivable that the same deeply conserved modules have been co-opted for similar functions many times, giving rise not only to the morphologically quite different types of animal segmentation but also to segmented tissues of different embryonic origin found among metazoans. Segmentation would then be a case of repeated evolution not implying the existence of a segmented Urbilateria.
The way in which Pax6 and its associated genes are involved in eye development across the Metazoa suggests a shared genetic potential for the occurrence of eyes [90]. Yet the phylogenetic pattern of the distribution of eye structures (adopting the concept that a simple photoreceptor is not an image-forming eye [91]) is clearly polyphyletic, which indicates multiple independent origins from forms lacking eye development [91,92] (Fig 4A). The current situation most likely reflects successive losses and gains of the use of the Pax6 network during the evolution of metazoan animals [6]. Some authors have suggested that Pax6 has become integrated into several independently evolved genetic programmes to regulate particular aspects of eye development [76,93,94], rather than being a master regulator of eye development [95]. Therefore, repeated utilization of similar genetic pathways involving pre-existing building blocks may emerge as a common theme in animal evolutionary history. Completely new morphological structures can likewise evolve by the integration of independent anatomical entities, e.g., cell populations, which differ in their structure and tissue of origin. According to recent evidence the vertebrate eye is a compound structure comprising two types of light-sensitive cells (rhabdomeric and ciliary receptors) with independent evolutionary histories [45]. The presence of ciliary photoreceptors containing an opsin similar to those of vertebrates in the brain of the ragworm Platynereis (Annelida) suggests that both receptor types were present in Urbilateria. Thus, it is straightforward to propose that vertebrate and invertebrate eyes are partially homologous since they contain homologous (e.g., rhabdomeric photoreceptors) and non-homologous cell types derived from different germ layers.
Testing Evolutionary Hypotheses
The study of character evolution has become one of the central aspects in modern phylogenetic analysis due to its power to reveal and test different evolutionary hypotheses [96,97]. One of the essential prerequisites of this approach, however, is that for those tests to be valid one needs to work within a highly resolved and robust phylogenetic framework. Not only can simple character changes and homology relationships be investigated but also the existence of controversial evolutionary mechanisms. This is particularly useful for assessing conceptual problems in evolutionary developmental biology.
Developmental bias
It has been suggested that evolution can be biased by development in that, starting with a particular ontogeny, some phenotypes might be easily produced while others are unlikely or even impossible [1,7]. This idea is also perfectly compatible with the parsimony principle, since an already existing structure will more likely be used for a new purpose rather than a whole new structure will evolve for the same purpose, i.e., instead of evolving gills whales adapted the respiratory mechanisms of their lungs [8]. Thus, it appears that mutation and selection are not the only forces that have shaped the 'possible creatures' inhabiting this planet. In order to assess the significance of DEVELOPMENTAL BIAS as an evolutionary mechanism one can devise hypothetical character distributions on a phylogeny that would support this notion. Two main categories of character evolution in which developmental bias could have played a major role can be perceived (Fig 6.). Firstly, the convergent evolution of a character towards a similar morphology, such as colour patterns in Lake Tanganyika and Lake Malawi cichlids, may be indicative of biased evolution or selection (Fig. 6A). Secondly, a tree topology that suggests phylogenetic constraints (also known as PHYLOGENETIC INERTIA) may imply that at least part of this pattern may be due to developmental constraints. For instance, several tribes of cichlids often show relatively constant ranges in the number of spines of their dorsal and anal fins [98] (Fig. 6B). In general, phylogenetic constraints refer to taxon-specific limitations forcing a taxon into certain combinations of characters regardless of where that taxon occurs, i.e., an explicitly non-adaptive interpretation [99,100]. There are several methods available to test for phylogenetic inertia, such as autocorrelational analysis and phylogenetic correlograms [100,101].
Figure 6 Phylogenetic character distributions that would support developmental bias. (A) Convergence – If certain characters appear repeatedly in a phylogeny despite the obvious absence of exogenous selective causes, this may suggest an evolution biased by development. A possible example are the horizontal bar patterns evolved in parallel in Lake Tanganyika and Lake Malawi cichlids (indicated by black hashmarks), for which similar developmental mechanisms might be used. (B) Phylogenetic inertia (or constraints) – If characters show significant resistance against evolutionary change despite substantial environmental heterogeneity, then at least part of these constraints may be developmentally induced. Different species-rich tribes of East African cichlids show a relatively conservative number of spines of the dorsal fin (Ectodini: 12–15 with the exception of Enantiopus having 16–19; Lamprologini: 17–24, indicated by red hashmarks) and of the anal fin (Ectodini and most other cichlids: 3; Lamprologini: 4 or more [98], indicated by grey and black hashmarks) despite having evolved manifold body shapes.
Once a phylogenetic pattern of character distribution that is suggestive of biased evolution has been established, the problem arises how to distinguish developmental bias from highly similar selection pressures. This is by no means an easy task and will certainly present a great challenge for the future. One possible test for this could be to rule out adaptive explanations for certain morphologies by employing character correlation analysis with environmental variables and to estimate fitness values to test for actual selection. Another approach could be based on the assumption that the likelihood of bias increases with the number of species showing that particular morphology in relation to evolutionary distance. Though theoretically well explored [8], available evidence for developmental bias is presently scarce, and probably the most convincing example is that of the 3,000 and more species of centipedes all have odd numbers of leg-bearing segments. To invoke selection for this phenomenon seems unreasonable [102]. Further, floral symmetry patterns in angiosperms have been proposed as an example for bias-led evolution [103].
Retention of genetic programmes
In evolution, certain structures may be lost and later re-appear yet the genetic potential to produce those structures may be retained, even though the structures are not continuously present in all ancestors (Fig 4B). This observation has become known as latent homology or re-awakening [66,69,73]. Re-awakening may represent a valid hypothesis in cases where phylogenetic character distributions suggest the reappearance of a character that would be considered homologous on morphological grounds. Many instances of reversals that appear in robust phylogenies might actually be hidden cases of latent homology [69]. There are, however, only a few not extremely well-supported examples of this phenomenon but methods could be established to test for this interesting case of silencing and re-expressing of genetic pathways more meticulously. If dormant genes can be re-activated after further speciation, we expect high similarities in the developmental systems between the lost and the regained structure, i.e., these structures should be generatively homologous. The most convincing way of testing for re-awakening after having established a reliable character evolution would be to demonstrate the functionality of the genetic programme by experimental induction of the putatively re-evolved trait in the species that do not exhibit the trait.
Several species in the genus Xiphophorus display a so-called sword, which is a sexually selected extension of the ventral tail fin. The molecular phylogeny shows that the sword was lost once and later re-evolved at least twice in different branches, which suggests re-awakening of the 'sword-developing' programme [66]. Further, swords could be induced in some of these naturally sword-less species through testosterone treatment, thereby making a much stronger case for this hypothesis. It should be noted, however, that swords could also be induced by artificial selection in more distantly related species and that the sword as a morphological structure is rather loosely defined.
In arthropods, there is molecular phylogenetic evidence for the independent origin of compound eyes (Fig. 4B). Myocopids are the only group within the Ostracoda (Crustacea) that have compound eyes, and these are nested phylogenetically within several groups that lack this kind of eye [104]. Maximum likelihood methods of ancestral-state reconstruction were highly significant in supporting the independent origin of compound eyes from eyeless ancestors. However, the ommatidia of many diverse groups of arthropods (including the Maxillopoda as outgroup to Ostracoda) have an arrangement of photoreceptive cells different from the regained eyes in ostracods, which casts some doubt on their homology from a morphological point of view. It will be interesting to know whether the same genetic pathways are used to produce these slightly different types of compound eyes, i.e., if they are generatively homologous, which would strengthen the case for re-awakening.
A recent example from stick insects suggests that wings have re-evolved as many as four times during the radiation of this group [105]. An alternative interpretation would involve 13 independent occasions of wing loss, which is the less parsimonious solution (requiring more evolutionary changes). But accepting the alternative hypothesis would not appear implausible if a different method of CHARACTER OPTIMIZATION were used, i.e., one that assumes loss to be more likely than re-appearance [34]. This example shows that care has to be taken while establishing the basic requirement for re-awakening, which is a robust phylogenetic hypothesis of character evolution. In addition, phylogenies based on morphological characters will be affected in a negative way by re-awakening because the reappearing character causes homoplasy and will be incongruent with other characters. Although re-appearance of a character using the same genetic machinery is evolutionarily truly parsimonious, conflicting hypotheses may arise when applying the parsimony principle for phylogenetic tree construction in such cases.
The importance of phylogenetic taxon sampling
To determine the mode and direction of character evolution in a phylogeny most commonly the outgroup comparison is used [96,97]. The outgroup constitutes a species set that is as closely related to the ingroup as possible but must not be part of the ingroup. Shared character states between ingroup and outgroup indicate the ancestral state of a character, for example using frogs as outgroup to reptiles shows that having four legs is an ancestral trait. Very often only a single species is used as outgroup, which can be misleading if this particular species is not representative for the whole group – it is obvious that caecilian amphibians would not be good candidates to infer ancestral character states for limb development. In order to draw firm conclusions on ancestral character states extensive taxon sampling should be performed so that ingroup and outgroup contain a broad range of evolutionarily informative (i.e., phenotypically diverse) species. As a consequence, reliably defining ancestral character states in a phylogeny is impossible without the aid of non-model organisms since the typical model systems are often not characteristic for an entire clade, i.e., often show a high proportion of advanced character states. A good example for this comes from the Hox genes of C. elegans [106], which, as a member of the ecdysozoan clade, was expected to have a relatively large number of these genes. However, the investigation of a representative range of taxa shows a prominent loss of Hox genes during nematode evolution in which the most derived state (at least five Hox genes fewer than most other Ecdysozoa) was observed in the model C. elegans.
The attempt to reconstruct more gradual sequences of change emerges as a general requirement to deduce reliable conclusions on evolutionary events. This is also the case when studying the relationship of molecular changes and phenotypic evolution. Hox gene expression in crustaceans was found to be tightly correlated with alterations in morphology in that a shift in the anterior-posterior border of Ubx expression coincides with the occurrence of feeding appendages (maxillipeds). The latter could be demonstrated by investigating different evolutionary stages represented by seven genera from six crustacean orders [57]. Without sampling those intermediates, which often are "minor" taxa, false correlations between molecular and morphological evolution are expected because multiple character changes will be compressed into a single event on a phylogenetic tree [34]. This highlights the importance of reaching confidence on ancestral nodes as the crucial point in interpreting character evolution.
Conclusion
Understanding the high degree of conservation in genes and gene cascades on one hand, and the large morphological diversity on the other will be a great challenge. Molecular evolutionary analyses will have to focus more on the interaction context of gene networks and the concept of modularity rather than on individual genes [49,107]. It also emphasizes the need for correctly assessing the degree of homology vs. homoplasy in defining common components of developmental pathways. For many questions in comparative biology, and also in evo-devo research, the availability of a well-supported phylogeny among the study organisms is of paramount importance. Phylogenomic approaches (using multiple gene loci sequence data) to display evolutionary relationships among model organisms will certainly appear more frequently [43,108]. One obvious drawback of these studies is that data are available for only a small number of taxonomic representatives. In many fields of research, especially in medical applications, conclusions still tend to be drawn from a small number of model systems, though there is clear evidence that integration of many species sheds new light onto old questions. The Cnidaria and Porifera are good candidates in the hunt for new genetic inventions [14,36,109], as these two phyla are basal to the Bilaterians. Sponges (Porifera) possess the most basic features of the metazoan bodyplan. Thus, comparative genomics including this animal group emerge as a promising tool to gain insights into the genetic architecture of the hypothetical Urmetazoa as the earliest common ancestor of all metazoans [110,111]. Further studies in this direction will finally be able to reveal the minimal toolkit assembly for metazoan animals, and open up new research avenues in the evo-devo field.
Authors' contributions
MS and M-BB carried out most of the initial literature search and drafted the first version of the manuscript. GB expanded the review and, together with MS, wrote the final version. AM revised the manuscript critically, which was read and approved by all authors.
Glossary
CHARACTER OPTIMIZATION
The most parsimonious reconstruction of the states of a character (e.g. 'winged' and 'wingless') when mapping that character onto a pre-established (usually molecular) phylogeny.
CO-OPTION
An evolutionary process in which existing features become adapted for new functions, e.g., the change of gene function to new pattern forming processes.
DEVELOPMENTAL REPROGRAMMING
Modification of the prevailing ontogenetic trajectory within an evolutionary lineage.
DEVELOPMENTAL BIAS
The concept that evolution may be biased by development because some ontogenetic changes may be more likely than others, which is also termed 'developmental drive'. The unlikely or impossible changes are referred to as 'developmental constraints'.
GENERATIVE HOMOLOGY
Generation of a character by employing shared genetic mechanisms or pathways that consist of a set of homologous genes, while the character itself need not be homologous on morphological grounds.
GENETIC TOOLKIT
The toolkit is composed of a small fraction of all genes that are widely conserved among different animal phyla, and which generally control the expression of other genes.
HISTORICAL HOMOLOGY
Biological structures or traits are homologous if they were inherited from a most recent common ancestor, which can be evaluated on a phylogenetic tree.
HOMOCRACY
Organs or structures characterized through the expression of functionally identical patterning genes but the homology relationships of these genes may be unresolved (and thus may include orthologues and paralogues).
LATENT HOMOLOGY OR RE-AWAKENING
Phylogenetic re-appearance of a morphologically very similar character (being homologous in the sense of the homology criteria of position and special qualities [69]) for which the genetic potential to produce that character is retained.
MODULARITY
Biological level of organization into a set of interconnected units in an organism.
PHYLOGENETIC INERTIA
The tendency of traits to resist evolutionary change despite environmental perturbations.
SYNAPOMORPHY
A shared derived character state that is indicative of a phylogenetic relationship among two or more taxa.
Acknowledgements
We thank J. Luo, D. Steinke, I. Braasch, N. Offen and W. Salzburger for valuable and critical comments on the manuscript. We are also grateful for helpful suggestions and stimulating criticisms from two anonymous reviewers.
==== Refs
Raff RA Evo-devo: the evolution of a new discipline Nat Rev Genet 2000 1 74 79 11262880 10.1038/35049594
Raff RA The Shape of Life: Genes, Development and the Evolution of Animal Form 1996 Chicago: Chicago University Press
Davidson EH Genomic Regulatory Systems. Development and Evolution 2001 San Diego: Academic Press
Arthur W The emerging conceptual framework of evolutionary developmental biology Nature 2002 415 757 764 11845200
Gould SJ The Structure of Evolutionary Theory 2002 Cambridge: Harvard University Press
Wilkins AS The Evolution of Developmental Pathways 2002 Sunderland: Sinauer Associates
Arthur W The concept of developmental reprogramming and the quest for an inclusive theory of evolutionary mechanisms Evol Dev 2000 2 49 57 11256417 10.1046/j.1525-142x.2000.00028.x
Arthur W Developmental drive: an important determinant of the direction of phenotypic evolution Evol Dev 2001 3 271 278 11478524 10.1046/j.1525-142x.2001.003004271.x
McGinnis W Garber RL Wirz J Kuroiwa A Gehring WJ A homologous protein-coding sequence in Drosophila homeotic genes and its conservation in other metazoans Cell 1984 37 403 408 6327065 10.1016/0092-8674(84)90370-2
Scott MP Weiner AJ Structural relationships among genes that control development: sequence homology between the Antennipedia, Ultrabithorax and fushu tarazu loci of Drosophila Proc Natl Acad Sci USA 1984 81 4115 4119 6330741
Finnerty JR Martindale MQ Ancient origins of axial patterning genes: Hox genes and para Hox genes in the Cnidaria Evol Dev 1999 1 16 23 11324016 10.1046/j.1525-142x.1999.99010.x
Miller DJ Hayward DC Reece-Hoyes JS Scholten I Catmull J Gehring WJ Callaerts P Larsen JE Ball EE Pax gene diversity in the basal cnidarian Acropora millepora (Cnidaria, Anthozoa): implications for the evolution of the pax gene family Proc Natl Acad Sci USA 2000 97 4475 4480 10781047 10.1073/pnas.97.9.4475
Thornton JW Need E Crews D Resurrecting the ancestral steroid receptor:ancient origin of estrogen signaling Science 2003 301 1714 1717 14500980 10.1126/science.1086185
Kortschak RD Samuel G Saint R Miller DJ EST analysis of the cnidarian, Acropora millepora, reveals extensive gene loss and rapid sequence divergence in the model invertebrates Curr Biol 2003 13 2190 2195 14680636 10.1016/j.cub.2003.11.030
Raible F Arendt D Metazoan evolution: some animals are more equal than others Curr Biol 2004 14 R106 108 14986636 10.1016/S0960-9822(04)00030-2
Christen B Slack J All limbs are not the same Nature 1998 395 230 231 9751047 10.1038/26133
Peterson KJ Davidson EH Regulatory evolution and the origin of the bilaterians Proc Natl Acad Sci USA 2000 97 4430 4433 10781037 10.1073/pnas.97.9.4430
Levine M Tjian R Transcription regulation and animal diversity Nature 2003 424 147 151 12853946 10.1038/nature01763
Caroll S Endless forms: the evolution of gene regulation and morphological diversity Cell 2000 101 577 580 10892643 10.1016/S0092-8674(00)80868-5
Kawasaki K Suzuki T Weiss KM Genetic basis for the evolution of vertebrate mineralized tissue Proc Natl Acad Sci USA 2004 101 11356 1161 15272073 10.1073/pnas.0404279101
Wang W Grimmer JF Van de Water TR Lufkin T Hmx2 and Hmx3 homeobox genes direct development of the murine inner ear and hypothalamus and can be functionally replaced by Drosophila Hmx Dev Cell 2004 7 439 453 15363417 10.1016/j.devcel.2004.06.016
Abouheif E Developmental genetics and homology: a hierarchical approach Trends Ecol Evol 1997 12 405 408 10.1016/S0169-5347(97)01125-7
Abouheif E Akam M Dickinson WJ Holland PWH Meyer A Patel NH Raff RA Roth VL Wray GA Homology and developmental genes Trends Genet 1997 13 432 433 9385839 10.1016/S0168-9525(97)01271-7
Fitch W Homology – a personal view on some of the problems Trends Genet 2000 16 227 231 10782117 10.1016/S0168-9525(00)02005-9
Mindell DP Meyer A Homology evolving Trends Ecol Evol 2001 16 434 440 10.1016/S0169-5347(01)02206-6
Prud'homme B de Rosa R Arendt D Julien JF Pajaziti R Dorresteijn AWC Adoutte A Wittbrodt J Balavoine G Arthropod-like expression patterns of engrailed and wingless in the annelid Platynereis dumerilii suggest a role in segment formation Curr Biol 2003 13 1876 1881 14588243 10.1016/j.cub.2003.10.006
Stollewerk A Schoppmeier M Damen WGM Involvement of Notch and Delta genes in spider segmentation Nature 2003 423 863 865 12815430 10.1038/nature01682
Davis GK Patel NH The origin and evolution of segmentation Trends Genet 1999 15 M68 M72 10.1016/S0168-9525(99)01875-2
Balavoine G Adoutte A The segmented Urbilateria: a testable scenario Int Comp Biol 2003 43 137 147
Seaver EC Segmentation: mono- or polyphyletic? Int J Dev Biol 2003 47 583 595 14756334
Holland P The ups and downs of a sea anemone Science 2004 304 1255 1256 15166351 10.1126/science.1099829
Locascio A Manzanares M Blanco MJ Nieto A Modularity and reshuffling of Snail and Slug expression during vertebrate evolution Proc Natl Acad Sci USA 2002 99 16841 16846 12482931 10.1073/pnas.262525399
Abouheif E Wray GA Evolution of the gene network underlying wing polyphenism in ants Science 2002 297 249 252 12114626 10.1126/science.1071468
Telford MJ Budd GE The place of phylogeny and cladistics in Evo-Devo research Int J Dev Biol 2003 47 479 490 14756323
Galliot B Miller DJ Origin of anterior patterning – how old is our head? Trends Genet 2000 16 1 5 10637621 10.1016/S0168-9525(99)01888-0
Kusserow A Pang K Sturm C Hrouda M Lentfer J Schmidt HA Technau U von Haeseler A Hobmeyer B Maretindale MQ Holstein TW Unexpected complexity of the Wnt gene family in a sea anemone Nature 2005 433 156 160 15650739 10.1038/nature03158
Jaillon O Genome duplication in the teleost fish Tetraodon nigroviridis reveals the early vertebrate proto-karyotype Nature 2004 431 946 957 15496914 10.1038/nature03025
Málaga-Trillo E Meyer A Genome duplications and accelerated evolution of Hox genes and cluster architecture in teleost fishes Am Zool 41 676 686
Bertrand S Brunet GF Escriva H Parmentier G Laudet V Robinson-Rechavi M Evolutionary genomics of nuclear receptors: from twenty-five ancestral genes to derived endocrine systems Mol Biol Evol 2004 21 1923 1937 15229292 10.1093/molbev/msh200
Hedges SB Blair JE Venturi ML Shoe JL A molecular timescale of eukaryote evolution and the rise of complex multicellular life BMC Evol Biol 2004 4 2 doi: 10.1186/1471-2148-4-2 15005799 10.1186/1471-2148-4-2
Hughes AL Friedmann R Shedding genomic ballast; extensive parallel loss of ancestral gene families in animals J Mol Evol 2004 59 827 833 15599514 10.1007/s00239-004-0115-7
Hedges SB The origin and evolution of model organisms Nat Rev Genet 2002 3 838 849 12415314 10.1038/nrg929
Wolf YI Rogozin IB Koonin EV Coelomata and not Ecdysozoa: evidence from genome-wide phylogenetic analyses Genome Res 2004 14 29 36 14707168 10.1101/gr.1347404
Philippe H Lartillot N Brinkmann H Multigene analysis of bilaterian animals corroborate the monophyly of Ecdysozoa, Lophotrochozoa, and Protostomia Mol Biol Evol 2005 published online February 9 th 15703236
Arendt D Tessmar-Raible K Snyman H Dorresteijn AW Wittbrodt J Ciliary photoreceptors with a vertebrate-type opsin in an invertebrate brain Science 2004 306 869 871 15514158 10.1126/science.1099955
True JR Carroll SB Gene co-option in physiological and morphological evolution Annu Rev Cell Dev Biol 2002 18 53 80 12142278 10.1146/annurev.cellbio.18.020402.140619
Lynch M Force A The probability of duplicate gene preservation by subfunctionalization Genetics 2000 154 459 473 10629003
Jacob F "Evolution and Tinkering" Science 1977 196 1161 1166 860134
Duboule D Wilkins AS The evolution of 'bricolage' Trends Genet 1998 14 54 59 9520598 10.1016/S0168-9525(97)01358-9
Riedl R Order in Living Systems: A Systems Analysis of Evolution 1978 New York: Wiley
Bonner JT The Evolution of Complexity 1988 Princeton: Princeton University Press
Gilbert SF Opitz J Raff RA Resynthesizing evolutionary and developmental biology Dev Biol 1996 173 357 372 8605997 10.1006/dbio.1996.0032
Carroll SB Grenier JK Weatherbee SD From DNA to Diversity: Molecular Genetics and the Evolution of Animal Design 2001 Malden, MA: Blackwell Science
Shimeld SM Holland PWH Vertebrate innovations Proc Natl Acad Sci USA 2000 97 4449 4452 10781042 10.1073/pnas.97.9.4449
Kawasaki K Weiss KM Mineralized tissue and vertebrate evolution: The secretory calcium-binding phosphoprotein gene cluster Proc Natl Acad Sci USA 2003 100 4060 4065 12646701 10.1073/pnas.0638023100
International Chicken Genome Sequencing Consortium Sequence and comparative analysis of the chicken genome provide unique perspectives on vertebrate evolution Nature 2004 432 695 716 15592404 10.1038/nature03154
Averof M Patel NH Crustacean appendage evolution associated with changes in Hox gene expression Nature 1997 388 682 686 9262403 10.1038/41786
Lee PN Callaerts P de Couet HG Martindale MQ Cephalopod Hox genes and the origin of morphological novelties Nature 2003 424 1061 1065 12944969 10.1038/nature01872
Hinman VF O'Brien EK Richards GS Degnan BM Expression of anterior Hox genes during larval development of the gastropod Haliotis asinina Evol Dev 2003 5 508 521 12950629 10.1046/j.1525-142X.2003.03056.x
Gompel N Prud'homme B Wittkopp PJ Kassner VA Carroll SB Chance caught on the wing: cis-regulatory evolution and the origin of pigment patterns in Drosophila Nature 2005 433 481 487 15690032 10.1038/nature03235
Carroll SB Evolution at two levels: on genes and form PLOS Biology 2005 3 e245 16000021 10.1371/journal.pbio.0030245
Raes J Van de Peer Y Functional divergence of proteins through frameshift mutations Trends Genet 2005 21 428 431 15951050 10.1016/j.tig.2005.05.013
Marcellini S Technau U Smith JC Lemaire P Evolution of Brachyury proteins: identification of a novel regulatory domain conserved within Bilateria Dev Biol 2003 260 352 361 12921737 10.1016/S0012-1606(03)00244-6
Messenger NJ Kabitschke C Andrews R Grimmer D Nunez Miguel R Blundell TL Smith JC Wardle FC Functional specificity of the Xenopus T-domain protein Brachyury is conferred by its ability to interact with Smadl Dev Cell 2005 8 599 610 15809041 10.1016/j.devcel.2005.03.001
Meyer A Ritchie PA Witte KE Predicting developmental processes from evolutionary patterns: a molecular phylogeny of the zebrafish (Danio rerio) and its relatives Phil Trans Roy Soc London B 1995 349 103 111
Meyer A Brock GR, Cardew G Homology and homoplasy: the retention of genetic programs Homology Symposium on Homology held at the Novartis Foundation (Symposium 222); London 1999 Wiley: Chichester, UK 141 157
Rudel D Sommer RJ The evolution of developmental mechanisms Dev Biol 2003 264 15 37 14623229 10.1016/S0012-1606(03)00353-1
Patterson C Homology in classical and molecular biology Mol Biol Evol 1988 5 603 625 3065587
Wake DB Edited by Hall BK, Olson WM Homology and homoplasy Keywords and Concepts in Evolutionary Developmental Biology 2003 Harvard: Harvard University Press 190 201
Abouheif E Brock GR, Cardew G Establishing homology criteria for regulatory gene networks: prospects and challenges Homology Symposium on Homology held at the Novartis Foundation (Symposium 222); London 1999 Wiley: Chichester, UK 207 225
Butler AB Saidel WM Defining sameness: historical, biological, and generative homology BioEssays 2000 22 846 853 10944587 10.1002/1521-1878(200009)22:9<846::AID-BIES10>3.0.CO;2-R
Van Valen LM Homology and causes J Morphol 1982 173 305 312 7186550 10.1002/jmor.1051730307
Tautz D Debatable homologies Nature 1998 395 17 19 9738490 10.1038/25604
Minelli A Molecules, developmental modules, and phenotypes: a combinatorial approach to homology Mol Phyl Evol 1998 9 340 347 10.1006/mpev.1997.0490
Wray GA Abouheif E When is homology not homology? Curr Opin Genet Dev 1998 8 675 680 9914205 10.1016/S0959-437X(98)80036-1
Nielsen C Martinez P Patterns of gene expression: homology or homocracy Dev Genes Evol 2003 213 149 154 12690454
Haszprunar G The types of homology and their significance for evolutionary biology and phylogenetics J Evol Biol 1992 5 13 25 10.1046/j.1420-9101.1992.5010013.x
Hillis DM Hall BK Homology in molecular biology The Hierarchical Basis of Comparative Biology 1994 San Diego, New York, Boston, London, Sydney, Tokyo, Toronto: Academic Press 339 366
Hinman VF Nguyen AT Cameron RA Davidson EH Developmental gene regulatory network architecture across 500 million years of echinoderm evolution Proc Natl Acad Sci USA 2003 100 13356 13361 14595011 10.1073/pnas.2235868100
Lowe CJ Wu M Salic A Evans L Lander E Stange-Thomann M Gruber CE Gerhart J Kirschner M Anteroposterior patterning in hemichordates and the origins of the chordate nervous system Cell 2003 113 853 865 12837244 10.1016/S0092-8674(03)00469-0
De Robertis EM Sasai Y A common plan for dorso- and the origin of chordates ventral patterning in bilateria Nature 1996 380 37 40 8598900 10.1038/380037a0
Finnerty JR Pang K Burton P Paulsen B Martindale MQ Origins of bilateral symmetry: Hox and Dpp expression in a sea anemone Science 2004 304 1335 1337 15131263 10.1126/science.1091946
Martindale MQ Finnerty JR Henry JQ The Radiata and the evolutionary origins of the bilaterian body plan Mol Phyl Evol 2002 24 358 65 10.1016/S1055-7903(02)00208-7
Svensson ME Homology and homocracy revisited: gene expression patterns and hypotheses of homology Dev Genes Evol 2004 214 418 421 15221379 10.1007/s00427-004-0416-2
Panganiban G Irvine SM Lowe C Roehl H Corley LS Sherbon B Grenier JK Fallon JF Kimble J Walker M Wray G Swalla BJ Martindale MQ Carroll SB The origin and evolution of animal appendages Proc Natl Acad Sci USA 1997 94 5162 5166 9144208 10.1073/pnas.94.10.5162
Kimmel CB Was Urbilateria segmented? Trends Genet 1996 12 329 331 8855654 10.1016/S0168-9525(96)80001-1
Erwin DH Davidson EH The last common bilaterian ancestor Dev 2002 129 3021 3032
Minelli A The Development of Animal Form: Ontogeny, Morphology, and Evolution 2003 Cambridge: Cambridge University Press
Jenner RA Evolution of animal body plans: the role of metazoan phylogeny at the interface between patterns and processes Evol Dev 2000 2 208 221 11252564 10.1046/j.1525-142x.2000.00060.x
Fernald RD Evolution of eyes Curr Opin Neurobiol 2000 10 444 450 10981612 10.1016/S0959-4388(00)00114-8
Dickinson WJ Seger J Eye evolution Science 1996 272 467 468 8614787
Salvini-Plawen LV Mayr E Hecht MK, Stene WC, Wallace B On the evolution of photoreceptors and eyes Evolutionary Biology 1977 10 New York: Plenum Press 207 263
Harris WA Pax-6 : Where to be conserved is not conservative Proc Natl Acad Sci USA 1997 94 2098 2100 9122153 10.1073/pnas.94.6.2098
Fernald RD Evolving eyes Int J Dev Biol 2004 48 701 705 15558462 10.1387/ijdb.041888rf
Haider G Callaerts P Gehring WJ Induction of ectopic eyes by targeted expression of the eyeless gene in Drosophila Science 1995 267 1788 1792 7892602
Maddison WP Donoghue MJ Maddison DR Outgroup analysis and parsimony Syst Zool 1994 33 83 103
Swofford DL Olsen GJ Waddell PJ Hillis DM Hillis DM, Moritz C, Mable BK Phylogenetic inference Molecular Systematics 1996 Sunderland: Sinauer Associates 407 514
Poll M Classification des Cichlidae du lac Tanganika. Tribus, genres et espèces Acad R Belg Mem Cl Sci 1986 45 1 163
Lord J Westoby M Leishman M Seed size and phylogeny in six temperate floras: constraints, niche conservatism, and adaptation Am Nat 1995 146 349 364 10.1086/285804
Morales E Estimating phylogenetic inertia in Tithonia (Asteraceae): a comparative approach Evolution 2000 54 475 484 10937224
Brooks DR McLennan DH Phylogeny, Ecology, and Behavior: a Research Program in Comparative Biology 1991 Chicago: University of Chicago Press
Arthur W Farrow M The pattern of variation in centipede segment number as an example of developmental constraint in evolution J Theor Biol 1999 200 183 191 10504284 10.1006/jtbi.1999.0986
Donoghue M Ree RH Homoplasy and developmental constraint: a model and an example from plants Am Zool 2000 40 759 769
Oakley TH Cunningham CW Molecular phylogenetic evidence for the independent evolutionary origin of an arthropod compound eye Proc Natl Acad Sci USA 2002 99 1426 1430 11818548 10.1073/pnas.032483599
Whiting MF Bradler S Maxwell T Loss and recovery of wings in stick insects Nature 2003 421 264 267 12529642 10.1038/nature01313
Aboobaker AA Blaxter ML Hox gene loss during dynamic evolution of the nematode cluster Curr Biol 2003 13 37 40 12526742 10.1016/S0960-9822(02)01399-4
Cork JM Purugganan D The evolution of molecular genetic pathways and networks BioEssays 2004 26 479 484 15112228 10.1002/bies.20026
Chen WJ Ortí G Meyer A Novel evolutionary relationships among four fish model systems Trends Genet 2004 20 424 431 15313551 10.1016/j.tig.2004.07.005
Hill AL Hill MS Liubicich DM Insights into early animal evolution: developmental genes in sponges Am Zool 2000 40 1056 1057
Müller WEG Schröder HC Skorokhod A Buenz C Müller IM Grebenjuk VA Contribution of sponge genes to unravel the genome of the hypothetical ancestor of Metazoa (Urmetazoa) Gene 2001 276 161 173 11591483 10.1016/S0378-1119(01)00669-2
Adell T Grebenjuk VA Wiens M Müller WEG Isolation and characterization of two T-box genes from sponges, the phylogenetically oldest metazoan taxon Dev, Genes Evol 2003 213 421 434 12898249 10.1007/s00427-003-0345-5
Adoutte A Balavoine G Lartillot N Lespinet O Prud'homme B de Rosa R The new animal phylogeny: reliability and implications Proc Natl Acad Sci USA 2000 97 4453 4456 10781043 10.1073/pnas.97.9.4453
Peterson KJ Eernisse DJ Animal phylogeny and the ancestry of bilaterians: inferences from morphology and 18S rDNA gene sequences Evol Dev 2001 3 170 205 11440251 10.1046/j.1525-142x.2001.003003170.x
Halanych KM The new view of animal phylogeny Annu Rev Ecol Evol Syst 2004 35 229 256 10.1146/annurev.ecolsys.35.112202.130124
|
16216118
|
PMC1282587
|
CC BY
|
2021-01-04 16:38:34
|
no
|
Front Zool. 2005 Oct 10; 2:15
|
utf-8
|
Front Zool
| 2,005 |
10.1186/1742-9994-2-15
|
oa_comm
|
==== Front
Int Semin Surg OncolInternational seminars in surgical oncology : ISSO1477-7800BioMed Central London 1477-7800-2-241625091410.1186/1477-7800-2-24Case ReportSmall bowel Gastrointestinal Stromal Tumors can physiologically alter gut motility before causing mechanical obstruction Kothari Manish S [email protected] Vasilis [email protected] John [email protected] Specialist Registrar, General Surgery, Charing Cross Hospital, London W6 8RF, UK2 Senior House Officer, General Surgery, Watford General Hospital, Watford, WD18 0HB, UK3 Consultant Surgeon, General Surgery, Watford General Hospital, Vicarage Road, Watford, WD18 0HB, UK2005 26 10 2005 2 24 24 7 9 2005 26 10 2005 Copyright © 2005 Kothari et al; licensee BioMed Central Ltd.2005Kothari et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
Gastro Intestinal Stromal Tumors (GISTs) are rare stromal neoplasms that represent the most common mesenchymal tumor of the G.I. tract, accounting for 5% of all sarcomas [1,2]. Originating from interstitial cells of Cajal, which are regulators of gut peristalsis, they are preferentially located in the stomach and the small intestine [3] and clinical presentation is variable, ranging from vague complaints to major G.I. bleeding. Surgical resection is the mainstay of treatment for patients with resectable GIST and 5-year survival ranges from 21% to 88% in different series depending on risk grading and completeness of surgical resection [4,5]. Imatinib mesylate, a tyrosine kinase inhibitor, provides an encouraging option for treating high risk GISTs.
Case presentation
We report the case of a 62-year-old lady who had been diagnosed and being treated unsuccessfully for Irritable bowel syndrome for 11 years and eventually found to have an obstructing small bowel GIST.
Conclusion
The symptoms from GIST may mimic those of irritable bowel syndrome. A physiological alteration in gut peristalsis resulting from neoplastic transformation of the interstitial cells of Cajal, is a hypothesis that could explain this presentation. An alternative diagnosis should be considered when treating patients with irritable bowel syndrome who fail to respond for a prolonged period.
==== Body
Case Report
A 69 year old lady presented 11 years ago with pain of mild to moderate severity on the right side of her abdomen. This pain was intermittent, described as 'no more than a dull ache' and associated with nausea, bloating and flatulence. She also suffered from constipation and experienced partial relief of her symptoms following defeacation. She reported worsening of her symptoms after a heavy meal, and experienced occasional bouts of steatorrhea. Physical examination revealed mild right-sided abdominal tenderness only.
Apart from the above she was a healthy lady without any co morbidity and no history of previous abdominal surgery. Initial investigations included an abdominal ultrasound scan (USS), a hepatobiliary iminodiacetic acid (HIDA) scan and an oesophago-gastro-duodenoscopy (OGD), which did not reveal any abnormality. Flexible sigmoidoscopy and colonoscopy were also unremarkable. Her pain was thought to be due to a slow transit colon and she was advised to take laxatives on a regular basis, which resulted in partial and temporary relief of her symptoms. Although her condition did not impact her lifestyle significantly, her symptoms never subsided completely and almost two years following initial presentation she underwent an endoscopic retrograde cholangiopancreatography (ERCP), a cream absorption test, wheat intolerance test and a CT scan of her abdomen. All of these did not reveal any abnormality and thus irritable bowel syndrome was thought to represent best her condition. She was also advised to take antispasmodics together with various combinations of laxatives and a course of Pancrelipase (Creon) tablets for presumed pancreatic insufficiency, which resulted in only temporary relief.
An opinion of a gastroenterology consultant physician was sought, resulting in another abdominal USS and OGD, as well as a barium follow through study which were all normal and thus she continued to take symptomatic treatment.
Finally, 11 years from presentation, the patient was admitted to hospital with acute small bowel obstruction and underwent a laparotomy within 24 hours of admission after adequate fluid and electrolyte replacement. At the operation a sizable mid-ileum, obstructing tumor was revealed, that was wedge resected with adequate gross margins followed by end to end hand sewn anastomosis. The rest of the laparotomy was unremarkable with no evidence of macroscopic spread of disease.
Histopathological analysis revealed a densely crowded spindle and polygonal cell tumour showing moderate nuclear pleomorphism (Figures 1 and 2) . No mitoses were found and the tumour spread into the mucosa and outwards into the serosa. There was no lymph nodal spread. Immunohistochemistry demonstrated positive staining for CD117 (cKit) (figure 3), and negative staining for broad range cytokeratins (MNF 116), CD34, desmin, S-100 and chromogranin. Although there were no mitoses, based on the size (80 × 40 × 50 mm), the tumour was classified as a GIST of intermediate malignant potential.
Figure 3 The tumour stained strongly for CD117 at immunostaining.
The patient made an uneventful recovery and remarkably, all her previous abdominal symptoms were completely resolved. A CT scan of the chest, abdomen and pelvis performed subsequently did not reveal the presence of metastatic disease. Her case was discussed in our multidisciplinary team meeting and it was decided not to proceed with any adjuvant treatment. At one year follow up our patient was a very happy lady completely free of her abdominal symptoms.
Discussion
Although morphologically similar to other benign and malignant smooth muscle and neural stromal tumors, GIST constitutes a distinct group of rare gastrointestinal tract tumors that originate from the interstitial cells of Cajal [6]. The latter are regulators of gut peristalsis and normally express CD117, which is a product of the c-kit proto-oncogene that encodes a tyrosine kinase receptor, which regulates cellular proliferation in GISTs [1,6].
GISTs arise from the muscularis mucosa or muscularis propria layers and most exhibit an endophytic growth pattern, growing within the bowel lumen. The overlying mucosa is usually preserved but larger and more aggressive tumors tend to ulcerate through this. In up to one third of patients the tumor invades an adjacent organ. The vast majority of GISTs (up to 70%) arise in the stomach, with 20–30% originating in the small intestine and the remainder 10% occurring in the oesophagus, colon and rectum [1,3].
The clinical presentation is variable and depends on tumor size and anatomic site. Their submucosal location can produce local obstructive symptoms, particularly when arising in the oesophagus or the small intestine. Vague upper abdominal pain, fullness, GI bleeding, palpable mass are other modes of presentation whereas sometimes they are found incidentally during barium studies, endoscopy or abdominal scans performed for other reasons [1]. According to some authors, visceral obstruction is a rare occurrence even in the presence of extensive peritoneal metastatic disease [7].
Our patient had a very long course of vague symptoms consisting of abdominal pain, nausea, bloating and constipation, which were initially attributed to impaired gut motility as part of the irritable bowel syndrome. Her symptoms, however, failed to improve sufficiently to any treatment approach.
As mentioned above GIST's arise from the interstitial cells of Cajal, which play an important role in the regulation of gut peristalsis. Neoplastic transformation of these cells to a GIST could possibly result in alterations of the normal, local regulation of the gut motility. This hypothesis could, at least in part, explain our patient's symptoms especially at the initial stages of her presentation when the tumor would be too small to cause mechanical obstruction. More research into the physiology of the gut motility in relation to a GIST tumor would be needed to support or reject such a hypothesis.
Conclusion
The occurrence of a GIST in the small bowel can present with vague symptoms, in our case symptoms thought to be from IBS for 11 years. We suggest that at least in the initial years our patients' symptoms may have been the result of functional alteration of gut peristalsis due to the increased number of CD117 positive cells in a slow growing GIST. That would suggest that GISTs can alter the motility of the G.I tract, even when the tumor is significantly small and thus difficult to detect on routine investigations. The symptoms may mimic those of irritable bowel syndrome and an alternative diagnosis should be considered when treating patients who fail to respond for a prolonged period. Resection of the tumour resulting in complete resolution of her symptoms strengthens the above suggestion. Although a diagnostic laparoscopy may not have helped detect the tumour when very small, it could have been considered at some stage during the years preceding her final presentation as acute intestinal obstruction.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
MSK has contributed towards conception, design, analysis and interpretation of data
VK has contributed towards conception, acquisition of data and preparation of the draft.
JMT has contributed towards revising the manuscript critically and has given final approval for the version to be published.
Figure 1 Histopathological appearance of the tumour under low and higher magnification respectively. The tumour consists of densely crowded spindle and polygonal cells with moderate nuclear pleomorphism. It has involved all the layers of the small bowel wall from mucosa outwards to the serosa.
Figure 2 Histopathological appearance of the tumour under low and higher magnification respectively. The tumour consists of densely crowded spindle and polygonal cells with moderate nuclear pleomorphism. It has involved all the layers of the small bowel wall from mucosa outwards to the serosa.
==== Refs
Roberto Logrono Dennie JonesV Sohaib Faruqi Manoop BhutaniS Recent advances in cell biology, diagnosis and therapy of gastrointestinal stromal tumor (GIST) Cancer Biology and Therapy 2004 33 251 258 14726714
Miettinen M Sobin LH Sarlomo-Ricala M Immunohistochemical spectrum of GISTs at different sites and their differential diagnosis with reference to CD117 (kit) Mod Pathol 2000 13 1134 42 11048809 10.1038/modpathol.3880210
DeMatteo RP Lewis JJ Leung D Mudan SS Woodriff JM Brennan MF Two hundred gastro-intestinal stromal tumors: recurrence patterns and prognostic factors for survival Ann Surg 2000 231 51 8 10636102 10.1097/00000658-200001000-00008
Besana-Ciani I Boni L Dionigi G Benevento A Dionigi R Outcome and long-term results of surgical resection for gastrointestinal stromal tumors (GIST) Scand J Surg 2003 92 195 199 14582540
Singer S Rubin BP Lux ML Chen CJ Demetri GD Fletcher CD Fletcher JA Prognostic value of KIT mutation type, mitotic activity and histologic subtype in gastrointestinal stromal tumors J Clin Oncol 2002 20 3898 3905 12228211 10.1200/JCO.2002.03.095
Sincar K Hewlett BR Huizinga JD Chorneyko K Berezin I Riddell RH Intestinal cells of Cajal as precursors of gastrointestinal stromal tumors Am J Surg Pathol 1999 23 377 89 10199467 10.1097/00000478-199904000-00002
Burkill GJC Badran M Thomas JM Judson IR Fisher C Moskovic EC Malignant gastrointestinal stromal tumor: Distribution, imaging features and pattern of metastatic spread Radiology 2003 226 527 532 12563150
|
16250914
|
PMC1282588
|
CC BY
|
2021-01-04 16:38:36
|
no
|
Int Semin Surg Oncol. 2005 Oct 26; 2:24
|
utf-8
|
Int Semin Surg Oncol
| 2,005 |
10.1186/1477-7800-2-24
|
oa_comm
|
==== Front
Nutr Metab (Lond)Nutrition & Metabolism1743-7075BioMed Central London 1743-7075-2-281622974410.1186/1743-7075-2-28ResearchA ketogenic diet reduces amyloid beta 40 and 42 in a mouse model of Alzheimer's disease Van der Auwera Ingrid [email protected] Stefaan [email protected] Leuven Fred [email protected] Samuel T [email protected] NV reMYND, Minderbroederstraat 12, 3000 Leuven, Belgium2 Experimental Genetics Group, K.U. Leuven – Campus Gasthuisberg O&N-06.602, B-3000 Leuven, Belgium3 Accera, Inc., 10901 W 120th Ave., Ste 340, Broomfield, CO, 80021, USA2005 17 10 2005 2 28 28 9 8 2005 17 10 2005 Copyright © 2005 Van der Auwera et al; licensee BioMed Central Ltd.2005Van der Auwera et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
Alzheimer's disease (AD) is a progressive neurodegenerative disorder that primarily strikes the elderly. Studies in both humans and animal models have linked the consumption of cholesterol and saturated fats with amyloid-β (Aβ) deposition and development of AD. Yet, these studies did not examine high fat diets in combination with reduced carbohydrate intake. Here we tested the effect of a high saturated fat/low carbohydrate diet on a transgenic mouse model of AD.
Results
Starting at three months of age, two groups of female transgenic mice carrying the "London" APP mutation (APP/V717I) were fed either, a standard diet (SD) composed of high carbohydrate/low fat chow, or a ketogenic diet (KD) composed of very low carbohydrate/high saturated fat chow for 43 days. Animals fed the KD exhibited greatly elevated serum ketone body levels, as measured by β-hydroxybutyrate (3.85 ± 2.6 mM), compared to SD fed animals (0.29 ± 0.06 mM). In addition, animals fed the KD lost body weight (SD 22.2 ± 0.6 g vs. KD 17.5 ± 1.4 g, p = 0.0067). In contrast to earlier studies, the brief KD feeding regime significantly reduced total brain Aβ levels by approximately 25%. Despite changes in ketone levels, body weight, and Aβ levels, the KD diet did not alter behavioral measures.
Conclusion
Previous studies have suggested that diets rich in cholesterol and saturated fats increased the deposition of Aβ and the risk of developing AD. Here we demonstrate that a diet rich in saturated fats and low in carbohydrates can actually reduce levels of Aβ. Therefore, dietary strategies aimed at reducing Aβ levels should take into account interactions of dietary components and the metabolic outcomes, in particular, levels of carbohydrates, total calories, and presence of ketone bodies should be considered.
==== Body
Background
Alzheimer's disease (AD) is an age-associated neurodegenerative disease that is very common in the US, affecting up to 50% of people between the ages of 75 to 84 years [1]. The number of cases of AD will increase dramatically in the next 50 years due to the aging population of the developed world and will present an increasing medical challenge [2]. Clinically, AD is characterized by progressive impairment in memory and language and is frequently accompanied by behavioral symptoms, such as anxiety and depression. Pathologically, AD is characterized by accumulation of senile plaques, dystrophic neurites, and neurofibrillar tangles. The plaques contain large amounts of the β-amyloid (Aβ) peptide derived from cleavage of the amyloid precursor protein (APP). Mutations in APP that result in increased generation of a particular form of Aβ (Aβ42) have been identified in familial cases of AD and this connection has led to the hypothesis that Aβ is central to the etiology of AD (for review see [3]). However, APP functions as a vesicular transport protein and the etiology of the disease may not be directly related to Aβ, but rather to abnormal cleavage of APP and failure to efficiently move vesicles in the axons [4]. While the precise role of Aβ in AD remains unresolved, it is clear that Aβ serves as a pathological marker for the disease.
The development of AD and the accumulation of Aβ have been linked to dietary factors. Diets rich in saturated fat have been repeatedly implicated in epidemiological studies [5-8], though they have been difficult to reproduce [9]. In addition, several experiments in mouse models seem to confirm the link between lipid rich diets and AD. Using transgenic mouse models of AD several groups have reported that high fat diets or diets with added cholesterol increased levels and deposition of the Aβ peptide [10-14]. However, these studies did not examine the effects of lipid rich diets in combination with low carbohydrate intake.
Diets that contain very low carbohydrate and high fat content are well known to induce the hepatic production of ketone bodies (β-hydroxybutyrate, acetoacetate and acetone) and are often referred to as ketogenic diets (KD). Ketogenic diets in some aspects mimic starvation and were developed for use in humans to treat epilepsy based on the long record of observations that fasting reduces seizures (for review see [15]). The experimental KD is calorie restricted and has fixed composition and is thus different from low carbohydrate diets used for weight loss, which are usually ad lib and variable in composition. Despite these differences, low carbohydrate diets may also be effective in preventing seizures and may work through similar mechanisms as a KD [16]. The precise mechanism for the anti-convulsant properties of these diets is still unknown. The low carbohydrate content of both diets induce many metabolic changes that may be protective, such as elevated circulating ketone body levels, increased oxidation of fats, changes in protein metabolism, and changes in gene expression [17,18].
Results
The present study tested experimentally the effects of a KD composed of extremely low carbohydrate and very high saturated fat content in a transgenic mouse model of AD. The mice express a human APP gene containing the "London" APP mutation (APP/V717I) driven by a thy-1 gene promoter. APP/V717I transgenic mice produce significant levels of soluble Aβ in the brain as early as 3 months of age and exhibit extensive plaque deposition by 12–14 months. The animals demonstrate early behavioral deficits and represent a model of early-onset familial AD [19].
Diet
Sixteen female APP/V717I mice were fed ad libitum on a standard diet (SD) comprised of a high carbohydrate/ low fat chow (Muracon-G chow: 35% carbohydrate, 21% protein, 4.5% fat, 39.5% water, fiber, and ash). The predominant fatty acid in Muracon-G is linoleic acid (18:2) and it comprises 1.4% of the chow by weight. At three months of age half the group (8 animals) was switched to a ketogenic diet (KD) comprised of very low carbohydrate/high fat chow while the remaining 8 animals remained on the SD. In both cases the animals had free access to chow at all times and intake was not experimentally limited. For the KD we used Bio-Serv Inc. F3666 chow: 0.76% carbohydrates, 8% protein, 79% fat, 12% water, fiber, and ash. F3666 is a ketogenic chow composed of lard, butter fat, dextrose, casein, fiber, corn oil, mineral mix, and a vitamin mix. F3666 is rich in saturated fats. Greater than 29% of the F3666 chow is composed of saturated fats by weight: 2.4% myristic acid (C14:0), 18.9% palmitic acid (C16:0), and 8.4% stearic (C18:0) (see Table 1). The animals fed the F3666 chow are referred to as the KD group. The mice that remained on the Muracon-G chow are referred to as the SD group.
Table 1 Fatty acid profile of F3666 (KD) chow
Fatty acid gm/kg gm/kg
C4 Butanoic 4.59 C16 Palmitic 189.03
C6 Hexanoic 3.19 C16:1 cis-9-Hexadecenoic 28.18
C8 Octanoic 2.99 C17 Heptadecanoic 2.38
C10 Decanoic 4.39 C17:1 Heptadecenoic 1.43
C10:1 Decenoic 0.80 C18 Stearic 84.79
C12 Lauric 6.41 C18.1 Oleic 298.34
C12:1 cis-9-Dodecenoic 0.40 C18:2 Linoleic 119.44
C14 Myristic 24.43 C18:3 Linolenic 6.33
C14:1 cis-9-Tetradecenoic 6.14 C20 Eicosanoic 1.18
C15 Pentadecanoic 0.48 C20:1 cis-11-Eicosenoic 3.80
KD diet and weight loss
During the first 7 days many of the animals in the KD group were reluctant to eat the new chow and lost weight (Figure 1). To improve consumption and mitigate weight loss, SD chow was mixed with KD chow at a ratio of 1:3 starting at day 16 until day 20. For the seven days following day 20 the amount of SD chow was reduced to a few crumbs sprinkled over the KD chow. After day 28 the animals were returned to KD chow only. The mixed chow restored body weights of the KD group to approximately the level of the SD group, at about 20 grams (Figure 1). When the animals were fed KD chow exclusively, body weights again dropped, yet tended to stabilize at approximately 18 grams (Figure 1). At the conclusion of the experiment mean weights were significantly different (SD 22.2 ± 0.6 g vs. KD 17.5 ± 1.4 g, p = 0.0067).
Figure 1 Average weight in grams of each group during the course of the experiment. Blue squares represent standard diet (SD) group. Red circles represent ketogenic diet (KD) group. Error bars represent standard error of the mean. Days signify time in days from start of diet change. Animals on KD lost weight. To mitigate weight loss and improve feeding a small amount of SD chow was mixed with KD chow during the second week and then removed on the third week, see methods.
KD diet elevates serum β-hydroxybutyrate levels
To measure the effectiveness of the chow to induce a ketogenic state, blood samples were taken weekly and examined for levels of β-hydroxybutyrate (BHB). Eight days after switching the chow animals in the KD group had greatly elevated BHB levels compared to the SD group (SD 0.26 ± 0.023 mM vs. KD 8.94 + 1.8 mM, p < 0.0001, Figure 2), possibly due to some animals not eating. As expected, feeding the mixed chow on days 16–28 reduced serum ketone bodies (Figure 2). Yet, at all time points examined after day 0 ketone levels were significantly greater in the KD fed group compared to the SD group (Figure 2). The average serum BHB concentration over the course of the experiment was elevated in the KD group compared to the SD group (SD 0.29 ± 0.06 mM vs. KD 3.85 ± 1.1 mM, p < 0.0078). The elevated ketone levels in the KD group suggested that a metabolic shift toward fat utilization had occurred in these animals.
Figure 2 Ketogenic diet induces ketone bodies production. Standard diet (SD) group shown in blue, ketogenic diet (KD) group shown in red, error bars represent standard error of the mean. Serum β-hydroxybutyrate (BHB) levels in mM. * indicates p < 0.05 between KD and SD group. Days signify time in days from start of diet change. Serum β-hydroxybutyrate levels were significantly elevated in KD group at all time points after day 0. Bar indicates period of mixed chow for KD group.
Cognitive testing
After 38 days on the diet animals were tested for behavioral deficits using object recognition tests as previously described [20], see methods. Despite the differences in chow, BHB levels, and weight loss, no difference in behavioral measures were detected between the groups (Table 2).
Table 2 Recognition Index
Test SD KD p value
RI by time 56.2 ± 5.7 56.7 ± 8.2 0.963
Curiosity by time 10.0 ± 2.7 11.6 ± 6.5 0.803
RI by frequency 49.3 ± 3.6 54.6 ± 9.7 0.628
Curiosity by frequency 14.4 ± 2.3 11.9 ± 3.9 0.530
Mean velocity 3.8 ± 0.4 4.5 ± 0.8 0.447
Aβ levels
At four months of age APP/V717I mice do not possess Aβ positive plaques and all the Aβ is present in the soluble fraction [19]. Therefore, 43 days after dietary change levels of soluble Aβ in the brain were measured in both groups of animals. Brain homogenate was isolated as previously described [20], see methods. One hemisphere from each animal was analyzed for both Aβ 40 and 42 levels using the Amyloid Aβ40 or Aβ42 ELISA High Sensitivity Kit (The Genetics Company, Zurich, Switzerland). Levels of both soluble Aβ 40 and 42 were found to be significantly lower in the KD fed group (Figure 3). In cases of familial AD excess Aβ42 is produced relative to Aβ40 thereby increasing the Aβ 42/40 ratio. We examined the ratio of Aβ42 to Aβ40 and found no difference between groups (SD 0.51 ± 0.024 vs. KD 0.56 ± 0.026, p= 0.2872), suggesting that the diet did not alter cleavage sites on APP, but instead promoted a general lowering of Aβ species.
Figure 3 Ketogenic diet reduces Aβ40 and Aβ42. Aβ levels as ng/g of brain tissue. Standard diet (SD) group shown in blue, ketogenic diet (KD) group shown in red, error bars represent standard error of the mean. SD chow Aβ40 1.72 ± 0.12 ng/g vs. KD chow Aβ40 1.28 ± 0.09 ng/g, p= 0.012. SD chow Aβ42 0.88 ± 0.05 ng/g vs. KD chow Aβ42 0.71 ± 0.0.4 ng/g, p = 0.016.
Total protein levels were examined to determine if a general decline in brain protein in the KD group could explain the decrease in Aβ levels. However, protein levels, measured as mg/ml of brain homogenate, did not differ between the two groups (SD 0.56 ± 0.035 mg/ml vs. KD 0.51 ± 0.017 mg/ml, p= 0.213). Since most (though not all) of the animals in the KD group lost weight, the animals with the greatest weight loss may have been expected to have the lowest Aβ levels. However, the levels of Aβ40 or Aβ42 did not correlate with weight change across all groups (Aβ40 r = 0.16, p = 0.20; Aβ42 r = 0.05, p = 0.49) or in the KD group alone (Aβ40 r < 0.0001, p = 0.96; Aβ42 r = 0.09, p = 0.62). Both of these measures suggest that the Aβ lowering effect is not the result of a general lowering of protein levels due to weight loss.
A better measure of the effectiveness of the ketogenic diet was serum BHB levels, since all the animals in the KD group exhibited elevated BHB levels. When Aβ40 and 42 levels from all animals were correlated with the average serum BHB over the course of the experiment, a significant correlation was observed (Aβ40 r = 0.42, p = 0.016; Aβ42 r = 0.42, p = 0.016). However, this correlation was most likely driven by the large differences in BHB and Aβ levels between the two groups, since there was no significant correlation between Aβ and BHB levels in the KD group alone (Aβ40 r = 0.083, p = 0.58; Aβ42 r = 0.15, p = 0.46).
Discussion
This study demonstrated the unexpected result that a brief treatment with a low carbohydrate/high saturated fat diet reduced total Aβ levels in a mouse model of Alzheimer's disease. Previous studies had suggested that diets rich in saturated fats or cholesterol increased both the production and deposition of Aβ in mouse models of AD, leading to the suggestion that diets rich in lipids were a factor in AD [10,12-14]. However, these diets were not low carbohydrate diets. In the high cholesterol diets, cholesterol was added to the diet without reduction in other components [10,14]. In the studies of high fat diets, carbohydrate content was still relatively high. For example, Ho et al. used a diet of 60% fat, 20% carbohydrate, 20% protein. This diet was sufficiently high in carbohydrate to cause large increases in body weight in the animals [11].
The interaction of different macronutrients, in particular fats and carbohydrates, is known to influence the metabolic state of the animal. For example, Marsset-Baglieri et al. examined if fat in the diet alone was sufficient to shift energy balance toward fat storage. Yet, rats fed ad libitum high fat (50%) diets devoid of carbohydrates did not increase energy intake and did not gain in body adiposity, while animals fed high fat diets (30%) in the presence of carbohydrates (56%) increased energy intake and gained in body fat [21]. Such studies support the view that when fat and carbohydrates are consumed simultaneously, the carbohydrates stimulate insulin secretion and thereby promote storage of fat (for recent review see [22]). Therefore, it is important to consider the macronutrient profile of the diet when examining the effects of dietary fat on biological processes.
In the present study, transgenic animals were fed ad libitum a very high fat (79%) diet that was practically devoid of carbohydrates (0.76%). The KD resulted in ketone body production, weight loss, and decreased Aβ levels. Hence, the data presented here suggests that it may not be fats in the diet that increases Aβ levels, but perhaps levels of total calories, carbohydrates, or the metabolic state of the animal.
Epidemiological studies in humans have implicated saturated fats in the diet as a risk factor for Alzheimer's disease. For example, Kalmijn et al. correlated eating habits with incidence of dementia after a two year follow-up in a large study of 5,386 subjects in Rotterdam, NL. The results from this analysis led the author to suggest that diets rich in saturated fats and cholesterol increased the risk of several types of dementia [5]. However, after a 6 year follow-up of this same population, no correlation between dementia and fat intake could be identified, leading the authors to conclude "High intake of total, saturated, and trans fat and cholesterol and low intake of MUFA, PUFA, n-6 PUFA, and n-3 PUFA were not associated with increased risk of dementia or its subtypes." [9]. More recent studies have also examined the link between dietary fat and cognitive decline. In a study of 2,560 participants ages 65 and older in the Chicago Health and Aging project, fat intake was measured by food questionnaire and correlated with cognitive testing examined after a 3 and 6 year follow- up. This large study found only weak trends between saturated fat and cholesterol intake and cognitive decline [7]. Both the rodent and human studies highlight the complications of trying to link complex environmental factors, such as eating habits or macronutrient intake, with dementia and Alzheimer's disease. In particular, one complicating factor in the human studies is the normal consumption of large amounts of carbohydrates in modern diets.
The present study demonstrates that, contrary to expectations, transgenic mice fed ab libitum a very low carbohydrate/high saturated fat diet present lower levels of Aβ after only 43 days of dietary change. The KD group exhibited low levels of both Aβ40 and the more amyloidic Aβ42, suggesting that the KD diet did not change or increase the efficiency of cleavage sites within APP. Instead the data suggests the KD regime either reduced processing of APP or increased degradation of Aβ species. Most of the animals administered the ketogenic diet lost body weight as well as exhibited reduced Aβ levels. However, the reduced Aβ levels may not have been due to a general lowering of protein content. Total brain protein levels did not differ between the groups and Aβ levels did not correlate with weight loss. Interestingly, despite change in diet, weight loss, and Aβ levels, no change in cognitive performance was observed (Table 2). This observation agrees with the general finding that KD diets are not harmful to mice [23]. Also, the finding that reduction in Aβ did not improve cognitive performance may be due to the modest lowering of levels under these conditions and longer treatment may be required.
The KD diet was developed to mimic a starvation response in animals without reducing calories to harmful levels [15]. In this way a KD is similar to caloric restriction (CR) regimes that have been used in many species to alter aging and increase some forms of stress resistance. CR typically reduces calories 30–40% compared to ad libitum fed animals and has numerous positive effects on animal health [24]. In the present study we did not attempt to restrict calories in any way and the animals had free access to the ketogenic chow at all times and intake was self limited. However, since the animals were reluctant at first to eat the KD chow and we observed weight loss in the KD group, we cannot rule out the possibility that the Aβ lowering effects were due to CR.
Yet, CR and KD may work through similar mechanisms. KD are well known to reduce insulin signaling and mimic starvation, thereby increasing fatty acid oxidation and promoting a catabolic state [17]. Similarly, CR is well known to reduce serum insulin and IGF levels and much of the benefit of CR may derive from this reduction in insulin/IGF signaling (for review see [25]). For example, decreased insulin/IGF-like signaling inhibits protein synthesis and promotes protein degradation, which may lead to the clearing of degradation-sensitive proteins, such as amyloidic peptides.
Increasing evidence suggests a role for insulin/IGF-1 in regulating APP and modulating Aβ levels. Receptors for both insulin and IGF-1 are highly expressed in brain, especially in hippocampus and cortex, where they may influence learning and memory [26]. Insulin signaling in the brain increases extracellular levels of Aβ by promoting secretion [27] and inhibiting degradation by insulin-degrading enzyme [28]. This view has also gained recent support in humans. Fishel et al. demonstrated that induced hyperinsulinemia in healthy elderly subjects elevated both serum and spinal fluid Aβ levels, suggesting insulin plays a role in elevating Aβ, especially under conditions such as type II diabetes [29].
Such an interpretation is consistent with recent studies demonstrating similar Aβ lowering effects of a low carbohydrate, caloric restriction (CR) regime in mice expressing the "Swedish" form of APP (Tg2576 mice). In these animals, lower levels of Aβ 40 and 42 were detected in animals fed 30% less carbohydrates than ad libitum fed animals. The Aβ lowering effect may have been due to increased α-secretase and insulin degrading enzyme activity in the CR animals [30].
Alternatively, other physiologic changes may have reduced Aβ levels in this study. The high levels of ketone bodies alone may have contributed to increased protein turnover. Ketone bodies added to cell culture have been shown to lead to increased oxidation of proteins prone to oxidative damage. The presence of damaged protein triggers proteolysis of normally long lived proteins via chaperone-mediated autophagy [31]. Such a mechanism may be at work in animals fed a ketogenic chow and exposed to high levels of ketone bodies. Some support for this model comes from the observation that elevated ketone body levels correlated better with lowering of Aβ species than did weight loss. In addition, ketone bodies may serve as an efficient substrate for neuronal metabolism. Previous studies have shown that acute elevation of ketone bodies may improve cognitive performance in some individuals with mild to moderate AD [32].
Conclusion
As the population of the developed world ages the incidence of AD is predicted to increase dramatically and will place a tremendous burden on health services [2]. Dietary intervention represents a relatively safe and readily available method to combat AD. Yet, the key dietary links remain unclear. Much of the earlier work has focused on the role of high fat or high cholesterol diets and their contribution to AD. However, evidence suggests that the primary genetic risk factor for late onset AD, the epsilon4 allele of apolipoprotein E, may have been selected against in populations with long historical exposure to agriculture [33]. In addition, foods rich in carbohydrates are relatively recent additions to the human diet and are likely to be more evolutionarily discordant than high fat diets [34]. Therefore, the recent evolutionary switch to high carbohydrate (HC) diets may play an important role in development of AD. HC diets are well known to stimulate insulin signaling and result in a suppression of lipid metabolism [22]. Thus, such diets may lead to inappropriate lipid environments in neurons, mis-cleavage of APP and the resulting inhibition of cellular trafficking, and ultimately increasing the risk of developing AD (for overview see [35]).
Methods
Animals
Sixteen APP [V717I] C57Bl × FVB female mice of 3 months of age were used for this study. Mice were housed under a reversed day-night rhythm: 14 hours light/10 hours darkness starting at 7 p.m. in standard metal cages type RVS T2 (area of 540 cm2). The number of mice per cage was limited in accordance with legislation on animal welfare. All mice were genotyped by polymerase chain reaction (PCR) at the age of 3 weeks. Mice were blind randomized and age-matched and had free access to pre-filtered and sterile water (UV-lamp). Mice had free access to either ketogenic (KD) (code F3666, Bio-Serv, Frenchtown, US) or standard (SD) chow (Muracon-G, Trouw Nutrition, Gent). The F3666 chow is a runny paste and was given in special designed liquid food suppliers and was refreshed daily. F3666 is a liquid chow and the animals frequently spilled the chow in the cage. Also, since all the animals in a given group were housed together in a single cage, measuring chow intake per animal was not possible and was not recorded. Due to some problems with weight loss in animals in the KD group, these animals were fed a mixed chow 1(SD):3(KD) starting at day 16 until day 20. From days 21–27 the amount of SD chow was reduced to a few crumbs sprinkled over the KD chow. After day 28 the animals were returned to KD chow only. However, one mouse in the KD group refused food intake and died despite attempts of feeding via gavage. One control animal died during blood draw.
Blood collection and analysis
Blood was collected from anesthetized mice from either the orbital plexus or via a heart puncture. β-hydroxybutyrate levels were measured spectrophotometrically using the Stanbio liquicolor β-hydroxybutyrate kit (Stanbio Inc., Boerne, Texas).
Novel object recognition test
The novel object recognition test was performed after 38 days of treatment using the method described by Dewachter et al. [20]. Briefly, mice were familiarized for one hour to a Plexiglas open-field box (52 × 52 × 40 cm) with black vertical walls and a translucent floor, dimly illuminated by a lamp placed underneath the box. The next day the animals were placed in the same box and submitted to a 10 minute acquisition trial. During this trial mice were placed individually in the open field in the presence of 2 × object A (orange barrel or green cube, similar sized of ± 4 cm), and the duration (time AA) and the frequency (Freq AA) exploring object A (when the animals snout was directed towards the object at a distance of < 1 cm and the mice were actively sniffing in the direction of the object) was recorded by a computerized system (Ethovision, Noldus information Technology, Wageningen, the Netherlands). During the 10 minute retention trial (second trial) performed 3 hours later, a novel object (object B, green cube or orange barrel) was placed together with the familiar object (object A) into the open field. (Freq A and Freq B and Time A and Time B, respectively). The recognition index (RI), defined as the ratio of the duration in which the novel object was explored over the duration in which both objects were explored (Time B/(Time A + Time B) × 100), was used to measure non-spatial memory. The duration and frequency object A was explored during the acquisition trial (Time AA and Freq AA) was used to measure curiosity.
Analytical techniques
The mice were anaesthetized with a mixture of Ketalar® (Ketamin), Rompun® (Xylazin 2%) and Atropine (2:1:1) and flushed trans-cardially with physiological serum at 4°C. This was performed to remove blood from the brain vessels, a procedure which has no influence on organ integrity. Blood was collected via a heart puncture and a 1 ml syringe in heparinized Eppendorf tubes. The brain was removed from the cranium and hindbrain and forebrain were separated with a cut in the coronal/frontal plane. The forebrain was divided evenly into left and right hemisphere by using a midline sagittal cut. One hemisphere of each animal was homogenized using a Potter, a glass tube (detergent free, 2 cm3) and a mechanical homogenizer (650 rpm). A volume of 6.5 × 1/2 brain weight of freshly prepared 20 mM Tris/HCl buffer (pH 8.5) with Proteinase Inhibitors (1 tablet per 50 ml Tris/HCl buffer, CompleteTM, Roche, Mannheim, Germany) was used as homogenization buffer. The homogenates were collected in Beckman centrifuge tubes TLX and collected on ice prior to centrifugation. Before the samples were centrifuged, 10 % of the sample was used for determination of the total protein concentration of the homogenate whereas 90 % of the samples were centrifuged to process further for biochemistry. The supernatant (soluble fraction containing secreted APP and amyloid peptides) was separated from the pellet by centrifugation (membrane fraction containing membrane-bound APP-fragments). The supernatant was processed further by column chromatography to concentrate the amyloid peptides using small reversed phase columns (C18-Sep-Pack Vac 3cc cartridges, Waters, Massachusetts, MA). Amyloid peptides were eluted with 75% A-TFA and the eluates were collected in 2 ml tubes on ice. Eluates were freeze-dried in a speedvac concentrator (Savant, Farmingdale, NY) overnight and resolved in 240 μl of the sample diluent furnished with the ELISA kits. To quantify the amount of human Aβ40 and human Aβ42 in the soluble fraction of the brain homogenates, commercially available Enzyme-Linked-Immunosorbent-Assay (ELISA) kits were used (h Amyloid Aβ40 or Aβ42 ELISA high sensitive, The Genetics Company, Zurich, Switzerland). Quantification of the Aβ content of the samples was obtained by comparing absorbance to a standard curve made with synthetic Aβ40 or Aβ42. Total protein concentration in the brain homogenate was measured using Bradford solution (Pierce Inc.).
List of abbreviations
AD, Alzheimer's disease; APP, amyloid precursor protein; Aβ, amyloid beta; SD, standard diet; KD, ketogenic diet; CR, caloric restriction; BHB, β-hydroxybutyrate.
Competing interests
As co-founder of Accera, Inc., STH holds shares in an organization and may gain or lose financially from the publication of this manuscript. In addition, STH has applied for patents relating to the content of the manuscript and may gain or lose financially from publication of this manuscript.
Acknowledgements
We thank the staff of Accera, Inc. for critical reading of this manuscript and helpful discussions.
==== Refs
Evans DA Funkenstein HH Albert MS Scherr PA Cook NR Chown MJ Hebert LE Hennekens CH Taylor JO Prevalence of Alzheimer's disease in a community population of older persons. Higher than previously reported Jama 1989 262 2551 2556 2810583 10.1001/jama.262.18.2551
Hebert LE Scherr PA Bienias JL Bennett DA Evans DA Alzheimer disease in the US population: prevalence estimates using the 2000 census Arch Neurol 2003 60 1119 1122 12925369 10.1001/archneur.60.8.1119
Selkoe DJ Alzheimer's disease: genes, proteins, and therapy Physiol Rev 2001 81 741 766 11274343
Stokin GB Lillo C Falzone TL Brusch RG Rockenstein E Mount SL Raman R Davies P Masliah E Williams DS Goldstein LS Axonopathy and transport deficits early in the pathogenesis of Alzheimer's disease Science 2005 307 1282 1288 15731448 10.1126/science.1105681
Kalmijn S Launer LJ Ott A Witteman JC Hofman A Breteler MM Dietary fat intake and the risk of incident dementia in the Rotterdam Study Ann Neurol 1997 42 776 782 9392577 10.1002/ana.410420514
Grant WB Dietary links to Alzheimer's disease: 1999 update J Alzheimers Dis 1999 1 197 201 12214118
Morris MC Evans DA Bienias JL Tangney CC Wilson RS Dietary fat intake and 6-year cognitive change in an older biracial community population Neurology 2004 62 1573 1579 15136684
Morris MC Evans DA Bienias JL Tangney CC Bennett DA Aggarwal N Schneider J Wilson RS Dietary fats and the risk of incident Alzheimer disease Arch Neurol 2003 60 194 200 12580703 10.1001/archneur.60.2.194
Engelhart MJ Geerlings MI Ruitenberg A Van Swieten JC Hofman A Witteman JC Breteler MM Diet and risk of dementia: Does fat matter?: The Rotterdam Study Neurology 2002 59 1915 1921 12499483
Refolo LM Malester B LaFrancois J Bryant-Thomas T Wang R Tint GS Sambamurti K Duff K Pappolla MA Hypercholesterolemia accelerates the Alzheimer's amyloid pathology in a transgenic mouse model Neurobiol Dis 2000 7 321 331 10964604 10.1006/nbdi.2000.0304
Ho L Qin W Pompl PN Xiang Z Wang J Zhao Z Peng Y Cambareri G Rocher A Mobbs CV Hof PR Pasinetti GM Diet-induced insulin resistance promotes amyloidosis in a transgenic mouse model of Alzheimer's disease Faseb J 2004 18 902 904 15033922
Levin-Allerhand JA Lominska CE Smith JD Increased amyloid- levels in APPSWE transgenic mice treated chronically with a physiological high-fat high-cholesterol diet J Nutr Health Aging 2002 6 315 319 12474021
Shie FS Jin LW Cook DG Leverenz JB LeBoeuf RC Diet-induced hypercholesterolemia enhances brain A beta accumulation in transgenic mice Neuroreport 2002 13 455 459 11930160 10.1097/00001756-200203250-00019
George AJ Holsinger RM McLean CA Laughton KM Beyreuther K Evin G Masters CL Li QX APP intracellular domain is increased and soluble Abeta is reduced with diet-induced hypercholesterolemia in a transgenic mouse model of Alzheimer disease Neurobiol Dis 2004 16 124 132 15207269 10.1016/j.nbd.2004.01.009
Thio LL Sitzwohl A Trevathan E Kleinman RE The ketogenic diet American Academy of Pediatrics Pediatric Nutrition Handbook 2002 5th Elk Grove Village IL, American Academy of Pediatrics
Kossoff EH Krauss GL McGrogan JR Freeman JM Efficacy of the Atkins diet as therapy for intractable epilepsy Neurology 2003 61 1789 1791 14694049
Stafstrom CE Dietary approaches to epilepsy treatment: old and new options on the menu Epilepsy Curr 2004 4 215 222 16059506 10.1111/j.1535-7597.2004.46001.x
Sullivan PG Rippy NA Dorenbos K Concepcion RC Agarwal AK Rho JM The ketogenic diet increases mitochondrial uncoupling protein levels and activity Ann Neurol 2004 55 576 580 15048898 10.1002/ana.20062
Moechars D Dewachter I Lorent K Reverse D Baekelandt V Naidu A Tesseur I Spittaels K Haute CV Checler F Godaux E Cordell B Van Leuven F Early phenotypic changes in transgenic mice that overexpress different mutants of amyloid precursor protein in brain J Biol Chem 1999 274 6483 6492 10037741 10.1074/jbc.274.10.6483
Dewachter I Reverse D Caluwaerts N Ris L Kuiperi C Van den Haute C Spittaels K Umans L Serneels L Thiry E Moechars D Mercken M Godaux E Van Leuven F Neuronal deficiency of presenilin 1 inhibits amyloid plaque formation and corrects hippocampal long-term potentiation but not a cognitive defect of amyloid precursor protein [V717I] transgenic mice J Neurosci 2002 22 3445 3453 11978821
Marsset-Baglieri A Fromentin G Tome D Bensaid A Makkarios L Even PC Increasing the protein content in a carbohydrate-free diet enhances fat loss during 35% but not 75% energy restriction in rats J Nutr 2004 134 2646 2652 15465761
Hellerstein MK Carbohydrate-induced hypertriglyceridemia: modifying factors and implications for cardiovascular risk Curr Opin Lipidol 2002 13 33 40 11790961 10.1097/00041433-200202000-00006
Rho JM Sarnat HB Sullivan PG Robbins CA Kim DW Lack of long-term histopathologic changes in brain and skeletal muscle of mice treated with a ketogenic diet J Child Neurol 2004 19 555 557 15526964
Masoro EJ Caloric restriction Aging (Milano) 1998 10 173 174 9666230
Bordone L Guarente L Calorie restriction, SIRT1 and metabolism: understanding longevity Nat Rev Mol Cell Biol 2005 6 298 305 15768047 10.1038/nrm1616
Wickelgren I Tracking insulin to the mind Science 1998 280 517 519 9575095 10.1126/science.280.5363.517
Gasparini L Gouras GK Wang R Gross RS Beal MF Greengard P Xu H Stimulation of beta-amyloid precursor protein trafficking by insulin reduces intraneuronal beta-amyloid and requires mitogen-activated protein kinase signaling J Neurosci 2001 21 2561 2570 11306609
Vekrellis K Ye Z Qiu WQ Walsh D Hartley D Chesneau V Rosner MR Selkoe DJ Neurons regulate extracellular levels of amyloid beta-protein via proteolysis by insulin-degrading enzyme J Neurosci 2000 20 1657 1665 10684867
Fishel MA Watson GS Montine TJ Wang Q Green PS Kulstad JJ Cook DG Peskind ER Baker LD Goldgaber D Nie W Asthana S Plymate SR Schwartz MW Craft S Hyperinsulinemia Provokes Synchronous Increases in Central Inflammation and {beta}-Amyloid in Normal Adults Arch Neurol 2005 62 1539 44 16216936 10.1001/archneur.62.10.noc50112
Wang J Ho L Qin W Rocher AB Seror I Humala N Maniar K Dolios G Wang R Hof PR Pasinetti GM Caloric restriction attenuates beta-amyloid neuropathology in a mouse model of Alzheimer's disease Faseb J 2005 19 659 661 15650008 10.1096/fj.04-2370com
Finn PF Dice JF Ketone bodies stimulate chaperone-mediated autophagy J Biol Chem 2005 280 25864 70 15883160 10.1074/jbc.M502456200
Reger MA Henderson ST Hale C Cholerton B Baker LD Watson GS Hyde K Chapman D Craft S Effects of beta-hydroxybutyrate on cognition in memory-impaired adults Neurobiol Aging 2004 25 311 314 15123336 10.1016/S0197-4580(03)00087-3
Corbo RM Scacchi R Apolipoprotein E (APOE) allele distribution in the world. Is APOE*4 a 'thrifty' allele? Ann Hum Genet 1999 63 ( Pt 4) 301 310 10738542 10.1046/j.1469-1809.1999.6340301.x
Cordain L Eades MR Eades MD Hyperinsulinemic diseases of civilization: more than just Syndrome X Comp Biochem Physiol A Mol Integr Physiol 2003 136 95 112 14527633 10.1016/S1095-6433(03)00011-4
Henderson ST High carbohydrate diets and Alzheimer's disease Med Hypotheses 2004 62 689 700 15082091 10.1016/j.mehy.2003.11.028
|
16229744
|
PMC1282589
|
CC BY
|
2021-01-04 16:37:46
|
no
|
Nutr Metab (Lond). 2005 Oct 17; 2:28
|
utf-8
|
Nutr Metab (Lond)
| 2,005 |
10.1186/1743-7075-2-28
|
oa_comm
|
==== Front
Respir ResRespiratory Research1465-99211465-993XBioMed Central London 1465-9921-6-1181623616310.1186/1465-9921-6-118ResearchPsoriasin, one of several new proteins identified in nasal lavage fluid from allergic and non-allergic individuals using 2-dimensional gel electrophoresis and mass spectrometry Bryborn Malin [email protected] Mikael [email protected] Lars-Olaf [email protected] Laboratory of Clinical and Experimental Allergy, Department of Otorhinolaryngology, Malmo University Hospital, Lund University, Malmo, Sweden2005 19 10 2005 6 1 118 118 5 4 2005 19 10 2005 Copyright © 2005 Bryborn et al; licensee BioMed Central Ltd.2005Bryborn et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
Extravasation and luminal entry of plasma occurs continuously in the nose. This process is markedly facilitated in patients with symptomatic allergic rhinitis, resulting in an increased secretion of proteins. Identification of these proteins is an important step in the understanding of the pathological mechanisms in allergic diseases. DNA microarrays have recently made it possible to compare mRNA profiles of lavage fluids from healthy and diseased patients, whereas information on the protein level is still lacking.
Methods
Nasal lavage fluid was collected from 11 patients with symptomatic allergic rhinitis and 11 healthy volunteers. 2-dimensional gel electrophoresis was used to separate proteins in the lavage fluids. Protein spots were picked from the gels and identified using mass spectrometry and database search. Selected proteins were confirmed with western blot.
Results
61 spots were identified, of which 21 were separate proteins. 6 of these proteins (psoriasin, galectin-3, alpha enolase, intersectin-2, Wnt-2B and hypothetical protein MGC33648) had not previously been described in nasal lavage fluids. The levels of psoriasin were markedly down-regulated in allergic individuals. Prolactin-inducible protein was also found to be down-regulated, whereas different fragments of albumin together with Ig gamma 2 chain c region, transthyretin and splice isoform 1 of Wnt-2B were up-regulated among the allergic patients.
Conclusion
The identification of proteins in nasal lavage fluid with 2-dimensional gelelectrophoresis in combination with mass spectrometry is a novel tool to profile protein expression in allergic rhinitis and it might prove useful in the hunt for new therapeutic targets or diagnostic markers for allergic diseases. Psoriasin is a potent chemotactic factor and its down-regulation during inflammation might be of importance for the outcome of the disease.
==== Body
Background
Increased vascular permeability and plasma exudation are important characteristics of allergic rhinitis leading to an increased amount of secreted proteins [1,2]. Earlier investigations with DNA microarray analysis [3] have described the gene expression in nasal mucosa. However, there is a considerable interest to identify some of the secreted proteins for a better understanding of the pathological processes and possibly to find new therapeutical targets or diagnostic markers for the disease.
Combining 2-dimensional gel electrophoresis (2-DE) with mass spectrometry (MS) have recently emerged as a method for identifying proteins in different biological samples. In short, proteins are separated in the first dimension according to their isoelectric points (pI) and then in the second dimension according to their molecular weight using SDS-PAGE. Each spot on the SDS-PAGE gel corresponds to one protein. The spots can be excised and further analysed and identified using mass spectrometry and database searching [4]. 2-DE together with MS has previously been used to investigate the protein content in nasal lavage fluid (NLF) [5] and a study of differences in the NLF protein content from smokers and non-smokers [6] is recently reported. However, changes in relation to allergic airway diseases have so far not been probed. The main purpose of this study was to use 2-DE in combination with MS and database search in order to map and identify the broad range of secreted proteins in NLF from individuals allergic to pollen (birch/timothy) and to compare that with NLF from non-allergic healthy individuals.
Materials and methods
Skin prick test
Skin prick tests (SPT) were performed with a standard panel of 10 common airborne allergens (ALK, Copenhagen, Denmark) including pollen (birch, timothy and artemisia), house dust mites (D. Pteronyssimus and D. Farinae), molds (Cladosporium and Alternaria) and animal allergens (cat, dog and horse). SPT were performed on the volar side of the forearm with saline buffer as negative and histamine chloride (10 mg/ml) as positive controls. The diameter of the wheal reactions were measured after 20 min with a ruler.
Subjects
The study included 11 patients (6 women) with symptomatic birch and/or grass pollen induced intermittent allergic rhinitis and 11 healthy volunteers (7 women), serving as controls. The mean age of patients and controls was 43 (26–55) and 41 (24–55) years, respectively. The diagnosis of birch and/or grass pollen induced allergic rhinitis was based on a positive history of intermittent allergic rhinitis for at least 2 years and positive SPT to birch and/or grass. All patients were classified as having severe symptoms (itchy nose and eyes, sneezing, nasal secretion and nasal blockage) during pollen season and they had all been treated with antihistamines and nasal steroids during pollen seasons previous years. Patients had no continuous symptoms of asthma and they did not take any asthma medication. All patients presented a wheal reaction diameter >3 mm towards birch or timothy in SPT (roughly correponding to a 3+ or 4+ reaction when compared with histamine [7]). Exclusion criteria included a history of perennial symptoms, upper airway infection for the last 2 weeks before the time of visit and treatment with local or systemic corticosteroids during the last 2 months. The controls were all symptom-free, had no history of allergic rhinitis and had negative SPT to the standard panel of allergens as described above. They had no history of upper airway infection for 2 weeks before the time of visit and they were all free of medication. The study was approved by the Ethics Committee of the Medical Faculty, Lund University.
Sample collection and preparation
Nasal lavage fluid was collected during either birch pollen (9 patients) or grass pollen season (2 patients). Patients were included when they had experienced substantial symptoms of rhinoconjunctivitis (itchy nose and eyes, sneezing, nasal secretion and nasal blockage) during at least 3 consecutive days. The majority of the patients were seen within 5–10 days after the first appearance of symptoms and a local pollen count.
Nasal lavage fluid was collected according to a previously described method [8]. After clearing of excess mucus from the nose sterile saline solution of room temperature was sprayed into both nostrils, respectively. The fluid was allowed to return passively and collected in a graded tube until 7 ml was recovered. NLFs were centrifuged at 1750 rpm at 4°C for 10 min to remove the cell content and the supernatants were stored at -70°C until sample preparation.
Before concentration of the samples NLFs were thawed and centrifuged at 12300 rpm at 4°C for 20 min to remove debris. Using Vivaspin 6 and Vivaspin 500 concentrators (Vivascience, Hannover, Germany) supernatants were concentrated and desalted. The protein concentration was determined using BCA Protein Assay Kit (Pierce Biotechnology, Rockford, USA) and resulted in a protein concentration of 1572–5625 μg/ml for healthy individuals and 1833–7867 μg/ml for allergic individuals. NLFs were stored at -70°C until analysed.
2-DE analysis
Samples were mixed with rehydration solution containing 8 M Urea, 2% CHAPS, 2.8 mg/ml DTT (Sigma-Aldrich, Steinheim, Germany), 0.5% IPG Buffer (pH 3–10) (Amersham Biosciences, Uppsala Sweden) and a small amount of bromophenol blue. For analytical gels 150 μg of protein was added to a final volume of 450 μl for each sample. For the preparative gels, one for healthy and one for allergic samples, 600 μg from a pool of samples was used. To be able to load as much as 600 μg on the preparative gels pooled samples were further concentrated using Microcon YM-3 (Millipore, Billerica, USA) before added to rehydration solution. Samples were incubated for approximately 15 min in room temperature in order to completely solubilize and denature the proteins. Samples were centrifuged at 13000 rpm for 10 min and thereafter loaded onto 24 cm 3–10 non linear IPG strips (Amersham Biosciences, Uppsala, Sweden). In-gel rehydration and isoelectric focusing (IEF) was performed over night (~60000 Vh) using Ettan IPGphor Isoelectric Focusing System (Amersham Biosciences, Uppsala, Sweden). After IEF strips were stored at -70°C until analysed. The IPG strips were equilibrated in SDS equilibration buffer (75 mM Tris, 6 M Urea, 30% glycerol, 2% SDS and 0.002% bromophenol blue (Sigma-Aldrich, Steinheim, Germany)) for 2 × 15 min. DTT (10 mg/ml) (Sigma-Aldrich, Steinheim, Germany) was added to the first and iodoacetamide (25 mg/ml) (Sigma- Aldrich, Steinheim, Germany) to the second equilibration step. After equilibration strips were loaded onto laboratory-made 12.5% acrylamide second dimension gels. SDS-PAGE was performed at constant effect (10 W/gel) for about 4 h and 30 min using the Ettan DALT II system (Amersham Biosciences).
Staining of gels and gel image analysis
Second dimension gels were fixed in 30% ethanol and 10% acetic acid over night, washed 4 × 30 min in 20% ethanol and stained with the fluorescent dye ruthenium II tris-bathophenantroline disulfonate (1 μM) for about 6 h. Thereafter gels were destained in 40% ethanol and 10% acetic acid over night and washed with double distilled water for about 4 × 30–60 min [9]. All incubation and washing steps were performed with gentle agitation. Gels were kept dark in double distilled water at 4°C until scanned. The gels were automatically scanned using a robotic system together with a 9410 Typhoon scanner (488 nm laser) from Amersham Biosciences [10] and the gel images were analysed using the computer softwares Image master 2D Platinum (Amersham Biosciences) and Ludesi 2D Interpreter (Ludesi AB, Lund, Sweden). The volume in each spot was calculated as integrated optical density over the spot's area. The amount of protein in each spot was expressed as %VOL (ppm), that is the volume for the spot divided with the total volume for all spots in the gel.
Spot picking, protein digestion and MALDI-TOF (Matrix Assisted Laser Desorption Ionization-Time Of Flight) analysis
Using the Ettan spot handling workstation (Amersham Biosciences) selected spots were automatically cut from the preparative gels, destained and enzymatically digested with trypsin (porcine Sequencing Grade Modified Trypsin, Promega, Madison, USA). The tryptic peptides were then spotted onto a MALDI target plate [11]. The MALDI target plates were loaded in a Micromass M@ldi MALDI-TOF mass spectrometer (Waters, Milford, USA) for analysis of the peptide masses.
Database search
Peptide masses retrieved from MALDI-TOF analysis spectra were submitted to a database (IPI human 1.38) [12] by using the search engine PIUMS [13]. The following matcher parameters were used: constant modification of cysteine by carbamidomethylation, variable modification of methionine by oxidation and maximum 1 missed cleavage for trypsin. A protein hit was considered significant if the PIUMS quality score was ≥ 4.7, which corresponds to an expectation value of 0.01. A search in IPI human 1.38 was also done using the search engine Mascot and the results from this search were compared with the results from PIUMS.
Western blot
NLFs were mixed with SDS sample buffer, heated at 95–100°C for 5 min and centrifuged at 10 000 rpm for 10 min. Equal amounts of the samples were loaded onto NuPAGE Bis-Tris 4–12% gel (Invitrogen, Carlsbad, USA), separated by electrophoresis (Mini vertical gel system, Thermo EC, Waltham, USA), and blotted to Immobilon-P PVDF membranes (Millipore, Billerica, USA). Membranes were blocked in buffer 1 (Tris-HCl 10 mM pH 7.4, NaCl 0.9% and dry milk 5%) and then incubated overnight with primary antibody (1 μg/ml) against psoriasin, galectin-3 (Abcam, Cambridge, UK), Wnt-2B (Zymed, South San Francisco, USA) and alpha enolase (Santa Cruz, Santa Cruz, USA), respectively. Membranes were washed 2 times with buffer 1 followed by incubation for approximately 2 h with HRP conjugated secondary antibody (50 ng/ml). After 2 washes with buffer 2 (Tris-HCl 10 mM pH 7.4, NaCl 0.9% and Tween 20 0.05%) membranes were incubated for 5 min in SuperSignal West Pico solution (Pierce Biotechnology, Rockford, USA). The chemiluminescence was detected using MAN-X X-ray system (Fujifilm Science Imaging systems, USA). Developed films for quantitative analysis were scanned and analysed in ImageQuant (Molecular dynamics, Sunnyvale, USA). There were no antibodies available against hypothetical protein MGC33648 and the relevant fragment of intersectin-2. Hence, these proteins could not be assessed with western blot.
Statistical analysis
All values were expressed as mean values ± SEM. Statistical analysis of the protein expression was performed in Ludesi 2D interpreter (Ludesi AB) using one-way ANOVA.
Results
Novel proteins in nasal lavage fluid
Of the spots picked from the 2D-gels and submitted to MALDI-TOF analysis 61 spots were identified (figure 1 and table 1). 21 of these were identified as separate proteins. The majority of the proteins has previously been identified in NLF [5,6,14] but this is the first study where psoriasin, galectin-3, alpha enolase, intersectin-2, Wnt-2B and hypothetical protein MGC33648 have been recognized. The occurrence of psoriasin, galectin-3, Wnt-2B and alpha enolase was confirmed with western blot (figure 2).
Figure 1 2-DE protein pattern for NLF from a healthy non-allergic individual. The protein name for each numbered spot is presented in table 1.
Table 1 Identified proteins in nasal lavage fluid from allergic and non-allergic individuals.
Gel no. Protein Accession no. (Swissprot/IPI) MW (kDa) (theoretical) pI (theoretical)
1 Albumin P02768 69.4 5.9
2 Albumin P02768 69.4 5.9
3 Albumin P02768 69.4 5.9
4 Albumin P02768 69.4 5.9
5 Albumin P02768 69.4 5.9
6 Albumin P02768 69.4 5.9
7 Albumin P02768 69.4 5.9
8 Albumin P02768 69.4 5.9
9 Albumin P02768 69.4 5.9
10 Albumin P02768 69.4 5.9
11 Albumin P02768 69.4 5.9
12 Albumin P02768 69.4 5.9
13 Albumin P02768 69.4 5.9
14 Albumin P02768 69.4 5.9
15 Albumin P02768 69.4 5.9
16 Albumin P02768 69.4 5.9
17 Albumin P02768 69.4 5.9
18 Albumin IPI00216773 45.2 6.0
19 Albumin IPI00384697 47.4 6.3
20 Albumin IPI00384697 47.4 6.3
21 Albumin IPI00384697 47.4 6.3
22 Albumin IPI00384697 47.4 6.3
23 Albumin IPI00384697 47.4 6.3
24 Albumin IPI00384697 47.4 6.3
25 Albumin IPI00384697 47.4 6.3
26 Albumin IPI00384697 47.4 6.3
27 Prolactin-inducible protein P12273 16.6 8.3
28 Prolactin-inducible protein P12273 16.6 8.3
29 Prolactin-inducible protein P12273 16.6 8.3
30 Prolactin-inducible protein P12273 16.6 8.3
31 Cystatin S P01036 16.2 4.9
32 Cystatin S P01036 16.2 4.9
33 Cystatin SN P01037 18.8 6.8
34 Hemoglobin beta chain P68871 16.0 7.3
35 Hemoglobin beta chain P68871 16.0 7.3
36 Intersectin 2,
(splice isoform 2) Q9NZM3-2 190.5 8.4
37 Lipocalin-1 P31025 19.3 5.4
38 Transthyretin P02766 15.9 5.5
39 Calgranulin B P06702 13.2 5.7
40 Psoriasin (S100A7) P31151 11.5 6.3
41 Psoriasin (S100A7) P31151 11.5 6.3
42 Galectin-3 P17931 26.2 8.6
43 Apolipoprotein A1 P02647 30.8 5.5
44 Apolipoprotein A1 P02647 30.8 5.5
45 Apolipoprotein A1 P02647 30.8 5.5
46 Alpha-2 glycoprotein 1, zink P25311 34.3 5.9
47 Alpha-2 glycoprotein 1, zink P25311 34.3 5.9
48 Serotransferrin P02787 77.1 6.8
49 Serotransferrin P02787 77.1 6.8
50 Serotransferrin P02787 77.1 6.8
51 Serotransferrin P02787 77.1 6.8
52 Alpha enolase P06733 47.0 7.0
53 Alpha enolase P06733 47.0 7.0
54 Hemopexin P02790 51.7 6.5
55 Hemopexin P02790 51.7 6.5
56 Hemopexin P02790 51.7 6.5
57 Ig gamma 2 chain c region P01859 35.9 7.7
58 Wnt-2B protein
(splice isoform 1) Q93097-1 43.8 9.3
59 Fibrinogen beta chain P02675 55.9 8.5
60 Hypothetical protein MGC33648 IPI00168581 34.2 9.0
61 Actin, cytoplasmic 1 or 2 P60709 41.7 5.3
Proteins in bold are newly identified in NLF.
Figure 2 Western blot analysis of NLF from healthy non-allergic individuals. (1) Psoriasin, (2) Wnt-2B, (3) Galectin-3, (4) Enolase.
Differences in protein expression between allergic and non-allergic individuals
14 spots exhibited a clear difference in the protein content when the material from allergic and non-allergic individuals was compared (table 2). 8 of these spots were identified as different fragments of albumin and all these were up-regulated in allergic individuals (1.8- to 2.6-fold). Wnt-2B (splice isoform 1), transthyretin and Ig gamma-2 chain c region were also found to be up-regulated in allergic compared to non-allergic individuals (2.5-, 1.6- and 2.1-fold, respectively). In contrast, prolactin-inducible protein and two forms of psoriasin were found to be down-regulated in allergic individuals (2.0-, 2.0- and 3.4-fold, respectively). The psoriasin levels in nasal lavage fluids from three patients with allergic rhinitis and three controls were also assessed using western blot analysis (figure 3A). Quantitative analysis revealed reduced levels among the allergic patients; 19962 ± 5410 for the non-allergic individuals compared to 6834 ± 2258 for the allergic individuals (figure 3B).
Table 2 Proteins differently expressed in allergic and non-allergic individuals.
Protein No. Non-allergica Allergica Fold changes
Psoriasin 40 1831 ± 425 543 ± 110* -3.4
Psoriasin 41 7410 ± 1675 3689 ± 650* -2.0
Prolactin-inducible protein 29 2322 ± 491 1130 ± 194* -2.0
Transthyretin 38 1182 ± 177 1902 ± 286* 1.6
Ig gamma 2 chain c region 57 3493 ± 590 7349 ± 1137* 2.1
Wnt-2B protein (splice isoform 1) 58 3360 ± 668 8405 ± 1761* 2.5
Albumin fragment 6 803 ± 210 1580 ± 258* 2.0
Albumin fragment 11 363 ± 83 957 ± 247* 2.6
Albumin fragment 15 386 ± 99 606 ± 116* 2.4
Albumin fragment 17 514 ± 115 927 ± 86* 1.8
Albumin fragment 21 407 ± 74 1008 ± 183* 2.5
Albumin fragment 23 139 ± 50 324 ± 72* 2.3
Albumin fragment 20 2135 ± 374 3991 ± 638* 1.9
Albumin fragment 4 219 ± 54 500 ± 92* 2.3
a The amount of protein is expressed as mean %VOL (ppm) ± SEM.
* p < 0.05
Figure 3 Expression analysis of psoriasin with western blot. A: Western blot analysis of NLF from three healthy non-allergic individuals (1–3) and three allergic individuals (4–6) demonstrating the levels of psoriasin. 1 μg of total protein was loaded in each lane. B: Quantitative analysis of western blot with ImageQuant (Molecular Dynamics, USA). Each data point represents the mean ± SEM.
Discussion
The present study is the first where 2-DE in combination with MS has been used to study differences between allergic and non-allergic individuals. It reveals the presence of six novel NLF proteins: psoriasin, galectin-3, alpha enolase, intersectin-2, Wnt-2B and hypothetical protein MGC33648. One of these novel proteins, psoriasin, was markedly down-regulated in allergic individuals. The same was found for prolactin-inducible protein, whereas different fragments of albumin together with Ig gamma 2 chain c region, transthyretin and splice isoform 1 of Wnt-2B were up-regulated among the allergic patients.
Increased vascular permeability and plasma exudation are important characteristics of allergic rhinitis resulting in an increased secretion of proteins. Several of the secreted proteins are collected in NLF and their identification is of importance for understanding pathological mechanisms in allergic rhinitis. DNA microarray technology is a relatively new method for analysing gene expression in different samples and it has recently been used in allergy research [15,16]. With DNA microarray technology it is possible to map genes that are up- or down-regulated in tissues or cells involved in allergic disease, something that might contribute to the identification of new pathological mechanisms or therapeutic targets [17]. However, all regulatory mechanisms are not operated at the transcriptional level. Hence, one of the disadvantages with DNA microarray technology is that the detected mRNA levels not always correlate with the actual protein levels in the sample. 2-DE together with MS-analysis is a powerful method to profile the protein expression in different samples. Two previous studies have used this methodology to analyse proteins in lavage fluids from the upper airways [6,14]. The present data now demonstrate that this approach also can be used to compare healthy and pathological samples in order to get an overview of which proteins that can be of importance for the development of allergic diseases.
Most of the 21 proteins identified in the present study correlate well with proteins found in NLF from normal, healthy individuals in previous studies [5,6,14]. However, 6 of the proteins (psoriasin, galectin-3, alpha enolase, intersectin-2, Wnt-2B and hypothetical protein MGC33648) have not previously been described in NLF. Psoriasin, also called S100A7, belongs to the S100 protein family and like other members in this family (for example calgranulin B) it has calcium-binding properties. It was first identified in psoriatic skin [18] where it is highly up-regulated. Psoriasin is thought to be involved in inflammation since it is a potent chemotactic factor for CD4+ T lymphocytes and neutrophils [19]. Galectin-3 belongs to a family of β-galactoside-binding animal lectins [20]. It is expressed in mast cells, monocytes/macrophages, neutrophils and eosinophils. Although galectin-3 lacks signal peptide, it can be secreted [21] and it functions as chemotactic factor for monocytes and macrophages [22]. Galectin-3 also has the ability to bind Ig-E and increased mRNA levels for galectin-3 have been found in neutrophils derived from the blood of allergic patients [23]. Alpha enolase is a ubiquitous multifunctional enzyme involved in many different processes [24]. It has been reported as an important allergen in inhalant allergies to fungi [25] and specific IgE antibodies have been found in patients allergic to fungi [26], all corroborating the notion that alpha enolase might play a role in allergic reactions. One spot was identified as splice isoform 2 of intersectin-2, a cytoplasmic protein involved in endocytosis [27]. The spot identified is probably a degradation product of intersectin-2 since it is present at a much lower MW than the theoretical value. Wnt-2B is a developmental protein that might play a role as hematopoietic growth factor [28].
The role of these newly identified proteins in allergic rhinitis is not known and they all render further investigation. However, special attention might be drawn to the two different forms of psoriasin [29] found to be down-regulated during allergic rhinitis in the present study. Since allergic rhinitis is an inflammatory disease and psoriasin is a chemotactic factor for inflammatory cells one could have expected the opposite. One explanation for this reversed condition is that psoriasin in addition to its chemotactic properties has an other not yet discovered role in the inflammation process. In this context, it is also essential to recognize that inflammation is normally a self-resolving process with the existence of both positive and negative regulators that ultimately allow complete resolution and homeostasis. In the absence of resolution and clearance or in the event of a dampened healing response, persistent inflammation can arise in the form of tissue damage as associated with chronic disease. Thus, the down-regulation of psoriasin during the allergic inflammation could be of importance for the natural resolution of the disease.
Previous findings [23] have suggested that galectin-3 is involved in the inflammatory reaction seen in allergic patients. However, in the present study no differences in the galectin-3 content were seen when material from allergic and non-allergic individuals were compared. Alpha enolase was identified in two spots but any quantitative difference between allergic and non-allergic individuals could not be detected. In contrast, splice isoform 1 of Wnt-2B was found to be up-regulated (2.5-fold) among allergic individuals. Such an increase of the Wnt-2B secretion might be related to the increased growth and maturation stimulation of eosinophils and neutrophils often seen during the allergic inflammation.
In addition to the novel NLF proteins psoriasin and Wnt-2B, a group of other proteins were also found to be differently expressed during the allergic inflammation. There was a 2.0-fold decrease of one form of prolactin-inducible protein (PIP) in allergic individuals. PIP is expressed in exocrine organs like sweat, salivary and lacrimal glands [30]. The functions of PIP is not completely known but it has CD4-binding properties and is a strong inhibitor of T lymphocyte apoptosis [31]. A down-regulation of PIP might therefore be associated with an increased apoptosis of T lymphocytes, something that might contribute to a limitation of the inflammatory process. The theoretical pI for PIP is 8.3 but it was detected in the gel at a pI around 4–5. Without its signal peptide the theoretical pI decreases to 5.4 which is closer to our observation. Transthyretin, also called prealbumin, is a plasma protein involved in the transport of thyroxine and retinol [32]. The small up-regulation of transthyretin detected in allergic individuals (1.6-fold) is probably due to the increased plasma exudation seen in allergic rhinitis [2].
Several of the spots were identified as the same protein. Hence, many proteins are present in different forms. The different forms may result from post-translational modifications like phosphorylation, glycosylation, acetylation or degradation of the proteins. All spots identified as albumin are probably different forms and fragments of albumin and since albumin is highly abundant in plasma this high amount of degradation products in NLF is expected. It is not surprising that a few of these fragments were up-regulated in allergic individuals since this only confirms previous findings that the secretion of albumin is increased in allergic individuals. Ig gamma 2 chain c region was also found to be up-regulated in allergic individuals which also confirms previous findings [33].
Conclusion
2-DE in combination with MS-analysis appears to be a powerful method to profile the protein expression and compare healthy samples with pathological samples. In this study both previously identified and newly identified proteins were detected in NLF using this method. Some of these proteins, like psoriasin, Wnt-2B and PIP were found to be differently expressed in allergic and non-allergic individuals. Further investigations are needed to explain the pathological significance of these proteins. It is possible that some of them can be defined as new therapeutic targets or diagnostic markers for allergic diseases.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
MB performed the sample preparation, 2-DE, gel image analysis, analysis of MALDI results, database search, western blot and drafted the manuscript. MA and LOC conceived the study, participated in its design and coordination and helped to draft the manuscript.
Acknowledgements
The present work was supported by the Swedish Medical Research Council, the Swedish Heart Lung Foundation, the Swedish Association for Allergology, the Swedish Foundation for Health Care Science and Allergic Research and the Royal Physiographic Society.
The 2-DE and MS-analysis took place at the SWEGENE Proteomics Platform in Lund, Sweden and the authors would like to thank professor Peter James (head of department) for the cooperation and Anna-Karin Påhlmann, Ulrika Brynnel and Liselotte Andersson for technical assistance and advice. Gustav Wallmark (Ludesi AB) is acknowledged for helping with the analysis in Ludesi 2D Interpreter.
==== Refs
Christodoulopoulos P Cameron L Durham S Hamid Q Molecular pathology of allergic disease. II: Upper airway disease J Allergy Clin Immunol 2000 105 211 223 10669839 10.1016/S0091-6749(00)90068-X
Svensson C Andersson M Greiff L Alkner U Persson CG Exudative hyperresponsiveness of the airway microcirculation in seasonal allergic rhinitis Clin Exp Allergy 1995 25 942 950 8556565
Benson M Svensson PA Carlsson B Jernas M Reinholdt J Cardell LO Carlsson L DNA microarrays to study gene expression in allergic airways Clin Exp Allergy 2002 32 301 308 11929497 10.1046/j.1365-2222.2002.01300.x
Gorg A Weiss W Dunn MJ Current two-dimensional electrophoresis technology for proteomics Proteomics 2004 4 12 3665 3685 15543535 10.1002/pmic.200590007
Lindahl M Stahlbom B Tagesson C Identification of a new potential airway irritation marker, palate lung nasal epithelial clone protein, in human nasal lavage fluid with two-dimensional electrophoresis and matrix-assisted laser desorption/ionization-time of flight Electrophoresis 2001 22 1795 1800 11425234 10.1002/1522-2683(200105)22:9<1795::AID-ELPS1795>3.0.CO;2-J
Ghafouri B Stahlbom B Tagesson C Lindahl M Newly identified proteins in human nasal lavage fluid from non-smokers and smokers using two-dimensional gel electrophoresis and peptide mass fingerprinting Proteomics 2002 2 112 120 11788998 10.1002/1615-9861(200201)2:1<112::AID-PROT112>3.3.CO;2-E
Aas K Belin L Standardization of diagnostic work in allergy Int Arch Allergy Appl Immunol 1973 45 57 60 4580380
Benson M Strannegard IL Wennergren G Strannegard O Interleukin-5 and interleukin-8 in relation to eosinophils and neutrophils in nasal fluids from school children with seasonal allergic rhinitis Pediatr Allergy Immunol 1999 10 178 185 10565558 10.1034/j.1399-3038.1999.00036.x
Lamanda A Zahn A Roder D Langen H Improved Ruthenium II tris (bathophenantroline disulfonate) staining and destaining protocol for a better signal-to-background ratio and improved baseline resolution Proteomics 2004 4 599 608 14997483 10.1002/pmic.200300587
Back P Nagard F Bolmsjo G Bengtsson S James P Automating gel image acquisition J Proteome Res 2003 2 662 664 14692461 10.1021/pr0340936
Levander F James P Automated Protein Identification by the Combination of MALDI MS and MS/MS Spectra from Different Instruments J Proteome Res 2005 4 71 74 15707359 10.1021/pr0498584
Kersey PJ Duarte J Williams A Karavidopoulou Y Birney E Apweiler R The International Protein Index: an integrated database for proteomics experiments Proteomics 2004 4 1985 1988 15221759 10.1002/pmic.200300721
Samuelsson J Dalevi D Levander F Rognvaldsson T Modular, scriptable and automated analysis tools for high-throughput peptide mass fingerprinting Bioinformatics 2004 20 3628 3635 15297302
Lindahl M Stahlbom B Svartz J Tagesson C Protein patterns of human nasal and bronchoalveolar lavage fluids analyzed with two-dimensional gel electrophoresis Electrophoresis 1998 19 3222 3229 9932819 10.1002/elps.1150191828
Benson M Svensson PA Adner M Caren H Carlsson B Carlsson LM Martinsson T Rudemo M Cardell LO DNA microarray analysis of chromosomal susceptibility regions to identify candidate genes for allergic disease: a pilot study Acta Otolaryngol 2004 124 813 819 15370566 10.1080/00016480410018025
Benson M Jansson J Adner M Luts A Uddman R Cardell LO Gene profiling reveals increased expression of uteroglobin and other anti-inflammatory genes in nasal fluid cells from patients with allergic rhinitis. Clin Exp Allergy 2005 35 4 473 478 15836756
Benson M Olsson M Rudemo M Wennergren G Cardell LO Pros and cons of microarray technology in allergy research Clin Exp Allergy 2004 34 1001 1006 15248841 10.1111/j.1365-2222.2004.02007.x
Madsen P Rasmussen HH Leffers H Honore B Dejgaard K Olsen E Kiil J Walbum E Andersen AH Basse B Molecular cloning, occurrence, and expression of a novel partially secreted protein "psoriasin" that is highly up-regulated in psoriatic skin J Invest Dermatol 1991 97 701 712 1940442 10.1111/1523-1747.ep12484041
Jinquan T Vorum H Larsen CG Madsen P Rasmussen HH Gesser B Etzerodt M Honore B Celis JE Thestrup-Pedersen K Psoriasin: a novel chemotactic protein J Invest Dermatol 1996 107 5 10 8752830 10.1111/1523-1747.ep12294284
Barondes SH Cooper DN Gitt MA Leffler H Galectins. Structure and function of a large family of animal lectins J Biol Chem 1994 269 20807 20810 8063692
Hughes RC Secretion of the galectin family of mammalian carbohydrate-binding proteins Biochim Biophys Acta 1999 1473 172 185 10580137
Sano H Hsu DK Yu L Apgar JR Kuwabara I Yamanaka T Hirashima M Liu FT Human galectin-3 is a novel chemoattractant for monocytes and macrophages J Immunol 2000 165 2156 2164 10925302
Truong MJ Gruart V Kusnierz JP Papin JP Loiseau S Capron A Capron M Human neutrophils express immunoglobulin E (IgE)-binding proteins (Mac-2/epsilon BP) of the S-type lectin family: role in IgE-dependent activation J Exp Med 1993 177 243 248 8418206 10.1084/jem.177.1.243
Pancholi V Multifunctional alpha-enolase: its role in diseases Cell Mol Life Sci 2001 58 902 920 11497239
Baldo BA Baker RS Inhalant allergies to fungi: reactions to bakers' yeast (Saccharomyces cerevisiae) and identification of bakers' yeast enolase as an important allergen Int Arch Allergy Appl Immunol 1988 86 201 208 3292441
Ishiguro A Homma M Torii S Tanaka K Identification of Candida albicans antigens reactive with immunoglobulin E antibody of human sera Infect Immun 1992 60 1550 1557 1548078
Pucharcos C Estivill X de la Luna S Intersectin 2, a new multimodular protein involved in clathrin-mediated endocytosis FEBS Lett 2000 478 43 51 10922467 10.1016/S0014-5793(00)01793-2
Van Den Berg DJ Sharma AK Bruno E Hoffman R Role of members of the Wnt gene family in human hematopoiesis Blood 1998 92 3189 3202 9787155
Celis JE Rasmussen HH Vorum H Madsen P Honore B Wolf H Orntoft TF Bladder squamous cell carcinomas express psoriasin and externalize it to the urine J Urol 1996 155 2105 2112 8618345 10.1097/00005392-199606000-00098
Myal Y Gregory C Wang H Hamerton JL Shiu RP The gene for prolactin-inducible protein (PIP), uniquely expressed in exocrine organs, maps to chromosome 7 Somat Cell Mol Genet 1989 15 265 270 2727805 10.1007/BF01534877
Gaubin M Autiero M Basmaciogullari S Metivier D Mis hal Z Culerrier R Oudin A Guardiola J Piatier-Tonneau D Potent inhibition of CD4/TCR-mediated T cell apoptosis by a CD4-binding glycoprotein secreted from breast tumor and seminal vesicle cells J Immunol 1999 162 2631 2638 10072505
Ingenbleek Y Young V Transthyretin (prealbumin) in health and disease: nutritional implications Annu Rev Nutr 1994 14 495 533 7946531 10.1146/annurev.nu.14.070194.002431
Johansson SG Deuschl H Immunoglobulins in nasal secretion with special reference to IgE. I. Methodological studies Int Arch Allergy Appl Immunol 1976 52 364 375 1017894
|
16236163
|
PMC1282590
|
CC BY
|
2021-01-04 16:36:26
|
no
|
Respir Res. 2005 Oct 19; 6(1):118
|
utf-8
|
Respir Res
| 2,005 |
10.1186/1465-9921-6-118
|
oa_comm
|
==== Front
Respir ResRespiratory Research1465-99211465-993XBioMed Central London 1465-9921-6-1241625090910.1186/1465-9921-6-124ResearchMultiple P2Y receptors couple to calcium-dependent, chloride channels in smooth muscle cells of the rat pulmonary artery Chootip Krongkarn [email protected] Alison M [email protected] Charles [email protected] Department of Physiology and Pharmacology, University of Strathclyde, Strathclyde Institute for Biomedical Sciences, John Arbuthnott Building, 27 Taylor Street, Glasgow G4 ONR, UK2 Department of Physiology, Faculty of Medical Science, Naresuan University, Phitsanulok 65000, Thailand2005 26 10 2005 6 1 124 124 20 7 2005 26 10 2005 Copyright © 2005 Chootip et al; licensee BioMed Central Ltd.2005Chootip et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
Uridine 5'-triphosphate (UTP) and uridine 5'-diphosphate (UDP) act via P2Y receptors to evoke contraction of rat pulmonary arteries, whilst adenosine 5'-triphosphate (ATP) acts via P2X and P2Y receptors. Pharmacological characterisation of these receptors in intact arteries is complicated by release and extracellular metabolism of nucleotides, so the aim of this study was to characterise the P2Y receptors under conditions that minimise these problems.
Methods
The perforated-patch clamp technique was used to record the Ca2+-dependent, Cl- current (ICl,Ca) activated by P2Y receptor agonists in acutely dissociated smooth muscle cells of rat small (SPA) and large (LPA) intrapulmonary arteries, held at -50 mV. Contractions to ATP were measured in isolated muscle rings. Data were compared by Student's t test or one way ANOVA.
Results
ATP, UTP and UDP (10-4M) evoked oscillating, inward currents (peak = 13–727 pA) in 71–93% of cells. The first current was usually the largest and in the SPA the response to ATP was significantly greater than those to UTP or UDP (P < 0.05). Subsequent currents tended to decrease in amplitude, with a variable time-course, to a level that was significantly smaller for ATP (P < 0.05), UTP (P < 0.001) and UDP (P < 0.05) in the SPA. The frequency of oscillations was similar for each agonist (mean≈6–11.min-1) and changed little during agonist application. The non-selective P2 receptor antagonist suramin (10-4M) abolished currents evoked by ATP in SPA (n = 4) and LPA (n = 4), but pyridoxalphosphate-6-azophenyl-2',4'-disulphonic acid (PPADS) (10-4M), also a non-selective P2 antagonist, had no effect (n = 4, 5 respectively). Currents elicited by UTP (n = 37) or UDP (n = 14) were unaffected by either antagonist. Contractions of SPA evoked by ATP were partially inhibited by PPADS (n = 4) and abolished by suramin (n = 5). Both antagonists abolished the contractions in LPA.
Conclusion
At least two P2Y subtypes couple to ICl,Ca in smooth muscle cells of rat SPA and LPA, with no apparent regional variation in their distribution. The suramin-sensitive, PPADS-resistant site activated by ATP most resembles the P2Y11 receptor. However, the suramin- and PPADS-insensitive receptor activated by UTP and UDP does not correspond to any of the known P2Y subtypes. These receptors likely play a significant role in nucleotide-induced vasoconstriction.
==== Body
Background
Uridine 5'-triphosphate (UTP) and uridine 5'-diphosphate (UDP) act via P2Y receptors, whilst adenosine 5'-triphosphate (ATP) acts via P2X as well as P2Y receptors, to modulate vascular tone [1-3]. P2X receptors are ligand-gated cation channels and the ability of the P2X1 subtype to mediate rapid, transient inward currents in pulmonary artery smooth muscle cells [4,5] and induce constriction of the pulmonary vasculature (see [6] and references therein) has been characterised in some depth. P2Y receptors are G protein-coupled receptors and P2Y agonists act at smooth muscle receptors to evoke vasoconstriction in the rat perfused lung at resting tone, but induce vasodilation via endothelial receptors if muscle tone is first raised [7-10]. Similarly, P2Y agonists are contractile at resting tone and relaxant at raised tone in isolated branches of rat intrapulmonary arteries [11-13]. Compared with P2X receptors much less is known about which of the eight mammalian P2Y subtypes (P2Y1,2,4,6,11,12,13,14) [14,15] are expressed in pulmonary vascular smooth muscle or about the signalling pathways through which they act.
In a previous study [6] we showed that UTP and UDP both act via two P2Y receptors to evoke contraction of rat isolated pulmonary arteries. For each agonist one site was insensitive to the antagonists suramin and pyridoxalphosphate-6-azophenyl-2',4'-disulphonic acid (PPADS), whilst the other was inhibited by suramin, but not PPADS. UTP is a potent agonist at the P2Y2 and P2Y4 receptors and a weaker agonist at the P2Y6 subtype [16,17]. Of these three receptors, only the P2Y2 is suramin-sensitive and PPADS-insensitive [18], so this is likely to be one of the sites of action of UTP. The molecular identity of the suramin-and PPADS-insensitive site of action of UTP is unclear as the P2Y4 and P2Y6 subtypes are both reported to be antagonised by PPADS, but not suramin [18-20]. UDP is a potent agonist at the P2Y6 receptor only [16,17]. mRNA for this subtype and suramin-insensitive contractions to UDP in pulmonary arteries have been demonstrated [12], but the lack of effect of PPADS against the contractions evoked by UDP in our previous study are inconsistent with the P2Y6 receptor.
A number of factors that can complicate the characterisation of P2Y receptors may have prevented the clear identification of the P2Y receptors mediating the contractions seen in previous studies. These include the release of nucleotides from cells, their breakdown by ecto-nucleotidases and their bioconversion by ecto-nucleoside diphosphokinase (eNDPK) [16,21-24]. Thus, as well as a direct action at the P2Y6 receptor, UTP can also act indirectly, after dephosphorylation to UDP. Likewise, UDP can be converted to UTP by eNDPK and so act indirectly at P2Y2 and P2Y4 receptors. At present, potent and selective inhibitors of the ecto-enzymes are not available, but one way to minimise these metabolic problems is to apply the agonists to rapidly perfused, dissociated cells.
The aim of the present study was to extend the pharamacological characterisation of the P2Y receptors mediating pulmonary vasoconstriction, in conditions that minimise the influence of the release and extracellular metabolism of nucleotides. We used the perforated-patch clamp technique to record the Ca2+-dependent, Cl- current (ICl,Ca) induced by nucleotides in single, acutely dissociated pulmonary artery smooth muscle cells [4,5]. The pulmonary vascular bed has well characterised regional differences in receptor and ion channel distribution, including ICl,Ca [8,25], so the responses of cells isolated from large and small pulmonary arteries were compared. In addition to UTP and UDP, we also applied ATP, an agonist at the P2Y1,2,4,11 & 12 subtypes [14] and determined the ability of the P2 antagonists suramin and PPADS to inhibit the responses evoked by each of the agonists. We found that at least two P2Y subtypes couple to ICl,Ca, with no apparent regional variation in their distribution. The suramin-sensitive, PPADS-resistant site activated by ATP most resembles the P2Y11 receptor. However, the suramin- and PPADS-insensitive receptor activated by UTP and UDP does not correspond to any of the known P2Y subtypes.
Methods
Isolated cell preparation
Male Sprague-Dawley rats (150 – 250 g) were killed by the approved Schedule 1 method of cervical dislocation and exsanguination. After thoracotomy, the heart and lungs were removed en bloc, the lungs separated and small (SPA, 200–500 μm id) and large (LPA, 1.0–1.5 mm id) intrapulmonary arteries dissected out. The arteries were cut open longitudinally and strips of smooth muscle bathed in a dissociation medium (DM) composed of (mM); NaCl 110; KCl 5; KH2PO4 0.5; NaH2PO4 0.5; NaHCO3 10; N-[2-hydroxyethyl]piperazine-N'-[2-ethane-sulfonic acid] (HEPES)10; phenol red 0.03; taurine 10; ethylenediaminetetraacetic acid (EDTA) 0.5; MgCl2 2; glucose 10 and CaCl2 0.16, titrated to pH 7.0 with KOH. After incubation in DM containing 0.6 – 0.8 mg.ml-1 papain, 0.04% BSA and 0.4 mM dithiothreitol at 37°C (15 min for LPA, 10 min for SPA), collagenase (0.6 – 0.8 mg.ml-1; type IA) was added and the tissues incubated for a further 10 (LPA) or 5 (SPA) min. Cells were then dispersed by mild trituration in enzyme-free solution and used within 7 hours.
Electrophysiological recording
Cells were placed in a 50 μl chamber and superfused at room temperature with physiological salt solution (PSS) composed of (mM): NaCl 122; KCl 5; HEPES 10; KH2PO4 0.5; NaH2PO4 0.5; MgCl2 1; glucose 11; CaCl2 1.8; titrated to pH 7.3 with NaOH. Electrophysiological responses of isolated smooth muscle cells were studied in the whole-cell, perforated-patch mode with amphotericin B (150 μg.ml-1) added to a pipette solution of the following composition (mM): KCl 125; MgCl2 4; HEPES 10; ethylene glycol-bis(2-aminoethylether)-N,N,N',N'-tetraacetic acid (EGTA) 0.02, titrated to pH 7.3 with KOH. Pipette resistance was 4–8 MΩ. The cells were voltage-clamped at -50 mV using an Axopatch 200A amplifier (Axon Instruments). Data were recorded and analysed with a personal computer interfaced with a Digidata 1200 A/D converter (Axon Instruments) using Axotape and pClamp (V5) software (Axon Instruments). Current responses to -10 mV hyperpolarizing steps were used to measure cell capacitance.
We have reported previously that 10-4 M ATP, UTP and UDP each evoked pronounced contractions of rat isolated SPA and LPA and that 10-4 M suramin and PPADS produced maximum inhibition of these responses [6]. Therefore, this concentration of these drugs was used here. All were applied to the cells using a gravity-feed perfusion system, for which the time for complete solution exchange was less than 2 s. Only one agonist was applied to each cell. P2Y receptor-mediated contractions develop slowly and take 5–10 min to reach a steady-state plateau, therefore, in most cases the agonists were applied to the cells for 5 min or more.
Electrophysiological analysis
The rat pulmonary artery is a relatively short vessel, with only a thin layer of smooth muscle cells. Enzymatic dissociation produces a lower yield of cells that are smaller and often less robust than those from systemic blood vessels. Consequently, although oscillating inward currents were observed in response to P2Y receptor agonists in the majority of cells studied, quantitative analysis was hampered by the short period of time that many cells could be maintained in the perforated-patch configuration or by the disappearance of the response during the recording. Quantitative analysis was applied only to cells that could be held for 5 min or more and in which the oscillating currents lasted more than 4 min. In these cells the following parameters were measured: a) the peak amplitude of each current (pA), which was normalised against the cell capacitance (pF) to control for variations in cell size; b) the rise time (ms) of the current at each oscillation from baseline holding current to peak; c) the width of each oscillation (ms) at the point where it reached 50% of its peak amplitude. For each, the average value during successive 30 s intervals over a 4 min period was calculated and compared. Finally, the frequency of oscillations (peak.min-1) was measured as the number of transient currents occurring during successive 1 min intervals over a 4 min period.
To investigate the effects of P2 receptor antagonists on the currents, some cells were preincubated with antagonist for 5 min before adding an agonist, but in most cases an agonist was applied for 2 min and then suramin or PPADS were co-applied for a further 2–3 min. The current amplitude and frequency were measured and average values compared for the 1 min periods immediately before and after antagonist addition. The data were compared with control cells where agonist alone was added and the currents measured over the same time course.
Tension recording
Rat SPA and LPA were dissected out as described above, cut into rings 5 mm long and mounted horizontally in 1 ml baths on a pair of intraluminal wires [6]. Tissues were allowed to equilibrate under a resting tension of 0.5 g (SPA) and 1.0 g (LPA) for 60 min at 37°C in PSS. Tension was recorded with Grass FT03 isometric force transducers connected to a MacLab/4e system, using Chart 3.3 software (AD Instruments). Cumulative concentration-response curves to ATP were obtained in rings in the absence of antagonist (control) or in the presence of a single concentration (3 × 10-5, 10-4 or 3 × 10-4 M) of suramin or PPADS. Contractions generally took 1–4 min to reach a plateau and are expressed as a percentage of the contraction induced in the same preparation by 4 × 10-2 M KCl, which was applied by replacement of the PSS solution with PSS in which the KCl concentration was raised by equimolar substitution for NaCl.
Data analysis
Values in the text and figures refer to mean ± S.E.M.. Data were compared by paired and unpaired t-tests, or one-way analysis of variance and Tukey's comparison as appropriate. Differences were considered significant when P < 0.05.
Drugs and solutions
ATP (magnesium salt), UDP (sodium salt), UTP (sodium salt), suramin hexasodium and PPADS tetrasodium (Sigma/RBI, UK) were dissolved in deionised water as 100 mM stock solutions and diluted in PSS before application to the cells.
Results
P2Y receptor agonists induce oscillating inward currents
ATP, UTP and UDP (10-4 M) each evoked inward currents (peak amplitude = 13 – 727 pA) in most SPA (n = 118) and LPA (n = 117) smooth muscle cells held at -50 mV (ATP-91%/88%, UTP-91%/93%, UDP-71%/81%, SPA/LPA respectively). Outward currents or no response were evoked in the remaining cells, which were not studied further. In most cells the inward currents activated in an oscillating manner (Figure 1). The first current was usually the largest and subsequent currents decreased in amplitude, with a variable time-course. The peak of the first current appeared to be larger for ATP than UTP or UDP (Figure 2), but this was significant only in the SPA (P < 0.05). For each agonist there was no significant difference in the amplitude of the first response between the small and large vessels. The initial current often had a "W-shaped" profile (Figure 1), but in most cells (85%) the biphasic profile disappeared by the second, third or fourth oscillation, such that subsequent currents were monophasic. ATP, UTP and UDP each evoked this profile of responses in a similar proportion of cells.
Figure 1 Oscillating currents induced by P2Y receptor agonists. (a) ATP, (b) UTP and (c) UDP (all 10-4 M), added as indicated by the horizontal bars, evoked oscillating inward currents in smooth muscle cells isolated from SPA (a & c) and LPA (b) and voltage-clamped at -50 mV. The insets show W-shape currents induced by (a) ATP and (b) UTP. The arrow on the left-hand side of each trace indicates zero holding current.
Figure 2 Amplitude of the first inward current induced by P2Y receptor agonists. The mean ± s.e.mean of the peak amplitude of the initial inward currents evoked by 10-4 M ATP, UTP and UDP in SPA and LPA isolated smooth muscle cells, voltage-clamped at -50 mV, are shown. The number of cells for each is indicated in parentheses. *P < 0.05 for responses to ATP versus UTP and UDP in SPA.
Quantitative analysis of oscillating currents
The decrease in the amplitude of the oscillating currents during agonist application complicated the quantification of the effects of P2Y antagonists, so it was necessary to first quantify the time-course of the currents. In order to be able to study the effects of both an agonist and antagonist on the same cell, the analysis was limited to a subpopulation of cells that were maintained under voltage-clamp for more than 5 min and in which the oscillating currents lasted for 4 min or more.
In both SPA (Figure 3a) and LPA (not shown), the amplitude of the oscillating currents tended to decrease over successive 30 s intervals, particularly within the first 2 min, and this was significant for ATP (P < 0.05), UTP (P < 0.01) and UDP (P < 0.05) in the SPA. The oscillations induced by ATP, UTP and UDP had similar frequencies (mean≈6–11.min-1), both in SPA (Figure 3b) and LPA (not shown) and showed no significant change over 4 min, apart from a small increase in the LPA between the first and second min after UTP application (P < 0.05). The currents evoked by ATP in the first 30 s had a rise time of 1.7 ± 0.3 s and width at 50% peak of 2.9 ± 0.4 s (n = 4) in SPA and 1.5 ± 0.6 s and 2.2 ± 0.3 s (n = 5) respectively in LPA. The rise time and width at 50% peak then decreased significantly (P < 0.01) over the next 60–90 s to a steady state of around 0.8 s in both SPA and LPA. The width at 50% peak of currents evoked by UTP in the first 30 s was 2.1 ± 0.5 s (n = 5) in SPA and 2.1 ± 0.5 s (n = 5) in LPA and both decreased significantly (P < 0.01) over the next 60–90 s to a steady state of also about 0.8 s. In contrast, currents evoked by UTP and UDP in SPA and LPA showed no significant change in rise time and those to UDP in SPA and LPA showed no significant change in width at 50% peak, all having a steady-state value of about 0.8 s.
Figure 3 The amplitude and frequency of oscillating currents induced by P2Y receptor agonists. (a) The mean amplitude of oscillating currents induced by ATP, UTP and UDP (10-4 M) was measured over successive 30 s intervals for 4 min in smooth muscle cells isolated from SPA and voltage-clamped at -50 mV. 0.5 = the first 30 s of agonist application, 1 = 30 – 60 s, and so on. (b) The frequency of the oscillations in the same cells was measured over successive 1 min intervals. 1 = the first min of agonist application, 2 = the second min, and so on. Vertical bars represent s.e.mean. The number of cells for each agonist is shown in parentheses. * P < 0.05 for amplitude of peak current at 30 s versus that at 2:30, 3:00, 3:30 and 4:00 min; ** P < 0.01 for 30 s versus 3:30 and 4:00 min.
Effects of P2 receptor antagonists
Having quantified the time-course of the agonist-induced oscillating currents, we then determined the effects of the P2Y antagonists suramin and PPADS (10-4 M). In most cells currents were initiated by an agonist, the antagonist was then coapplied and the currents compared for 1 min before and after antagonist addition. To take into account the decline in current amplitude normally seen over this time-course (~20–40%), the % decrease in amplitude over the 2 analysis periods was calculated and compared with that in control cells where agonist alone was added.
Suramin rapidly and reversibly abolished the currents evoked by ATP in SPA (Figure 4a, 5) and LPA (not shown). Additionally, ATP did not elicit currents if cells were preincubated with suramin for 5 min (n = 2, not shown). In contrast, PPADS (10-4 M) had no effect on the amplitude or frequency of the ATP-induced currents in SPA (Figure 4b, 5) or LPA (not shown). The rise-time and width at 50% peak were also unaffected (not shown). PPADS was also ineffective if applied for 5 min prior to ATP (n = 2). Neither suramin nor PPADS had any effect on the amplitude or frequency of the oscillating currents elicited by UTP or UDP in either SPA or LPA (Figure 4c, 5). The rise-time and width at 50% peak were also unaffected (not shown). PPADS (n = 12) and suramin (n = 7) (Figure 4d) were also ineffective if applied for 5 min before UTP.
Figure 4 The effects of P2 receptor antagonists on oscillating currents. The P2 receptor antagonists suramin and PPADS (10-4 M) were applied either 2 min after oscillations were induced by continuous application of 10-4 M (a, b) ATP or (c) UDP or (d) 5 min before application of 10-4 M UTP to SPA (a, b) or LPA (c, d) dissociated smooth muscle cells voltage-clamped at -50 mV. The horizontal bars indicate agonist and antagonist applications. The arrow on the left-hand side of each trace indicates zero holding current.
Figure 5 The effects of P2 receptor antagonists on oscillating current amplitude and frequency. The effects of suramin and PPADS (10-4 M) on (a) the amplitude and (b) the frequency of oscillating inward currents induced by ATP, UTP and UDP (10-4 M) in SPA dissociated smooth muscle cells, voltage-clamped at -50 mV, are shown. The agonist was applied for 2 min and then suramin or PPADS was coapplied for a further 2–3 min. The current amplitude was measured for 1 min immediately before and after antagonist addition and average values calculated. The % decrease in amplitude was then calculated as the difference in the 2 average values. The control data were obtained over the same time-course in cells where agonist alone was added. The average frequency of oscillations was also measured for 1 min immediately before and after antagonist addition and compared directly. Vertical lines show s.e.mean. The number of cells is shown in parentheses.
Effects of suramin and PPADS on contractions evoked by ATP
We have reported previously the effects of suramin and PPADS on contractions of rat pulmonary arteries induced by UTP and UDP [6]. Since suramin abolished current oscillations induced by ATP, but not UTP or UDP, we investigated if nucleotide-induced contractions showed the same differential sensitivity. We report that ATP (10-7 - 3 × 10-4 M) evoked concentration-dependent contractions of the rat SPA (Figure 6). Suramin (3 × 10-5 - 10-4 M) caused a progressive rightward shift of the ATP concentration-response curve and the responses were abolished by the highest concentration of the antagonist (Figure 6a). PPADS (3 × 10-5 M) also shifted the ATP concentration-response curve to the right, but increasing its concentration to 10-4 M and 3 × 10-4 M) produced no further inhibition (Figure 6b). In rat LPA ATP induced contractions only at 10-4 M and above [6] and these small contractions were abolished by 3 × 10-4 M suramin or PPADS (not shown).
Figure 6 Effects of suramin and PPADS on contractions evoked by ATP. The effects of (a) suramin and (b) PPADS on contractions of rat isolated SPA induced by ATP are shown. Cumulative concentration-response curves to ATP (10-7 - 3 × 10-4 M) were obtained in rings in the absence of antagonist (control) or in the presence of 3 × 10-5, 10-4 or 3 × 10-4 M of antagonist. Contractions are expressed as a percentage of the contraction induced by 4 × 10-2 M KCl. Vertical lines show s.e.mean. n = 5 for suramin and 4 for PPADS.
Discussion
The present study shows that ATP, UTP and UDP induce oscillating inward currents with similar amplitudes and frequencies in smooth muscle cells of rat pulmonary arteries. Such Cl- currents have been reported previously in these cells and are dependent upon nucleotide-evoked release of Ca2+ from sarcoplasmic reticulum stores [4,5,26]. The P2Y1, P2Y2, P2Y4, P2Y6 and P2Y11 receptors all couple to the Gq/11 G proteins, leading to the release of IP3-sensitive Ca2+ stores [14] and so could, if present in the tissue, mediate activation of ICl,Ca. UTP and UDP both acted at a site that was insensitive to the antagonists suramin and PPADS, which may be the P2Y6 receptor or perhaps a novel receptor. ATP clearly acted via a different subtype, which most resembles the P2Y11 receptor. There were few differences apparent between the SPA and LPA, consistent with our previous conclusion from contractile studies that there is no regional variation in the P2Y subtype distribution. Thus, multiple subtypes of P2Y receptor are widely expressed in pulmonary artery smooth muscle and are likely to play a role in nucleotide-induced vasoconstriction.
P2Y receptors in SPA and LPA
In these experiments, the currents evoked by ATP in cells from the SPA and LPA were abolished by suramin, but unaffected by PPADS. ATP is an agonist at the P2Y1,2,4 & 11 receptors [15] and the P2Y1 and P2Y4 subtypes can be ruled out, because PPADS antagonises both of these [4]. The P2Y2 receptor can also be discounted as responses to the P2Y2 agonist UTP were not inhibited by suramin. The remaining P2Y11 receptor is antagonised by suramin, but not PPADS [27], consistent with a role in mediating the ATP-induced ICl,Ca. This is problematic, however, as the rat P2Y11 receptor has yet to be cloned. Indeed, it is not clear that it is present in the rodent genome, although a previous pharmacological study is also consistent with its expression in rat blood vessels [28]. Further studies are required to address this issue. Note that ATP has been reported to be an agonist at the P2Y12 receptor [29] and that a contractile P2Y12 receptor was recently reported in human blood vessels [30]. However, the agonist action of ATP has been questioned [31] and the P2Y12 receptor couples to Gi and so is unlikely to induce the release of Ca2+ stores needed to activate ICl,Ca in rat pulmonary arteries [5].
UTP and UDP also activated ICl,Ca, but the responses were unaffected by suramin or PPADS. This is consistent with the lack of effect of the antagonists on UTP- and UDP-induced vasoconstriction in the rat perfused pulmonary vascular bed [10] and with the antagonist-insensitive component of UTP and UDP contraction of isolated pulmonary artery [6], but contrasts with the abolition by suramin of UTP-elicited oscillating currents seen previously in single cells [5,12]. The reason for these differences in suramin activity is not clear. Which receptor(s) mediated the actions of UTP and UDP in the present study is also unclear. UTP is an agonist at the P2Y2,4 &6 subtypes, whilst UDP is only active at the P2Y6 receptor [2,15]. Detailed studies show clearly that the P2Y2 receptor is antagonised by suramin and the P2Y4 receptor by PPADS (2,20). So, these subtypes do not mediate the effects of UTP (or UDP) seen here.
If the P2Y2 and P2Y4 receptors are ruled out, then the P2Y6 receptor is the prime candidate for the site of action of UTP and UDP. Indeed, its mRNA is present in rat pulmonary artery smooth muscle and it has been proposed to underlie the UDP-induced ICl,Ca [12]. However, the effects of suramin and PPADS at this site are not well characterised. In the only study on the cloned rat P2Y6 receptor, 10-4 M suramin (the same concentration used in the present study) depressed the agonist response by 20% [32]. Similar inhibition (27%) was seen at the cloned human P2Y6 receptor [18]. PPADS was not tested at the rat receptor, but at 10-4 M it inhibited the response to UDP at the human subtype by 69%. This pronounced effect of PPADS is inconsistent with the P2Y6 receptor being the receptor through which UTP and UDP activated ICl,Ca in the present study. Further characterisation of the effects of suramin and PPADS at the recombinant rat P2Y6 receptor is, however, needed to substantiate this conclusion.
If the P2Y2, P2Y4 and P2Y6 receptors are not the site(s) of action of UTP and UDP, then what is? One possibility is that UTP and UDP activated ICl,Ca in rat SPA and LPA smooth muscle via a novel, as yet uncloned P2Y receptor or another, non-P2Y receptor. For example, UDP has been proposed to interact with cysteinyl leukotriene receptors in human mast cells [33,34]. Alternatively, one of the known P2Y receptors may interact with another P2Y subtype, or with a non-P2Y receptor, to form a dimer with novel pharmacological properties. Indeed, the P2Y1 and P2Y2 receptors both appear to form dimers with the A1 adenosine receptor [35]. Further studies are needed to investigate these possibilities.
P2X receptors in SPA and LPA
In this study, the currents evoked by ATP, UTP and UDP had similar time-courses, as measured by rise time and width at 50% peak. This may appear surprising as ATP, but not UTP or UDP, is also an agonist at the P2X1 receptor and so might be expected to activate an initial, rapid, transient inward current, in addition to the slower, longer lasting, P2Y-mediated oscillations, as has been reported previously in rat pulmonary artery smooth muscle cells [4,5]. The apparent absence of the transient response may be due to the relatively slow speed of application of ATP used here. The P2X1 receptor desensitizes rapidly and slow agonist administration elicits much slower and smaller currents in vascular smooth muscle cells [36]. Although this would be disadvantageous if studying P2X receptors, by minimizing the P2X response it is in fact an advantage when P2Y receptors are under study. The initial current evoked by ATP in SPA and LPA may well be a mixture of P2X1 and P2Y receptor-induced responses, which would explain the larger amplitude of the initial ATP-induced current, compared with UTP and UDP. Nevertheless, any P2X1 component appears to play a relatively minor role and would not contribute to the sustained phase of oscillations.
Contribution of P2Y subtypes to contractions
Although the receptors that mediate activation of ICl,Ca by nucleotides in the rat pulmonary artery have not been identified unequivocally, we can still consider their role in vasoconstriction of the rat pulmonary vascular bed [10] and isolated arteries [6,12,13]. In this study, contractions of the SPA elicited by ATP were abolished by suramin, but only partially inhibited by PPADS. The PPADS-resistant contractions likely reflect release of Ca2+ stores, causing the ICl,Ca recorded here. They may also involve Ca2+ influx via L-type Ca2+ channels, opened by depolarisation due to ICl,Ca. Further experiments using channel blockers are needed to confirm this. The P2X1 receptor in SPA smooth muscle [6] is most likely to underlie the remaining suramin- and PPADS-sensitive component. Interestingly, contractions of the rat LPA were abolished by both suramin and PPADS, suggesting that only one receptor, probably the P2X1 receptor, mediates the contractile actions of ATP here. This is consistent with the much lower contractile potency of ATP in LPA [6], but it suggests that the similar suramin- and PPADS-insensitive ICl,Ca observed in response to ATP in LPA and SPA may serve different functions. In our previous study [6] contractions of rat SPA induced by UTP and UDP were not inhibited by PPADS and were only partially suppressed by suramin. These antagonist-resistant contractions again likely reflect release of Ca2+ stores and activation of ICl,Ca. The identity of the suramin-sensitive receptors remains to be determined.
Advantages of the patch clamp technique
This study shows that recording ion currents in single cells can be useful in characterising the receptors expressed in tissues where multiple subtypes are present. A particular problem with P2Y receptors is ecto-nucleotidases, which are inhibited by PPADS in smooth muscle [37] and other tissues [38,39]. Recording from rapidly perfused, single cells minimises the problems created by extracellular metabolism in whole tissues, which may explain why PPADS potentiated contractions to UTP and UDP in the intact artery [6], but had no effect on activation of ICl,Ca in single cells. Such studies also allow the regional variation in ion channel expression to be studied. Interestingly, we recorded ICl,Ca in a similar proportion of rat SPA and LPA smooth muscle cells, whereas in rabbits it is more predominant in smaller pulmonary arteries [25]. Limitations of the patch clamp technique encountered here were short recording times, wide variation in current amplitude between cells and a decline in the amplitude of ICl,Ca over the recording period, all of which hampered quantitative analysis of antagonist action. It is not clear why rundown occurred, as loss of diffusible cytosolic factors into the recording pipette should have been minimised with the perforated-patch technique. Similar rundown was seen in previous patch clamp studies in these cells [4,5] and with ATP- and UTP-induced oscillations in cytosolic [Ca2+] [26]. Thus, the decline in ICl,Ca may in fact reflect a physiological mechanism of signalling whereby the P2Y receptors become desensitised and/or intracellular stores release progressively less Ca2+ during maintained activation of P2Y receptors.
Conclusion
The results of the present study indicate the presence of at least two different subtypes of P2Y receptors mediating oscillating inward currents in rat SPA and LPA smooth muscle cells. ATP acts via a suramin-sensitive, PPADS-insensitive site, which most resembles the P2Y11 receptor. The site of action of UTP and UDP is less clear. Its pharmacology is inconsistent with our present understanding of P2Y2,4 & 6 receptors, so a novel receptor or receptor complex may be involved. These different P2Y receptors are likely to play a significant role in nucleotide-induced pulmonary vasoconstriction as ATP, UTP and UDP each induce contractions of the rat pulmonary artery with matching pharmacological profiles.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
KC was involved in the planning of the experiments described and carried them out. She also analysed the data, drafted the manuscript and was involved in its revision. AMG was involved in the planning of the experiments and revision of the manuscript. CK was involved in the planning of the experiments, the analysis of the data and revision of the manuscript. All authors read and approved the final manuscript.
Acknowledgements
This work was supported by funding from the Faculty of Medical Science, Naresuan University, Thailand and The Royal Society of Edinburgh.
==== Refs
Burnstock G Kennedy C A dual function for adenosine triphosphate in the regulation of vascular tone: excitatory cotransmitter with noradrenaline from perivascular nerves and locally released inhibitory intravascular agent Circ Res 1986 58 319 330 3013455
Boarder MR Hourani MO The regulation of vascular function by P2 receptors: multiple sites and multiple receptors Trends Pharmacol Sci 1998 19 99 107 9584626 10.1016/S0165-6147(98)01170-5
Ralevic V Burnstock G Receptors for purines and pyrimidines Pharmacol Rev 1998 50 413 491 9755289
Bakhramov A Hartley SA Salter KJ Kozlowski RZ Contractile agonists preferentially activate Cl- over K+ currents in arterial myocytes Biochem Biophys Res Comm 1996 227 168 175 8858120 10.1006/bbrc.1996.1484
Hartley SA Kozlowski RZ Electrophysiological consequences of purinergic receptor stimulation in isolated rat pulmonary arterial myocytes Circ Res 1997 80 170 178 9012739
Chootip K Ness KF Wang Y Gurney AM Kennedy C Regional variation in P2 receptor expression in the rat pulmonary arterial circulation Br J Pharmacol 2002 137 637 646 12381677 10.1038/sj.bjp.0704915
Mc Cormack DG Barnes PJ Evans TW Purinoceptors in the pulmonary circulation of the rat and their role in hypoxic vasoconstriction Br J Pharmacol 1989 98 367 372 2819324
Barnes PJ Liu SF Regulation of pulmonary vascular tone Pharmacol Rev 1995 47 87 131 7784481
Hasséssian H Burnstock G Interacting roles of nitric oxide and ATP in the pulmonary circulation of the rat Br J Pharmacol 1995 114 846 850 7773545
Rubino A Burnstock G Evidence for a P2 purinoceptor mediator mediation vasoconstriction by UTP, ATP and related nucleotides in the isolated pulmonary vascular bed of the rat Br J Pharmacol 1996 118 1415 1420 8832066
Liu SF Mc Cormack DG Evans TW Barnes PJ Characterization and distribution of P2-purinoceptor subtypes in rat pulmonary vessels J Pharmacol Exp Ther 1989 251 1204 1210 2600811
Hartley SA Kato K Salter KJ Kozlowski RZ Functional evidence for a novel suramin-insensitive pyrimidine receptor in rat small pulmonary arteries Circ Res 1998 83 940 946 9797343
Rubino A Ziabary L Burnstock G Regulation of vascular tone by UTP and UDP in isolated rat intrapulmonary arteries Eur J Pharmacol 1999 370 139 143 10323262 10.1016/S0014-2999(99)00150-8
Boarder MR Webb TE Abbracchio MP, Williams M P2Y receptors: structure and function Handbook of Experimental Pharmacology 151/1; Purinergic and Pyrimidinergic Signalling I: Molecular, Nervous and Urogenitary System Function 2001 Springer 65 88
Abbracchio MP Boeynaems JM Barnard EA Boyer JL Kennedy C Miras-Portugal M King BF Gachet C Jacobson KA Weisman GA Burnstock G Characterization of the UDP-glucose receptor (re-named here the P2Y14 receptor) adds diversity to the P2Y receptor family Trends Pharmacol Sci 2003 24 52 55 12559763 10.1016/S0165-6147(02)00038-X
Nicholas RA Watt WC Lazarowski ER Li Q Harden TK Uridine nucleotide selectivity of three phospholipase C-activating P2 receptors: Identification of a UDP-selective, a UTP-selective, and an ATP- and UTP-specific receptor Mol Pharmacol 1996 50 224 229 8700127
Filippov AK Webb TE Barnard EA Brown DA Dual coupling of heterologously-expressed rat P2Y6 nucleotide receptors to N-type Ca2+ and M-type K+ currents in rat sympathetic neurones Br J Pharmacol 1999 126 1009 1017 10193782 10.1038/sj.bjp.0702356
Robaye B Boeynaems JM Communi D Slow desensitization of the human P2Y6 receptor Eur J Pharmacol 1997 329 231 236 9226417
Bogdanov YD Wildman SS Clements MP King BF Burnstock G Molecular cloning and characterisation of rat P2Y4 nucleotide receptor Br J Pharmacol 1988 124 428 439 9647463 10.1038/sj.bjp.0701880
Suarez-Huerta N Pouillon V Boeynaems JM Robaye B Molecular cloning and characterization of the mouse P2Y4 nucleotide receptor Eur J Pharmacol 2001 416 197 202 11290369 10.1016/S0014-2999(01)00875-5
Kennedy C Leff P How should P2X-purinoceptors be characterised pharmacologically? Trends Pharmacol Sci 1995 16 168 174 7624972 10.1016/S0165-6147(00)89010-0
Lazarowski ER Boucher RC Harden TK Constitutive release of ATP and evidence for major contribution of ecto-nucleotide pyrophosphatase and nucleoside diphosphokinase to extracellular nucleotide concentrations J Biol Chem 2000 275 31061 31068 10913128 10.1074/jbc.M003255200
Lazarowski ER Boucher RC Harden TK Mechanisms of release of nucleotides and integration of their actions as P2X- and P2Y-receptor activating molecules Mol Pharmacol 2003 64 785 795 14500734 10.1124/mol.64.4.785
Zimmermann H Extracellular metabolism of ATP and other nucleotides Naunyn-Schmied Arch Pharmacol 2000 362 299 309 10.1007/s002100000309
Smani T Iwabuchi S López-Barneo J Ureña J Differential segmental activation of Ca2+-dependent Cl- and K+ channels in pulmonary arterial myocytes Cell Calcium 2001 29 369 377 11352503 10.1054/ceca.2001.0199
Guibert C Pacaud P Loirand G Effect of extracellular ATP on cytosolic Ca2+ concentration in rat pulmonary artery myocytes Am J Physiol 1996 271 L450 L458 8843794
Communi D Robaye B Boeynaems JM Pharmacological characterization of the human P2Y11 receptor Br J Pharmacol 1999 128 1199 1206 10578132 10.1038/sj.bjp.0702909
Gitlin JM Zanesco A Stanford SJ Evans TW Anning PB Mitchell JA The second phase of ATP mediated relaxation is mediated through P2Y11 receptors Br J Pharmacol 2002 135 210P
Takasaki J Kamohara M Saito T Matsumoto M Matsumoto SI Ohishi T Soga T Matsushime H Furuichi K Molecular cloning of the platelet P2TAC ADP receptor: pharmacological comparison with another ADP receptor, the P2Y1 receptor Mol Pharmacol 2001 60 432 439 11502873
Wihlborg AK Wang L Braun OO Eyjolfsson A Gustafsson R Gudbjartsson T Erlinge D ADP receptor P2Y12 is expressed in vascular smooth muscle cells and stimulates contraction in human blood vessels Arterioscler Thromb Vasc Biol 2005 24 1 7
Kauffenstein G Hechler B Cazenave JP Gachet C Adenosine triphosphate nucleotides are antagonists at the P2Y12 receptor J Thromb Haem 2004 2 980 1988
Chang K Hanaoka K Kumada M Takuwa Y Molecular cloning and functional analysis of a novel P2 nucleotide receptor J Biol Chem 1995 270 26152 26158 7592819 10.1074/jbc.270.44.26152
Mellor EA Maekawa A Austen KF Boyce JA Cysteinyl leukotriene receptor 1 is also a pyrimidergic receptor and is expressed by human mast cells Proc Natl Acad Sci USA 2001 98 7964 7969 11438743 10.1073/pnas.141221498
Mellor EA Austen KF Boyce JA Cysteinyl leukotrienes and uridine diphosphate induce cytokine generation by human mast cells through an interleukin 4-regulated pathway that is inhibited by leukotrienne receptor antagonists J Exp Med 2002 195 583 592 11877481 10.1084/jem.20020044
Yoshioka K Saitoh O Nakata H Heteromeric association creates a P2Y-like adenosine receptor Proc Natl Acad Sci USA 2001 98 7617 7622 11390975 10.1073/pnas.121587098
Evans RJ Kennedy C Characterisation of P2-purinoceptors in the smooth muscle of the rat tail artery: a comparison between contractile and electrophysiological responses Br J Pharmacol 1994 113 853 860 7858877
Khakh BS Michel AD Humphrey PPA Inhibition of ectoATPase and Ca-ATPase in rat vas deferens by P2-purinoceptor antagonists Br J Pharmacol 1995 115 2P
Chen BC Lee CM Lin WW Inhibition of ecto-ATPase by PPADS, suramin and reactive blue in endothelial cells, C6 glioma cells and RAW 264.7 macrophages Br J Pharmacol 1996 119 1628 1634 8982511
Grobben B Claes P Roymans D Esmans EL Van Onckelen H Slegers H Ecto-nucleotide pyrophosphatase modulates the purinoceptor-mediated signal transduction and is inhibited by purinoceptor antagonists Br J Pharmacol 2000 130 139 145 10781009 10.1038/sj.bjp.0703289
|
16250909
|
PMC1282591
|
CC BY
|
2021-01-04 16:36:25
|
no
|
Respir Res. 2005 Oct 26; 6(1):124
|
utf-8
|
Respir Res
| 2,005 |
10.1186/1465-9921-6-124
|
oa_comm
|
==== Front
Respir ResRespiratory Research1465-99211465-993XBioMed Central London 1465-9921-6-1261625577510.1186/1465-9921-6-126ResearchThe role of surfactant protein D in the colonisation of the respiratory tract and onset of bacteraemia during pneumococcal pneumonia Jounblat R [email protected] H [email protected] P [email protected] S [email protected] PW [email protected] A [email protected] Department of Infection Immunity and Inflammation, University of Leicester, Leicester, LE1 9HN, UK2 MRC Immunochemistry Unit, University of Oxford, South Parks Road, Oxford, OX1 3QU, UK3 Institute of Biomedical and Clinical Sciences, Peninsula Medical School, Exeter, EX1 2LU, UK4 Cardiovascular Research Institute and Department of Paediatrics, University of California, San Francisco, San Francisco, California, USA2005 28 10 2005 6 1 126 126 11 7 2005 28 10 2005 Copyright © 2005 Jounblat et al; licensee BioMed Central Ltd.2005Jounblat et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
We have shown previously that surfactant protein D (SP-D) binds and agglutinates Streptococcus pneumoniae in vitro. In this study, the role of SP-D in innate immunity against S. pneumoniae was investigated in vivo, by comparing the outcome of intranasal infection in surfactant protein D deficient (SP-D-/-) to wildtype mice (SP-D+/+). Deficiency of SP-D was associated with enhanced colonisation and infection of the upper and lower respiratory tract and earlier onset and longer persistence of bacteraemia. Recruitment of neutrophils to inflammatory sites in the lung was similar in both strains mice in the first 24 hrs post-infection, but different by 48 hrs. T cell influx was greatly enhanced in SP-D-/- mice as compared to SP-D+/+ mice. Our data provides evidence that SP-D has a significant role to play in the clearance of pneumococci during the early stages of infection in both pulmonary sites and blood.
Streptococcus pneumoniaesurfactant protein Drespiratory tract
==== Body
Introduction
Streptococcus pneumoniae is a major human pathogen responsible for respiratory tract infections, septicaemia and meningitis. The pneumococcus is particularly well adapted to colonising the mucosal surfaces of the nasopharynx and the combination of bacterial virulence factors and the manipulation of host tissue components allow the pneumococcus to spread from the nasopharynx to sterile regions of the lower respiratory tract, leading to infections such as pneumonia. In the early stages after infection, natural pulmonary defence mechanisms are required for efficient clearance of the pneumococcus. Recent studies have drawn attention to the important role of lung surfactant protein D (SP-D) as the first line of defence in natural innate immunity to microbial invasion of the respiratory tract, involved in the binding, aggregation, and phagocytic uptake of invading micro-organisms [1-4]. In addition, SP-D has also been shown to be involved in binding to apoptotic polymorphonuclear leukocytes and alveolar macrophages to enhance their clearance by healthy resident macrophages [5].
SP-D, is a member of the collectin family that also includes mannose binding lectin (MBL), conglutinin, collectin-43 and surfactant protein A (SP-A). It is predominantly found in the respiratory tract, but is also detected at other non-pulmonary mucosal surfaces such as the salivary and lachrymal gland, ovary, uterus, oesophagus, stomach, testes, thyroid, heart and kidney [4,6,7]. In the lung, SP-D is secreted by alveolar type II cells and by non-ciliated Clara cells as dodecamers consisting of four collagenous trimers cross-linked by disulphide bonds, to create a cruciform structure. Each trimer of the molecule consists of three polypeptide chains and each subunit consists of four domains: a short amino acid terminal end, a collagen-like region followed by a short α-helical region and a C-type carbohydrate recognition domain (CRD) responsible for its lectin activity [1,2,8,9].
A number of pulmonary pathogens, including Streptococcus pneumoniae, have been reported to be agglutinated by lung surfactant protein D in vitro [10-13]. In one such study using SP-D knockout mice (SP-D-/-), the in vivo requirement for SP-D in the early pulmonary clearance and modulation of the inflammatory response to bacterial pathogens was shown. Although increased inflammation, oxidant production and decreased macrophage phagocytosis were associated with SP-D deficiency in the lungs of mice, killing of Gram-negative (Haemophilus influenzae) and Gram-positive (group B streptococcus) bacteria was unaltered [14]. In another study, a decrease in viral clearance and an increase in production of inflammatory cytokines were detected in response to viral challenge in SP-D-deficient mice when compared to control mice [15]. Furthermore, treatment of wild-type mice with native full length SP-D or recombinant SP-D substantially increased their survival rate in mice challenged intranasally with Aspergillus fumigatus spores [16] and recombinant SP-D promoted the clearance of fungal spores from the mouse lung (Howard Clark et al., unpublished).
Another study reported that highly multimerised SP-D molecules bound to strains of serotype 4, 19 and 23 S. pneumoniae, causing their agglutination and enhancing their uptake by neutrophils [17]. More recently, we showed that recombinant human SP-D, expressed in Escherichia coli, consisting of the head and neck regions of the native molecule, bound to all strains of S. pneumoniae that were tested, but the extent of binding varied between strains. Full-length native SP-D aggregated pneumococci in a calcium-dependent manner in vitro, but the aggregation of pneumococci varied not only between strains of the same multilocus sequence type (but different serotypes), but also between strains of the same serotype. Neither recombinant truncated SP-D nor native full-length SP-D enhanced killing of pneumococci by human neutrophils in the absence of serum however [11].
Given the above findings, we hypothesise that SP-D has an important role to play in the innate immune defence of the upper and lower respiratory tract against pneumococcal infection in vivo, by promoting the agglutination and subsequent clearance of S. pneumoniae. This would prevent the colonisation of the nasopharynx and subsequently limit the spread of pneumococci from the upper to the lower respiratory tract by enhancing clearance via the mucocilliary system, thus allowing enough time for other components of both the innate and adaptive immune system to come into play. In the present study we assessed the in vivo contribution of SP-D to host defence by intranasally infecting SP-D-deficient and sufficient mice with S. pneumoniae. Bacterial growth kinetics in the nasopharynx, trachea, lungs and blood, development of lung pathology and host inflammatory leukocyte infiltration into lungs was compared in both strains of mice following infection.
Methods
Source of mice
Wild-type control C57BL/6 mice were obtained from Harlan Olac (Bicester, UK) and SP-D genes were ablated by gene targeting of embryonic stem cells, backcrossed 10 generations into the C57BL/6 genetic background, and maintained at the animal house of the Department of Biochemistry, Oxford University under barrier facilities [5,18]. All mice were at least 8 weeks old at use and did not have detectable levels of anti-type 2 antibodies. All experimental protocols were approved by appropriate U.K. Home Office licensing authorities and by the University of Leicester Ethical Committee.
Bacteria
Streptococcus pneumoniae serotype 2, strain D39 was obtained from the National Collection of Type Cultures, London, UK (NCTC 7466). Bacteria were identified as pneumococci prior to experiments by Gram stain, catalase test, α-haemolysis on blood agar plates and by optochin sensitivity. To obtain virulent pneumococci, bacteria were cultured and passaged through mice as described previously [19] and subsequently recovered and stored at -80°C. When required, suspensions were thawed at room temperature and bacteria harvested by centrifugation before re-suspension in sterile phosphate buffered saline (PBS).
Intranasal challenge of mice with S. pneumoniae
As previously described, [19] mice were infected intranasally with 1 × 106 CFU S. pneumoniae. At pre-chosen intervals following infection, groups of mice were deeply anaesthetised with 5% (v/v) fluothane (Astra-Zeneca, Macclesfield, UK) and blood was collected by cardiac puncture. Mice were killed by cervical dislocation, and the lungs, trachea and nasopharynx were removed separately into 10 ml of sterile PBS, weighed and then homogenised in a hand held homogeniser (Fischer Scientific, UK). Viable counts in homogenates and blood were determined by serial dilution in sterile PBS and plating onto blood agar plates as previously described [19].
Pathology
At intervals following infection, lungs were excised, embedded in Tissue-Tec OCT (Sakura), and frozen in liquid nitrogen with an isopentane heat buffer to prevent snap freezing and tissue damage. Samples were stored at -80°C. Sections (10 μm) were taken at -18°C on a Bright cryostat and then allowed to dry at room temperature. Sections from throughout the lung were taken with at least thirty sections per lung being analysed. Following acetone fixation, the sections were stained with haematoxylin and eosin and fixed with DPX mountant (BDH) for permanent storage [19]. Lung pathology was scored blind on the following criteria; cellular infiltration around bronchioles, perivascular and peribronchial areas, hypertrophy of bronchiole walls, and oedema.
Immunohistochemistry
As described previously [19], leukocyte recruitment into lung tissue was analysed by an alkaline phosphatase anti-alkaline phosphatase (APAAP) antibody staining method. Rat anti-mouse monoclonal antibodies to T cells (anti-CD3), B cells (anti-CD19), macrophages (anti-F4/80) and neutrophils (anti-Gr-1) (Serotec, Oxford, UK) and secondary rabbit anti-rat antibody (Dako, Denmark) and rat APAAP antibody were used as previously described on infected lung tissue sections. Sections from throughout the lung were taken with at least twenty sections per lung being analysed. Tissue sections (approximately twenty sections from each lung at chosen time points) were used for each antibody to be tested, along with 3 sections for negative controls which consisted of using an isotype matched control antibody; excluding the primary antibody (or the secondary enzyme conjugated antibody); or not incubating with the substrate-chromogen solution. Finally, the sections were washed and counterstained briefly with haematoxylin and mounted in aqueous mounting medium (Aquamount, DAKO). Once stained, each section was quantified double blind by two observers (RJ and AK). Positively stained cells within the vicinity of inflamed bronchioles were enumerated within the 1 mm2 area of a counting grid. Twenty individual grids per each tissue section were quantified, making a total of 400 grids counted per lung per each time point (20 tissue sections in total per each antibody tested). A total of four lungs per time point were analysed.
Statistical analysis
Comparisons of bacterial loads between mouse strains or treatments were made with unpaired Students t tests. Statistical significance was considered at P values <0.05.
Results
The role of SP-D in upper and lower respiratory tract pneumococcal colonisation Nasopharynx
S. pneumoniae successfully colonised the nasopharynx of SP-D-/- mice, but were cleared from SP-D+/+ mice. Pneumococcal numbers in the nasopharynx of SP-D-/- mice remained unchanged over the 48 hr period post infection (Fig. 1) whereas pneumococcal numbers were significantly reduced in SP-D+/+ mice by 48 hrs post infection as compared to SP-D-/- mice (P < 0.01). SP-D+/+ mice eventually cleared the pneumococci in their nasopharynx by 72 hrs (by which time-point the experiment was ended) while SP-D-/- mice remained colonised at the same rate at this time-point (data not shown).
Figure 1 Time course of the change in numbers of S. pneumoniae in the nasopharynx (figure 1), trachea (figure 2), lungs (figure 3) and blood (figure 4) of SP-D-/- (◆) and SP-D+/+ (■) mice infected intranasally with 106 CFU (n = 10 mice at each time point, error bars indicate SEM). * denotes P < 0.01, ** denotes P < 0.05 for SP-D-/- when compared to wildtype at equivalent time point.
Trachea
Differences between SP-D+/+ and SP-D-/- mice were also apparent in the colonisation of the trachea by pneumococci. Numbers of pneumococci remained constant over the 48 hr period post-infection period in SP-D-/- mice, whereas bacteria were cleared from the trachea of SP-D+/+ mice by 48 hrs post-infection (P < 0.01, compared to SP-D-/- mice) (Fig. 2).
Figure 2 Time course of the change in numbers of S. pneumoniae in the nasopharynx (figure 1), trachea (figure 2), lungs (figure 3) and blood (figure 4) of SP-D-/- (◆) and SP-D+/+ (■) mice infected intranasally with 106 CFU (n = 10 mice at each time point, error bars indicate SEM). * denotes P < 0.01, ** denotes P < 0.05 for SP-D-/- when compared to wildtype at equivalent time point.
Lungs
Pneumococcal growth in the lungs of SP-D-/- mice was significantly greater at 6, 24 and 48 hrs (P < 0.05) post-infection when compared to SP-D+/+ mice (Fig. 3). The number of pneumococci in SP-D-/- lungs increased over the 24 hr period post infection, whereas numbers of pneumococci in the lungs of SP-D+/+ mice decreased over this same period (P < 0.05 compared to time zero). Thereafter, numbers of pneumococci recovered from lungs declined significantly (P < 0.05 compared to 24 hrs) in both mice by 48 hrs post-infection (Fig. 3). Pneumococci were cleared from SP-D+/+ mice by 48 hrs post-infection and by 54 hrs for SP-D-/- mice (data not shown for this time-point).
Figure 3 Time course of the change in numbers of S. pneumoniae in the nasopharynx (figure 1), trachea (figure 2), lungs (figure 3) and blood (figure 4) of SP-D-/- (◆) and SP-D+/+ (■) mice infected intranasally with 106 CFU (n = 10 mice at each time point, error bars indicate SEM). * denotes P < 0.01, ** denotes P < 0.05 for SP-D-/- when compared to wildtype at equivalent time point.
The role of SP-D in bacteraemia
In the blood of SP-D-/- mice (Fig. 4), pneumococci were recovered as early as 6 hrs after infection and bacterial numbers were further increased by 24 hrs (P < 0.05, compared to 6 and 12 hrs). In contrast, pneumococci were not detected in blood of SP-D+/+ mice at 6 and 12 hrs post infection and by 24 hrs was present in significantly lower numbers (P < 0.01) as compared to SP-D-/- mice at equivalent time-point. These bacteria were eventually cleared in SP-D+/+ mice by 48 hr post-infection whereas they were still present in the blood of SP-D-/- mice by 48 hrs, albeit at a lower level (Fig. 4).
Figure 4 Time course of the change in numbers of S. pneumoniae in the nasopharynx (figure 1), trachea (figure 2), lungs (figure 3) and blood (figure 4) of SP-D-/- (◆) and SP-D+/+ (■) mice infected intranasally with 106 CFU (n = 10 mice at each time point, error bars indicate SEM). * denotes P < 0.01, ** denotes P < 0.05 for SP-D-/- when compared to wildtype at equivalent time point.
Development of pathology in SP-D-/- and SP-D+/+ lungs infected with S. pneumoniae
Histopathological examination of lung tissue of SP-D+/+ and SP-D-/- mice infected with S. pneumoniae was done at time zero and at 24 and 48 hrs post infection. We have previously described in detail, the lung histology in non-infected SP-D-/- mice [5,18]. The histology of SP-D-/- mice used in the infection studies at time zero was the same as non-infected SP-D-/- mice. Briefly, histological changes in both non-infected SP-D-/- and infected SP-D-/- mice at time zero included increases in the size of alveolar type-II cells and scattered accumulation of material in the alveolar lumen (although many alveoli still remained unaffected), a marked increase in alveolar macrophage size, with many macrophages exhibiting a foamy appearance. Other than these well-documented features however, these mice exhibited no further histological evidence of lung inflammation or injury, consistent with the apparent health of these mice (data not shown as we have extensively described these features before) [5,18]. By 24 hrs post-infection however, SP-D-/- lungs exhibited features that were not apparent at time zero. These included heavy cellular infiltration visible around infected bronchioles and perivascular areas (Fig. 5, arrows-1) and increased inflammation characterised by exudate and thickening of the bronchiolar walls secondary to inflammation (Fig. 5, arrows-2). The same extent of bronchiole wall thickening was seen in the lungs of both SP-D-/- and SP-D+/+ mice at 24 hrs post infection but there was considerably less cellular infiltration into peribronchial and perivascular areas of SP-D+/+ lungs when compared to SP-D-/- lungs (Fig. 6, arrows 1 for bronchiole wall thickening, arrows 2 for cellular infiltration). However, by 48 hrs post-infection, peribronchial and perivascular cellular infiltration into SP-D-/- lungs had decreased significantly (Fig. 7, arrows 1 & 2) but had increased in SP-D+/+ lungs as compared to SP-D-/- mice (Fig. 8, arrows 1 for cellular infiltration & arrows 2 for bronchial inflammation).
Figure 5 Light microscopy of lung tissue from mice infected with 106CFU of S. pneumoniae. SP-D-/- 24 h post-infection (figure 5), SP-D+/+ 24 h post-infection (figure 6), SP-D-/- 48 h post-infection (figure 7) and SP-D+/+ 48 h post-infection (figure 8). Magnification ×250 for figures 5 and 8, ×400 for figures 6 and 7. See results for description of arrows.
Figure 6 Light microscopy of lung tissue from mice infected with 106CFU of S. pneumoniae. SP-D-/- 24 h post-infection (figure 5), SP-D+/+ 24 h post-infection (figure 6), SP-D-/- 48 h post-infection (figure 7) and SP-D+/+ 48 h post-infection (figure 8). Magnification ×250 for figures 5 and 8, ×400 for figures 6 and 7. See results for description of arrows.
Figure 7 Light microscopy of lung tissue from mice infected with 106CFU of S. pneumoniae. SP-D-/- 24 h post-infection (figure 5), SP-D+/+ 24 h post-infection (figure 6), SP-D-/- 48 h post-infection (figure 7) and SP-D+/+ 48 h post-infection (figure 8). Magnification ×250 for figures 5 and 8, ×400 for figures 6 and 7. See results for description of arrows.
Figure 8 Light microscopy of lung tissue from mice infected with 106CFU of S. pneumoniae. SP-D-/- 24 h post-infection (figure 5), SP-D+/+ 24 h post-infection (figure 6), SP-D-/- 48 h post-infection (figure 7) and SP-D+/+ 48 h post-infection (figure 8). Magnification ×250 for figures 5 and 8, ×400 for figures 6 and 7. See results for description of arrows.
Analysis of leukocyte infiltration into lungs of SP-D+/+ and SP-D-/- mice following pneumococcal infection
After intranasal challenge, leukocyte infiltration patterns into SP-D-/- and SP-D+/+ lungs were analysed at time zero, 24 and 48 hrs post-infection (Table-1A). In both strains, increased recruitment of neutrophils into inflamed areas of lung tissue was detected within the bronchiolar lumen, in the bronchiole wall and also in the perivascular areas within the vicinity of inflamed bronchioles. At 24 hrs post-infection, in areas of inflamed bronchioles of both mouse strains, numbers of neutrophils increased in significant numbers (P < 0.01, compared to time zero values for both strains, Table-1A). There was no significant difference between the strains at this time-point when compared to each other. However, by 48 hrs post-infection, there was a significant decrease in SP-D-/- mouse lung neutrophil numbers (P < 0.05 compared to 24 hrs) whereas the number of neutrophils in SP-D+/+ lungs at 48 hrs further increased as compared to 24 hrs and as compared to SP-D-/- mice at equivalent timepoint (P < 0.05, Table-1A). Overall, there was a 7.7 fold increase in neutrophil numbers by 24 hrs in SP-D-/- mice as compared to time zero, which dropped to a 5.2 fold increase by 48 hrs post-infection. In SP-D+/+ mice, there was a smaller 5.4 fold increase in neutrophil numbers by 24 hrs as compared to time zero, however the proportion of neutrophils in these mice at 24 hrs (70% of total leukocyte population) was greater than that of SP-D-/- mice at equivalent timepoint (53% of total leukocyte population) and also so by 48 hrs post-infection (73% to 59%, SP-D+/+ to SP-D-/- respectively). Importantly, in contrast to SP-D-/- mice, the neutrophil influx in SP-D+/+ mice continued to increase by 6.5 fold by 48 hrs post-infection compared to time zero. In SP-D-/- mice the neutrophils influx had declined by this time point.
Table 1 Lung leukocyte populations in SP-D-/- and SP-D+/+ mice at time zero, 24 and 48 hrs post-intranasal pneumococcal challenge. Leukocyte subpopulations (neutrophils, T cells, macrophages and B cells) numerated in the vicinity of inflamed bronchioles were expressed as cells per mm2 lung tissue. Leukocyte subpopulations expressed as the percentage of total lung leukocytes are shown in parenthesis. Fold increases in leukocytes subpopulations compared to time zero levels. N = 4 mice per each time point analysed for all samples. "a" denotes P < 0.01, "b" denotes P < 0.05 as compared to time zero values. "c" denotes P < 0.01, "d" denotes P < 0.05, SP-D-/- mice compared to SP-D+/+ mice at equivalent time-point.
A) Neutrophils SP-D+/+ mice SP-D-/- mice
Time Cells/mm2 Fold increase Cells/mm2 Fold increase
Zero: 11 +/- 2 9 +/-2
24: 60 +/-5 a (70%) 5.4 70+/-8a (53%) 7.7
48: 72 +/-7 a (73%) 6.5 47 +/-4 b, d (59%) 5.2
B) T cells SP-D+/+ mice SP-D-/- mice
Time Cells/mm2 Fold increase Cells/mm2 Fold increase
Zero: 9 +/-1 8 +/-2
24: 9 +/-1 (11%) no change 48 +/-8 a, c (36%) 6
48: 8 +/-1 (8%) no change 16 +/-4 b, d (20%) 2
C) Macrophage SP-D+/+ mice SP-D-/- mice
Time Cells/mm2 Fold increase Cells/mm2 Fold increase
Zero: 7 +/-1 no change 7 +/-1
24: 8 +/-2 (9%) no change 8 +/-3 (6%) no change
48: 9 +/-2 (9%) no change 8 +/-2 (10%) no change
D) B cells SP-D+/+ mice SP-D-/- mice
Time Cells/mm2 Fold increase Cells/mm2 Fold increase
Zero: 8 +/-1 6 +/-1
24: 8 +/-2 (9%) no change 7 +/-3 (5%) no change
48: 9 +/-2 (9%) no change 9 +/-1 (11%) no change
There were, dramatic differences in lung tissue T cell accumulation between SP-D-/- and SP-D+/+ mice. In SP-D-/- lungs, T cell numbers showed a sharp 6-fold increase around inflamed bronchioles by 24 hrs post-infection (P < 0.01, when compared to time zero, see table-1B). T cell numbers then decreased to a 2-fold increase by 48 hrs post-infection (P < 0.05 as compared to time zero, see table-1B). In contrast, in the lungs of SP-D+/+ mice, there was no increase in the numbers of T cells in inflamed areas throughout the 48 hr post-infection period (P > 0.05, when compared to time zero). T cell numbers in SP-D+/+ were significantly lower than in SP-D-/- lungs at 24 and 48 hr post-infection (P < 0.01 for 24 hr and P < 0.05 for 48 hrs).
Macrophage and B cells numbers remained unchanged in the lungs of both SP-D-/- or SP-D+/+ (P > 0.05 as compared to time zero) over the 48 hr post-infection period (Table-1C &1D). PBS alone challenged mice had minimal leukocyte numbers in lungs, with each leukocyte population counted below 10 cells/mm2 (data not shown).
Discussion
Previous evidence has shown that SP-D interacts with S. pneumoniae in vitro [11,17]. The results of the current study are the first to demonstrate in vivo, that SP-D has an important role to play in pneumococcal clearance. Pneumococcal colonisation of the upper and lower respiratory tract, and infiltration patterns of leukocytes into the lungs of infected mice were affected by the absence of SP-D. Pulmonary clearance of intranasally administered S. pneumoniae was significantly reduced in SP-D deficient mice as compared to SP-D sufficient controls. Furthermore, our results clearly demonstrate that lack of SP-D allows persistent pneumococcal colonisation of the nasopharynx and trachea and early onset and increased levels of bacteraemia in colonised mice. Our results also indicate that SP-D influences the accumulation of T cells within the vicinity of inflamed bronchioles, whereby increased levels of T cell infiltration into SP-D deficient lungs was observed. This is the first report to demonstrate in vivo, that SP-D deficiency leads to increased pneumococcal colonisation of the nasopharynx and trachea, hastens the onset and development of bacteraemia, and affects leukocyte infiltration patterns into infected lungs.
SP-D is synthesised and secreted not only by pulmonary epithelial cells but also by epithelial cells and submucosal glands of the trachea of the normal adult mouse [20] and has been detected at low concentration (56 ng/ml) in nasopharyngeal washings of normal mice [21]. Based on our results in the nasopharynx and trachea it is clear that SP-D has a crucial role to play in these sites during pneumococcal infection. Consequently, it is clear therefore that SP-D prevents persistent upper airway colonisation by pneumococci and helps protect against invasion of the lower airways. However, it is also conceivable that lack of SP-D may affect resident leukocyte populations involved in host response or alters host tissue sites as to make them more suitable for pneumococcal adherence and colonisation. We are currently investigating these possibilities.
Our results also indicate that of lack of SP-D contributes to the early onset and increased levels of bacteraemia during pneumococcal pneumonia. It is important to note that SP-D+/+ mice cleared bacteria from their blood by 48 hrs post infection and that the numbers of pneumococci in the blood of both strains of mice reflected their levels in the lung. These results strongly suggest that lung surfactant protein D plays an important role in delaying the appearance of pneumococci in the blood and in limiting their numbers in the bloodstream.
SP-D binds and agglutinates S. pneumoniae in the presence of calcium and is thought to enhance mucociliary and phagocytic clearance [11,17]. In addition, binding of SP-D to lipoteichoic acid and peptidoglycan [22] may suggest a role for SP-D in the prevention of bacterial colonisation of the alveolar epithelium. Elimination of these SP-D functions could explain the colonisation of the trachea and nasopharynx, the decreased pneumococcal clearance from lungs and the early onset of pneumococcal bacteraemia observed in SP-D deficient mice in our study.
As reported for other strains of mice [19,23,24], pneumococcal infection was coupled with an influx of neutrophils into the lung tissue of both SP-D+/+ and SP-D-/- mice. This is consistent with the data of LeVine and colleagues [14,15] who also showed that neutrophil accumulation was similar in the lungs of SP-D-/- and SP-D+/+ mice after H. influenzae and group B streptococcal infection. In our study, the recruitment of neutrophils in the first 24 hrs post-infection was not affected by the absence of SP-D. However, our results also indicate that the neutrophil response in SP-D deficient mice was not maintained for as long as in wild-type mice. SP-D has been reported as a chemotactic factor for neutrophils in vitro [25], and although our data demonstrates that the lack of SP-D does not effect early neutrophil infiltration into lungs, it does clearly affect the longer-term influx of neutrophils as demonstrated by the significant drop in neutrophil infiltration by 48 hrs in SP-D-/- mice. This is not a simple reflection of lung pneumococcal numbers either, as by 24 hrs although there is a significant difference in bacterial CFUs in mice (see figure-3, SP-D+/+ compared to SP-D-/-), the neutrophil numbers in these mice at 24 hrs is not significantly different. Although a similar accumulation of neutrophils was observed in the lungs of both SP-D+/+ and SP-D-/- mice by 24 h after infection, there were significantly greater numbers of pneumococci in the lungs of SP-D-/- mice at this timepoint. This could have resulted in decreased levels of phagocytosis due to the deficiency in the binding and opsonisation of the pneumococcus due to the lack of SP-D, but also could be due to other factors affecting neutrophil activity. For example, as others and we have previously shown, SP-D deficient mice, despite their healthy appearance, develop progressive alveolar proteinosis and have increased numbers of foamy alveolar macrophages [5,18,26]. Thus, it is possible that the excess lipid in SP-D-/- lungs may inhibit the neutrophil respiratory burst, as previously demonstrated in vitro [27].
Together with others we have also previously shown that SP-D deficient mice have a 5- to 10-fold increase in the number of apoptotic and necrotic alveolar macrophages compared to wild-type mice, suggesting a contribution of SP-D to immune homeostasis by recognising and promoting removal of apoptotic cells in vivo [28,29]. It will be of value to assess the clearance of infected apoptotic neutrophils during pneumococcal infection in SP-D deficient and sufficient mice. We are currently in the process of examining this.
Previous studies have also reported that SP-D inhibits T lymphocyte proliferation and local T cell responses in vitro [30,31]. It is therefore noteworthy that we found a heavy infiltration of T lymphocytes in the vicinity of inflamed bronchioles in SP-D deficient lungs at 24 hrs post pneumococcal infection, in contrast to infected SP-D+/+ mice, which exhibited minimal numbers of T cell infiltration. Thus, it appears that SP-D influences T cell infiltration patterns in lungs during pneumococcal infection. It is unclear however, whether SP-D influences T lymphocyte recruitment directly or whether the enhanced T cell infiltration is a consequence of the stimulus of bacteria persisting longer in the respiratory tract of the SP-D deficient mouse. Our previous studies would indicate however, that T cell infiltration is not directly dependant upon pneumococcal numbers as similar colony forming units of pneumococci in lungs and blood of mice can result in totally different T cell infiltration patterns [32]. In addition, in this study we have shown that significantly different pneumococcal numbers can result in significantly different leukocyte infiltration patterns and vice versa.
It has been suggested that SP-D might provide an important link between innate and adaptive immunity, by modulation of antigen presenting cells and T cell function [33] whereby SP-D would enhance the uptake of respiratory pathogens in the alveolar space by recruited antigen presenting cells, whilst suppressing T cell activation in the alveolar space in order to prevent an inflammatory cascade that could damage the local lung airspaces and impair gas exchange [4,33]. Our findings also support an important anti-inflammatory role for SP-D in pneumoccocal infection in vivo. Indeed, previous studies have also shown increased pulmonary inflammation, cellular recruitment, oxidant production and decreased macrophage phagocytosis in SP-D deficient mice infected with Haemophilus influenzae and group B streptococcus. No decrease in bacterial killing in the lungs of these mice were observed in this study [14], suggesting that other aspects of immunity compensated for the lack of SP-D and cleared the infection effectively. However, after intranasal infection with influenza A virus, SP-D deficient mice showed decreased viral clearance and uptake by alveolar macrophages and increased production of inflammatory cytokines in response to viral challenge [15]. Additional studies are clearly required to further elucidate the role of SP-D in regulating adaptive immune responses in vivo.
The potential of truncated recombinant forms of SP-D as a new therapy for infectious and inflammatory diseases has recently been investigated [[34]-35]. Treatment by intranasal administration of SP-D and a 60-KDa recombinant fragment of human SP-D (rSP-D) had a protective effect in a murine model of fungal infection and allergy caused by Aspergillus fumigatus [16]. The survival rate of mice increased to 60 and 80% after treatment with SP-D and rSP-D, respectively [16]. In addition, intrapulmonary administration of rSP-D reduced the number of apoptotic and necrotic alveolar macrophages and partially corrected lipid accumulation in SP-D-/- mice [28]. Thus, it would be of a great interest to investigate whether the co-administration of SP-D or truncated rSP-D with S. pneumoniae would correct the defects observed in SP-D deficient mice during pneumococcal bronchopneumonia. Administration of SP-D at intervals after infection may also indicate at what stages in the disease process, the protein is most heavily involved. We are currently in the process of investigating these questions.
In summary, the absence of lung surfactant protein D increases the persistence of pneumococcal colonisation and infection in the upper and lower respiratory tract, as well as leading to earlier onset and increased levels of bacteraemia. In addition, the pattern of cellular infiltration into the lungs of SP-D-/- mice following pneumococcal infection is different from SP-D+/+ mice, as characterised by shorter-term neutrophil influx and increased levels of T cell infiltration. SP-D clearly has an important function in the early stages of infection as part of the host immune response to pneumococcal invasion and warrants further study.
Acknowledgements
The Wellcome Trust supported the work in Leicester. RJ was in receipt of a studentship from The Lebanese University.
==== Refs
Crouch EC Surfactant protein D and pulmonary host defence Respiratory Research 2000 1 93 108 11667972 10.1186/rr19
Crouch E Wright JR Surfactant proteins A and D and pulmonary host defense Annu Rev Physiol 2001 63 521 554 11181966 10.1146/annurev.physiol.63.1.521
Wright JR Immunomodulatory functions of surfactant Physiol Rev 1997 77 931 962 9354809
Wright JR Immunoregulatory functions of surfactant proteins Nature Reviews Immunology 2005 5 58 68 15630429 10.1038/nri1528
Clark H Palaniyar N Strong P Surfactant protein D reduces alveolar macrophage apoptosis in vivo Journal of Immunology 2002 169 2892 2899
Drickamer K Dordal MS Reynolds L Mannose binding proteins isolated from rat liver contain carbohydrate recognition domains linked to collagenous tails J Biol Chem 1986 261 6878 6887 3009480
Holmskov UJ Mollenhauer J Madsen L Cloning of gp-340, a putative opsonin receptor for lung surfactant protein D Proc Natl Acad Sci 1999 96 10794 10799 10485905 10.1073/pnas.96.19.10794
Kishore U Wang JY Hoppe HJ The alpha-helical neck region of human lung surfactant protein D is essential for the binding of the carbohydrate recognition domains to lipopolysaccharides and phospholipids Biochem J 1996 318 505 511 8809039
Ferguson J Voelker D Ufnar J Surfactant protein D inhibition of human macrophage uptake of Mycobacterium tuberculosis is independent of bacterial agglutination J Immunol 2002 168 1309 1314 11801670
Jounblat R Kadioglu A Iannelli F Binding and agglutination of Streptococcus pneumoniae by human surfactant protein D (Sp-D) varies between strains but Sp-D fails to enhance killing by neutrophils Infect Immun 2004 72 709 716 14742512 10.1128/IAI.72.2.709-716.2004
Madan T Eggleton P Kishore U Binding of pulmonary surfactant proteins A and D to Aspergillus fumigatus conidia enhances phagocytosis and killing by human neutrophils and alveolar macrophages Infect Immun 1997 65 3171 3179 9234771
Murray E Khamri W Walker M Expression of surfactant protein D in the human gastric mucosa and during Helicobacter pylori infection Infect Immun 2002 70 1481 1487 11854236 10.1128/IAI.70.3.1481-1487.2002
LeVine AM Whitsett JA Gwozdz JA Distinct effects of surfactant protein A or D deficiency during bacterial infection on the lung J Immunol 2000 165 3934 3940 11034401
LeVine AM Whitsett JA Hartshorn KL Surfactant protein D enhances clearance of influenza A virus from the lung in vivo J Immunol 2001 167 5868 5873 11698462
Madan T Kishore U Singh M Protective role of lung surfactant protein D in a murine model of invasive pulmonary aspergillosis Infect Immun 2001 69 2728 2731 11254642 10.1128/IAI.69.4.2728-2731.2001
Hartshorn KL Crouch E White MR Pulmonary surfactant proteins A and D enhance neutrophil uptake of bacteria Amer J Physiol 1998 274 L958 969 9609735
Botas C Poulain F Akiyama J Altered surfactant homeostasis and alveolar type II cell morphology in mice lacking surfactant protein D Proc Natl Acad Sci 1998 95 11869 11874 9751757 10.1073/pnas.95.20.11869
Kadioglu A Gingles NA Grattan K Host cellular immune response to pneumococcal lung infection in mice Infect Immun 2000 68 1557 1562 10678974 10.1128/IAI.68.3.1557-1562.2000
Wong CJ Akiyama J Allen L Localization and developmental expression of surfactant proteins D and A in the respiratory tract of the mouse Pediatr Res 1996 39 930 937 8725251
Reading PC Morey LS Crouch EC Collectin-mediated antiviral host defense of the lung: Evidence from Influenza virus infection of mice J Virol 1997 11 8204 8212 9343171
VandeWetering JK vanEijk M vanGolde LM Characteristics of surfactant protein A and D binding to lipoteichoic acid and peptidoglycan, 2 major cell wall components of gram-positive bacteria J Infect Dis 2001 184 1143 1151 11598837 10.1086/323746
Gingles NA Alexander JE Kadioglu A Role of genetic resistance in invasive pneumococcal infection: identification and study of susceptibility and resistance in inbred mouse strains Infect Immun 2001 69 426 434 11119534 10.1128/IAI.69.1.426-434.2001
Kerr AR Irvine JJ Search NA Role of inflammatory mediators in resistance and susceptibility to pneumococcal infection Infect Immun 2002 70 1547 1557 11854243 10.1128/IAI.70.3.1547-1557.2002
Crouch EC Persson A Griffin GL Interactions of pulmonary surfactant protein D (SP-D) with human blood leukocytes Amer J Respir Cell Molec Biol 1995 12 410 415 7695920
Korfhagen TR Sheftelyevich V Burhans MS Surfactant protein-D regulates surfactant phospholipid homeostasis in vivo J Biol Chem 1998 273 28438 284343 9774472 10.1074/jbc.273.43.28438
Ahuja A Oh N Chao W RG Inhibition of the human neutrophil respiratory burst by native and synthetic surfactant Amer J Resp Cell Molec Biol 1996 14 496 503
Clark H Palaniyar N Strong P Surfactant protein D reduces alveolar macrophage apoptosis in vivo J Immunol 2002 169 2892 2899 12218102
Vandivier RW Ogden CA Fadok VA Role of surfactant proteins A, D, and C1q in the clearance of apoptotic cells in vivo and in vitro: calreticulin and CD91 as a common collectin receptor complex J Immunol 2002 169 3978 3986 12244199
Borron PJ Crouch EC Lewis JF Recombinant rat surfactant-associated protein D inhibits human T lymphocyte proliferation and IL-2 production J Immunol 1998 161 4599 4603 9794387
Fisher JH Larson J Cool C Lymphocyte activation in the lungs of SP-D null mice Amer J Respir Cell Molec Biol 2002 27 24 33 12091242
Jounblat R Kadioglu A Mitchell TJ Pneumococcal Behaviour And Host Response During Bronchopneumonia Are Affected By the Cytolytic and Complement Activating Activities of Pneumolysin Infect Immun 2003 71 1813 1819 12654795 10.1128/IAI.71.4.1813-1819.2003
Shepherd VL Pulmonary surfactant protein D: a novel link between innate and adaptive immunity Am J Physiol Lung Cell Mol Physiol 2002 2 L516 L517 11839546
Clark H Reid K The potential of recombinant surfactant protein D therapy to reduce inflammation in neonatal chronic lung disease, cystic fibrosis, and emphysema Arch Dis Child 2003 88 981 984 14612363 10.1136/adc.88.11.981
Clark H Reid KBM Structural requirements for SP-D function in vitro and in vivo: Therapeutic potential of recombinant SP-D Immunobiol 2002 205 619 631 10.1078/0171-2985-00159
|
16255775
|
PMC1282592
|
CC BY
|
2021-01-04 16:36:26
|
no
|
Respir Res. 2005 Oct 28; 6(1):126
|
utf-8
|
Respir Res
| 2,005 |
10.1186/1465-9921-6-126
|
oa_comm
|
==== Front
BMC CancerBMC Cancer1471-2407BioMed Central London 1471-2407-5-1401625577810.1186/1471-2407-5-140Research ArticleAntimetastatic gene expression profiles mediated by retinoic acid receptor beta 2 in MDA-MB-435 breast cancer cells Wallden Brett [email protected] Mary [email protected] Mari E [email protected] Mary L [email protected] Karen [email protected] Department of Pathology, Box 357470, University of Washington, Seattle, WA, USA2 Department of Biostatistics, Box 357232, University of Washington, Seattle, WA, USA3 Division of Oncology, Box 358050, University of Washington, Seattle, WA, USA2005 28 10 2005 5 140 140 12 6 2005 28 10 2005 Copyright © 2005 Wallden et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
The retinoic acid receptor beta 2 (RARβ2) gene modulates proliferation and survival of cultured human breast cancer cells. Previously we showed that ectopic expression of RARβ2 in a mouse xenograft model prevented metastasis, even in the absence of the ligand, all-trans retinoic acid. We investigated both cultured cells and xenograft tumors in order to delineate the gene expression profiles responsible for an antimetastatic phenotype.
Methods
RNA from MDA-MB-435 human breast cancer cells transduced with RARβ2 or empty retroviral vector (LXSN) was analyzed using Agilent Human 1A Oligo microarrays. The one hundred probes with the greatest differential intensity (p < 0.004, jointly) were determined by selecting the top median log ratios from eight-paired microarrays. Validation of differences in expression was done using Northern blot analysis and quantitative RT-PCR (qRT-PCR). We determined expression of selected genes in xenograft tumors.
Results
RARβ2 cells exhibit gene profiles with overrepresentation of genes from Xq28 (p = 2 × 10-8), a cytogenetic region that contains a large portion of the cancer/testis antigen gene family. Other functions or factors impacted by the presence of exogenous RARβ2 include mediators of the immune response and transcriptional regulatory mechanisms. Thirteen of fifteen (87%) of the genes evaluated in xenograft tumors were consistent with differences we found in the cell cultures (p = 0.007).
Conclusion
Antimetastatic RARβ2 signalling, direct or indirect, results in an elevation of expression for genes such as tumor-cell antigens (CTAG1 and CTAG2), those involved in innate immune response (e.g., RIG-I/DDX58), and tumor suppressor functions (e.g., TYRP1). Genes whose expression is diminished by RARβ2 signalling include cell adhesion functions (e.g, CD164) nutritional or metabolic processes (e.g., FABP6), and the transcription factor, JUN.
==== Body
Background
It is estimated that approximately 15–25% of women with node-negative breast cancer will eventually succumb to the disease due to distant metastases [1]. While important and promising anti-hormonal therapies – tamoxifen and aromatase inhibitors – exist for postmenopausal breast cancer patients with estrogen receptor positive tumors [2], chemo- and radiation-therapy remain the major options for all other women with progressive disease. Elucidating gene expression alterations in metastatic lesions compared to non-metastatic tumors is compelling, and orthotopic xenograft models provide an avenue for tackling this challenge. The results of such studies could lead to rapid translational research and drug discovery [3,4]. Several candidate metastasis-promoting proteins have been identified in differential screening of cell cultures and primary invasive tumors and these include ERB2/Her2/neu, VEGF and stromelysin [5]. An exciting new field of investigation, the search for proteins responsible for inhibition of the metastatic cascade, "metastasis suppressor" genes, has revealed a set of molecules that include: NM23 (histidine kinase), KAI1 (a tetraspanin integral membrane protein that responds to NFκB), BRMS1 (gap junction function), and MKK4 (mitogen-activated protein kinase) [4,6,7]. Functions regulated by metastasis suppressors include transcription, signal transduction, cell adhesion, and inflammation. We have demonstrated that RARβ2, with known tumor suppressor functions, also confers antimetastatic properties in a xenograft model of human breast cancer [8].
RARs are members of the nuclear hormone receptor superfamily, and in conjunction with their heterodimeric partners, the retinoid X receptors (RXRs), they positively or negatively regulate (most commonly) genes containing a direct repeat five (DR5) retinoic acid response element (RARE) within promoters. Transactivation via these heterodimeric partners is generally conferred by physiological or pharmacologic levels of retinoid-derived ligands, including all-trans retinoic acid (AT-RA) and 9-cis-retinoic acid, for RARs and RXRs, respectively. We and others have shown that RARβ2 or RARβ4 mRNA is diminished in most breast cancer cell lines [9,10]. Several laboratories have also demonstrated loss or diminished expression of the RARβ2 mRNA in primary breast cancers [11,12]. The mechanism(s) for diminished RARβ2 expression include transcriptional repression by epigenetic silencing [13-15]. A truncated, oncogenic RARβ protein (RARβ-prime) is exclusively expressed in breast cancer cell lines, and the presence of this isoform likely obstructs tumor suppressor functions of RARβ2 and RARβ4 protein isoforms [16,17].
RARβ2 activates both tumor suppressor and antimetastatic programs. In cell cultures in which RARβ2 has been introduced via a retroviral vector, we found that all breast cancer cells could be inhibited in their proliferative capacity, even in the absence of the natural ligand, AT-RA. Moreover, there is a dichotomy in RARβ2-transduced tumor cells in response to AT-RA: ER-positive cells undergo apoptosis, while ER-negative cells are further reduced in their proliferative potential, in comparison to culture conditions in the absence of AT-RA. In our xenograft model, we engrafted MDA-MB-435 breast cancer cells containing either LXSN-vector or LXSN-RARβ2 into the mammary fat pad (mfp) of nude or SCID mice [8]. Our model was designed to mimic the treatment of patients with advanced local disease. After 12–15 weeks, the mfp-encapsulated, primary tumors were fully resected. Following an additional 7–11 weeks, the animals were necropsied and analyzed for multiple parameters, including incidence of metastases – the most common sites being lungs and pleura. The overall incidence of metastasis was 37% (19 of 52 mice) in the vector-control animals compared to 1.8% (1 of 55 mice) in the RARβ2-expressing tumor cell implants (p < 0.00001). The one metastatic lesion from the RARβ2-tumor implant was a micro-metastasis in a nude recipient. SCID recipients showed the highest metastatic incidence in the control implants (55%) compared to none in the animals receiving RARβ2-bearing xenografts.
Very few experiments have been conducted to determine the downstream factors regulated by ectopic RARβ2 expression. One such study with F9 teratocarcinoma cells, employing expression microarrays and subtractive hybridization, found differential expression of transcription factors, signalling molecules and metabolic enzymes [18]. Moreover, these investigators found that RARβ2 altered gene expression patterns even in the absence of AT-RA. Toulouse and colleagues transfected RARβ2 into two lung cancer cell lines that lacked endogenous RARβ2 expression [19]. Using the Clontech Atlas human cDNA array, they identified a significant number of genes involved in immune responses. It is likely, however, that unique expression profiles will be found with ectopic RARβ2 expression, depending upon the cell of origin. We used cultured cells, identical to those in the xenograft studies, to conduct expression microarray experiments in order to uncover gene functions or physiologic activities that may prevent metastasis due to overexpression of RARβ2. Due to the hierarchical transcriptional regulation by RARβ in embryogenesis and development, we hypothesized that multiple regulatory pathways, direct or indirect, could be mediated by overexpression of RARβ2 in breast cancer cells.
Methods
Cells and xenograft tumors
MDA-MB-435 human breast cancer cells were transduced with either the LXSN-vector or LXSN-RARβ2. Plasmid production, generation of retroviral vectors, and cell culture have been described previously [8,20]. For the microarrays, cells were grown to confluency in single Corning 150 mm cell culture-treated dishes in breast tumor cell media [16]. Two independent culture sets (pairs) were used among the eight arrays. Cells and RNA for validation (below) came from the same lineage (within one passage) as the cells used in the arrays. Tissues were primary, resected tumors from our previous study [8]. Primary xenograft tumors were resected and snap frozen in cryopreservation media in liquid nitrogen. RNA was extracted directly from archived primary tumor tissue. Additional LXSN-vector and LXSN-RARβ2 transductions were performed using the parental MDA-MB-435 cells from a passage near that of the original transduction and cloning. These latter cells were grown under selection (500 μg/mL G-418) on Corning 150 mm cell culture-treated dishes for 10 days when the cells were near confluent, at which time RNA was collected, as below, for qRT-PCR.
RNA isolation
Total RNA was isolated with Invitrogen's Micro to Midi kit (Carlsbad, CA) according to the manufacturer's protocol for cells or tissue, respectively. The purity and integrity of the RNA was evaluated by measuring the 260/280 nm optical density ratio and either by use of the Agilent Bioanalyzer (Agilent Technologies, Palo Alto, CA) or by denaturing agarose gel as for Northern blots [10].
RNA labelling, array hybridization, and feature extraction
Five μg of RNA from each transduced cell lineage (LXSN-vector or LXSN-RARβ2) was labelled using the Agilent Fluorescent Linear Amplification kit (Palo Alto, CA). Labelled RNA was further purified using the Qiagen RNeasy Mini kit protocol for liquid samples (Valencia, CA). Four pairs of labelled cRNA were prepared from two independent RNA extractions, hybridized to the Agilent Human 1A Oligo Array, and washed using the Agilent In situ Hybridization Kit Plus. cRNA was labelled with Cy5 and Cy3 for "swapped"-labelling for co-hybridizations on different chips. Hence, every chip had a complimentary chip with a swapped dye configuration, for a total of eight chips. Microarrays were scanned in an Agilent DNA Microarray Scanner and expression data were obtained using the Agilent Feature Extraction software (version 6.1.1), using defaults for all parameters.
Statistics and analysis
The image file was processed using Agilent's Feature Extraction software. This Feature Extraction program was used to identify pixels corresponding to fluorescent signal (as opposed to background) and to remove pixels with intensities that met the default criteria for outliers. No background subtraction was used nor were any of the available error models used, since the value of these has not been proven and our replicate data were more highly correlated without the use of these features. For each identified area of signal and each of the two dyes, the basic measure of RNA abundance was taken to be the mean intensity over pixels in the identified signal area. The log ratio of the red to green intensities for each signal area (the mean over approximately 60 pixels per oligonucleotide) was used for statistical analyses, with all subsequent analyses done using the R statistical software package [21]. The unit of analysis for the array data was a sample from a single cell line with parallel preparations for all samples and a sample size of eight.
Since the purpose of this study is partially hypothesis formation (as opposed to strict hypothesis testing), we elected a priori to examine the 100 most extreme median log ratios (over the eight cell culture arrays) for large differences between the treatments as well as biological patterns that could lead to more explicit hypothesis testing in future experiments using primary xenograft tumors. The median was chosen as the summary measure of expression across the eight arrays (for each oligonucleotide) because the median has been shown to have better performance than mean-based statistics, such as the t-statistic or the SAM statistic, in identifying spiked-in RNAs using small sample sizes [22]. This method of analysis is also similar to that suggested by Breitling et al. [23], with both biological and statistical considerations used to choose the analysis method, except that especially low or high ranks are given less weight in the current method than in the rank product method suggested therein. Before median log ratios were calculated, data from each array were normalized separately using the R function lowess, with a window size of approximately 360 points. The joint significance level (p-value) for finding 100 median log ratios as extreme or more extreme than observed was determined using the permutation test under the null hypothesis of no differential expression in any of the genes between the two groups, as in Pollard and van der Laan, comparing the 100th absolute value order statistic to its permutation reference distribution [24].
Standard binomial probabilities were used to determine the significance of concordance between quantitative real-time PCR (qRT-PCR) results [see below] using RNA from cultures and tissue and the apparent over-representation of Xq28 genes in the top 100 list.
Finally, since an unusually large number of top 100 ratios corresponded to genes on Xq28, we searched the sequence of Xq28 (positions Xq28:146,767,469–154,804,778 from human assembly, hg17, using Known Gene data downloaded from the UCSC Genome Browser [25], for matches to the conserved portions of the RARβ DR5 DNA binding element (βRARE) using the UNIX utility "agrep" [26]. Six βRARE sequences were found within the 89 listed "Known Genes" (exons, introns, plus 1000 upstream and downstream base pairs) on Xq28. In addition, the genomic regions for three other non Xq28 genes of interest (ZADH1, ZNF387, FABP6), including the leading and ending 100 base pairs, were searched for RAREs.
Validation: quantitative RT-PCR, northern blot and western blot
Total RNA from cells and tissue was treated with 1 U of RNase-free deoxyribonuclease 1 (DNase 1) (Ambion, Austin, TX) per μg of RNA to eliminate genomic DNA contamination. DNase 1 was inactivated with the addition of 2.5 mM EDTA (pH 8.0), followed by 10 minutes at 65°C. cDNA synthesis and PCR analysis were performed using the Invitrogen SuperScript III Platinum Two-Step qRT-PCR Kit with SYBR Green. Quantitative gene expression was detected using the LightCycler (Roche Molecular Biochemicals, Mannheim, Germany). The Invitrogen protocol includes a step using uracil DNA glycosylase and dUTP to prevent amplification of carryover PCR products.
All primers (Supplemental Table 1) were designed with Primer3 [27] using sequences contained in the 60-mer oligonucleotide probes found on the Agilent Human 1A Oligo Array. The arrays contain probes synthesized in situ on glass slides by phosphoramidite chemistry. Probes were designed with a 3' bias from collective database sources representing 17,086 genes [28]. Primers were synthesized by Qiagen [aka Operon] (Valencia, CA). qRT-PCR reactions for all samples were carried out in triplicate for each gene. In some cases, additional qRT-PCR was performed with independent RNA samples, as indicated in Tables 1 and 5. Genomic DNA from MDA-MB-435 parental cells was used to generate a standard curve, as this consistently gave values for slope, error and regression coefficient recommended for optimal results by Roche and based on calculations by Pfaffl [29]. All samples were normalized with respect to the constitutively expressed large ribosomal phosphoprotein P0 (RPLP0, aka 36B4) gene. The normalized data were then used to generate a differential expression ratio to compare to the expression microarray data. In cases where we were unable to confirm the expression differential obtained from the microarray analyses, we interrogated Map Viewer [30] or AceView [31] to determine if the 60-mer overlapped mRNA isoforms (e.g., splice variants) known for the gene of interest.
Northern blot methods are as described [32]. Northern probe production for SPP1 and CTAG1 is as follows: the 1384 bp SPP1 probe was generated from vector control cells using Qiagen's One step RT-PCR mix (with primers: Fwd CATCACCTGTGCCATACCAG and Rev CCGTGGGAAAACAAATAAGC). The 463 bp CTAG1 probe was generated from both RARβ2 and vector control cells using Qiagen's One step RT-PCR mix (with primers: Fwd CTTCAGGGCTGAATGGATG and Rev AACAAACATGTAAGCCGTCCT). An example of the difference in the level of RARβ2 mRNA in the transduced cells is shown in Figure 1A.
Two independent samples of the RARβ2- and vector-transduced cells were analyzed for SPP1/osteopontin using methods from Tuck et al. [33]. Briefly, 5 × 105 cells were plated on Corning 100 mm cell culture-treated dishes in breast cancer cell media [32] and incubated for 18 hours. Cells were then grown in 3 ml of serum-free breast cancer cell media for 24 hours. The media was centrifuged briefly to eliminate cellular debris, and then concentrated in an Amicon Ultra-4 (Millipore, Billerica, MA) with a 10 kDa cut off at 1500 rpm for 1.5 hrs, resulting in a final volume of ~60 μL each. The adherent cells were trypsinized and counted using a hemocytometer. Cells were lysed, using a mild lysis buffer (50 mM Tris-HCI pH 8, 300 mM NaCI,10 mM MgCI2,1 mM EDTA, 25 mM β-glycerophosphate, 0.5% lgepal-CA 630 [Sigma I3021]) [34] with a protease inhibitor cocktail (Roche, Mannheim, Germany). Protein from ~3 × 105 cell equivalents of both media and cell lysates were separated, along with recombinant SPP1 (gift from Dr. CM Giachelli), with SDS-PAGE and then blotted to PVDF membrane (BioRad, Hercules, CA). The membrane was probed using a 1:3,500 dilution of a goat anti-SPP1 polyclonal antibody (gift from Dr. CM Giachelli) followed by a horseradish peroxidase-conjugated rabbit anti-goat antibody at 1:20,000 (Pierce, Rockland, IL). The blot was visualized by chemiluminescent detection (Pierce).
Results & discussion
Differential expression level changes comparing cells with metastatic potential to RARβ2-overexpressing cells
RARβ2 is overexpressed in MDA-MB-435 cells transduced with a retroviral vector containing the full-length RARβ2 cDNA (Figure 1A). We hypothesized that the antimetastatic functions of RARβ2 were transcriptionally mediated, independent of excess or therapeutic concentrations of ligand, AT-RA, as it has been demonstrated that RARβ2 can modulate gene expression even in the absence of exogenous ligand [18,35].
Of the top 100 candidates (Additional File 2: Supplemental Table 2), which includes rank, fold change (RARβ2/vector), chromosome localization, and URL links to GeneCards), eighty-six comprised unique gene candidates, the remaining being duplicate probes on the arrays. The p-value for finding 100 median ratios this extreme or more extreme was p = 0.004. Fifty-three of the eighty-six unique probes (62%) indicated induced expression in RARβ2-transduced cells.
The relative ratios in expression levels between the RARβ2-transduced cells/vector control cells and the magnitude of the differences were confirmed in fourteen of the top twenty ranked median candidates (70%) by qRT-PCR (Table 1). In addition, CTAG1 expression differences were confirmed by northern blot analysis (Figure 1B). For two of the fourteen confirmed candidates, FABP6 and NNMT, the magnitude detected by qRT-PCR was greater than that detected with the expression microarrays. Overall, there was excellent concordance as to whether the gene candidate was induced or diminished between the array and qRT-PCR results.
Five genes were not confirmed (NC) by qRT-PCR. However, for each of these candidates, we hypothesize that there is potential overlap of an mRNA isoform with the Agilent 60-mer probe, as determined in Map Viewer or AceView for each gene. The remaining single probe, OR52P1, is a pseudogene, which could not be validated.
The top twenty genes obtained from the statistical analysis and subsequent validation provide an interesting window into the antimetastatic cellular functions mediated by RARβ2 in MDA-MB-435 cells. Three candidates, all reduced in RARβ2 cells, are implicated in cell adhesive properties of tumors (LSAMP [rank 1], PCDH11Y [rank 5], and CD164 [rank 8]). CD164 is a sialomucin, and in general, both transmembrane and secreted mucins can regulate growth factor receptor activity. Furthermore, cancer cells may subvert the normal role of the extensively post-translationally modified mucins or sialomucins in order to sense the environment and regulate metastasis [36]. Another category of genes, all with diminished expression in RARβ2 cells, are implicated in nutritional regulation or sensing, and hormone metabolism: SLC38A2 [rank 2], PCSK4 [rank 6], SR-BP1 [rank 11], and FABP6 [rank 12]. Two other genes of note in the top twenty ranked candidates that are elevated in RARβ2 cells are NPDC1 [rank 15] and ZNF387/ST18 [rank 20]. Little has been reported on these genes, but both are likely transcriptional regulatory molecules. NPDC1 has been associated with cessation of proliferation in the nervous system [37], and ZNF387/ST18 is hypothesized to be a tumor suppressor gene in breast cancer [38].
In order to rule out a clonal influence on the antimetasatic phenotype of RARβ2, we also performed qRT-PCR on mixed clones from our original transduction in order to look at a more heterogeneous culture. The pools consisted of equal numbers of cells from five RARβ2 clones and three vector-control clones. RNA extracted from these mixed populations underwent qRT-PCR, using primers from the top seven ranked genes. As shown in Additional File 3: Supplemental Table 3, all seven genes were validated. Additionally, we introduced LXSN-control and LXSN-RARβ2 vectors into MDA-MD-435 cells, and following a ten-day selection in G-418, collected RNA from the newly transduced cell populations. Also shown in Supplemental Table 3, six of the seven top genes were validated.
Genes in the cancer/testis antigen family localized to Xq28 are elevated in RARβ2 overexpressing cells
Eleven of the top 100 (87 unique) genes reside on the X chromosome, including CTAG1 [rank 3] (Supplemental Table 2 & Table 2). Of these, eight map to Xq28, the most telomeric cytogenetic long-arm band (Table 2). Given that there are seventy-five Xq28 genes on the array, p = 2 × 10-8 for finding eight Xq28 genes among the most differentially expressed 86 unique genes, using an a priori null hypothesis of no preference for Xq28 genes. Five of the eight probes or genes belong to the large cancer/testis (C/T) antigen gene family, which include CTAG1, CTAG2, LOC389903, MAGEA2, and TRAG3. The array experiments suggested that the C/T genes are induced either directly or indirectly by the presence of RARβ2 compared to the empty vector cells. We originally evaluated and validated CTAG1 by Northern blot analysis (Figure 1B) and secondarily by qRT-PCR, both of which showed an approximately two-fold or greater expression in RARβ2-transduced cells/vector control cells. The mRNAs for two C/T genes, LOC389903 and TRAG3, share significant sequence overlap within their respective 60-mer probes, which may account for our inability to confirm expression rations for these genes by qRT-PCR. Similarly, the MAGEA2 60-mer probe differs by 1 or 2 nucleotides compared to MAGEA3, MAGEA6, or MAGEA12.
C/T genes consist of over forty families, and their expression may not be limited to germ line cells as a small subset are expressed in somatic cells, particularly the pancreas [39,40]. As discussed by Scanlon et al., many of these proteins are expressed in breast cancers [40]. Recently, Montel et al. performed expression profiling using two well-characterized clones derived from MDA-MB-435 cells, one that metastasized in xenograft experiments, and the other that failed to metastasize. They found that a number of C/T genes (including CTAG1, CTAG2, a colon cancer antigen 16, and MAGEA1) were elevated in the clone that did not metastasize [41]. Theoretically, an adaptive immune response may be responsible for biologic effect; however, members of the C/T family may have as yet unknown functions, independent of tumor antigenicity. For example, MAGE-11 was recently shown to be a co-activator of the androgen receptor in androgen dependent tissue. This latter discovery was conducted with yeast two-hybrid interactions and functional analysis [42]. Cronwright and colleagues found that the C/T genes are expressed in mesenchymal stem cells but are down-regulated following differentiation [43]. This latter observation supports the interesting hypothesis that cancer cells are stem cell derivatives.
Additional genes on Xq28
Three additional genes on Xq28 with apparent up-regulation by RARβ2 were confirmed by qRT-PCR (Table 2). ARHGAP4 is a GTPase-activating protein that may modulate stress fiber organization and function [44]. RPL10 is a 60S ribosomal protein that may possess other functions, including the ability to bind cYES kinase and affect SH3-mediated signal transduction cascades, resulting in tumor suppressor functionality [45]. Finally, SSR4, signal sequence receptor delta 4, was also elevated in RARβ2 cells. SSR4 encodes a protein subunit for a complex involved in protein secretion [OMIM:300090].
Is Xq28 enriched with beta retinoic acid response elements?
Six potential RAR binding elements were found during the search of the genomic coding sequences, including 1000 base pairs upstream and downstream from the start and stop codons, corresponding to known genes on Xq28. In this analysis, we simply searched for previously reported RARβ response elements [46], since this has the potential to provide additional information regarding regulatory pathways. Additional File 4: Supplemental Table 4 lists the sequences found and their locations. In comparison, no known RAREs were found in the regions of ZADH1, ZNF387, nor FABP6, none of which are Xq28 genes, though this does not rule out the possibility that novel RAREs or more distant RAREs contribute to the expression of these genes. These are intriguing results, given the large number of Xq28 genes with differential expression in this experiment. More research is needed to determine the significance of these results, both statistically (after determining the appropriate control for comparison of the number of elements found) and biologically (by testing of authentic RARβ2 binding within the larger context of the sites). We note that the genes in Supplemental Table 4 were all on the array but were not in top 100 ranked genes (Supplemental Table 2). One caveat to consider is that even if the putative RAREs are nearest to the noted genes, the RARE sites could act as distant transcriptional regulatory elements, as is the case for the HOX gene cluster, where RARE activation may occur at distances from 6 to more than 20 kilo bases [47]. In addition, RAREs may be found in 3' regions of genes [48]. If future expression profiling experiments using tumors should reveal more genes at Xq28, we could test as a control the possibility of retroviral integration at this site by fluorescence in situ hybridization and molecular mapping. Despite the preliminary nature of these results, they provide a basis for further research by us and others as well as demonstrate a potentially simple and powerful auxiliary analysis to array experiments that makes use of the growing amount of knowledge of consensus sequences.
Candidate genes involved in immune response and interferon signalling
As introduced above with the C/T genes on Xq28, a high proportion of the most extreme 100 median-log genes are associated with immune function. Table 3 lists additional immune function candidates. Of the nine genes in Table 3, five are known to be regulated through interferon or are involved in interferon signalling: IFI27 [rank 81], IFIT1 [rank 47], IFITM1 [rank 28], RIG-I [rank 79], and USP18 [rank 35].
Two genes, IFIT1 and IFITM1, are elevated in RARβ2 cells compared to control cells. IFIT1, which is induced by interferon alpha in response to viral infection, contains ten tetratricopeptide repeats, and these repeats are hypothesized to confer stability to the protein [49]. IFITM1 is induced by both alpha and gamma interferons. Based on the protein sequence, IFITM1 is likely an integral membrane protein, and exogenous expression in COS cells confers an antiproliferative, non-migrating phenotype [50]. A recent study of chronic myelocytic leukaemia patients determined that expression of IFITM1 in neoplastic cells predicted improved survival [51].
Two other interferon regulated genes are diminished in RARβ2 cells compared to control cells, IFI27 and USP18/UPB43. In one study, ~50% of human breast cancers expressed elevated IFI27 RNA, which was inversely correlated with estrogen receptor status [52]. Moreover, retinoids have been shown to suppress IFI27's expression [53]. Interferon signalling pathways are multi-nodal and involve molecules such as AKT, PCK, various STATs and NFκB [54]. The results of our expression microarray studies suggest that exogenous RARβ2 may modulate interferon signalling in MDA-MB-435 cells, leading to enhanced immune surveillance 'in the host.' Expression profiles indicated that RIG-I/DDX58 [rank 79] was elevated in RARβ2-transduced cells. This observation was confirmed with qRT-PCR for both the cell cultures (Table 3) and tumor tissue (Table 5). RIG-I is a retinoic acid inducible gene that encodes an RNA helicase/adaptor protein that is critical in sensing virus and propagating interferon gamma innate responses [55].
Candidate genes involved in transcription regulation
RARs are requisite transcription factors for a vast array of biological functions, such as embryogenesis, development, homeostasis, vision, and the immune system [56]. There are likely thousands of genes regulated directly through RAR transcriptional activation through canonical binding elements in gene promoters. One of the more well-characterized systems in development is the interplay between RARs and HOX gene expression [57]. Table 4 lists candidate genes that may have a transcriptional function in either permitting or abrogating metastasis. HOXB7, which has higher expression in the control cells, may have a role in neoplasia, as it was shown to block differentiation of myelocytic cells [58]. ZNF387/ST18 (see also Table 1) likely exhibits tumor suppressor function of as yet unknown mechanism.
The finding of a potential down-regulation of the cJUN mRNA is very intriguing, as RARs and members of the AP1 family such as JUN have major antagonistic interactions and both gene families are master regulators of many physiologic functions. JUN and AP1-associated family members are key transcription factors that block the antiproliferative functions of p53, p21, and p16 [59]. The mechanisms of cross-repression by RARs [60], and particularly by RARβ2 [61], which inhibit JUN and AP1 activity, include sequestering of co-activators and blockage of signalling pathways that recruit transcriptional activators such as CBP [62]. We demonstrated metastasis suppression in the absence of pharmacologic ligand, AT-RA [8]. Moreover, it has been independently demonstrated that RARβ can abrogate AP1 activity and inhibit anchorage independent colony formation of RARβ-stably transfected MDA-MB-231 breast cancer cells [63]. Two other gene candidates apparently regulated by the transcription factor NFκB, SECTM1 [Table 1, Rank 17] and SLC20A1 [Rank 29], are also elevated in RARβ2 cells, and NFκB signalling is likely an important pathway mediating metastasis suppressor functions, including RIG-I [7,64].
A cautionary note concerning unexpected findings in gene array validation
Ten of the top 100 gene candidates were SPP1 (osteopontin), and there were ten independent SPP1 probes of the same sequence on the Agilent arrays (Supplemental Table 2). The statistical results indicated that SPP1/osteopontin was elevated (~1.4 fold) in RARβ2-transduced cells. We carefully investigated this unexpected finding, as SPP1/osteopontin may be a key player in promoting metastasis [65] or breast cancer progression [66]. Moreover, SPP1/osteopontin may be a critical protein component in breast cancer's ability to metastasize to bone [67]. Northern blot analysis also suggested that SPP1/osteopontin was marginally elevated at the steady state level in RARβ2 cells (Figure 2A). qRT-PCR findings also showed an increase in the level of SPP1/osteopontin message comparing RARβ2- to vector control-cells. Finally, we analyzed the media of the vector control and RARβ2 cells by western blot for possible differences in protein levels. Interestingly, we detected both slightly higher protein levels plus a higher molecular weight form of SPP1/osteopontin protein in the vector control cells. There was no evidence for SPP1 protein in the western blot of cellular lysates, only from the concentrated media. (Figure 2B). SPP1/osteopontin can be post-translationally modified via a vast variety of moieties, including N- or O-linked glycosylation, phosphorylation, cross-linking with other proteins, deamidation, oxidation and carbamylation. Recently mass spectrometry experiments demonstrated that SPP1 has over thirty potential phosphorylation residues [68,69]. Whatever the sum of post-translational changes, the modified protein is likely a better substrate for possible interactions with fibronectin that may confer enhanced cellular migration.
Comparison of candidates derived from cell culture with primary xenograft tumors
We evaluated expression of selected genes of interest in two pairs of xenograft tumors from nude mice. We chose the gene candidates using the following criteria: 1) they had been validated by qRT-PCR in the cell cultures; and 2) they sampled the full rank range (Table 5). One candidate, CA14 [carbonic anhydrase 14, rank 86], with evidence for RARβ2-mediated diminished expression, is elevated in a variety of cancers, particularly hypoxic tumors [70]. Another candidate, NDRG1 [N-Myc down-stream regulated gene 1, rank 42], is regulated by the PTEN protein and may control metastasis in colon, prostate, and breast cancers [71,72]. Although the precise function(s) for NDRG1 has not been elucidated, Kim et al. demonstrated p53-dependent microtubule check point function for the protein in breast cancer cells [73]. Stein et al. found that NDRG1 was transcriptionally regulated by p53 and is necessary, but not sufficient, for p53 modulated apoptosis [74]. Thirteen of fifteen selected genes are consistent between cell cultures and the randomly selected tumors (p = 0.007). Thus our hypothesis testing, using cell lines from the xenograft studies, suggests that RARβ2-manifest anti-metastasis functions preexist in the transduced cells.
Conclusion
Our expression profiling experiments revealed both potential activated and repressed cellular activities in response to overexpression of RARβ2 (Figure 3). RARβ2 induces expression of several members of the C/T antigen family found on Xq28 as well as interferon signalling genes (Tables 2 and 3), suggesting that therapeutic modulation of RARβ2 in tumor cells may enhance endogenous immunity or potentially be additive to therapeutic immunomodulation. Our profiles indicate that RARβ2 induces a number of tumor suppressor functions (NDRG1, RPL10, and ST18) and known metastasis suppressors (NDRG1 and TYRP1). We find that the expression of a number of genes involved in cell adhesion (LSAMP, PCDH11Y, and CD64); nutrient availability (FABP6, SLC38A2, PCSK4, SRPB1); and transcription/AP1 activity (HOXB7 and JUN) are repressed in RARβ2 cells. Our findings suggest new arenas of complimentary or synergistic pathways in the regulation of metastasis.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
KS, BW, and ME designed the project. BW performed the microarray and validation experiments. ME performed statistical analysis and consensus sequence matching. MES performed the genome annotation for response elements on Xq28. MLD reviewed and contributed to the discussion of the immune function data. All authors contributed to the writing of the manuscript and have read and approved its revised and final drafts.
Pre-publication history
The pre-publication history for this paper can be accessed here:
Supplementary Material
Additional File 2
Supplemental Table 2 – Curated top 100 median log ratios
Click here for file
Additional File 3
Supplemental Table 3 – Comparison of expression of pooled clones, individual clones, and additional short term transduction
Click here for file
Additional File 4
Supplemental Table 4 – Binding elements detected on chromosome Xq28
Click here for file
Additional File 1
Supplemental Table 1 – Oligonucleotide sequences of primers for qRT-PCR
Click here for file
Acknowledgements
We thank Cecilia M. Giachelli for the gift of SPP1 antibody and recombinant protein as well as excellent discussions. This work was funded through the National Institutes of Health, NCI RO1CA82455 (KS), 1R29CA77607 (ME), and RO1CA101190 (MLD). This work was supported in part by the University of Washington NIEHS sponsored Center for Ecogenetics and Environmental Health, Grant: NIEHS P30ES07033.
Figures and Tables
Figure 1 Expression of RARβ2 and CTAG1 by Northern analysis. Northern analysis of RARβ2 (A) and CTAG1 (B) in RARβ2-transduced MDA-MB-435 cells compared to vector-control cells. Using total RNA used in the arrays, Northern blots were probed with 32P-dCTP labelled RARβ2 or CTAG1 cDNA. The RARβ2 probe consists of ~1.2 kb KpnI-BamH1 digest fragment of pSG RARβ2 ([75], from Pierre Chambon). The CTAG1 probe was generated from a 463 bp reverse transcriptase PCR product that encompasses the Agilent 60-mer. Blots were sequentially probed with a cDNA for RPLP0 (36B4) as a loading and transfer control. Quantitation of relative transcript levels was normalized to RPLP0 by phosphorimaging using exposure levels within the linear range of detection, avoiding saturation. The RARβ2 mRNA includes elements transcribed from the retroviral vector [8].
Figure 2 SPP1/Osteopontin expression and a validation dilemma. SPP1 expression comparison of RNA and protein. A. Northern blot analysis. Eight μg of whole cell total RNA extracts were probed with a 1384 bp fragment of SPP1 (see methods) or 800 bp, Pst1 fragment of RPLP0 (36B4). Phosphorimaging was used for detection and quantitation. B. Western immunoblot analysis. Two independent preparations of each clonal cell line were used. Three mL of serum free-media from 3 × 105 cell equivalents was immunoblotted, using a goat anti-SPP1/osteopontin polyclonal antibody (antibody and recombinant protein were a gift from Dr. CM Giachelli) and a secondary rabbit anti-goat antibody (Pierce).
Figure 3 Model of antimetastatic functions modulated by RARβ2. Expression of RARβ2 in metastatic breast cancer cells confers multiple antimetastatic properties, including induction of cancer antigens, tumor suppressors, and genes involved in interferon signalling. Other gene activities are suppressed through RARβ2 action: AP1, cell adhesion, and nutrient processes. All gene activities in this diagram have been confirmed by cell culture qRT-PCR. Gene names in italics (red), have also been confirmed using randomly selected xenograft primary tumors, comparing two pairs of vector control- and RARβ2-resected tumors.
Table 1 Top twenty ranked genes
Gene Symbol (Probe) Name †Fold change (qRT-PCR) Rank Function/Other information
LSAMP Limbic system associated membrane protein -3.2 (-3.1) 1 Immunoglobulin superfamily member; cell adhesion
SLC38A2 Solute carrier family 38 -2.2 (-2.1) 2 Amino acid transporter A2, regulated by growth factors & hormones
CTAG1 Cancer/testis antigen 1 2.2 *(1.6, 1.5) 3 (see also Table 2)
MGC2780 Hypothetical protein -2.2 (-5.5) 4 Unknown
§PCDH11Y Proto-cadherin on the Y -2.1 (-1.9) 5 Cell adhesion X chromosome homologue
PCSK4 Convertase; subtilisin/kexin 4 -2.0 (-1.6) 6 Cleave pro-hormones and pro-growth factors
ZADH1 Zinc-binding alcohol dehydrogenase domain containing protein 1.9 *(1.3, 1.2) 7 Alcohol dehydrogenase; possible retinoid metabolism
CD164 Cluster designation protein 164 -1.9 (-1.5) 8 Sialomucin, AKA, MGC-24
SFTPA2 Surfactant pulmonary associated protein 1.8 NC 9 Calcium dependent signalling innate immune response
FLJ23577 Hypothetical Kruppel-like protein -1.8 NC 10 unknown
SR-BP1, OPRS1 Sterol binding protein Opoid receptor sigma -1.7 NC 11 Sterol isomerase, ergosterol biosynthesis
FABP6 Fatty acid binding protein 6 -1.7 (-6.6) 12 Bile acid binding;
OR52P1 Olfactory receptor 52 pseudogene 1.7 NC 13 Pseudogene
CSTA Cystatin A -1.7 (-2.3) 14 Cysteine proteinase inhibitor
NPDC1 Neural differentiation & control protein 1.7 *(1.5, 1.3) 15 Inhibits proliferation
NXT2, P15-2 Nuclear transport factor 2-like export factor 1.7 NC 16 Exports RNA from nucleus
SECTM1 Secreted and transmembrane protein 1 -1.6 (-2.8) 17 Signal transduction activity with the NF-κB cascade
HKE2 HLA class II region KE2 gene 1.6 NC 18 Protein folding
NNMT Nicotinamide N-methyltransferase -1.6 (-4.6) 19 Methylates pyridines; biotransformation of drugs; enhanced in cancer cells
ZNF387, ST18 Zinc finger protein 387; breast cancer tumor suppressor 1.6 (2.7) 20 Transcription (see also Table 4)
† "Fold change," top number is the relative expression of RARβ2 transduced cells/empty vector control cells from Agilent arrays. The number in parentheses is the relative expression of the same cells from qRT-PCR analysis, normalized by respective qRT-PCR RPLP0 expression.
NC = not confirmed. In all cases, these genes have known isoform(s) [with exception OR52P1, which is a pseudogene] that overlap(s) the 60-mer probes on the arrays, which could account for the discrepancy between array finding and qRT-PCR.
* two qRT-PCRs (each in triplicate) of independent RNA samples
§shares near identical homology to homologue on X chromosome
Table 2 Gene cluster on Xq28
Gene Symbol (Probe) Name †Fold change (qRT-PCR) Rank Function/Other information
ARHGAP4 Rho-GAP GTPase 1.4 (2.2) 44 Possible inhibitory role on stress fiber organization and microtubules
§CTAG1 Cancer/testis antigen 1 2.2 (1.6) 3
§CTAG2 Cancer/testis antigen 2 1.3 (1.6) 93
§ *LOC389903 Cancer antigen 1.3 (NC) 75
§ **MAGEA2 Cancer antigen 1.3 (NC) 73
RPL10 Ribosomal protein 1.6 (1.4) 22 Tumor suppressor
SSR4 Signal sequence receptor/translocon associated protein delta 1.4 (1.2) 52 Type I receptor in endoplasmic reticulum/intracellular protein transport
§ *TRAG3 Tumor antigen 1.3 (NC) 77
† "Fold change," top number is the relative expression of RARβ2 transduced cells/empty vector control cells from Agilent arrays. The number in parentheses is the relative expression of the same cells from qRT-PCR analysis, normalized by respective qRT-PCR RPLP0 expression.
NC = differential expression was not confirmed by qRT-PCR
§ Genes in the cancer/testis antigen superfamily
* Shares significant overlapping sequence within their respective 60 mer probes, differing by 2–3 nucleotides
** MAGEA2 60 mer probe shares significant sequence identity with MAGEA-3, -6, and -12 (differs by 1–2 nucleotides)
Table 3 Immune response and interferon signalling
Gene Symbol (Probe) Function/family †Fold change (qRT-PCR) Rank Function/Other information
APP Amyloid precursor protein 1.6 (ND) 21 Proinflammatory
HLA-DRB1 Histocompatability complex, class II, DR beta 1 -1.3 (ND) 89 Signal transduction; antigen processing
IFI27 Interferon (IFN)-alpha inducible Protein 27 -1.3 (-1.9) 81 Integral membrane protein; unknown function
IFIT1 IFN-induced protein with tetratricopeptide repeats 1.4 (1.3) 47 Transmembrane protein; Interferon, retinoic acid inducible
IFITM1 IFN-induced transmembrane protein 1 1.5 (2.9) 28 Growth inhibition
PTMA/TMSA Prothymosin A -1.4 (ND) 53 Innate immunity; nuclear; associated with proliferation
RIG-I/DDX58 Retinoic acid inducible gene I/DEAD box 58 1.3 (1.7) 79 IFN signalling; RNA helicase
TRA1/GP96 Tumor rejection antigen 1 1.3 (NC) 98 Glycoprotein; stress response
USP18/UBP 43 Ubiquitin specific protease 18 -1.5 (-1.6) 35 Protease/ubiquitinase regulated by IFN
† "Fold change," top number is the relative expression of RARβ2 transduced cells/empty vector control cells from Agilent arrays. The number in parentheses is the relative expression of the same cells from qRT-PCR analysis, normalized by respective qRT-PCR RPLP0 expression.
ND = not done
NC = differential expression was not confirmed by qRT-PCR, and this gene has mRNA isoform that overlaps the 60-mer probe
Table 4 Transcriptional regulation: global and targeted
Gene Symbol (Probe) Name †Fold change (qRT-PCR) Rank Function/Other information
ANC 2H01/ZNF639/ZASC1 Zinc finger protein 639 1.3 (ND) 76 Kruppel domain
CITED4 CPB/p300-binding protein & interacting 4 1.3 (ND) 74 Transcriptional coactivator
FJL23577 Hypothetical, Kruppel-like protein -1.8 (NC) 10 Unknown
HIPK2 Homeodomain-p53-interacting protein -1.4 (ND) 51 Homeodomain interacting protein kinase 2
HOXB7 Homeobox gene B7 (ANTP family) -1.4 (-1.2) 38 Inhibits differentiation
*JUN cJUN -1.3 (-1.5) 90 AP1 complex
KIAA115 SIT4 phosphatase-associated protein-like -1.3 (ND) 82 Cyclin/G1 transcription
MGC15737/TCEAL3 Transcription elongation factor A-like 3 1.3 (ND) 100 Transcriptional elongation
ZNF387, ST18 Zinc finger protein 387; breast cancer suppressor of tumorigenicity 1.6 (2.7) 20 Transcription/Tumor suppressor (see also Table 2)
†"Fold change," top number is the relative expression of RARβ2 transduced cells/empty vector control cells from Agilent arrays. The number in parentheses is the relative expression of the same cells from qRT-PCR analysis, normalized by respective qRT-PCR RPLP0 expression
ND = not done
NC = not confirmed, has at least two isoforms that overlap the 60-mer probe
* "cross-talks" with RARs
Table 5 Comparison of expression: xenograft tumors and cultured cells
Gene symbol (Rank) Array fold change Cell fold change †Tissue fold change
ARGAP4 (44) 1.4 2.2 1.1
CA14 (86) -1.3 -1.9 -1.2
CTAG1 (3) 2.2 1.6 2.2
CTAG2 (93) 1.3 1.6 1.2
IFI27 (81) -1.3 -1.9 -1.4
‡IFIT1 (47) 1.4 1.3 -1.2
‡IFITM1 (28) 1.5 2.9 -1.5
JUN (90) -l.3 -1.5 -1.2
NDRG1 (42) 1.4 4.5 2.9
RPL10 (22) 1.6 1.4 1.2
RIG-I (79) 1.3 1.7 1.1
RNF28 (37) -1.5 *-2.2/-1.7 -1.4
SSR4 (52) 1.4 1.2 1.6
ST18 (20) 1.6 2.7 1.8
USP18 (35) -1.5 -1.6 -1.3
†Fold change for tissue was determined by dividing the normalized (to RPLP0) qRT-PCR value of two tumors that contained exogenous RARβ2 by the normalized value of two tumors that contained the empty vector.
‡Map Viewer or AceView indicate that there are overlapping mRNA isoforms with the 60 mer probe.
*two qRT-PCRs (each in triplicate) of independent RNA samples
==== Refs
Heimann R Hellman S Clinical progression of breast cancer malignant behavior: what to expect and when to expect it J Clin Oncol 2000 18 591 599 10653874
Morrow M Gradishar W Breast cancer Brit Med J 2002 324 410 414 11850376
Steeg PS Molecular biology of breast cancer metastasis. 'Has it spread?': disarming one of the most terrifying questions Breast Cancer Res 2000 2 396 399 11250732 10.1186/bcr85
Steeg PS Metastasis suppressors alter the signal transduction of cancer cells Nat Rev Cancer 2003 3 55 63 12509767 10.1038/nrc967
Singletary SE A working model for the time sequence of genetic changes in breast tumorigenesis J Am Coll Surg 2002 194 202 216 11848636 10.1016/S1072-7515(01)01108-5
Yoshida BA Sokoloff MM Welch DR Rinker-Schaeffer CW Metastasis-suppressor genes: a review and perspective on an emerging field J Natl Cancer Inst 2000 92 1717 1730 11058615 10.1093/jnci/92.21.1717
Shevde LA Welch DR Metastasis suppressor pathways – an evolving paradigm Cancer Lett 2003 198 1 20 12893425 10.1016/S0304-3835(03)00304-5
Treuting PM Chen LI Buetow BS Zeng W Birkebak TA Seewaldt VL Sommer KM Emond M Maggio-Price L Swisshelm K Retinoic acid receptor beta 2 inhibition of metastasis in mouse mammary gland xenografts Breast Cancer Res Treat 2002 72 79 88 12000222 10.1023/A:1014906529407
Roman SD Clarke CL Hall RE Alexander IE Sutherland RL Expression and regulation of retinoic acid receptors in human breast cancer cells Cancer Res 1992 52 2236 2242 1313739
Swisshelm K Ryan K Lee X Tsou HC Peacocke M Sager R Down-regulation of retinoic acid receptor beta in mammary carcinoma cell lines and its up-regulation in senescing normal mammary epithelial cells Cell Growth Differ 1994 5 133 141 8180126
Widschwendter M Berger J Daxenbichler G Muller-Holzner E Widschwendter A Mayr A Marth C Zeimet AG Loss of retinoic acid receptor beta expression in breast cancer and morphologically normal adjacent tissue but not in the normal breast tissue distant from the cancer Cancer Res 1997 57 4158 4161 9331065
Xu XC Sneige N Liu X Nandagiri R Lee JJ Lukmanji F Hortobagyi G Lippman SM Dhingra K Lotan R Progressive decrease in nuclear retinoic acid receptor beta messenger RNA level during breast carcinogenesis Cancer Res 1997 57 4992 4996 9371489
Sirchia SM Ferguson AT Sironi E Subramanyan S Orlandi R Sukumar S Sacchi N Evidence of epigenetic changes affecting the chromatin state of the retinoic acid receptor beta 2 promoter in breast cancer cells Oncogene 2000 19 1556 1563 10734315 10.1038/sj.onc.1203456
Sirchia SM Ren M Pili R Sironi E Somenzi G Ghidoni R Toma S Nicolo G Sacchi N Endogenous reactivation of the RAR beta 2 tumor suppressor gene epigenetically silenced in breast cancer Cancer Res 2002 62 2455 2461 11980632
Mehrotra J Vali M McVeigh M Kominsky SL Fackler MJ Lahti-Domenici J Polyak K Sacchi N Garrett-Mayer E Argani P Sukumar S Very high frequency of hypermethylated genes in breast cancer metastasis to the bone, brain, and lung Clin Cancer Res 2004 10 3104 3109 15131050 10.1158/1078-0432.CCR-03-0118
Sommer KM Chen LI Treuting PM Smith LT Swisshelm K Elevated retinoic acid receptor beta(4) protein in human breast tumor cells with nuclear and cytoplasmic localization Proc Natl Acad Sci U S A 1999 96 8651 8656 10411930 10.1073/pnas.96.15.8651
Chen LI Sommer KM Swisshelm K Downstream codons in the retinoic acid receptor beta -2 and beta -4 mRNAs initiate translation of a protein isoform that disrupts retinoid-activated transcription J Biol Chem 2002 277 35411 35421 12118004 10.1074/jbc.M202717200
Zhuang Y Faria TN Chambon P Gudas LJ Identification and characterization of retinoic acid receptor beta 2 target genes in F9 teratocarcinoma cells Mol Cancer Res 2003 1 619 630 12805409
Toulouse A Loubeau M Morin J Pappas JJ Wu J Bradley WE RAR beta involvement in enhancement of lung tumor cell immunogenicity revealed by array analysis Faseb J 2000 14 1224 1232 10834944
Seewaldt VL Johnson BS Parker MB Collins SJ Swisshelm K Expression of retinoic acid receptor β mediates retinoic acid- induced growth arrest and apoptosis in breast cancer cells Cell Growth Differ 1995 6 1077 1088 8519684
R Development Core Team R statistical software package (R: A language and environment for statistical computing) 2005 Current Version: 2.2.0
Qin LX Kerr KF Empirical evaluation of data transformations and ranking statistics for microarray analysis Nucleic Acids Res 2004 32 5471 5479 15479783 10.1093/nar/gkh866
Breitling R Armengaud P Amtmann A Herzyk P Rank products: a simple, yet powerful, new method to detect differentially regulated genes in replicated microarray experiments FEBS Lett 2004 573 83 92 15327980 10.1016/j.febslet.2004.07.055
Pollard KS van der Laan MJ Resampling-based multiple testing: asymptotic control of type I error and applications to gene expression data UC Berkeley Division of Biostatistics Working Paper Series 2003 Working Paper 121
Genome Bioinformatics Group UCSC Human Reference Sequence, NCBI Build 35 (hg17)
Wu S Manber U AGREP – A fast approximate pattern-matching tool Proceedings of the Winter USENIX Technical Conference: San Francisco 1992 153 162
Rozen S Skaletsky HJ Krawetz S, Misener S Primer3 and the WWW for general users and for biologist programers Bioinformatics Methods and Protocols: Methods in Molecular Biology 2000 Totawa, NJ: Humana Press 365 386
Hughes TR Mao M Jones AR Burchard J Marton MJ Shannon KW Lefkowitz SM Ziman M Schelter JM Meyer MR Kobayashi S Davis C Dai H He YD Stephaniants SB Cavet G Walker WL West A Coffey E Shoemaker DD Stoughton R Blanchard AP Friend SH Linsley PS Expression profiling using microarrays fabricated by an ink-jet oligonucleotide synthesizer Nat Biotechnol 2001 19 342 347 11283592 10.1038/86730
Pfaffl MW A new mathematical model for relative quantification in real-time RT-PCR Nucleic Acids Res 2001 29 e45 11328886 10.1093/nar/29.9.e45
National Center for Biotechnology Information Map Viewer
National Center for Biotechnology Information AceView
Swisshelm K Machl A Planitzer S Robertson R Kubbies M Hosier S SEMP1, a senescence-associated cDNA isolated from human mammary epithelial cells, is a member of an epithelial membrane protein superfamily Gene 1999 226 285 295 9931503 10.1016/S0378-1119(98)00553-8
Tuck AB Arsenault DM O'Malley FP Hota C Ling MC Wilson SM Chambers AF Osteopontin induces increased invasiveness and plasminogen activator expression of human mammary epithelial cells Oncogene 1999 18 4237 4246 10435636 10.1038/sj.onc.1202799
Pichierri P Rosselli F Franchitto A Werner's syndrome protein is phosphorylated in an ATR/ATM-dependent manner following replication arrest and DNA damage induced during the S phase of the cell cycle Oncogene 2003 22 1491 1500 12629512 10.1038/sj.onc.1206169
Hauksdottir H Farboud B Privalsky ML Retinoic acid receptors beta and gamma do not repress, but instead activate target gene transcription in both the absence and presence of hormone ligand Mol Endocrinol 2003 17 373 385 12554770 10.1210/me.2002-0340
Hollingsworth MA Swanson BJ Mucins in cancer: protection and control of the cell surface Nat Rev Cancer 2004 4 45 60 14681689 10.1038/nrc1251
Spencer ML Theodosiou M Noonan DJ NPDC-1, a novel regulator of neuronal proliferation, is degraded by the ubiquitin/proteasome system through a PEST degradation motif J Biol Chem 2004 279 37069 37078 15229225 10.1074/jbc.M402507200
Jandrig B Seitz S Hinzmann B Arnold W Micheel B Koelble K Siebert R Schwartz A Ruecker K Schlag PM Scherneck S Rosenthal A ST18 is a breast cancer tumor suppressor gene at human chromosome 8q11.2 Oncogene 2004 23 9295 9302 15489893
Zendman AJ Ruiter DJ Van Muijen GN Cancer/testis-associated genes: identification, expression profile, and putative function J Cell Physiol 2003 194 272 288 12548548 10.1002/jcp.10215
Scanlan MJ Simpson AJ Old LJ The cancer/testis genes: review, standardization, and commentary Cancer Immun 2004 4 1 14738373
Montel V Huang TY Mose E Pestonjamasp K Tarin D Expression profiling of primary tumors and matched lymphatic and lung metastases in a xenogeneic breast cancer model Am J Pathol 2005 166 1565 1579 15855655
Bai S He B Wilson EM Melanoma antigen gene protein MAGE-11 regulates androgen receptor function by modulating the interdomain interaction Mol Cell Biol 2005 25 1238 1257 15684378 10.1128/MCB.25.4.1238-1257.2005
Cronwright G Le Blanc K Gotherstrom C Darcy P Ehnman M Brodin B Cancer/testis antigen expression in human mesenchymal stem cells: down-regulation of SSX impairs cell migration and matrix metalloproteinase 2 expression Cancer Res 2005 65 2207 2215 15781633 10.1158/0008-5472.CAN-04-1882
Tribioli C Droetto S Bione S Cesareni G Torrisi MR Lotti LV Lanfrancone L Toniolo D Pelicci P An X chromosome-linked gene encoding a protein with characteristics of a rhoGAP predominantly expressed in hematopoietic cells Proc Natl Acad Sci U S A 1996 93 695 699 8570618 10.1073/pnas.93.2.695
Oh HS Kwon H Sun SK Yang CH QM, a putative tumor suppressor, regulates proto-oncogene c-yes J Biol Chem 2002 277 36489 36498 12138090 10.1074/jbc.M201859200
Mangelsdorf DJ Umesona K Evans RM Sporn MB, Roberts AB, Goodman DS The retinoid receptors The Retinoids: Biology, Chemistry, and Medicine 1994 2 New York: Raven Press 319 349
Oosterveen T van Vliet P Deschamps J Meijlink F The direct context of a hox retinoic acid response element is crucial for its activity J Biol Chem 2003 278 24103 24107 12697756 10.1074/jbc.M300774200
Zhang F Popperl H Morrison A Kovacs EN Prideaux V Schwarz L Krumlauf R Rossant J Featherstone MS Elements both 5' and 3' to the murine Hoxd4 gene establish anterior borders of expression in mesoderm and neurectoderm Mech Dev 1997 67 49 58 9347914 10.1016/S0925-4773(97)00104-4
Main ER Stott K Jackson SE Regan L Local and long-range stability in tandemly arrayed tetratricopeptide repeats Proc Natl Acad Sci U S A 2005 102 5721 5726 15824314 10.1073/pnas.0404530102
Deblandre GA Marinx OP Evans SS Majjaj S Leo O Caput D Huez GA Wathelet MG Expression cloning of an interferon-inducible 17-kDa membrane protein implicated in the control of cell growth J Biol Chem 1995 270 23860 23866 7559564 10.1074/jbc.270.40.23860
Akyerli CB Beksac M Holko M Frevel M Dalva K Ozbek U Soydan E Ozcan M Ozet G Ilhan O Gurman G Akan H Williams BR Ozcelik T Expression of IFITM1 in chronic myeloid leukemia patients Leuk Res 2005 29 283 286 15661263 10.1016/j.leukres.2004.07.007
Rasmussen UB Wolf C Mattei MG Chenard MP Bellocq JP Chambon P Rio MC Basset P Identification of a new interferon-alpha-inducible gene (p27) on human chromosome 14q32 and its expression in breast carcinoma Cancer Res 1993 53 4096 4101 8358738
Suomela S Cao L Bowcock A Saarialho-Kere U Interferon alpha-inducible protein 27 (IFI27) is upregulated in psoriatic skin and certain epithelial cancers J Invest Dermatol 2004 122 717 721 15086558 10.1111/j.0022-202X.2004.22322.x
Caraglia M Marra M Pelaia G Maselli R Caputi M Marsico SA Abbruzzese A Alpha-interferon and its effects on signal transduction pathways J Cell Physiol 2005 202 323 335 15389589 10.1002/jcp.20137
Yoneyama M Kikuchi M Natsukawa T Shinobu N Imaizumi T Miyagishi M Taira K Akira S Fujita T The RNA helicase RIG-I has an essential function in double-stranded RNA-induced innate antiviral responses Nat Immunol 2004 5 730 737 15208624 10.1038/ni1087
Lefebvre P Martin PJ Flajollet S Dedieu S Billaut X Lefebvre B Transcriptional activities of retinoic acid receptors Vitam Horm 2005 70 199 264 15727806 10.1016/S0083-6729(05)70007-8
Serpente P Tumpel S Ghyselinck NB Niederreither K Wiedemann LM Dolle P Chambon P Krumlauf R Gould AP Direct crossregulation between retinoic acid receptor {beta} and Hox genes during hindbrain segmentation Development 2005 132 503 513 15634700 10.1242/dev.01593
Yaron Y McAdara JK Lynch M Hughes E Gasson JC Identification of novel functional regions important for the activity of HOXB7 in mammalian cells J Immunol 2001 166 5058 5067 11290787
Shaulian E Karin M AP-1 as a regulator of cell life and death Nat Cell Biol 2002 4 E131 136 11988758 10.1038/ncb0502-e131
Caelles C Gonzalez-Sancho JM Munoz A Nuclear hormone receptor antagonism with AP-1 by inhibition of the JNK pathway Genes Dev 1997 11 3351 3364 9407028
Soprano DR Scanlon E Shukri M Zhang ZP Soprano KJ Murine RARbeta4 displays reduced transactivation activity, lower affinity for retinoic acid, and no anti-AP1 activity J Cell Biochem 2000 77 604 614 10771516 10.1002/(SICI)1097-4644(20000615)77:4<604::AID-JCB8>3.0.CO;2-T
Benkoussa M Brand C Delmotte MH Formstecher P Lefebvre P Retinoic acid receptors inhibit AP1 activation by regulating extracellular signal-regulated kinase and CBP recruitment to an AP1-responsive promoter Mol Cell Biol 2002 22 4522 4534 12052862 10.1128/MCB.22.13.4522-4534.2002
Lin F Xiao D Kolluri SK Zhang X Unique anti-activator protein-1 activity of retinoic acid receptor beta Cancer Res 2000 60 3271 3280 10866321
Kim JH Kim B Cai L Choi HJ Ohgi KA Tran C Chen C Chung CH Huber O Rose DW Sawyers CL Rosenfeld MG Baek SH Transcriptional regulation of a metastasis suppressor gene by Tip60 and beta-catenin complexes Nature 2005 434 921 926 15829968 10.1038/nature03452
Wai PY Kuo PC The role of osteopontin in tumor metastasis J Surg Res 2004 121 228 241 15501463 10.1016/j.jss.2004.03.028
Cook AC Tuck AB McCarthy S Turner JG Irby RB Bloom GC Yeatman TJ Chambers AF Osteopontin induces multiple changes in gene expression that reflect the six "hallmarks of cancer" in a model of breast cancer progression Mol Carcinog 2005 43 225 236 15864800 10.1002/mc.20105
Kang Y Siegel PM Shu W Drobnjak M Kakonen SM Cordon-Cardo C Guise TA Massague J A multigenic program mediating breast cancer metastasis to bone Cancer Cell 2003 3 537 549 12842083 10.1016/S1535-6108(03)00132-6
Christensen B Nielsen MS Haselmann KF Petersen TE Sorensen ES Post-translationally modified residues of native human osteopontin are located in clusters: identification of 36 phosphorylation and five O-glycosylation sites and their biological implications Biochem J 2005 390 285 292 15869464
Keykhosravani M Doherty-Kirby A Zhang C Brewer D Goldberg HA Hunter GK Lajoie G Comprehensive identification of post-translational modifications of rat bone osteopontin by mass spectrometry Biochemistry 2005 44 6990 7003 15865444 10.1021/bi050109p
Potter CP Harris AL Diagnostic, prognostic and therapeutic implications of carbonic anhydrases in cancer Br J Cancer 2003 89 2 7 12838292 10.1038/sj.bjc.6600936
Guan RJ Ford HL Fu Y Li Y Shaw LM Pardee AB Drg-1 as a differentiation-related, putative metastatic suppressor gene in human colon cancer Cancer Res 2000 60 749 755 10676663
Bandyopadhyay S Pai SK Hirota S Hosobe S Tsukada T Miura K Takano Y Saito K Commes T Piquemal D Watabe M Gross S Wang Y Huggenvik J Watabe K PTEN up-regulates the tumor metastasis suppressor gene Drg-1 in prostate and breast cancer Cancer Res 2004 64 7655 7660 15520163 10.1158/0008-5472.CAN-04-1623
Kim KT Ongusaha PP Hong YK Kurdistani SK Nakamura M Lu KP Lee SW Function of Drg1/Rit42 in p53-dependent mitotic spindle checkpoint J Biol Chem 2004 279 38597 38602 15247272 10.1074/jbc.M400781200
Stein S Thomas EK Herzog B Westfall MD Rocheleau JV Jackson RS 2ndWang M Liang P NDRG1 is necessary for p53-dependent apoptosis J Biol Chem 2004 279 48930 48940 15377670 10.1074/jbc.M400386200
Brand N Petkovich M Krust A Chambon P de The H Marchio A Tiollais P Dejean A Identification of a second human retinoic acid receptor Nature 1988 332 850 853 2833708 10.1038/332850a0
|
16255778
|
PMC1283145
|
CC BY
|
2021-01-04 16:03:05
|
no
|
BMC Cancer. 2005 Oct 28; 5:140
|
utf-8
|
BMC Cancer
| 2,005 |
10.1186/1471-2407-5-140
|
oa_comm
|
==== Front
Biomed Eng OnlineBioMedical Engineering OnLine1475-925XBioMed Central London 1475-925X-4-601625313410.1186/1475-925X-4-60ResearchReliability of old and new ventricular fibrillation detection algorithms for automated external defibrillators Amann Anton [email protected] Robert [email protected] Karl [email protected] Innsbruck Medical University, Department of Anesthesia and General Intensive Care, Anichstr. 35, A-6020 Innsbruck, Austria and Department of Environmental Sciences, ETH-Hönggerberg, CH-8093 Zürich, Switzerland2 Research Center PPE, FH-Vorarlberg, Achstr. 1, A-6850 Dornbirn, Austria2005 27 10 2005 4 60 60 6 5 2005 27 10 2005 Copyright © 2005 Amann et al; licensee BioMed Central Ltd.2005Amann et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
A pivotal component in automated external defibrillators (AEDs) is the detection of ventricular fibrillation by means of appropriate detection algorithms. In scientific literature there exists a wide variety of methods and ideas for handling this task. These algorithms should have a high detection quality, be easily implementable, and work in real time in an AED. Testing of these algorithms should be done by using a large amount of annotated data under equal conditions.
Methods
For our investigation we simulated a continuous analysis by selecting the data in steps of one second without any preselection. We used the complete BIH-MIT arrhythmia database, the CU database, and the files 7001 – 8210 of the AHA database. All algorithms were tested under equal conditions.
Results
For 5 well-known standard and 5 new ventricular fibrillation detection algorithms we calculated the sensitivity, specificity, and the area under their receiver operating characteristic. In addition, two QRS detection algorithms were included. These results are based on approximately 330 000 decisions (per algorithm).
Conclusion
Our values for sensitivity and specificity differ from earlier investigations since we used no preselection. The best algorithm is a new one, presented here for the first time.
ventricular fibrillation detectionautomated external defibrillator (AED)ventricular fibrillation (VF)sinus rhythm (SR)ECG analysis
==== Body
Background
Sudden cardiac arrest is a major public health problem and one of the leading causes of mortality in the western world. In most cases, the mechanism of onset is a ventricular tachycardia that rapidly progresses to ventricular fibrillation [1]. Approximately one third of these patients could survive with the timely employment of a defibrillator.
Besides manual defibrillation by an emergency paramedic, bystander defibrillation with (semi-)automatic external defibrillators (AEDs) has also been recommended for resuscitation. These devices analyze the electrocardiogram (ECG) of the patient and recognize whether a shock should be delivered or not, as, e.g., in case of ventricular fibrillation (VF). It is of vital importance that the ECG analysis system used by AEDs differentiates well between VF and a stable but fast sinus rhythm (SR). An AED should not deliver a shock to a collapsed patient not in cardiac arrest. On the other hand, a successfully defibrillated patient should not be defibrillated again.
The basis of such ECG analysis systems of AEDs is one or several mathematical ECG analysis algorithms. The main purpose of this paper is to compare various old and new algorithms in a standardized way. To gain insight into the quality of an algorithm for ECG analysis, it is essential to test the algorithms under equal conditions with a large amount of data, which has already been commented on by qualified cardiologists.
Commonly used annotated databases are Boston's Beth Israel Hospital and MIT arrhythmia database (BIH-MIT), the Creighton University ventricular tachyarrhythmia database (CU), and the American Heart Association database (AHA).
We used the complete CU and BIH-MIT arrhythmia database, and the files 7001 – 8210 of the AHA database, [2], [3,4]. For each algorithm approximately 330 000 decisions had been calculated. No preselection of certain ECG episodes was made, which mimics the situation of a bystander more accurately. In this investigation we analyzed 5 well-known standard and 5 new ventricular fibrillation detection algorithms. In addition, two QRS detection algorithms were included. The results are expressed in the quality parameters Sensitivity and Specificity. In addition to these two parameters, we calculated the Positive Predictivity and Accuracy of the investigated algorithms. Furthermore, the calculation time in comparison to the duration of the real data was calculated for the different algorithms. The calculation times were obtained by analyzing data from the CU database only.
The quality parameters were obtained by comparing the VF/no VF decisions suggested by the algorithm with the annotated decisions suggested by cardiologists. The cardiologists' decisions are considered to be correct. We distinguished only between ventricular fibrillation and no ventricular fibrillation, since the annotations do not include a differentiation between ventricular fibrillation and ventricular tachycardia. The closer the quality parameters are to 100%, the better the algorithm works. Since an AED has to differentiate between VF and no VF, the sensitivity and specificity are the appropriate parameters. To represent the quality of an algorithm by its sensitivity and specificity bears some problems. A special algorithm can have a high sensitivity, but a low specificity, or conversely. Which one is better? To arrive at a common and single quality parameter, we use the receiver operating characteristic (ROC). The sensitivity is plotted in dependence of (1 - specificity), where different points in the plot are obtained by varying the critical threshold parameter in the decision stage of the algorithm. By calculating the area under the ROC curve (we call this value "integrated receiver operating characteristic", and denote it by IROC), it is possible to compare different algorithms by one single value. Figure 1 shows a typical example of an ROC curve.
Figure 1 Receiver operating characteristic for the algorithm "complexity measure" described in the introduction, for a window length of 8 s. The calculated value for the area under the curve, IROC, is 0.87.
Section 1 provides the necessary background for the algorithms under investigation. Section 2 describes the methods of evaluation and represents our results in Table 1, 2, 3 and Figure 5 and 6. A discussion of the results follows in Section 3. Appendix A recalls the basic definitions of the quality parameters Sensitivity, Specificity, Positive Predictivity, Accuracy, and ROC curve. In Appendix B we provide more details on one of the new algorithms.
Table 1 >Quality of ventricular fibrillation detection algorithms (sensitivity (Sns), specificity (Spc), integrated receiver operating characteristic (IROC)) in per cent, rounded on 3 significant digits, wl = window length in seconds. (*) ... no appropriate parameter exists.
Data Source MIT DB CU DB AHA DB overall result
Parameter wl Sns. Spc. Sns. Spc. Sns. Spc. Sns. Spc. IROC
TCI 3 82.5 78.1 73.5 62.6 75.2 78.3 75.0 77.5 82
TCI 8 74.5 83.9 71.0 70.5 75.7 86.9 75.1 84.4 82
ACF95 8 33.2 45.9 38.1 58.9 51.5 52.2 49.6 49.0 49
ACF99 8 59.4 30.1 54.7 49.3 71.5 40.3 69.2 35.0 49
VF 4 36.7 100 32.2 99.5 17.6 99.9 19.6 99.9 85
VF 8 29.4 100 30.8 99.5 16.9 100 18.8 100 87
SPEC 8 23.1 100 29.0 99.3 29.2 99.8 29.1 99.9 89
CPLX 8 6.3 92.4 56.4 86.6 60.2 91.9 59.2 92.0 87
STE 8 54.5 83.4 52.9 66.6 49.6 81.0 50.1 81.7 67
MEA 8 62.9 80.8 60.1 87.5 49.8 88.6 51.2 84.1 82
SCA 8 72.4 98.0 67.7 94.9 71.7 99.7 71.2 98.5 92
WVL1 8 28.7 99.9 26.2 99.4 26.8 99.5 26.7 99.7 80
WVL2 8 81.1 89.0 61.0 72.1 73.5 89.6 72.0 88.4 (*)
LI 8 3.1 95.1 7.5 94.8 9.3 92.0 9.0 93.9 58
TOMP 8 68.5 40.6 71.3 48.4 95.9 39.7 92.5 40.6 67
Table 2 Quality of ventricular fibrillation detection algorithms (positive predictivity (PP), accuracy (Ac), calculation time(ct)) for a window length of 8 seconds. Positive predictivity and accuracy in per cent, rounded on 3 digits; calculation time in per cent of the real time of the data, rounded on 2 digits, wl = window length in seconds.
Data Source MIT DB CU DB AHA DB overall result
Parameter wl PP. Ac. PP. Ac. PP. Ac. PP. Ac. ct.
TCI 3 0.6 78.1 33.9 64.8 41.7 77.8 23.7 77.2 1.5
TCI 8 0.8 83.9 38.9 70.6 54.4 84.9 31.1 83.6 2.1
ACF95 8 0.1 45.9 19.7 54.5 18.2 52.1 8.3 49.0 3.6
ACF99 8 0.1 30.2 22.2 50.4 19.9 45.7 9.1 37.9 3.6
VF 4 91.3 99.9 94.0 85.5 98.0 85.8 97.0 93.1 1.4
VF 8 82.4 99.9 94.5 85.2 98.9 85.7 97.7 93.0 1.9
SPEC 8 60.6 99.8 92.0 84.6 97.3 87.7 96.1 93.8 1.9
CPLX 8 0.1 92.3 52.7 80.3 60.7 86.5 40.8 89.2 2.5
STE 8 0.5 83.4 29.5 63.8 35.1 75.6 20.4 79.0 1.9
MEA 8 0.5 80.8 56.0 81.8 47.5 81.9 23.2 81.3 2.5
SCA 8 5.6 97.9 77.8 89.2 98.0 94.9 81.6 96.2 5.9
WVL1 8 38.9 99.8 92.1 84.1 91.8 87.0 90.5 93.5 1.9
WVL2 8 1.2 88.9 36.6 69.8 59.3 86.8 36.8 87.0 40
LI 8 0.1 94.9 27.5 76.5 19.4 77.8 12.1 86.6 15
TOMP 8 0.2 40.6 26.7 53.2 24.8 49.4 12.7 45.0 0.84
Table 3 Sensitivity of ventricular fibrillation detection algorithms in per cent, wl = window length in seconds.
Parameter wl Sns. if Spc. = 95 Sns. if Spc. = 99
TCI 3 15.0 1.0
TCI 8 25.3 1.3
ACF95 8 3.0 0.6
ACF99 8 3.0 0.6
VF 4 71.0 59.2
VF 8 73.4 59.7
SPEC 8 69.8 58.9
CPLX 8 38.8 5.8
STE 8 29.4 10.8
MEA 8 7.0 0.5
SCA 8 79.0 66.4
WVL1 8 56.7 35.2
LI 8 7.3 1.4
TOMP 8 9.1 1.8
Figure 5 ROC curve for the algorithms ACF, STE, MEA, WVL1, LI, TOMP.
Figure 6 ROC curve for the algorithms TCI, VF, SPEC, CPLX, SCA.
1 Methods
The ventricular fibrillation detection algorithms considered here are partly taken from the scientific literature, five of them are new. Some of them have been evaluated in [5] and [6].
For all algorithms we used the same prefiltering process. First, a moving average filter of order 5 is applied to the signal. (Note 1: Fixing the order of 5 seems to bear a problem. Data with different frequencies (AHA, CU ... 250 Hz, BIH-MIT ... 360 Hz) are filtered in a different way. But this is neglectable, when the Butterworth filter is applied afterwards.) This filter removes high frequency noise like interspersions and muscle noise. Then, a drift suppression is applied to the resulting signal. This is done by a high pass filter with a cut off frequency of one Hz. Finally, a low pass Butterworth filter with a limiting frequency of 30 Hz is applied to the signal in order to suppress needless high-frequency information even more. This filtering process is carried out in a MATLAB routine. It uses functions from the "Signal Processing Toolbox". (Note 2: We used MATLAB R13 – R14 and "Signal Processing Toolbox" version 6.1 – 6.3 on a Power Mac G5, 2 GHz.) In order to obtain the ROC curve we have to change a parameter which we call "critical threshold parameter" below.
TCI algorithm
The threshold crossing intervals algorithm (TCI) [7] operates in the time domain. Decisions are based on the number and position of signal crossings through a certain threshold.
First, the digitized ECG signal is filtered by the procedure mentioned above. Then a binary signal is generated from the preprocessed ECG data according to the position of the signal above or below a given threshold. The threshold value is set to 20% of the maximum value within each one-second segment S and recalculated every second. Subsequent data analysis takes place over successive one-second stages. The ECG signal may cross the detection threshold one or more times, and the number of pulses is counted. For each stage, the threshold crossing interval TCI is the average interval between threshold crossings and is calculated as follows
Figure 2 illustrates the situation.
Figure 2 Binary signal with 2 pulses in threshold crossing intervals algorithm.
Here, N is the number of pulses in S. t1 is the time interval from the beginning of S back to the falling edge of the preceding pulse. t2 is the time interval from the beginning of S to the start of the next pulse. t3 is the interval between the end of the last pulse and the end of S and t4 is the time from the end of S to the start of the next pulse.
If TCI ≥ TCI0 = 400 ms, SR is diagnosed. Otherwise sequential hypothesis testing [7] is used to separate ventricular tachycardia (VT) from VF.
As stated above, the original algorithm works with single one-second time segments, (see [7], page 841). To achieve this the algorithm picks a 3-second episode. The first second and the third second are used to determine t1 and t4. The 2nd second yields the value for TCI. When picking an 8-second episode we can hence evaluate 6 consecutive TCI values. Final SR decision is taken if diagnosed in four or more segments otherwise the signal is classified as VF.
The critical threshold parameter to obtain the ROC is TCI0.
ACF algorithm
The autocorrelation algorithms ACF95 (Note 3: Probability of 95% in the Fisher distribution → α = 0.05 in F(α, k1, k2) with k1 = 1, k2 = 5) and ACF99 (Note 4: Probability of 99% in the Fisher distribution → α = 0.01 in F(α, k1, k2) with k1 = 1, k2 = 5) [8] analyze the periodicities within the ECG. Given a discrete signal x(m), the short-term autocorrelation function (ACF) of x(m) with a rectangular window is calculated by
Here, this technique is used to separate VT and SR from VF. It is assumed that VF signals are more or less aperiodic and SR signals are approximately periodic.
This assumption is however questionable since VF signals may have a cosine like shape. Compare this assumption with the assumption made by the algorithm in the next subsection.
Note that the autocorrelation function of a function f is connected to the Power Spectrum of by the "Wiener-Khinchin Theorem".
The detection algorithm performs a linear regression analysis of ACF peaks. An order number i is given to each peak according to its amplitude. So, the highest peak is called P0, etc., ranged by decreasing amplitudes. In a SR signal, which is considered to be periodic or nearly periodic, a linear relationship should exist between the peaks lag and their index number i. No such relationship should exist in VF signals. The linear regression equation of the peak order and its corresponding lag of m peaks in the ACF is described as
yi = a + bxi, (3)
where xi is the peak number (from 0 to (m - 1)), and yi is the lag of Pi.
In this study, m = 7. The variance ratio VR is defined by
where
If VR ≥ VR0 is greater than the Fisher statistics for degrees of freedom k1 = 1 and k2 = m - 2 = 5 with 95% (99%) probability, the rhythm is classified by ACF95 (ACF99) to be SR, otherwise it is VF. The critical threshold parameter to obtain the ROC is VR0, (VR0 ≈ 6.61(16.3) at 95% (99%)).
VF filter algorithm
The VF filter algorithm (VF) [9] applies a narrow band elimination filter in the region of the mean frequency of the considered ECG signal.
After preprocessing, a narrow band-stop filter is applied to the signal, with central frequency being equivalent to the mean signal frequency fm. Its calculated output is the "VF filter leakage". The VF signal is considered to be approximately of sinusoidal waveform.
The number N of data points in an average half period N = T/2 = 1/(2fm) is given by
where Vi are the signal samples, m is the number of data points in one mean period, and ⌊...⌋ denotes the floor function. The narrow band-stop filter is simulated by combining the ECG data with a copy of the data shifted by a half period. The VF-filter leakage l is computed as
In the original paper [9] this algorithm is invoked only if no QRS complexes or beats are detected. This is done by other methods. Since we employ no prior QRS detection, we use the thresholds suggested by [5]. If the signal is higher than a third of the amplitude of the last by the VF-filter detected QRS complex in a previous segment and the leakage is smaller than l0 = 0.406, VF is identified. Otherwise the leakage must be smaller than l0 = 0.625 in order to be classified as VF.
The critical threshold parameter to obtain the ROC is the leakage l0.
Spectral algorithm
The spectral algorithm (SPEC) [10] works in the frequency domain and analyses the energy content in different frequency bands by means of Fourier analysis.
The ECG of most normal heart rhythms is a broadband signal with major harmonics up to about 25 Hz. During VF, the ECG becomes concentrated in a band of frequencies between 3 and 10 Hz (cf. [11,12], with particularly low frequencies of undercooled victims).
After preprocessing, each data segment is multiplied by a Hamming window and then the ECG signal is transformed into the frequency domain by fast Fourier transform (FFT). The amplitude is approximated in accordance with ref. [10] by the sum of the absolute value of the real and imaginary parts of the complex coefficients. (Note 5: Normally one would take the modulus of complex amplitudes.) Let Ω be the frequency of the component with the largest amplitude (called the peak frequency) in the range 0.5 – 9 Hz. Then amplitudes whose value is less than 5% of the amplitude of Ω are set to zero. Four spectrum parameters are calculated, the normalized first spectral moment M
jmax being the index of the highest investigated frequency, and A1, A2, A3. Here wj denotes the j-th frequency in the FFT between 0 Hz and the minimum of (20Ω, 100 Hz) and aj is the corresponding amplitude. A1 is the sum of amplitudes between 0.5 Hz and Ω/2, divided by the sum of amplitudes between 0.5 Hz and the minimum of (20Ω, 100 Hz). A2 is the sum of amplitudes between 0.7Ω and 1.4Ω divided by the sum of amplitudes between 0.5 Hz and the minimum of (20Ω, 100 Hz). A3 is the sum of amplitudes in 0.6 Hz bands around the second to eighth harmonics (2Ω – 8Ω), divided by the sum of amplitudes in the range of 0.5 Hz to the minimum of (20Ω, 100 Hz).
VF is detected if M ≤ M0 = 1.55, A1 <A1,0 = 0.19, A2 ≥ A2,0 = 0.45, and A3 ≤ a3,0 = 0.09.
The critical threshold parameter to obtain the ROC is A2,0, the other threshold parameters (A1,0, A3,0, M0) being kept constant.
Complexity measure algorithm
The complexity measure algorithm (CPLX) [13] transforms the ECG signal into a binary sequence and searches for repeating patterns.
Lempel and Ziv [14] have introduced a complexity measure c(n), which quantitatively characterizes the complexity of a dynamical system.
After preprocessing, a 0 – 1 string is generated by comparing the ECG data xi (i = 1...n, n being the number of data points) to a suitably selected threshold. The mean value xm of the signal in the selected window is calculated. Then xm is subtracted from each signal sample xi. The positive peak value Vp, and the negative peak value Vn of the data are determined.
By counting, the quantities Pc and Nc are obtained. Pc represents the number of data xi with range 0 <xi < 0.1Vp and Nc the number of data xi with range 0.1Vn <xi < 0. If (Pc + Nc) < 0.4n, then the threshold is selected as Td = 0. Else, if Pc <Nc, then Td = 0.2Vp, otherwise Td = 0.2Vn. Finally, xi is compared with the threshold Td to turn the ECG data into a 0 – 1 string s1s2s3 ... sn. If xi <Td, then si = 0, otherwise si = 1. Now, from this string a complexity measure c(n) is calculated by the following method, according to [14].
If S and Q represent two strings then SQ is their concatenation. SQπ is the string SQ when the last element is deleted. Let v(SQπ) denote the vocabulary of all different substrings of SQπ. At the beginning, c(n) = 1, S = s1, Q = s2, and therefore SQπ = s1. For generalization, now suppose S = s1s2 ... sr and Q = sr+1. If Q ∊ v(SQπ), then sr+1 is a substring of s1s2 ... sr, therefore S does not change. Q has to be renewed to be sr+1sr+2. Then it has to be judged if Q belongs to v(SQπ) or not. This procedure has to be carried out until Q ∉ v(SQπ), now Q = sr+1 sr+2 ... sr+i, which is not a substring of s1s2 ... srsr+1 ... sr+i-1, thus c(n) is increased by one. Thereafter S is combined with Q, and S is renewed to be S = s1s2 ... srsr+1 ... sr+i, and at the same time Q has to be renewed to be Q = sr+i+1. The above procedures are repeated until Q contains the last character. At this time the number of different substrings of s1, s2, ..., sn is c(n), i.e., the measure of complexity, which reflects the rate of new pattern arising with the increase of the pattern length n.
The normalized C(n) is computed:
where b(n) gives the asymptotic behavior of c(n) for a random string:
Evidently, 0 ≤ C(n) ≤ 1. In order to obtain results that are independent of n, n must be larger than 1000. Since n is given by window length wl times sampling rate, we choose wl = 8s.
If C <C0 = 0.173 the ECG is classified as SR, if 0.173 ≤ C ≤ 0.426 as VT, and if C > C1 = 0.426 as VF. A shock is recommended only if C > C1.
The critical threshold parameter to obtain the ROC is C0.
Standard exponential algorithm
The standard exponential (STE) algorithm counts the number of crossing points of the ECG signal with an exponential curve decreasing on both sides. The decision for the defibrillation is made by counting the number of crossings. This simple algorithm is probably well-known, but we did not find any description of it in the literature.
The ECG signal is investigated in the time domain. First, the absolute maximum value of the investigated sequence of the signal is searched. An exponential like function Es(t) is put through this point. This function is decreasing in both directions. Hence, it has the representation:
Here, M is the amplitude of the signal maximum, tm is the corresponding time, τ is a time constant. In our investigation, τ is set to 3 seconds. The number of intersections n of this curve with the ECG signal is counted and a number N is calculated by
where T is the time length of the investigated signal part. If N > N0 = 250 crossings per minute (cpm), the ECG-signal is classified as VF. If N <N1 = 180 cpm, SR is identified. Otherwise the signal is classified as VT. Figure 3 illustrates the situation (note that each QRS complex intersects the exponential function two times).
Figure 3 A 8 second episode of SR rhythm is investigated with the standard exponential algorithm (STE). The exponential function intersects the signal 12 times.
A shock is recommended only if N > N0.
The critical threshold parameter to obtain the ROC is N0.
Modified exponential algorithm
A modified version of STE, called MEA lifts the decreasing exponential curve at the crossing points onto the following relative maximum. This modification gives rise to better detection results.
This algorithm works in the time domain. First, the first relative maximum value of the investigated sequence of the signal is searched and an exponential like function En,1(t) is put through this point. Here, it has the representation:
with Mj being the value of the j-th relative maximum of the signal, tm,j the corresponding time and τ the time constant. Here, τ is set to 0.2 seconds. tc,j is the time value, where the exponential function crosses the ECG signal.
The difference to STE is, that here the function does not have the above representation over the whole investigated signal part, but only in the region from the first relative maximum to the first intersection with the ECG signal. Then, the function En,j(t) coincides with the ECG signal until it reaches a new relative maximum. In some way one can say that the function MEA(t) is "lifted" here from a lower value to a peak. From that peak on it has again the above representation with M being the value of the next relative maximum. This is done until the curve reaches the end of the investigated ECG sequence.
The number of the liftings n of this curve with the ECG signal is counted and a number N is calculated by
where T is the time length of the investigated signal part. If N > N0 = 230 crossings per minute (cpm), the ECG-signal is classified as VF. If N <N1 = 180 cpm, SR is identified.
Otherwise the signal is classified as VT. Figure 4 illustrates the situation. A shock is recommended only if N > N0.
Figure 4 A 8 second episode of SR rhythm is investigated with the modified exponential algorithm (MEA). The exponential function is lifted 7 times.
The critical threshold parameter to obtain the ROC is N0.
Signal comparison algorithm
This new algorithm (SCA) compares the ECG with four predefined reference signals (three sinus rhythms containing one PQRST segment and one ventricular fibrillation signal) and makes its decision by calculation of the residuals in the L1 norm.
The algorithm works in the time domain. After preprocessing, all relative maxima of a modified ECG signal are searched. The relative positions in time tj and amplitude aj of these points are considered. We call this set M0, with M0 = {(tj, aj)|aj is a local maximum}. With this information a probability test for being the peak of a possible QRS complex is performed. For a detailed description of this test see steps 1 and 2 in Appendix B. In a normal ECG, most of the relative maxima M0 of the ECG signal, which are not the peaks of an QRS complex, are sorted out and omitted by this procedure. On the other hand, in an ECG signal with ventricular fibrillation only such peaks are preserved, which are peaks of a fibrillation period.
In other words: Most of the relative maxima, which exist due to noise in the ECG signal are deleted. Furthermore, nearly all relative maxima, which are peaks of insignificant elevations (in this algorithm also P waves and T waves) are deleted as well. This selection procedure is carried out by setting adaptive thresholds. The value of the thresholds is calculated by use of different parameters, that were selected by experiments with ECG signals. The result is a set of points X, which is a subset of M0. In fact, the temporal appearance of the points in X is related to the frequency of the heart beat. The average frequency found by this points is related to a certain probability factor. This factor, together with other results, is finally used to make a decision whether the signal is VF or not.
Now, the central idea of the algorithm is applied. The points in X are used to generate four artificial signals. The first signal looks like a normal sinus rhythm, that has its QRS peaks exactly at the points of X. A reference signal which consists of one PQRST segment is fitted from one maximum of X to the next. To fit the different size of the peaks it is scaled linearly. It has all features that a normal ECG signal should have (narrow QRS complex, P wave, T wave). The second artificial signal is the average of about 700 normal sinus rhythm signals found in 16 files of the MIT database and 16 files of the CU database. The third artificial signal has QRS complexes and an elevated T wave. The fourth signal, which we use as a reference for a ventricular fibrillation signal, has the shape of a cosine like function, which has its peaks at the points of X and therefore simulates ventricular fibrillation.
The next step is the calculation of the residuals with respect to the reference signals. We call the ECG signal E(t), the reference signals that simulate a healthy heart Sj(t), j = 1,2,3, and the ventricular fibrillation signal F(t). The following parameters are calculated
where I = [t0, t1] with t0 = min {tj|tj ∊ X} and t1 = max {tj|tj ∊ X}. Thus, I is the temporal interval from the smallest tj in X to the largest tj in X. Now, four further values are calculated
c1 and c2 are two constants that were suitably chosen by tests. Finally, VRF and VRS are compared. If all tj, j = 1, 2, 3 are smaller than 1, the signal is classified as VF, otherwise it is considered to be SR. (Note 6: Using an L2 norm did not improve the quality.)
The critical threshold parameter to obtain the ROC are tj,0.
Wavelet based algorithms
The continuous wavelet transform of a signal f ∊ L2 is defined by
where ψ is the mother wavelet, ψ ∊ L2, and admissible, i.e.,
Here, denotes the Fourier transform of ψ
The wavelet transform Lψf contains information about the frequency distribution as well as information on the time distribution of a signal.
According to Lemma 1.1.7 from [15], the Fourier transform of Lψf is given by
WVL1
A new simple wavelet based algorithms (WVL1) operates like SPEC in the frequency domain. The idea of this first wavelet algorithm is the following: First, a continuous wavelet transform of the ECG signal is carried out using a Mexican hat as mother wavelet. Then a Fourier transform is performed. Now, the maximum absolute values are investigated in order to make the decisions for the defibrillation process. However, one can show that these maximum values are located on a hyperbola in the (a, w) plane of the Fourier transform of the wavelet transform of the ECG signal, i.e., on a curve that has the representation aw = C, C being a constant. The values on this curve in the (a, w) plane are the Fourier transform of the ECG signal multiplied by a weight function g(w). Therefore, if one searches for the maximum values of in the (a, w) plane of the wavelet transform, it is sufficient to search for the maxima of the weighted Fourier transform of the ECG signal .
Since we are looking for maxima of the modulus of , we need to consider the maxima of only. In WVL1 the function is handled exactly like the spectrum in the algorithm SPEC. The same spectrum parameters are calculated and also the thresholds for the decision have the same values like the algorithm in SPEC.
In WVL1, the critical threshold parameter to obtain the ROC is A2,0.
WVL2
This new method of detecting ventricular fibrillation uses a discrete wavelet transform. It is split into two parts.
(i) Finding VF
The first part uses the algorithm SPEC to search for typical VF properties in the ECG. If the algorithm decides that the ECG part contains VF, then the result is accepted as true and no further investigation is carried out. This procedure can be justified by the high specificity of the SPEC algorithm. If the algorithm yields that the ECG part is "no VF", a further investigation is carried out to confirm this result or to disprove it.
(ii) Discrete Wavelet Transform (DWT)
This part is only carried out, if the first part of the algorithm considers the ECG episode to be "no VF". In this case a discrete wavelet transform is applied, that searches for QRS complexes in the following way:
The third scale of a discrete wavelet transform with 12 scales and a "Daubechies8" wavelet family is used. Numerical tests have shown that this scale makes it easiest to distinguish VF from "no VF". If the signal in the third scale has a value higher than a certain threshold, the according ECG part is considered as QRS complex. The threshold used in this investigation is set to 0.14 max(ECG signal). Multiple peaks belonging to the same QRS complex are removed.
If more than two but less than 40 QRS complexes are found within an 8 second episode, "no VF" is diagnosed. Otherwise the two spectral parameters FSMN and A2 from the first part are investigated again. If FSMN < 2.5 and A2 > 0.2, the considered ECG part is diagnosed as VF.
The mentioned range for the number of found QRS complexes has the following reason: Sometimes, especially in ECGs with a high amount of noise, the DWT part makes wrong interpretations and "finds" QRS complexes also in QRS free episodes. Therefore, a minimal number of three QRS complexes is demanded to confirm the existence of QRS complexes. On the other side, if the DWT part "finds" more than 40 QRS complexes (equal to a pulse of 300 beats per minute), the signal is likely to be VF, since such high sinus rhythms do not appear. The limits of the range were chosen from experiments with data.
In WVL2 no IROC is calculated due to the special structure of the algorithm. Since it consists of two parts and the second part is not executed always, we do not have a single parameter that includes the calculations of both algorithm parts in every ECG segment. Hence we cannot calculate an IROC value. Using the parameters of the SPEC algorithm as an IROC parameter does not yield an ROC curve over the full range.
Finally we want to compare the VF detection algorithms with two algorithms, that are originally used for QRS detection. The decision thresholds of these algorithms have been optimized to be suitable for VF detection.
Li algorithm
The Li algorithm (LI) [16] is based on wavelet analysis, too.
The wavelet transform of an ECG signal is calculated using the following equations
Here, is a smoothing operator and being the ECG signal. hk and gk are coefficients of a lowpass filter H(w) and a highpass filter G(w), respectively. Scales 21 to 24 are selected to carry out the search for QRS complexes. QRS complexes are found by comparing energies from the ECG signal in the scale 23 with the energies in the scale 24. Redundant modulus maximum lines are eliminated and the R peaks detected. Different methods from [17] are used to improve the detection quality:
Method 1
Blanking, where events immediately following a QRS detection are ignored for a period of 200 ms.
Method 2
Searching back, where previously rejected events are reevaluated when a significant time has passed without finding a QRS complex. If no QRS complex was detected within 150% of the latest average RR interval, then the modulus maxima are detected again at scale 23 with a new threshold.
If the number of found QRS complexes is 0 or higher than 5 times the window length in seconds, the ECG segment is classified as VF.
The critical threshold parameter to obtain the ROC is the number of found QRS complexes.
Tompkins algorithm
This algorithm is based on a QRS complex search (TOMP) [18]. It uses slope, amplitude and width information to carry out this task.
After preprocessing, the ECG signal is band filtered by a low pass filter and a high pass filter to reduce interference and high frequency noise. Then, the signal is differentiated to provide the QRS complex slope information. The difference equation for the slope y(j) of the ECG data x(j) reads
where T is the sampling period of the ECG signal. Afterwards the signal is squared to make all data points positive. A moving window integration with a window width of 150 ms (e.g., 54 points at a sampling rate of 360 Hz) is applied. Thresholds are set up to detect QRS complexes.
This algorithm uses a dual threshold technique and a searchback for missed beats. If the number of found QRS complexes is smaller than l0 = 2 or higher than l1 = 32, the ECG segment is classified as VF.
The critical threshold parameter to obtain the ROC is l0.
2 Results
For all algorithms tested in this paper we used the same prefiltering process described at the beginning of the previous section. The function filtering.m for preprocessing can be found on the website .
First, we investigated ECG episodes of window length according to the original papers, and then of window length of 8 seconds since that yielded the best results. For the investigation we simulated a continuous analysis by selecting the data in steps of one second. The decision of an algorithm analyzing an episode of a certain window length was assigned to the endpoint of that interval. By its very nature this continuous monitoring of an ECG signal contains transitions of different rhythms. All algorithms were tested under equal conditions. Finally, we recorded the results together with the annotations in an output file.
The quality parameters are presented in the following tables and figures. The perfect algorithm would have values for sensitivity, specificity, positive predictivity, accuracy and IROC of 100%, assuming that the annotations are 100% correct.
The data sets were taken from the BIH-MIT database (48 files, 2 channels per file, each channel 1805 seconds long), the CU database (35 files, 1 channel per file, each channel 508 seconds long), and the AHA database (files 7001 – 8210, 40 files, 2 channels per file, each channel 1800 seconds long).
Note 7: ANSI/AAMI EC38:1998 Ambulatory electrocardiographs: "The incidence and variety of VF in the AHA and MIT databases are not sufficient to allow those databases to serve as substitutes for the CU DB for the purposes of 5.2.14.5. An evaluation of VF detection using the 80 records of the AHA DB and the 48 records of the MIT DB should supplement the required CU DB evaluation, however, as the CU DB does not contain a sufficient sample of signals likely to provoke false VF detections."
Thus, the total number of decisions per algorithm (window length = 8s) was 2·48·(1805 - 7) + 35·(508 - 7) + 2·40·(1800 - 7) = 333 583.
The annotations of these databases are on a beat to beat level. When taking an arbitrary 8 second episode which includes a VF sequence at the end, it was assumed that the overall classification is VF.
The testing was done automatically by an application written with MATLAB, since there is no chance to inspect 330 000 ECG episodes by hand.
Numerical results
Table 1 shows the values for the sensitivity, the specificity and the integrated receiver operating characteristic of the investigated algorithms. The great differences in performance on different databases can be easily explained by the different nature of this databases (see Note 7). The overall results were directly calculated from the 333 583 decisions of VF/no VF.
Table 2 shows the values for the positive predictivity, the accuracy and the calculation time of the investigated algorithms.
Table 3 shows the values for the sensitivity of the investigated algorithms, if, due to an appropriate adaption of the threshold parameters, the specificity were 95% and 99%, respectively.
Figure 5 and 6 show the ROC curves for all algorithms.
For the computation of the ROC curves, we used 64 nodes. Since some critical threshold parameters are discrete the points of the ROC curve are not equidistant.
3 Discussion and Conclusion
In real applications of AEDs the specificity is more important than the sensitivity, since no patient should be defibrillated due to an analysis error which might cause cardiac arrest. Therefore, a low number of false positive decisions should be achieved, even if this process makes the number of false negative decisions higher. But one has to distinguish between our calculated values for specificity and sensitivity and the values in [19]. Our values were determined for the basic mathematical algorithms, whereas this paper gives recommendations for whole ECG analysis systems. It also does not consider an analysis without preselection.
Our results show that no algorithm achieves its proclaimed values for the sensitivity or specificity as described in the original papers or in [5] and [6] when applied to an arbitrary ECG episode. The main reason for this is the following: Whereas all other researchers made a preselection of signals, we simulated the situation of a bystander, who is supposed to use an AED, more accurately. Hence no preselection of ECG episodes were made.
The best algorithm SCA, which yields the best value for the integrated receiver operating characteristic (IROC) is a new algorithm followed by the algorithms SPEC and VF. Studying the ROC curves in Figure 5 and Figure 6 we see that the relevant part of the ROC curves lies at the left side. The ROC curve also enables us to compare different algorithms given a specified specificity.
All other algorithms yielded only mixed results in our simulations. We also conclude that algorithms developed for QRS detection, like LI and TOMP, are not suitable for VF detection even when the thresholds are suitably adapted.
Outlook
The currently best algorithm works in the time domain. The two algorithms SPEC and VF use information on the energy distribution from the frequency domain but do not use any corresponding phase information. Whereas the algorithm CPLX which uses methods from chaos theory has a poor performance in the region where Specificity > 80% our current investigations indicate a promising good performance for new algorithms based on other methods coming from chaos theory which are currently under development. When finished these algorithms will be presented elsewhere.
Appendix A: Sensitivity, Specificity, Positive Predictivity, Accuracy, and ROC
Sensitivity is the ability (probability) to detect ventricular fibrillation. It is given by the quotient
with TP being the number of true positive decisions, FN the number of false negative decisions. Specificity is the probability to identify "no VF" correctly.
It is given by the quotient
where TN is the number of true negative decisions, and FP is the number of false positive decisions. This means that if a defibrillator has a sensitivity of 90% and a specificity of 99%, it is able 90% of the time to detect a rhythm that should be defibrillated, and 99% of the time to recommend not shocking when defibrillation is not indicated.
Remark: a trivial algorithm which classifies every ECG episode as "no VF" will reach a specificity of 100%, but will have sensitivity 0%. On the other hand, a trivial algorithm which classifies every ECG episode as VF will reach a sensitivity of 100%, but will have specificity 0%. The ROC curve (see below) describes this inherent tradeoff between sensitivity and specificity.
Furthermore, we calculated the Positive Predictivity and the Accuracy of the investigated algorithms.
Positive predictivity is defined by
Positive predictivity is the probability, that classified VF is truly VF:
Accuracy is defined by
Accuracy is the probability to obtain a correct decision.
Specificity and sensitivity always depend on the chosen critical threshold parameters which depend, on the other hand, on the databases used for evaluation (see Note 7).
To get rid of at least of the dependence on the chosen critical threshold parameter one uses the ROC curve. The sensitivity is plotted in dependence of (1 - specificity), where different points in the plot are obtained by varying the critical threshold parameter in the decision stage of an algorithm.
The ROC curves enables us to compare different algorithms when choosing a specified specificity. For more information on ROC curves see [20,21].
Appendix B: Algorithm details: Signal Comparison Algorithm (SCA)
Here we describe the search for relative maxima and the appropriate choice among them, used in Section 1 in more detail.
An offset is added to the ECG signal to make its mean value to zero. We construct a set Z containing the values aj and temporal positions tj of this new signal, i.e., Z = {(tj, aj)|aj is the value of the ECG signal at time tj}.
All further steps are executed both with the set Z and the set -Z, where -Z = {(tj, bj)|bj = -aj is the value of the negative ECG signal at time tj} with the help of the reference signals rECGℓ, ℓ being VF, SR1, SR2 or SR3, or, equivalently, ℓ = 0,1, 2, 3. Note, that the maxima of Z correspond to the minima of -Z. So we get 2 * 4 = 8 tests to find out whether a signal is VF or SR. If any of the 8 tests yields SR, the signal is considered to be SR.
Step 1
All relative maxima aj of Z and their corresponding times tj are determined. The resulting set is called M0, i.e., M0 = {(tj, aj)|aj is a local maximum}, so M0 ⊂ Z. All aj in M0, that are smaller than A, where A = Δ·max(aj) and Δ is a threshold, are deleted. The threshold Δ is set to 0.1 for the VF reference signal and to 0.2 for the SR reference signals. We call the reduced set M1. In Figure 7 we see an ECG episode from the CU database (cu21, from t = 148 s until t = 156 s) together with its selected relative maxima according to the status after processing step 1.
Figure 7 ECG signal with relative maxima (indicated by stars left) after applying step 1 of Signal Comparison Algorithm (SCA). This ECG episode is annotated as no VF, a.u. ... arbitrary units.
Now, we introduce an index l and set it to l = 1.
Step 2
Ml is reduced further: The maximum aj in Ml is searched. Here, we call it amax. amax has a corresponding temporal position tmax. Then, the largest possible temporal interval Il in Z around tmax is searched, so that all values aj in this interval are equal or smaller than amax and larger than 0.2amax. All pairs (aj, tj) except (amax, tmax) in Ml, that are referred to the found interval Il, are deleted. We get a set that we call Ml+1. This procedure is repeated with all untreated aj in Ml, until every aj has been considered and afterwards either been deleted or kept. After each step, l is increased by 1. This means, first we consider M1, then M2 = M1\I1, then M3 = M1\{I1 ∪ I2} and so on, until we reach a highest l, called lmax. In the end, we get a set that we call M, with .
In the end, the aj in M are the relative maxima in Z, that are higher than A and are the only ones in certain subintervals of Z. Two different aj in M can only be neighbors in Z, if they are separated by a valley that is deeper than 20% of the higher peak of the two. In Figure 8 we again see the ECG episode, together with its newly selected relative maxima according to the status after processing step 2.
Figure 8 ECG signal with relative maxima left (indicated by stars) after applying step 2 of Signal Comparison Algorithm (SCA).
Step 3
A value Ω is calculated from M. The frequency Ω of "peaks" is given by
where NM is the number of points in M and tmax - tmin is the maximum temporal range of the elements in M.
Step 4
Now, if two different elements (ai, ti) and (aj, tj) of M are separated by a temporal distance |ti - tj| smaller than , the element with the smaller a is deleted from M. This final set is called X. In Figure 9 we again see the ECG episode as in the figures above and together with its newly selected relative maxima according to the status after processing step 4.
Figure 9 ECG signal with relative maxima (indicated by stars) after applying step 4 of Signal Comparison Algorithm (SCA).
Step 5
Ω is recalculated by Equation (28) with the help of the recalculated set X. If Ω > 280, r is set to 2, if Ω < 180, r is set to 0.9, else r is set to 1.
Step 6
The decision is calculated by Equation (16). VRF is calculated for the ventricular fibrillation reference signal, VRS for the sinus rhythm reference signal.
In Figure 10 we see the ECG episode together with the corresponding VF reference signal.
Figure 10 ECG signal with relative maxima and VF reference signal after applying step 6 of Signal Comparison Algorithm (SCA).
In Figure 11 we see the ECG episode together with the first corresponding SR reference signal. c1 is set to 2/r, c2 is set to 1.
Figure 11 ECG signal with relative maxima and SR reference signal after applying step 6 of Signal Comparison Algorithm (SCA).
Acknowledgements
We are indebted to the referees for numerous helpful suggestions. A. A. gratefully appreciates continuous support by the Bernhard-Lang Research Association.
==== Refs
Zheng Z Croft J Giles W Mensah G Sudden cardiac death in the United States, 1989 to 1998 Circulation 2001 104 2158 63 11684624
American Heart Association, AHA database
Massachusetts Institute of Technology, MIT-BIH arrhythmia database
Massachusetts Institute of Technology, CU database
Clayton R Murray A Campbell R Comparison of four techniques for recognition of ventricular fibrillation from the surface ECG Med Biol Eng Comput 1993 31 111 7 8331990
Jekova I Comparison of five algorithms for the detection of ventricular fibrillation from the surface ECG Physiol Meas 2000 21 429 39 11110242
Thakor N Zhu Y Pan K Ventricular tachycardia and fibrillation detection by a sequential hypothesis testing algorithm IEEE Trans Biomed Eng 1990 37 837 43 2227970
Chen S Thakor N Mower M Ventricular fibrillation detection by a regression test on the autocorrelation function Med Biol Eng Comput 1987 25 241 9 3329694
Kuo S Dillman R Computer Detection of Ventricular Fibrillation Computers in Cardiology, IEEE Computer Society 1978 347 349
Barro S Ruiz R Cabello D Mira J Algorithmic sequential decision-making in the frequency domain for life threatening ventricular arrhythmias and imitative artefacts: a diagnostic system J Biomed Eng 1989 11 320 8 2755113
Clayton R Murray A Campbell R Changes in the surface ECG frequency spectrum during the onset of ventricular fibrillation Proc Computers in Cardiology 1990 1991 Los Alamitos, CA: IEEE Computer Society Press 515 518
Murray A Campbell R Julian D Characteristics of the ventricular fibrillation waveform Proc Computers in Cardiology 1985 Washington, DC: IEEE Computer Society Press 275 278
Zhang X Zhu Y Thakor N Wang Z Detecting ventricular tachycardia and fibrillation by complexity measure IEEE Trans Biomed Eng 1999 46 548 55 10230133
Lempel A Ziv J On the complexity of finite sequences IEEE Trans Inform Theory 1976 22 75 81
Louis A Maaß P Rieder A Wavelets 1998 Stuttgart: Teubner
Li C Zheng C Tai C Detection of ECG characteristic points using wavelet transforms IEEE Trans Biomed Eng 1995 42 21 8 7851927
Hamilton P Tompkins W Quantitative investigation of QRS detection rules using the MIT/BIH arrhythmia database IEEE Trans Biomed Eng 1986 33 1157 65 3817849
Pan J Tompkins W A real-time QRS detection algorithm IEEE Trans Biomed Eng 1985 32 230 6 3997178
Kerber R Becker L Bourland J Cummins R Hallstrom A Michos M Nichol G Ornato J Thies W White R Zuckerman B Automatic External Deflbrillators for Public Access Defibrillation: Recommendations for Specifying and Reporting Arrhythmia Analysis Algorithm Performance, Incorporating New Waveforms, and Enhancing Safety Circulation 1997 95 1677 1682 9118556
Swets J Pickett R Evaluation of diagnostic systems: Methods from signal detection theory 1992 New York: Academic Press
The magnificent ROC
|
16253134
|
PMC1283146
|
CC BY
|
2021-01-04 16:37:36
|
no
|
Biomed Eng Online. 2005 Oct 27; 4:60
|
utf-8
|
Biomed Eng Online
| 2,005 |
10.1186/1475-925X-4-60
|
oa_comm
|
==== Front
Curr Control Trials Cardiovasc MedCurrent Controlled Trials in Cardiovascular Medicine1468-67081468-6694BioMed Central 1468-6708-6-161624204910.1186/1468-6708-6-16ResearchDoes the Angiotensin-converting enzyme (ACE) gene insertion/deletion polymorphism modify the response to ACE inhibitor therapy? – A systematic review Scharplatz Madlaina [email protected] Milo A [email protected] Johann [email protected] Annalisa [email protected] Lucas M [email protected] Horten Centre for patient oriented research, University of Zurich, Switzerland2 Mario Negri Institute for Pharmacological Research, Clinical Research Center for Rare Diseases, "Aldo e Cele Daccò" Villa Camozzi, Bergamo, Italy3 Division of Epidemiology and Biostatistics, Department of Social and Preventive Medicine, University of Bern, Switzerland2005 24 10 2005 6 1 16 16 31 8 2005 24 10 2005 Copyright © 2005 Scharplatz et al; licensee BioMed Central Ltd.2005Scharplatz et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
Pharmacogenetic testing to individualize ACE inhibitor therapy remains controversial. We conducted a systematic review to assess the effect modification of the insertion/deletion (I/D) polymorphism of the ACE gene on any outcome in patients treated with ACE inhibitors for cardiovascular and/or renal disease.
Methods
Our systematic review involved searching six electronic databases, then contacting the investigators (and pharmaceutical industry representatives) responsible for the creation of these databases. Two reviewers independently selected relevant randomized, placebo-controlled trials and abstracted from each study details on characteristics and quality.
Results
Eleven studies met our inclusion criteria. Despite repeated efforts to contact authors, only four of the eleven studies provided sufficient data to quantify the effect modification by genotypes. We observed a trend towards better response to ACE inhibitors in Caucasian DD carriers compared to II carriers, in terms of blood pressure, proteinuria, glomerular filtration rate, ACE activity and progression to end-stage renal failure. Pooling of the results was inappropriate, due to heterogeneity in ethnicity, clinical domains and outcomes.
Conclusion
Lack of sufficient genetic data from the reviewed studies precluded drawing any convincing conclusions. Better reporting of genetic data are needed to confirm our preliminary observations concerning better response to ACE inhibitors among Caucasian DD carriers as compared to II carriers.
Angiotensin-converting enzymepharmacogeneticsACE I/D polymorphism
==== Body
Background
Angiotensin-converting enzyme (ACE) inhibitors have become a cornerstone in the management of patients with cardiovascular disorders [1]. A large number of trials have demonstrated important clinical benefits for this class of drugs in patients with arterial hypertension [2], heart failure [3-6], diabetic and non-diabetic nephropathy [7-11] and in patients who have undergone renal transplantation [12,13]. Despite the generally positive effects of ACE inhibitors, response to equivalent doses of these drugs varies considerably among individuals [14]. As an example, up to one third of patients with congestive heart failure may not tolerate or respond to ACE inhibitor therapy [15,16].
In the late 1980s, researchers began investigating genetic factors to determine the origins of inter-individual variability in patients' responses to ACE inhibitor therapy. Rigat et al. [17] identified the insertion/deletion (I/D) polymorphism of the ACE gene, which is based on the presence (insertion) or absence (deletion) of a 287-bp element on intron 16 on chromosome 17. They noted that this polymorphism accounted for about 47% of the phenotypic variance for serum ACE levels. This led to the hypothesis that the I/D polymorphism may influence the effect of ACE inhibitors on clinical outcomes.
The ACE I/D polymorphism has received much attention since its discovery, and many pharmacogenetic studies have been conducted in different patient groups to assess its effect modification. Debate, however, continues regarding the extent and direction of its effect modification [8,14,18]. The objective of our study was to systematically review all randomised, placebo-controlled trials that had evaluated to what extent the ACE gene insertion/deletion polymorphism influences treatment effects of ACE inhibitors on any surrogate and on any clinically relevant parameters in patients with cardiovascular diseases, diabetes, renal transplantation and/or renal disease.
Methods
A previous publication details our study methodology [19].
Search Strategy
We performed extensive literature searches in (Pre-) MEDLINE (DataStar® version Cary North Carolina from inception to 2005), EMBASE (DataStar® version from inception to 2005, Cary North Carolina), Biosis (Ovid® version "Previews 1989 to 2005", New York), the Cochrane Controlled Trials Register (CCTR <2rd Quarter 2005>, Oxford, United Kingdom) and the Science Citation Index. In collaboration with an information specialist, we carried out a preliminary literature search in Medline to design the final search strategy [19]. We used the following final search terms "peptidyl-dipeptidase A," "dipeptidyl-carboxypeptidase-inhibitor," "ACE inhibitor," "genetics," "pharmacogenetic" and "polymorphism" without language restriction. The last search was performed in July, 2005.
We also contacted authors who had published pharmacogenetic analyses in the area of cardiovascular disease, diabetes, renal transplantation and/or renal diseases, as well as relevant collaborative review groups of the Cochrane Collaboration and pharmaceutical companies for additional published or unpublished data. Finally, we reviewed bibliographies of all included studies to identify additional relevant articles, using the "related articles" function of PubMed and the citation index of ISI Web of Science by entering all studies included in the review.
Inclusion criteria
Two reviewers (MS, MAP) independently assessed all obtained titles and abstracts stored in Reference Manager® files (Professional Edition Version 11, ISI ResearchSoft, Berkeley, California) and ordered the full text of all potential articles. The two reviewers then examined the full texts of all retrieved citations and included studies if they 1) were randomised controlled trials comparing ACE inhibitors to placebo or to a non-active treatment 2) included patients with heart failure, primary and secondary hypertension, coronary artery disease, diabetic nephropathy, primary nephropathy and patients who had undergone renal transplantation and 3) had determined the insertion/deletion polymorphism.
Data extraction and quality assessment
Two reviewers independently abstracted the data and assessed the quality of the included studies. We corrected discordant items based on obvious reading errors and resolved through consensus discordance based on real differences in interpretation. A third reviewer (LMB) resolved any remaining discrepancies.
We used a pre-designed and pilot-tested data extraction form [19] and recorded details on study design, treatments, patients, pharmacogenetic tests, outcome parameters and results. We also focused our efforts on obtaining additional unpublished data on genetic test information and effect measures from authors of included studies. We sent our requests and subsequent reminders for these data to the first and last authors. To assess internal validity of the included studies, we utilized a detailed list of quality items [20], adding other items that we considered to be important for pharmacogenetic studies (e.g., blinding of laboratory assessor of study outcomes, blinding of outcome assessor for genotypes and blinding of treatment provider for genotypes).
Methods of analysis and synthesis
Results of the data extraction and assessment of study validity were summarized in structured tables. In comparing the intervention group (ACE inhibitor) with control groups (placebo or no active control), treatment effects for each genetic subgroup (DD, DI and II) were assessed by calculating mean differences for continuous outcomes and relative risks for dichotomous outcomes. We also calculated mean differences and relative risks between intervention and control groups for the whole study population, including all three genetic subgroups (DD/DI/II), in order to provide an overall treatment effect of the study. All statistical analyses were performed using the Stata® statistical software package (StataCorp. 2004. Stata® Statistical Software: Release 8.2 College Station, Texas, USA).
Results
Figure 1 summarizes the selection process for the 656 identified abstracts. The eleven placebo-controlled, randomized clinical trials (RCTs) that met the final inclusion criteria involved post-myocardial infarction participants as well as subjects with congestive heart failure (n = 3), arterial hypertension (n = 1), chronic proteinuric nephropathies (n = 1), diabetes (n = 2) and patients who had undergone coronary stent implantation or coronary angioplasty (n = 3) or renal transplantation (n = 1). Only three trials reported sufficient data [21-23] and the authors of one trial sent additional requested data [24] for the analysis of effect modification. Despite repeated efforts to obtain additional unpublished data from the remaining trials, [23,25-31], we were unable to secure this information.
Table 1 summarizes the methodological quality of all included trials. Disagreement on quality assessment was resolved through consensus. In general, the quality of trials was poor to moderate. Trials scored poorly for blinding, which was considered only for interventions but not for genotypes. The four trials included in the analysis of the effect modification were of moderate methodological quality. All four studies controlled for imbalances of baseline characteristics between genotypes, whereas only three studies [21,23,24] controlled for co-interventions that might affect study outcomes.
Figure 1 Study flow from identification to final inclusion of studies. We identified 656 studies from which 11 studies eventually met the inclusion criteria.
Table 1 Internal validity of included randomized controlled trials [20]
Homogenous study group Blinding of outcome assessor for genotype Blinding of outcome assessor for treatment Blinding of laboratory assessor for outcome Blinding of treatmen t provider (doctor) for genotype Blinding of treatment treatment Blinding of patients for treatment Check to what extent blinding was successful Registration of loss to follow-up Random allocation (description of procedure) Treatment allocation concealed
Hernandez 2000 [25] + + + - - - + - + - -
Hingorani 1997 [26] + - - - - - - - - +/- -
Kventy 2000 [27] +/- - - - - + + - + - -
Meurice 2001 [28] +/- - - - - + + - + - +/-
Okamura 1999 [21] - - - - - - - - + - -
Okumura 2002 [29] + - - + - - - - +/- - -
Pedersen 1997 [30] - + + + - + + - + +/- -
Penno 1998 [23] +/- - - +/- - + + - + + +
Perna 1999 [24] +/- - - - - + + - - + +
Pinto1995 [31] - + + + - + + - - - -
Van Geel 2003 [22] + - - - - - + - - +/- -
Methods for dealing with missing values Compliance checked Analysis of CI Control for possible clinical and other confounders between genotypes Control of co- interventions that bear on outcome for each genotype Per protocol analysis Intention to treat analysis
Hernandez 2000 [25] - +/- - + - - +
Hingorani 1997 [26] +/- - - + - +/- -
Kventy 2000 [27] - - - - - - -
Meurice 2001 [28] + + - - + + +
Okamura1999 [21] - - - +/- +/- - -
Okumura 2002 [29] - - - - - + -
Pedersen 1997 [30] - +/- +/- - - +/- -
Penno 1998 [23] +/- +/- + +/- +/- +/- +
Perna 1999 [24] - - - +/- - - +
Pinto 1995 [31] - + - - - +/- +
Van Geel 2003 [22] - - - +/- - +/- -
+
: Fulfilled;
+/-
: Partially fulfilled;
-
: Not fulfilled or no information provided
Patient characteristics of included studies
Details about the individual trials and participant profiles are provided in Table 2. Mean age of patients with cardiovascular disease and renal disease plus diabetes was 66 and 48 years, respectively. Nine of the 11 studies included Caucasians, while two studies investigated Asian populations.
Table 2 Study characteristics of RCTs
Study Population ACE intervention Control intervention Mean follow up Additional interventions Outcomes
Perna 1999 [24] 212 (87 DD/ 99DI /26 II)
Caucasians, mean age of 50 years, with non- diabetic proteinuric chronic nephropathies: (urinary protein excretion> 1 g/24 h for last 3 months and creatinine clearance of 20 – 70 mL/min/1.73m2) Ramipril 1.25, increased to 2.5 or 5 mg/d Placebo or conventional treatment 30.3 months Conventional treatment for chronic nephropathy
Blood pressure, proteinuria, glomerular filtration rate, end stage renal disease
Van Geel 2003 [22] 86 (20 DD/ 43 DI/ 23 II)
Caucasians, mean age of 62 years, undergoing elective coronary artery bypass graft surgery; able to take the study drug for at least seven days before surgery Quinapril 40 mg/d Placebo 12 months Aspirin, coumarin-derivatives (anticoagulants), β- blocker, Ca+2 channel, nitrates
Plasma ACE activity
Okamura 1999 [21] 97 (16 DD/26 DI/36 II)
Asians, mean age of 63 years, with stable angina pectoris undergoing percutaneous transluminal coronary angioplasty Imidapril 5 mg/d Placebo 3–6 months Aspirin and warfarin
Minimal lumen diameter of coronary artery (net gain (mm), diameter stenosis %, late loss (mm) and loss index)
Penno 1998 [23] N = 530 (137 DD/296 DI /77 II)
Caucasians, mean age of 53, diabetes with normo- and microalbuminuria, diastolic blood pressure < 90 mmHg and > 75 mmHg, systolic blood pressure < 155 mmHg Lisinopril, 10–20mg/d Placebo 24 months Glycaemic control
Proteinuria
Hernandez 2000 [25] N= 52 (25 DD/DI and 17 II)
Caucasians, mean age of 47, patients with renal transplantation, Stable renal function (creatinine clearance < 2.5 mg/dL for more than 6 months), absence of renal artery stenosis or chronic allograft nephropathy, no renovascular hypertension, absence of proteinuria Lisinopril, 10 mg/d Placebo 12 months Antithymocyte globulin, prednisone, cyclosporine, azathioprine
Serum creatinine, left ventricular ejection fraction, left ventricular isovolumetric relaxation time, left ventricular mass index
Hingorani 1997 [26] N = 125 (37 DD/70 DI/17 II)
Caucasians, mean age of 54, with untreated essential hypertension >160/90 (mainly primary care referrals Captopril (50 mg/d), Enalapril (10 mg/d), Lisinopril (10 mg/d), Perindopril (4 mg/d) Placebo 4 weeks Ca+2-channel blocker
Blood pressure
Kventy 2000 [27] N = 57 (17/DD/ 24 DI /16 II)
Caucasians, mean age of 45, with diabetes > 5 years, no microalbuminuria < 20 ug/min, albumin creatinine ratio < 2.5 mg/mmol, normal serum creatinine and urine sediment Perindopril, 4 mg/d Placebo 24 months None
Albumin/creatinine ratio, blood pressure, glomerular filtration rate
Meurice 2001 [28] N = 91 (only DD)
Caucasians, mean age of 85, after coronary stent implantation Quinapril 40 mg/d Placebo 6 months Aspirin (75–300 mg for 6 months), ticlopidine (500mg for 1 months)
Minimal luminal diameter of coronary artery (restenosis)
Pedersen 1997 [30] N = 56 (14 DD, 26 DI,16 II)
Caucasians, mean age of 68, after acute myocardial infarction and moderate to severe left ventricular dysfunction (2–6 days after myocardial infarction) Trandolapril, 4 mg/d Placebo 12 months None
Tissue-type plasminogen activator [30]32 and plasminogen activator inhibitor (PAI) ACE activity
Pinto 1995 [31] N = 96 (34 DD/DI and 62 II)
Caucasians with first anterior myocardial infarction, treated with thrombolytic therapy (streptokinase intravenously) Captopril 75 mg/d Placebo 12 months Streptokinase administration (1500000U)
Nitrates, Ca+2 channel blocker, β- blocker, aspirin, diuretics
End-systolic and end-diastolic volume
Okumura 2002 [29] N = 100 (21 DD/22 DI/49 II)
Asians, mean age of 64, after coronary artery stent implantation Quinapril 10–20 mg/d Placebo 6 months Ca+2 channel blocker, β- blocker, nitrates, aspirin, ticlopidine
Minimal lumen diameter of coronary artery (restenosis)
Of the studies included in the analysis of the effect modification, three had enrolled Caucasians with non-diabetic chronic nephropathies [24], insulin-dependent diabetes mellitus [23] or patients undergoing elective coronary artery bypass graft surgery [22]. One study included Asians who had undergone percutanous transluminal coronary angioplasty [21]. In all four studies, ACE inhibitors were assigned as add-on therapy to slow disease progression.
Comparisons between genotypes across clinical domains
The studies of Perna [24] and Van Geel [22], which were based on previously published trials [32-34], provided sufficient data to assess the overall ACE inhibitor effects on different outcomes. Perna et al. assessed systolic and diastolic blood pressure, proteinuria, glomerular filtration rate and end-stage renal disease (defined as need for dialysis), and Van Geel et al. assessed the plasma ACE activity for each genotype. In these two studies, which included Caucasians with the DD genotype, ACE inhibitors reduced systolic blood pressure more effectively (mean difference 5.6 mmHg, 95% confidence intervals [0.96 to 10.97]) compared to the overall treatment effect (3.1 mmHg [-2.60 to 8.88]), while in II carriers, ACE inhibitors increased systolic blood pressure (-3.8 mmHg [-12.92 to 5.32]) compared to placebo (Figure 2). Effect modification with benefit for DD carriers and treatment failure for II carriers was also present for diastolic blood pressure, decline of glomerular filtration rate, ACE activity and progression to end-stage renal disease. Differences between genotypes failed to achieve statistical significance. Details on baseline characteristics and differences in treatment effects across genotypes are shown in Table 3. The risk reduction of the incidence of end-stage renal disease was 22% (4% to 39%) for DD carriers compared to 2% (-14% to 18%) for DI carriers and 1% (-21% to 41%) for II carriers (see Figure 2).
Figure 2 Comparison of treatment effects between genotypes in Caucasians (results of the study of Van Geel [22] and Perna [24] on different outcomes). Treatment effects for each genotype (DD/DI/II) and overall treatment effects are presented as mean differences (from baseline to follow up) or relative risks with 95% confidence intervals.
Table 3 Data of effect modification for continuous clinical outcomes
Outcome Study
Genotype
N (ACE)
Baseline
Follow up
Mean (SD) ACE inhibitor
N (Control)
Baseline
Follow up
Mean (SD) Control group
Mean difference (estimated SD)
Overall effect (weighted mean and SD)
Reduction in ACE activity (U/I)
Van Geel [22] DD 8 31† 25† 6.29
(11.20) 12 27† 35† -7.82
(11.22) 14.11
(11.2) 9.84
(0.83)
DI 23 25.5† 22† 3.14
(10.31) 20 23† 20† -6.76
(10.28) 9.9
(10.3)
II 11 18† 13† 4.61
(7.73) 12 18† 20† -1.42
(3.98)) 6.03
(6.1)
Reduction in end-systolic blood pressure (mmHg)
Perna [24] DD 42 147.8
(19.1) 139.8
(15.1) 8.8
(14.58) 45 146
(17.7) 142.8
(13.9) 3.2
(10.68) 5.6
(12.8) 3.1
(12.8)
DI 47 145.1
(18.9) 139
(17.5) 6.1
(13.88) 52 146.5
(17.2) 143.2
(12.72) 3.3
(12.45) 2.8
(13.08)
II 16 145.1
(18.4) 139.7
(17.8) 5.4
(11.29) 10 153.2
(15.9) 144
(14.7) 9.2
(11.75) -3.8
(11.85)
Reduction in end-diastolic blood pressure (mmHg)
Perna [24] DD 42 89.9
(11) 86.2
(6.9) 3.8
(8.76) 45 88.6
(10.2) 87.4
(7.6) 1.2
(6.42) 2.6
(7.8) 1.4
(9.1)
DI 47 88.9
(13.9) 86.1
(9.64) 2.8
(12.10) 52 90.6
(12.6) 88.6
(8.86) 2.0
(9.7) 0.8
(10.9)
II 16 90.8
(11.6) 85.3
(5.8) 4.9
(7.61) 10 97.5
(6.1) 92.1
(6.25) 5.4
(5.67) -0.5
(7.0)
Reduction of proteinuria (g/24 h)
Perna [24] DD 42 2.85
(1.69) 2.28
(1.5) 0.57
(1.18) 45 2.8
(1.7) 2.73
(1.4) 0.07
(1.10) 0.5
(1.1) 0.33
(1.2)
DI 47 3.3
(2.2) 2.97
(1.97) 0.51
(1.31) 52 3.8
(2.5) 3.65
(1.9) 0.15
(1.20) 0.36
(1.2)
II 16 3.45
(2.37) 3.41
(3.3) 0.05
(1.57) 10 3.4
(1.7) 3.08
(0.32) 0.32
(1.15) -0.37
(1.4)
Decline of glomerular filtration rate (per months)
Perna [24] DD 42 44.2
(19.1) n.a. -0.28
(0.46) 45 40.2
(17) n.a. -0.43
(0.63) 0.15
(0.54) 0.08
(0.63)
DI 47 45.3
(20.7) n.a. -0.48
(0.83) 52 40.6
(17.1) n.a. -0.52
(0.68) 0.04
(0.75)
II 16 47.3
(23.5) n.a. -0.38
(0.41) 10 47.8
(20.5) n.a. -0.37
(0.53) -0.01
(0.43)
Albumin excretion rate(μg/min)
Penno [23] DD 71 8.1* 17.1† 9.00† 66 8.1* 17.2† 9.78† -0.78 -7.33
DI 154 7.6* 17.4† 9.80† 142 7.6* 19.1† 11.5† -1.70
II 29 9.2* 16.2† 7.00† 48 9.2* 24.3† 15.11† -8.11
Net gain (mm)
Okamura [21] DD 9 0.64
(0.27) 1.09
(0.66)) 0.45
(0.23) 7 0.49
(0.27) 1.64
(0.96) 1.16
(0.26) -0.71
(0.2) 0.03
(0.28)
DI 13 0.58
(0.32) 1.62
(0.79) 1.04
(0.24) 13 0.53
(0.18) 1.42
(0.97) 0.89
(0.26) 0.15
(0.3)
II 10 0.55
(0.25) 1.82
(0.69) 1.27
(0.21) 26 0.65
(0.16) 1.42
(0.54) 0.77
(0.18) 0.5
(0.2)
Late Loss (mm)
Okamura [21] DD 9 2.43
(0.54) 1.09
(0.66)) 1.34
(0.23) 7 2.46
(0.57) 1.64
(0.96) 0.82
(0.24) -0.52
(0.67) 0.15
(0.74)
DI 13 2.58
(0.55) 1.62
(0.79) 0.96
(0.21) 13 2.32
(0.49) 1.42
(0.97) 0.90
(0.23) -0.06
(0.79)
I24I 10 2.44
(0.51) 1.82
(0.69) 0.62
(0.16) 26 2.65
(0.25) 1.42
(0.54) 1.23
(0.16) 0.61
(0.75)
Loss index (% at follow up)
Okamura [21] DD 9 n.a. n.a. 0.78
(0.13) 7 n.a. n.a. 0.42
(0.12) -0.36
(0.36) 0.03
(0.39)
DI 13 n.a. n.a. 0.53
(0.12) 13 n.a. n.a. 0.49
(0.13) -0.04
(0.45)
II 10 n.a. n.a. 0.38
(0.09) 26 n.a. n.a. 0.63
(0.08) 0.25
(0.37)
Diameter stenosis
Okamura [21] DD 9 n.a. n.a. 57.4
(23.4) 7 n.a. n.a. 40.08
(26.4) -0.17
(0.25) 0.03
(0.39)
DI 13 n.a. n.a. 43.8
(25.2) 13 n.a. n.a. 45.1
(33.9) 0.01
(0.30)
II 10 n.a. n.a. 34.3
(19.9) 26 n.a. n.a. 48.2
(30.6) 0.14
(0.28)
Values are presented as means and SD; n.a.: not available; * indicates combined baseline data for intervention and control group; † indicates extracted values of figures (no SD)
The study of Penno et al. [23] was also based on a previously published trial [11]. Caucasian patients with insulin-dependent diabetes mellitus and normoalbuminuria (83% in DD, 87% in DI and 81% in II) at baseline showed comparable treatment effects from ACE inhibitors between genotypes within the first 12 months. Thereafter, II carriers had an enhanced response to ACE inhibitors with regard to reduction of albumin excretion rate (8.1 μg/min) compared to DI (1.7μg/min) and DD carriers (0.8μg/min). Patients with II genotypes also exhibited the largest benefit in terms of progression from normoalbuminuria to micro- or macroalbuminuria (Risk ratio 0.36 [0.05 to 2.74]), whereas in DD carriers, ACE inhibitors tended to have a negative effect (risk ratio of 1.18 [0.33 to 4.20]). There were, however, baseline imbalances in important prognostic variables between genotypes. Compared to DI and DD carriers, II carriers had pronounced albuminuria at baseline, with placebo group participants experiencing the greatest progression of albumin excretion rate.
In the study of Okamura [21], which included a Japanese population, only the II subgroup had an enhanced response to ACE inhibitors (manifested by prevention of restenosis, as defined by most indexes after percutaneous transluminal coronary angioplasty). Otherwise, the overall treatment effects were not significant (Figure 3). II carriers showed 1) an increased net gain in minimal lumen diameter of 0.5 mm (-1.04 to 1.04) compared to the overall effect of 0.14 mm (-0.49 to 0.76), 2) a higher percentage of change in diameter stenosis 0.14 (-0.06 to 0.33) compared to the overall effect of 0.03 (-0.16 to 0.22), 3) an improved late loss of lumen diameter of 0.61 mm (0.13 to 1.09) compared to the overall effect of 0.15 mm (-0.33 to 0.64) and 4) a better loss index (the ratio of late loss to acute gain) of 0.25 (-0.02 to 0.52) compared to the overall effect of 0.03 (-0.24 to 0.29). In the DD subgroup, ACE inhibitors showed a negative effect on changes in minimal lumen diameter of coronary arteries. Thus, Japanese DD carriers did not benefit from ACE inhibitor therapy, while DI carriers demonstrated moderate responses and II carriers showed the greatest treatment response.
Figure 3 Comparison of treatment effects between genotypes in Asians (results of the study of Okamura [21] on differences in minimal luminal diameter). Effects for each genotype (DD/DI/II) and the overall treatment effect are presented as mean differences (from follow-up to baseline) with 95% confidence intervals.
Discussion
Based on our systematic review, evidence quantifying the extent of effect modification related to the I/D polymorphism is sparse. We did note a trend towards better response to ACE inhibitor therapy in DD Caucasians as compared to II carriers, who seemed not to benefit.
The strengths of this review include the comprehensive literature search and strict adherence to systematic review methodology. We restricted our analyses to trials with placebo controls, as studies without a placebo control group do not allow for estimation of the ACE inhibitor effect and are likely to be confounded [35,36].
Although we identified 11 randomised controlled trials that assessed differences in treatment effects among genotypes, only the results of four trials studying 925 patients contributed to our analyses. The authors of the remaining seven studies that included 577 patients did not provide data about genetic subgroups in the intervention and control arms [26-28,30]. Others presented combined results for two different genotypes [25,29,31]. We made substantial efforts to contact the authors of these seven trials as well as researchers known to be active in the field of pharmacogenetics, but we did not succeed in receiving additional unpublished data. Thus, our analysis might be subject to publication bias. While publication bias is a common problem in systematic reviews, the situation might be aggravated in reviews of genetic data. Reporting of genetic data is in general poor and most "negative" results of association studies do not even reach conference proceedings [37].
Despite an overall lack of evidence, some of our findings merit attention. For example, the second largest study of Caucasians with chronic nephropathies [24] showed a consistent trend towards a beneficial effect for various surrogate and clinical outcomes, whereas II carriers appeared to be unresponsive to treatment. Comparable results have been observed in the reduction of plasma ACE activity for patients after coronary artery bypass surgery [22]. On the other hand, the findings of the largest study [23], including diabetic patients with normo- or microalbuminuria, yielded conflicting results to the above-discussed effect modification in Caucasians. In terms of baseline values of the main outcome (level of albumin extraction rates), differences between genotypes limit the interpretation of these results. Looking at the results of the Asian study [21], DD carriers did not benefit from ACE inhibitor therapy while DI carriers showed moderate and II carriers showed large treatment responses. Arguably, these conflicting results for treatment success of ACE inhibitors between the genetic subgroups in Asians and Caucasians might be attributed to a different genotype-phenotype relationship. In Asians, the prevalence of the D allele frequency ranges from 27 to 40 percent, whereas in Caucasians, it ranges from 50 to 63 percent. Additional ethnic factors might also affect these genotype-phenotype relationships. For example, the level of circulating ACE is 60% higher in Caucasian DD carriers than in II carriers, whereas for Asians, there are no differences [8,38]
From the patient and clinician's perspective, it is still too early to draw solid conclusions about the optimal treatment among different genotypes. We can, however, speculate that an effect modification exists and that pharmacogenetic testing of the I/D polymorphism might provide additional information about the adequate treatment for these patients. From the public health perspective, it remains unclear whether pharmacogenetic testing would be justifiable in clinical practice. Before investing additional resources to reevaluate our preliminary observations in a primary study of high methodological quality, it might be informative to assess whether screening patients for the I/D polymorphism would have potential economic value. One recent economic analysis showed, for example, that screening men for the I/D polymorphism before starting lipid-lowering therapy with statins would result in considerable cost savings [39]. Thus, estimations of the potential cost-effectiveness of this pharmacogenetic test might be worth considering before starting ACE inhibitor treatment.
Conclusion
We conclude that evidence is still scarce as there are few pharmacogenetic studies of high methodological quality with comprehensive reporting of their results. Nevertheless we did note a trend towards better response to ACE inhibitors in Caucasian DD carriers compared to II carriers. Future efforts should focus on conducting high-quality pharmacogenetic studies, and reporting of genetic data should be improved.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
All authors approved the manuscript.
Funding disclosure
The study was supported by unrestricted educational grants from AstraZeneca, Switzerland, and from Pfizer, Switzerland. Dr Bachmann's work (grants no. 3233B0-103182 and 3200B0-103183) was supported by the Swiss National Science Foundation.
Acknowledgements
We would like to thank Dr. Pius Estermann, Information Specialist of the University Hospital of Zurich, for carrying out the literature searches.
==== Refs
Masoudi FA Rathore SS Wang Y Havranek EP Curtis JP Foody JM Krumholz HM National Patterns of Use and Effectiveness of Angiotensin-Converting Enzyme Inhibitors in Older Patients With Heart Failure and Left Ventricular Systolic Dysfunction Circulation 2004 110 724 731 15289383 10.1161/01.CIR.0000138934.28340.ED
Angeli F Verdecchia P Reboldi GP Gattobigio R Bentivoglio M Staessen JA Porcellati C Meta-Analysis of effectiveness or lack thereof of angiotensin-converting enzyme inhibitors for prevention of heart failure in patients with systemic hypertension Am J Cardiol 2004 93 240 243 14715360 10.1016/j.amjcard.2003.09.052
A placebo-controlled trial of captopril in refractory chronic congestive heart failure. Captopril Multicenter Research Group J Am Coll Cardiol 1983 2 755 763 6350401
Effects of enalapril on mortality in severe congestive heart failure. Results of the Cooperative North Scandinavian Enalapril Survival Study (CONSENSUS). The CONSENSUS Trial Study Group N Engl J Med 1987 316 1429 1435 2883575
Effect of enalapril on survival in patients with reduced left ventricular ejection fractions and congestive heart failure. The SOLVD Investigators N Engl J Med 1991 325 293 302 2057034
Garg R Yusuf S Overview of randomized trials of angiotensin-converting enzyme inhibitors on mortality and morbidity in patients with heart failure. Collaborative Group on ACE Inhibitor Trials JAMA 1995 273 1450 1456 7654275 10.1001/jama.273.18.1450
Jafar TH Schmid CH Landa M Giatras I Toto R Remuzzi G Maschio G Brenner BM Kamper A Zucchelli P Becker G Himmelmann A Bannister K Landais P Shahinfar S de Jong PE de ZD Lau J Levey AS Angiotensin-converting enzyme inhibitors and progression of nondiabetic renal disease. A meta-analysis of patient-level data Ann Intern Med 2001 135 73 87 11453706
Navis G Van Der Kleij FGH de Zeeuw D de Jong PE Angiotensin-converting enzyme gene I/D polymorphism and renal disease J Mol Med 1999 77 781 791 10619438 10.1007/s001099900063
Ruggenenti P Perna A Gherardi G Gaspari F Benini R Remuzzi G Renal function and requirement for dialysis in chronic nephropathy patients on long-term ramipril: REIN follow-up trial. Gruppo Italiano di Studi Epidemiologici in Nefrologia (GISEN). Ramipril Efficacy in Nephropathy Lancet 1998 352 1252 1256 9788454 10.1016/S0140-6736(98)04433-X
Strippoli GF Craig M Deeks JJ Schena FP Craig JC Effects of angiotensin converting enzyme inhibitors and angiotensin II receptor antagonists on mortality and renal outcomes in diabetic nephropathy: systematic review BMJ 2004 329 828 15459003 10.1136/bmj.38237.585000.7C
Randomised placebo-controlled trial of lisinopril in normotensive patients with insulin-dependent diabetes and normoalbuminuria or microalbuminuria. The EUCLID Study Group Lancet 1997 349 1787 1792 9269212 10.1016/S0140-6736(96)10244-0
Suwelack B Kempkes-Koch M Kobelt V Hillebrand U Matzkies F Gerhardt U Hohage H Impact of ACE polymorphism on renal allograft function, blood pressure, and proteinuria under ACE inhibition Transplant Proc 2002 34 1763 1766 12176567 10.1016/S0041-1345(02)03058-0
Suwelack B Kobelt V Erfmann M Hausberg M Gerhardt U Rahn KH Hohage H Long-term follow-up of ACE-inhibitor versus beta-blocker treatment and their effects on blood pressure and kidney function in renal transplant recipients Transpl Int 2003 16 313 320 12759722 10.1111/j.1432-2277.2003.tb00306.x
Haas M Yilmaz N Schmidt A Neyer U Arneitz K Stummvoll HK Wallner M Auinger M Arias I Schneider B Mayer G Angiotensin-converting enzyme gene polymorphism determines the antiproteinuric and systemic hemodynamic effect of enalapril in patients with proteinuric renal disease. Austrian Study Group of the Effects of Enalapril Treatment in Proteinuric Renal Disease 156 Kidney blood press res 1998 21 66 69 9661139 10.1159/000025845
O'Toole L Stewart M Padfield P Channer K Effect of the insertion/deletion polymorphism of the angiotensin-converting enzyme gene on response to angiotensin-converting enzyme inhibitors in patients with heart failure J Cardiovasc Pharmacol 1998 32 988 994 9869506 10.1097/00005344-199812000-00017
Opie LH Przybojewski JZ Angiotensin-converting enzyme inhibitor therapy S Afr Med J 1992 81 183 185 1738902
Rigat B Hubert C Alhencgelas F Cambien F Corvol P Soubrier F An Insertion Deletion Polymorphism in the Angiotensin I-Converting Enzyme Gene Accounting for Half the Variance of Serum Enzyme Levels J Clin Invest 1990 86 1343 1346 1976655
Mayer G ACE genotype and ACE inhibitor response in kidney disease: a perspective Am J Kidney Dis 2002 40 227 235 12148094 10.1053/ajkd.2002.34489
Scharplatz M Puhan MA Steurer J Bachmann LM What is the impact of the ACE gene insertion/deletion (I/D) polymorphism on the clinical effectiveness and adverse events of ACE inhibitors?--Protocol of a systematic review BMC Med Genet 2004 5 23 15361261 10.1186/1471-2350-5-23
ter-Riet G Kessels AGH Commentary on Rampes et al 'Does electroacupuncture reduce craving for alcohol? A randomized controlled study' Complement Ther Med 1997 116 118 10.1016/S0965-2299(97)80010-6
Okamura A Ohishi M Rakugi H Katsuya T Yanagitani Y Takiuchi S Taniyama Y Moriguchi K Ito H Higashino Y Fujii K Higaki J Ogihara T Pharmacogenetic analysis of the effect of angiotensin-converting enzyme inhibitor on restenosis after percutaneous transluminal coronary angioplasty Angiology 1999 50 811 822 10535720
van Geel PP Buikema H Rouleau JL van Gilst WH Genotype-dependent increase in plasma ACE activity after CABG is prevented by ACE inhibition Can J Cardiol 2003 19 701 704 12774785
Penno G Chaturvedi N Talmud PJ Cotroneo P Manto A Nannipieri M Luong LA Fuller JH Effect of angiotensin-converting enzyme (ACE) gene polymorphism on progression of renal disease and the influence of ACE inhibition in IDDM patients: findings from the EUCLID Randomized Controlled Trial. EURODIAB Controlled Trial of Lisinopril in IDDM Diabetes 1998 47 1507 1511 9726242
Perna A Ruggenenti P Testa A Spoto B Benini R Misefari V Remuzzi G Zoccali C ACE genotype and ACE inhibitors induced renoprotection in chronic proteinuric nephropathies Kidney Int 2000 57 274 281 10620209 10.1046/j.1523-1755.2000.00818.x
Hernandez D Lacalzada J Salido E Linares J Barragan A Lorenzo V Higueras L Martin B Rodriguez A Laynez I Gonzalez-Posada JM Torres A Regression of left ventricular hypertrophy by lisinopril after renal transplantation: role of ACE gene polymorphism Kidney Int 2000 58 889 897 10916115 10.1046/j.1523-1755.2000.00239.x
Hingorani AD Jia H Stevens PA Hopper R Dickerson JE Brown M Renin-angiotensin system gene polymorphisms influence blood pressure and the response to angiotensin converting enzyme inhibition J Hypertens 1995 13 1602 1609 8903618
Kvetny J Gregersen G Pedersen RS Randomized placebo-controlled trial of perindopril in normotensive, normoalbuminuric patients with type 1 diabetes mellitus QJM 2001 94 89 94 11181984 10.1093/qjmed/94.2.89
Meurice T Bauters C Hermant X Codron V VanBelle E Mc-Fadden EP Lablanche J Bertrand ME Amouyel P Effect of ACE inhibitors on angiographic restenosis after coronary stenting (PARIS): a randomised, double-blind, placebo-controlled trial Lancet 2001 357 1321 1324 11343737 10.1016/S0140-6736(00)04518-9
Okumura K Sone T Kondo J Tsuboi H Mukawa H Tsuzuki M Imai H Kamiya H Mabuchi Y Matsui H Hayakawa T Quinapril prevents restenosis after coronary stenting in patients with angiotensin-converting enzyme D allele Circ J 2002 66 311 316 11954942 10.1253/circj.66.311
Pedersen OD Gram J Jeunemaitre X Billaud E Jespersen J Does long-term angiotensin converting enzyme inhibition affect the concentration of tissue-type plasminogen activator-plasminogen activator inhibitor-1 in the blood of patients with a previous myocardial infarction Coron Artery Dis 1997 8 283 291 9285181
Pinto YM van Gilst WH Kingma JH Schunkert H Deletion-type allele of the angiotensin-converting enzyme gene is associated with progressive ventricular dilation after anterior myocardial infarction. Captopril and Thrombolysis Study Investigators J Am Coll Cardiol 1995 25 1622 1626 7759715 10.1016/0735-1097(95)00090-Q
Randomised placebo-controlled trial of effect of ramipril on decline in glomerular filtration rate and risk of terminal renal failure in proteinuric, non-diabetic nephropathy. The GISEN Group (Gruppo Italiano di Studi Epidemiologici in Nefrologia) Lancet 1997 349 1857 1863 9217756 10.1016/S0140-6736(96)11445-8
Oosterga M Voors AA Buikema H Pinto YM Haber HE Ebels T Morshuis WJ Kingma JH Crijns HJ van Gilst WH Angiotensin II formation in human vasculature after chronic ACE inhibition: a prospective, randomized, placebo-controlled study. QUO VADIS Investigators Cardiovasc Drugs Ther 2000 14 55 60 10755201 10.1023/A:1007843205311
Ruggenenti P Perna A Gherardi G Gaspari F Benini R Remuzzi G Renal function and requirement for dialysis in chronic nephropathy patients on long-term ramipril: REIN follow-up trial. Gruppo Italiano di Studi Epidemiologici in Nefrologia (GISEN). Ramipril Efficacy in Nephropathy Lancet 1998 352 1252 1256 9788454 10.1016/S0140-6736(98)04433-X
Altman DG Randomisation BMJ 1991 302 1481 1482 1855013
Schulz KF Chalmers I Hayes RJ Altman DG Empirical evidence of bias. Dimensions of methodological quality associated with estimates of treatment effects in controlled trials JAMA 1995 273 408 412 7823387 10.1001/jama.273.5.408
Colhoun HM McKeigue PM Davey SG Problems of reporting genetic associations with complex outcomes Lancet 2003 361 865 872 12642066 10.1016/S0140-6736(03)12715-8
Staessen JA Wang JG Ginocchio G Petrov V Saavedra AP Soubrier F Vlietinck R Fagard R The deletion/insertion polymorphism of the angiotensin converting enzyme gene and cardiovascular-renal risk J Hypertens 1997 15 1579 1592 9488209 10.1097/00004872-199715120-00059
Maitland-van der Zee AH Klungel OH Stricker BH Veenstra DL Kastelein JJ Hofman A Witteman JC Leufkens HG van Duijn CM de BA Pharmacoeconomic evaluation of testing for angiotensin-converting enzyme genotype before starting beta-hydroxy-beta-methylglutaryl coenzyme A reductase inhibitor therapy in men Pharmacogenetics 2004 14 53 60 15128051 10.1097/00008571-200401000-00006
|
16242049
|
PMC1283147
|
CC BY
|
2021-01-04 16:47:32
|
no
|
Curr Control Trials Cardiovasc Med. 2005 Oct 24; 6(1):16
|
utf-8
|
Curr Control Trials Cardiovasc Med
| 2,005 |
10.1186/1468-6708-6-16
|
oa_comm
|
==== Front
Curr Control Trials Cardiovasc MedCurrent Controlled Trials in Cardiovascular Medicine1468-67081468-6694BioMed Central 1468-6708-6-171625578010.1186/1468-6708-6-17ResearchCompliance of a cobalt chromium coronary stent alloy – the COVIS trial Hagemeister Jens [email protected] Frank M [email protected] Robert HG [email protected]öpp Hans W [email protected] Department of Medicine III, University of Cologne, Kerpener Str. 62, 50924 Cologne, Germany2005 28 10 2005 6 1 17 17 18 8 2005 28 10 2005 Copyright © 2005 Hagemeister et al; licensee BioMed Central Ltd.2005Hagemeister et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
Cobalt chromium coronary stents are increasingly being used in percutaneous coronary interventions. There are, however, no reliable data about the characteristics of unfolding and visibility of this stent alloy in vivo. The aim of this study is to compare cobalt chromium coronary stents with conventional stainless steel stents using intracoronary ultrasound.
Methods
Twenty de novo native coronary stenoses ≤ 20 mm in length (target vessel reference diameter ≥ 2.5 and ≤ 4.0 mm) received under sequential intracoronary ultrasound either a cobalt chromium stent (Multi-Link Vision®; n = 10) or a stainless steel stent (Multi-Link Zeta®; n = 10).
Results
For optimal unfolding, the cobalt chromium stent requires a higher balloon deployment pressure (13.90 ± 2.03 atm) than the stainless steel stent (11.50 ± 2.12 atm). Furthermore, the achieved target vessel diameter of the cobalt chromium stent (Visibility-Index QCA/IVUS Multi-Link Vision®1.13 / Multi-Link Zeta® 1.04) is more easily overrated by Quantitative Coronary Analysis.
Conclusion
These data indicate that stent material-specific recommendations for optimal implantation pressure and different stent material with an equal design should both be considered in interpreting QCA-analysis.
==== Body
Background
With the addition of coronary stents to percutaneous coronary intervention (PCI), the incidence of re-stenosis has been significantly reduced. Unfortunately, re-stenosis rates still range from 16% to 32 %[1]. Efforts to reduce re-stenosis include coating of conventional stents and use of alternative materials and design. Drug Eluting Stents (DES) are already established in clinical practice[2], whereas little data are available with respect to innovative stent material.
Cobalt chromium represents a more biocompatible material that is being increasingly used in coronary stents (Guidant Multi-Link Vision®/Guidant Corporation, Driver-Stent®/Medtronic, Costar-Stent®/Biotronic). In comparison with stainless steel, cobalt chromium has a higher radial strength and radiopacity for similar electronegativity. This allows for the production of thinner struts with a similar radiological visibility[3].
Although, results of two cobalt chromium registries[3,4] are already published, there are no data describing the basic characteristics of unfolding of a cobalt chromium stent. This information would be important to know in developing clinical recommendations for different alloys, since optimal inflation and complete adherence to vessel wall are key factors affecting the incidence of re-stenosis.
We investigated the balloon deployment pressure-related behaviour of a cobalt chromium stent (Multi-Link Vision®), comparing it to a similarly designed conventional stainless steel stent (Multi-Link Zeta®) using intravascular ultrasound. We also evaluated radiological visibility of both stents and the influence of radiological visibility on QCA-analysis.
Methods
Eighteen consecutive patients (14 men, 4 women, mean age 61 ± 9 years) with twenty single, de novo native coronary stenoses ≤ 20 mm in length were blindly randomised to either a Multi-Link Vision® cobalt chromium stent (n = 10) or to a Multi-Link Zeta® stainless steel stent (n = 10). Additional criteria for inclusion were age ≥ 18 years, clinical angina and/or a positive functional study and a target vessel reference diameter ≥ 2.5 mm and ≤ 4 mm.
Patients were excluded from the study if they presented with cardiogenic shock, acute coronary syndrome, intracoronary thrombus, vessel occlusion, target lesions in the left main artery, ostial or bifurcational stenosis, calcification of ≥ 180° of vessel circumference by intravascular ultrasound (IVUS), diabetes and/or known hypersensitivity to aspirin and clopidogrel. The study was approved by the ethics committee of the University of Cologne and all patients signed written informed consent before participating.
Percutaneous coronary intervention was performed in accordance with standard clinical procedures, and stent implantation without predilatation ("primary stenting") was encouraged. Pretreatment included an oral clopidogrel loading dose of 300 mg and aspirin 500 mg the day before intervention. Peri-interventional, weight-adapted heparin was given intravenously with consecutive control of activated clotting time. Implantation was performed with a primary balloon deployment pressure of 10 atmospheres (atm). If IVUS target criteria were not reached, a standardised further inflation with 13 atm and possibly 16 atm including additional IVUS control, followed. Post-procedural an oral antiplatelet therapy with clopidogrel 75 mg/day for at least 4 weeks and aspirin 100 mg/day as standard medication was obligatory.
Intravascular ultrasound imaging was performed after administering 0.2 mg of intracoronary nitroglycerin using a 30 MHz transducer within a 3.2 Fr imaging sheath (SCIMED/BSC, Maple Grove, Minnesota) and automatic transducer pullback of 0.5 mm/s. The distal and proximal reference segment was within 3 – 5 mm of the lesion or stent without a relevant stenosis (<20%). Quantitative Coronary Analysis (QCA, Pie Medical Imaging) was done for proximal and distal reference diameter (RD), minimal lumen diameter (MLD), diameter of stenosis and acute lumen gain.
The primary endpoint was balloon deployment pressure once IVUS criteria were reached. IVUS target criteria were similar to preceding studies[5], with a minimal lumen area (MLA) after stenting > 90% (for reference lumen area (RLA) ≤ 9.0 mm2), > 80% (for RLA > 9.0 mm2), a MLA > 90% of proximal RLA at the proximal end of the stent and a complete adherence to vessel wall. A secondary endpoint was angiographic visibility as generated by minimal lumen diameter in QCA and IVUS-analysis, respectively.
Statistical analysis was performed with SPSS, version 12.0 (SPSS Inc., 2003). Continuous variables are expressed as mean ± SD.
Results
Clinical data were comparable for both groups, except for the number of smokers (Vision® n = 7 / Zeta® n = 3), and were comparable to prior stent studies. Twenty five stents (13 ML Vision®, 12 ML Zeta®) were implanted with a 100% procedural success. An intraprocedural dissection in 3 patients required implantation of additional stents. In case of stent overlap, the same stent type was used.
The angiographic mean reference vessel diameter was 3.02 ± 0.40 mm (ML Vision® 3.04 ± 0.34 mm / ML Zeta® 2.99 ± 0.47 mm). Mean minimum lumen diameter was 0.70 ± 0.32 mm, which corresponds to a mean diameter stenosis of 76.5 ± 9.8% (Table 1).
Table 1 Baseline Lesion Characteristics (QCA).
ML Vision (n = 10) ML Zeta (n = 10)
Preprocedure reference vessel diameter (mm) 3.04 ± 0.34 2.99 ± 0.47
Preprocedure MLD (mm) 0.77 ± 0.32 0.63 ± 0.32
Preprocedure diameter stenosis (%) 73.50 ± 10.30 79.60 ± 8.80
Target vessel
Left anterior descending 20% 50%
Circumflex 20% 20%
Right coronary artery 60% 30%
Postprocedure MLD (mm) 3.34 ± 0.34 3.31 ± 0.48
Postprocedure diameter stenosis (%) 6.00 ± 1.80 5.10 ± 3.80
Pre-dilatation was necessary in one case, in which a 1.5 × 20 mm balloon was used and dilatation pressure was 8 atm.
Mean balloon deployment pressure when IVUS criteria were reached was 13.90 ± 2.03 atm for Multi-Link Vision® and 11.50 ± 2.12 atm for Multi-Link Zeta® (Figure 1).
Figure 1 Ballon pressure by IVUS-criteria: balloon deployment pressure when IVUS criteria were reached for Multi-Link Vision® and Multi-Link Zeta® (n = number of lesions).
A mean post-procedural QCA-MLD of 3.32 ± 0.41 mm and a mean MLD of 3.07 ± 0.41 mm by intravascular ultrasound generated a mean Visibility-Index of 1.08. For Multi-Link Vision® (3.34 ± 0.34 mm [QCA] vs. 2.95 ± 0.17 mm [IVUS]), mean Visibility-Index was 1.13. The mean Visibility-Index for Multi-Link Zeta® (3.31 ± 0.48 mm [QCA] vs. 3.19 ± 0.55 mm [IVUS]) was 1.04 (Table 2).
Table 2 Visibility-Index: generated from QCA- and IVUS-data.
All ML Vision ML Zeta
Postprocedure MLD QCA (mm) 3.32 ± 0.41 3.34 ± 0.34 3.31 ± 0.48
Postprocedure MLD IVUS (mm) 3.07 ± 0.41 2.95 ± 0.17 3.19 ± 0.55
Visibility [QCA/IVUS] 1.08 1.13 1.04
Conclusion
Different material properties of alloys used in a specific stent design represent a key consideration in clinical practice that should be taken into account, particularly with regard to balloon deployment pressure. Furthermore, the achieved post-procedure lumen diameter as measured by QCA is more easily overestimated for the cobalt chromium stent than for the stainless steel stent due to material properties.
Based on these results, further investigation of material-specific changes in stent unfolding is necessary to guide and optimize the clinical implantation practice.
Limitations
The number of twenty de novo native coronary stenoses is small, but relevant preliminary differences were found that should be further explored in future investigations.
Three standardised balloon deployment pressures (10, 13, 16 atm) were chosen because every change between balloon and IVUS-catheter increases the risk of vascular damage. Thus, a continuous, incremental escalation of balloon deployment pressure is not practical in vivo.
Acknowledgements
This study was supported by a grant from Guidant Corporation (Indianapolis, IN, USA).
==== Refs
Kastrati A Hall D Schömig A Long-term outcome after coronary stenting Curr Control Trials Cardiovasc Med 2000 1 48 54 11714409 10.1186/CVM-1-1-048
Babapulle MN Joseph L Belisle P Brophy JM Eisenberg MJ A hierarchical Bayesian meta-analysis of randomised clinical trials of drug-eluting stents Lancet 2004 364 558 9 15313338 10.1016/S0140-6736(04)16850-5
Kereiakes DJ Cox DA Hermiller JB Midei MG Bachinsky WB Nukta ED Leon MB Fink S Marin L Lansky AJ Guidant Multi-Link Vision Stent Registry Investigators Usefulness of a cobalt chromium coronary stent alloy Am J Cardiol 2003 92 463 6 12914881 10.1016/S0002-9149(03)00669-6
Sketch MH JrBall M Rutherford B Popma JJ Russell C Kereiakes DJ on behalf of the Driver Investigators Evaluation of the Medtronic (Driver) cobalt-chromium alloy coronary stent system Am J Cardiol 2005 95 8 12 15619386 10.1016/j.amjcard.2004.08.055
Mudra H di Mario C de Jaegere P Figulla HR Macaya C Zahn R Wennerblom B Rutsch W Voudris V Regar E Henneke KH Schachinger V Zeiher A OPTICUS (OPTimization with ICUS to reduce stent restenosis) Study Investigators Randomized comparison of coronary stent implantation under ultrasound or angiographic guidance to reduce stent restenosis (OPTICUS Study) Circulation 2001 104 1343 9 11560848
|
16255780
|
PMC1283148
|
CC BY
|
2021-01-04 16:47:32
|
no
|
Curr Control Trials Cardiovasc Med. 2005 Oct 28; 6(1):17
|
utf-8
|
Curr Control Trials Cardiovasc Med
| 2,005 |
10.1186/1468-6708-6-17
|
oa_comm
|
==== Front
Int J Health GeogrInternational Journal of Health Geographics1476-072XBioMed Central London 1476-072X-4-241623617410.1186/1476-072X-4-24MethodologyA suite of methods for representing activity space in a healthcare accessibility study Sherman Jill E [email protected] John [email protected] John S [email protected] Wilbert M [email protected] Thomas A [email protected] Department of Geography, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA2 Carolina Population Center, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA3 Department of Biostatistics, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA4 Department of Family and Community Medicine, Wake Forest University School of Medicine, Winston Salem, NC, USA2005 19 10 2005 4 24 24 26 8 2005 19 10 2005 Copyright © 2005 Sherman et al; licensee BioMed Central Ltd.2005Sherman et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
"Activity space" has been used to examine how people's habitual movements interact with their environment, and can be used to examine accessibility to healthcare opportunities. Traditionally, the standard deviational ellipse (SDE), a Euclidean measure, has been used to represent activity space. We describe the construction and application of the SDE at one and two standard deviations, and three additional network-based measures of activity space using common tools in GIS: the road network buffer (RNB), the 30-minute standard travel time polygon (STT), and the relative travel time polygon (RTT). We compare the theoretical and methodological assumptions of each measure, and evaluate the measures by examining access to primary care services, using data from western North Carolina.
Results
Individual accessibility is defined as the availability of healthcare opportunities within that individual's activity space. Access is influenced by the shape and area of an individual's activity space, the spatial distribution of opportunities, and by the spatial structures that constrain and direct movement through space; the shape and area of the activity space is partly a product of how it is conceptualized and measured. Network-derived measures improve upon the SDE by incorporating the spatial structures (roads) that channel movement. The area of the STT is primarily influenced by the location of a respondent's residence within the road network hierarchy, with residents living near primary roads having the largest activity spaces. The RNB was most descriptive of actual opportunities and can be used to examine bypassing. The area of the RTT had the strongest correlation with a healthcare destination being located inside the activity space.
Conclusion
The availability of geospatial technologies and data create multiple options for representing and operationalizing the construct of activity space. Each approach has its strengths and limitations, and presents a different view of accessibility. While the choice of method ultimately lies in the research question, interpretation of results must consider the interrelated issues of method, representation, and application. Triangulation aids this interpretation and provides a more complete and nuanced understanding of accessibility.
==== Body
Background
Researchers studying healthcare accessibility and utilization have long attempted to understand the influence of geography on individuals' use of healthcare services. Geographic access is usually operationalized as some measure of distance to care. Distance can be measured either from the supply perspective (distance from a clinic or hospital) or from the individual perspective (how far an individual has to travel to a healthcare provider) [1,2], or both [3]. However, these measures typically do not account for differences in individual mobility, spatial habits, and subjective meanings of distance, as well as differences in travel environment. Activity space, defined as "the local areas within which people move or travel in the course of their daily activities" [4], is a measure of individual spatial behavior that theoretically accounts for these individual and environmental differences and offers an alternative approach to studying geographic accessibility.
From a methodological standpoint, however, measuring activity space is more data and computationally intensive than distance, and its complexity has resulted in its underutilization. Prior to the introduction of Geographic Information Systems (GIS), approximations of activity space typically made use of Euclidean measures like the Standard Deviational Ellipse (SDE) [5-7], and place-based proxies for household locations (such as zip code centroid). The limited availability and expense of collecting spatially referenced data and the computational burden involved in generating SDEs restricted such studies to small samples. With advances in GIS and increasing availability of spatially referenced data, activity space has become a more viable tool for studying accessibility. These same technological advances enable researchers to develop new measures of activity space that improve on the precision of the SDE and better represent actual spatial behavior [8,9].
Although activity space is not a new concept, few studies have compared alternative methods of measuring activity space. Just as different methods of measuring distance (Euclidean distance, network distance, travel time distance) can yield different information and possibly different conclusions [2], different approaches to calculating activity space may yield different types of information and results [10]. By comparing these measures, we "triangulate" or view accessibility from multiple perspectives, and so arrive at a more nuanced understanding of accessibility [11,12]. Moreover, we develop a better understanding of how the measures used influence the results.
Addressing the interrelated dimensions of representation, method, and application are important for advancing accessibility research [13]. This paper describes five different measures of routine activity space, applies them in a rural mountain region, and evaluates their relative usefulness in the study of geographic access to health services. The five measures are (1) the standard deviational ellipse at 1 standard deviation (SDE1); (2) the standard deviational ellipse at 2 standard deviations (SDE2); (3) the road network buffer (RNB); (4) the 30-minute standard travel time polygon (STT); and (5) the relative travel time polygon (RTT).
Defining activity space
Activity space has been defined and theorized in different ways by researchers working in various traditions, including medical geography, spatial behavior, time-space studies, planning, travel and transportation studies, and human-environment interactions. Activity space represents the spatial movement component of an individual's day-to-day lived experience [14], and thus "experience of place" [15]. This experience of place is thought to mediate between the role of distance and the distribution of healthcare resources in the perception of healthcare accessibility. Activity space is also described as a measure of an individual's degree of mobility [7], incorporating constraints, needs, preferences and resources for movement. Research in space-time geography has also produced a body of work that uses methods and concepts similar to activity space, but uses different terminology. Kwan's "daily potential path area," for example, is used to measure individual access to urban opportunities [9].
Another perspective on activity space has recently emerged from Geographic Information Science (GISc), largely driven by research in human-environment interactions. One of the main challenges in linking social data to physical (environmental) data is determining how to represent people in space: mapping a person is different from mapping a stationary object, as people are not fixed to a single location. The key challenge in "linking people to pixels" lies partly in the difference between fixed and mobile features [16]. Typically, a person's residence – a single, non-dimensional point – is used to mark an individual's location in space. But as people are mobile, a single fixed point does not adequately represent an individual's location. Because activity space also represents "direct contact between individuals and their social and physical environments"[12], it is a potential solution to this problem. By obtaining locations of routine destinations, a two-dimensional space can be developed to represent a person's location. Once these data are available, however, the challenge becomes how to turn these points into a meaningful representation of activity space, and this can vary depending on the geographical context and the objectives of the research.
Distance, activity space and accessibility
Distance is related to access and utilization; the farther the distance required to travel, the less likely an individual is to use a service, all else being equal. Distance decay – or the attenuation of a pattern or process with distance – is a well-studied geographical phenomenon [12,17]. However, distance is typically a one-dimensional measure that requires knowing between which nodes to measure. For example, some studies have found a greater propensity for individuals to utilize health services that are near place of employment rather than residence [2,7].
Another limitation of distance is that it does not typically account for individual preferences or other factors that channel movement in a specific direction. Directional bias is related to preferences for a particular place over other places of equal distance due to some perceived quality of the preferred place [12]. Both the distribution of opportunities and individual characteristics are critical factors in determining geographical access and utilization. Because activity space is comprised of directional and temporal components of spatial movement in addition to distance, activity space supplies more information than a distance-only measure by demonstrating point patterns and degree of eccentricity, and is suggestive of how boundaries and transportation networks influence activity patterns. For example, Gesler and Meade's Savannah study found a correspondence between activity space and healthcare-seeking space, suggesting that urban structure was more important than demographic characteristics in influencing activity space [7].
Activity space can also account for what we call "relative distance," or individual tolerances for travel and distance to care. Both absolute distance and travel time can have different subjective meanings [15]. Age, ethnicity, income, social status, and health status and type of care sought all contribute to different distance tolerances [18-21]. Mode of transportation is also clearly important, in that 10 miles to someone with access to an automobile is a different burden than 10 miles to someone without access to a car. Where some tolerate a daily commute of one hour or longer, others may find half an hour to be a disincentive. In rural populations, individuals may expect to routinely travel father distances than their urban counterparts; where some may interpret this as a burden, others may interpret this as a benefit of "getting away from it all" [22,23].
Measuring activity space
Historically, activity space has been operationalized using the standard deviational ellipse (SDE) [5,6,24]. Analogous to the univariate statistical measure of standard deviation, the SDE is a bivariate statistical measure that provides a comparable estimate of an individual's activity space [6]. Gesler and Albert define it as "an ellipse whose major and minor axes are drawn to represent the magnitude of the minimum and maximum dispersion of a set of points from their mean center" [4]. By mapping daily trips and determining the locations of regular activities, SDEs are calculated based on distance and direction of these locations from the home. Activity frequencies and duration of activity can be used to weight the relative importance of each point [7]. The main limitation of the SDE is that it is an abstract representation of where people go. As a Euclidean measure, it does not account for actual spatial arrangements of geographic or human features. However, the SDE provides a better indicator of individual access than distance alone, and is now relatively easy to generate with available software.
Alternatives to Euclidean representations of activity space are largely network-based measures that utilize road network data to construct paths between points. This approach too has limitations, in that it assumes all movement is channeled by paved roads (unpaved roads are rarely available in commercial road data). Data quality of road network databases varies, and errors in the network, or out-of-date data will induce errors in the activity space. And, unless the actual route or path that a person takes to travel to each destination is known, decisions must be made on whether to connect the activity destinations to the household only or whether to construct paths between all of the destinations. Weighting destinations by frequency of visit is possible, yet requires a degree of arbitrary decision-making instead of the statistically determined weights of the SDE. Yet the network shape that results is a much more realistic representation of the space through which a person travels than the SDE. One example of a network-based approach is Kwan's "daily potential path area," constructed by calculating all the routes connecting the residence and activity nodes and creating a buffer around the nodes and paths [9].
Kwan published a comparative analysis of 30 network-based measures of individual accessibility, distinguishing between integral and space-time measures [10]. Integral measures were those that calculated "cumulative opportunity" based on counting the number of opportunities accessible within a given distance parameter (such as travel time) from the location of a single point (usually residence). Space-time measures such as the "daily potential path area," in contrast, were used to calculate "feasible opportunities" based on accessibility from multiple points representing residence and daily activities.
Activity space and healthcare accessibility
Providers or services located within the spatial and temporal bounds of a person's routine activity space can be considered more accessible than those located outside the activity space. A service "inside the activity space" is interpreted as one where a person could potentially obtain care with relative ease (not accounting for other dimensions of accessibility), while those opportunities "outside the activity space" are interpreted as requiring the person to deviate from normal routine or expend extra resources to access those opportunities. This fits Joseph and Phillips' (1984) concept of "effective accessibility" which meant, among other things, that the services were available within an individual's space-time budget [25,26], as well as Kwan's "feasible opportunity set" [9,10]. Thus, as a measure, the number of healthcare opportunities inside the activity space is interpreted as an indicator of individual healthcare accessibility.
Both the activity space literature and the space-time geography literature ultimately indicate that, while both the physical and social environments contribute to the spatial structuring of opportunity, individuals do not limit themselves to the "nearest" opportunity. Kwan concluded that spatial accessibility to jobs had no relationship with the length of commute, and questioned the appropriateness of using commuting distance as a measure of job access [9]. The corollary in health services research is that distance to the service actually used may not be an appropriate measure of geographic accessibility, as indeed studies of provider and facility choice, bypassing behavior, and health-seeking behavior have long indicated [27-32]. However, the location of a service actually used relative to activity space may be significant. Nemet and Bailey found that the presence of a respondent's actual provider within the individual's activity space was significantly associated with number of visits [15]. Thus, a provider located within an individual's activity space was "near" and one located outside the activity space was "far," regardless of the absolute distance.
Methods
The MAP survey
This research is part of the larger Mountain Accessibility Project, or MAP, which aims to study the effects of rurality and other geographic factors on healthcare accessibility and utilization in a mountain rural region of western North Carolina. The study area consists of 12 counties in the Mountain Area Health Education Center (MAHEC) region, selected to represent the most rural end of the rural-urban continuum as measured using 1993 Beale codes 6–9 (Figure 1). The metropolitan area of Asheville is the major service center of the region; interstate 40 runs east-west through the region, connecting the Asheville and Knoxville MSAs, while I-26 runs south from Asheville to the Greenville-Spartanburg MSA. Over the past decade, the region has seen high rates of retirement, recreation, and natural amenities-related development and in-migration [33-35].
Figure 1 MAP project study area in western North Carolina. Twelve rural counties were selected from the Mountain Area Health Education Center (MAHEC) region, based on degree of rurality as measured using 1993 Beale codes. The city of Asheville in metropolitan Buncombe County is the major service center of the region. Interstate highways connect residents of the region with other nearby metropolitan areas in the state, as well as in the bordering states of Virginia, Tennessee, Georgia, and South Carolina.
The MAP survey is the primary data source for this analysis. A total of 1,059 adult interviews were completed from June 1999 to January 2000, yielding a rich dataset consisting of individual- and household level data, including variables on healthcare accessibility and utilization; sociodemographic characteristics; transportation and spatial behavior; health status and health behavior; and cultural and attitudinal variables. Spatial data were collected for household locations, routine activity destinations, and locations where respondents reported obtaining health services during the year preceding the interview. The study protocol was approved by the Institutional Review Boards of the University of North Carolina at Chapel Hill, Wake Forest University School of Medicine, and Research Triangle Institute. Details of the study design are reported elsewhere [36,37].
Although the objectives of this paper are primarily methodological, the characteristics of the study area exert important influences on the results. Apart from the physical features and transportation networks noted above, the project documented the finding that the region is well supplied with primary care services [38]. While there are spatial variations in service supply, we were unable to identify sub-regions or subpopulations where the geographic accessibility of primary care services failed to meet federal guidelines. On the one hand, this implies that we are unable to compare the performance of these measures in an environment with poor availability and accessibility. On the other hand, this context allows us to compare these measures to a kind of "gold standard" of primary care accessibility.
Collecting activity data
Activity spaces are constructed from "the subset of all locations within which an individual has direct contact as a result of his or her day-to-day activities" [12]. Three types of locational variables were collected in this survey: residence, routine activity destinations, and healthcare destinations. The primary node in each activity space is the location of residence. Interviewers were trained in the use of GPS receivers and recorded the latitude and longitude of the household at the time of the survey. For households where valid GPS coordinates were not collected, two backup methods of georeferencing the residence location were utilized. First, the project produced maps of the entire study area divided into grid cells of one square kilometer; interviewers were to record the coordinates of the map grid cell in which the residence was located, and coordinates of the grid cell centroid were used to georeference the household. Where neither valid GPS nor map coordinates were recorded, street addresses provided the option of address geocoding using a commercially available database. All households were thus able to be georeferenced.
The other nodes in the activity space were the locations of the routine activity destinations. Using the gridded maps, respondents were asked to identify the locations of where they usually went for a total of 34 possible routine activities, including grocery shopping; to pick up items from a convenience mart; buy gasoline; get the car serviced; clothes or retail shopping; visit friend or family member; eat out; to see movies; go to church or other religious activities; play sports or exercise; watch sports; bank; buy stamps and send letters and packages; go to school; work; and any other common activity not specifically prompted. Interviewers were to prompt respondents for up to two locations per activity type.
Respondents also reported the locations of healthcare providers and facilities visited in the year preceding the interview. Regardless of the frequency of visits, these health destinations were not treated as "routine activity destinations" and were not used in constructing activity spaces. The goal was to compare the degree of overlap or correspondence between routine activity spaces (e.g. non healthcare) and healthcare destinations. Locational data on healthcare destinations were used to test whether the different measures of activity space captured the actual healthcare destination for each respondent. We suggest that the best measure of routine activity space for studying healthcare accessibility will effectively capture not only potential healthcare opportunities, but also the locations of services actually used.
In order to construct the activity space measures, it was necessary to place spatial and temporal bounds on activity destinations reported by respondents. We spatially limited the destinations to those points located within North Carolina (inside or outside of the study area) and the neighboring states of Virginia, Tennessee, Georgia, and South Carolina. This spatial bounding was necessary because some respondents reported traveling great distances on a routine basis. While these destinations may have very important impacts on respondents' lives, we interpreted them as "non-local" activities that did not interact with the local healthcare environment. Healthcare destinations were similarly bounded.
Frequency of activity is also an important data component of activity space. The frequency of "daily" or "routine" has not been defined in any consistent manner in previous studies. Nemet and Bailey defined their activity space as the spatial extent of an individual's weekly activities, but response categories on the questionnaire were "never," "less than once a week," and "more than once a week" which were then weighted in the calculation of the SDE [15]. Kwan used a detailed two-day diary to collect daily activity destinations in sequence [9]. In the MAP survey, respondents were asked to indicate frequency in terms of the number of times per day, week, month or year that they visited each routine activity destination, and these frequencies were converted into a fraction of 365. For this analysis, we limited the construction of routine activity spaces to those activities with a frequency of 12 or more (at least once a month).
Of 1,059 completed interviews, 1,047 respondents reported at least one activity destination with a frequency of at least once a month. Ten of these reported only one destination, and 20 reported only two destinations (in nine of these cases, the destinations had the same coordinates, resulting in effectively only one point). Generating the two axes of an ellipse requires at least three unique points (including the household location); thus, for 19 cases, we had insufficient data to generate ellipses. To avoid dropping these cases and inducing bias toward larger activity spaces, we constructed 1 km circle buffers around single unique points, and likewise connected 2-point cases with a 1 km linear buffer around the two points. These procedures only affected the two SDE measures.
Other data sources
Two additional data sources were required for this study. First, in order to construct the network-based measures (RNB, STT, RTT), a road network database was required. The road network for the region was obtained from ESRI's StreetMap 2000 product [39]. In addition to providing a spatial representation of the road network for the region, it also contains the Census Feature Class Codes (CFCC) for the road segments. CFCCs are standardized descriptions of the type of road and can be used, among other purposes, to assign speed limits to roads; these speed limits are used in the calculation of travel time. For this project a standardized set of speed limits by CFCC provided by ESRI were used.
The third data source provided the locations for healthcare practitioners and facilities in western North Carolina. A database provided courtesy of McMillan & Moss Research, Inc., was constructed from a 2000 survey using GPS receivers to collect the spatial locations of all healthcare delivery sites in the region [40]. From this dataset we subset the locations where primary care services were available. The database did not provide the kind of data that would enable us to classify a service delivery point based on the type of service or facility (private practitioner, clinic, hospital), health workforce (number or type of providers per location), or capacity (e.g. number of patients in practice). However, the point locations do represent "primary care opportunities," or PCOs, where a respondent could potentially seek care.
Another limitation of this dataset is that it includes only 15 counties within the MAHEC service region. All study counties shared a border with counties not covered by the database. Patients, of course, do not necessarily observe county or state borders when seeking health services; in our survey, 48% percent of reported health destinations were in a different county than county of residence. Respondents could potentially have more primary care opportunities available within their activity space than we were able to count because of services available on the other side of the border. This "edge effect" occurs when the administrative boundaries of a dataset artificially bound what is in reality a continuous surface [2]. The project did make efforts to obtain locational data for health services in bordering counties in North Carolina, Tennessee, Georgia and South Carolina, however comparable data were not obtainable. An assessment of the extent of the edge effect revealed that, out of 2,872 reported health destinations (multiple destinations per respondent were possible), 13% were outside the counties covered by the McMillan & Moss database, with 5% out of North Carolina. Half of these out-of-state destinations (n = 66) were reported by residents of Polk County, which has a high rate of commuting to work to the Greenville-Spartanburg MSA in South Carolina. Clay and Cherokee counties accounted for 24% and 17% of out-of-state health destinations, with negligible numbers attributed to the other counties. Thus, while the boundary crossing problem potentially affects residents of all counties in the study area (resulting in an underestimate of availability of health services), results of the survey data suggests that the problem affects only a small proportion of respondents in 3 of the 12 counties.
To determine the routes between the points of the activity space and perform network calculations, points need to be located on the road network. Neither the household coordinates nor the activity destination coordinates were necessarily on the road network. Some households could be a considerable distance from the road network, for example, if they lived on an unpaved road or had a long driveway. Use of grid centroids to determine locations may have placed some points off the network. Unknown errors in the road network data could also introduce a gap between the network and the location. Points not located close enough to a road network segment were "snapped," or moved to the nearest segment of the road network by the GIS. This new snapped location was used when calculating the network-based travel time polygons and road buffers.
Snapping is a standard procedure in GIS, but it does induce a small amount of error. The mean Euclidean distance that household points were snapped was 131 meters, with the maximum being 1.2 kilometers. The mean distance the activity destination points were snapped was 154 meters, with the maximum being 5.2 kilometers from its original location. We concluded that snapping had a minimal impact on the distance estimates. This snapping error was calculated to assess the impact of snapping on the travel estimates; they cannot be used to adjust travel distance or time to "account for snapping error" because of the difference between Euclidean and road distance, and because snapping could move a point closer or farther from the actual travel path. Because it was not possible to determine exactly how snapping affected each household or activity location, no "adjustment" to travel distance/time was made to counteract the effect of snapping; any such procedure would have induced more error.
Constructing the activity space measures
In this section we will describe how each measure is constructed, using five measures of the activity space of a single respondent to illustrate. At the same time we will describe the hypothesized advantages and disadvantages of each measure prior to analyzing and interpreting the results.
The standard deviational ellipse (SDE)
For the routine activity SDE, the location of the respondent's residence and each of the routine activity destinations was mapped in the GIS. Each point was weighted based on the number of times per year the respondent went to that destination, with the respondent's residence given a value of 365, assuming the respondent was there every day. Once the spread of points was mapped, an ellipse was generated using the formula found in the spatial statistics program CRIMESTAT [41]. The standard deviation of the distances between each point and the mean center are calculated for the X direction and the Y direction. This distance is used as the major and minor axis of the ellipse [41].
Because of the structure of the data table, it was necessary to write a script which could accommodate the data file, rather than relying on a pre-existing application such as CRIMESTAT to generate the ellipses. The script developed by the authors provides the following output: the area of the ellipse, the lengths of the ellipse's X and Y axes, its theta angle (the angle between the major axis and the horizontal or X axis), and the location of the weighted mean center of all points. Additionally a graphic representation of the ellipse was generated to be included in the GIS.
Because the literature is inconsistent on the use of one or two standard deviations in generating the ellipses, we chose to do both for comparative purposes. The SDE at one standard deviation (SDE1) contains approximately 68% of the points within the boundary of the ellipse, while the SDE at two standard deviations (SDE2) encompasses approximately 95% of the points (if all destinations are given equal weight; weighting destinations based on frequency of visit will alter these proportions).
To illustrate, Figure 2 shows the SDE of one respondent, at 1 and 2 standard deviations, along with the residence and activity destinations. This respondent has a total of 15 routine activity destinations; however, only nine unique destinations in terms of location are visible because of shared coordinates. Also shown are the primary care opportunities in the region. For this respondent, three opportunities lie within the SDE1, and four within the SDE2, and are suggestive of reasonably good access to primary care services. However, this respondent's actual health destination is not inside either ellipse.
Figure 2 Example of the one and two standard deviational ellipse (SDE1, SDE2) measures. The figure shows the activity space of a single respondent as represented by the two Euclidean measures. The activity destination furthest from the household is the respondent's place of employment. The high frequency of trips to the workplace exerts a strong influence on the shape and orientation of the ellipses. Note, however, that each ellipse encompasses large areas without any activity destinations. Also note that there is no route between the household and the workplace that lies within the boundaries of either ellipse. Finally, while there are 3 and 4 primary care opportunities within the SDE1 and SDE2 respectively, neither ellipse captures the respondent's actual health destination.
Road network buffer (RNB)
The activity space of the RNB is essentially the area around the roads that an individual is likely to travel from home to all destinations; this measure is the closest to Kwan's (1998) time-space measures (although the temporal component is treated quite differently). Within the GIS, the shortest road distance between the respondent's household and each destination was calculated using the ArcView 3.3 Network Analyst extension [42]. Then a buffer was calculated around each route to create a "polygon" in a GIS. Routes were aggregated into a single shape for each respondent using the dissolve function in ArcView to erase the boundaries between buffered routes. A program written in Avenue (ArcView's programming language) was used to automate the process.
The size of the buffer was set to 1 km, meaning that an area of 1 km on each side of the road is encompassed in the polygon. The rationale for choosing a buffer size of 1 km was that (a) given the use of the centroids of 1 km map grid cells to geocode the activity destinations, we have approximately 1 km of error built into the distance between two points, and (b) 1 km off the path would not be considered an additional burden in terms of travel distance, even if walking, for most people.
There are limitations in the calculation of the RNB measure. First, it requires the use of a GIS to calculate, and requires an adequate road dataset. The quality of the measure is dependent on the quality of the road data. Also, the method chosen to calculate the RNB was by calculating the shortest route (a function of road segment length) between a respondent's house and each activity destination, based on the assumption of the most likely route. However, it is possible the respondent may have chosen a different route for their journey. In the absence of data on the actual paths or routes the individual travels to each activity destination, any number of choices could be made. Kwan's 2-day travel diaries included data on trip-chaining and times at each location, and she chose to calculate the paths between each node [9]; however, these paths are limited to activities in a two-day period and may include non-routine as well as routine activities, thus limiting their generalizability to a longer time frame or broader definition of "routine."
In Figure 3, we show the RNB activity space for the same respondent as shown in Figure 2. The shape, area, and extent of the activity space using the RNB are quite different from the SDE, illustrating one of the major limitations of the SDE. For this respondent, not only is there a different number of primary care SDPs inside the activity space (5), but also the set of primary care opportunities identified as "accessible" within each measure – although overlapping – are different (this is not necessarily true for all respondents). Rather than hypothetical access, the RNB representation of activity space identifies the actual SDPs that a person presumably passes. By comparing the points that lie within the buffer to the points that the respondent actually used, we could construct an index of bypassing, although that is beyond the scope of this paper.
Figure 3 Example of the road network buffer (RNB) measure. This is the activity space of the same respondent as represented using the RNB measure. This activity space captures a different set of primary care opportunities based travel patterns structured by road networks. It also eliminates the excess space captured by the SDEs. However, the actual health destination is still outside of the activity space.
Of the five measures, the RNB might be considered the best representation of the spatial movement component of a respondent's activity space, because the area is limited to the likely routes and locations that a person travels. It is descriptive of what the person actually does, not what the person could or should do. The RNB measure is also one of only two measures presented here that encompass 100% of the activity destinations. This measure is the most useful for understanding issues of bypassing and provider/facility choice.
Standard travel time polygon (STT)
Like the RNB, the two travel time polygon measures are based on network calculations. They are constructed by determining a travel time limit, traveling outward from the household location to that limit on all roads within the network leading away from the household, and bounding the area to create a polygon. For the STT, a GIS was employed to determine how far each respondent could travel from their house in thirty minutes. Every road available to them, within thirty minutes, was included in this analysis. This procedure was performed in ArcView 3.3 using the "service area" command in the Network Analyst Extension. A polygon was then created which encompassed all roads within that 30-minute threshold and this serves as the STT activity space used for analysis. The standard of 30 minutes was chosen because of its use in other studies of healthcare accessibility and distance [1,2,43], and because it serves as the federal guideline for maximum distance to primary care under the Health Professional Shortage Area guidelines [44].
The STT is thus based strictly on location of residence and the road network; the routine activity destinations are not used in the construction of this measure. Because the size and shape of the polygon is independent of the activity destinations, the measure does not represent the actual activity space reported by respondents, but is a measure of "potential" activity space because it captures where the person could go within a specified level of effort. It is a normative representation of activity space comparable to Kwan's integral measure of "cumulative opportunity" [10].
Figure 4 displays the STT measure for the example respondent. The scale of the travel time measures greatly exceeds the SDE and RNB measures, as do their areas. The area of the STT is 2384 km2, in comparison with 73 km2 and 293 km2 for the SDE1 and SDE2, and 187 km2 for the RNB measures. In this example, 13 of 15 routine activity destinations were captured by the STT, indicating that the respondent travels further than 30 minutes in reaching two routine activity destinations. This respondent has 11 primary care opportunities within 30 minutes, clearly meeting federal guidelines for accessibility to primary care. Importantly, this respondent's actual health destination is also within the 30-minute travel time polygon. This is not necessarily true for all respondents.
Figure 4 Example of the standard travel time polygon (STT) and relative travel time polygon (RTT) measures. For the same respondent, the two potential activity spaces as represented by travel time polygons. In this example, the RTT is larger than the 30-minute STT, indicating that the respondent routinely travels farther than 30 minutes (two activity destinations lie outside of the STT polygon). The smaller STT does capture the actual health destination, however, along with numerous primary care opportunities.
Relative travel time polygon (RTT)
This measure was calculated in the same way as the STT, however, the travel time threshold was individually determined and therefore different for each respondent. To determine the relative travel time threshold, the GIS used the most direct route (in travel time) between a respondent's home and each of their activity destinations. In order to determine the most direct route, the GIS traversed all possible routes between the household location and activity destination, and then determined the route with the shortest travel time. The travel time from the household to each routine activity destination was calculated, and the longest time for that respondent was used to set the relative threshold distance and generate the relative travel time polygon (again, the process was automated using an Avenue script). For instance, the travel time to the farthest destination in this respondent's activity space was 36 minutes. The GIS determined how far that respondent could travel in 36 minutes in all directions; this maximum extent was used to create the RTT polygon (Figure 4).
The shape of the RTT is similar to the STT, however it is a little bit larger for this respondent (the reverse is equally possible). The RTT by definition captures all 15 of the routine activity destinations, and is similar to the RNB in this characteristic. In this example, the RTT has the largest area of the activity space measures at 3351 km2, although this is unlikely to be true for all or even most respondents. In this case, the extra distance did not capture any more opportunities than the STT, but like the STT it did capture the actual health destination.
Conceptually, this measure is a hybrid of Kwan's integral cumulative opportunity measure and the space-time measures. It is constructed like the integral measure based on proximity to a single location; however, the threshold is empirically determined for each individual based on the routine activity destinations. We suggest that the strength of this measure is it takes into account the amount of travel burden that is acceptable to the individual while not excluding opportunities that may be equally accessible. Like the STT, it represents a potential or normative situation; a person could go to a destination within these bounds without undue travel burden, even if that person normally does not.
There are some limitations in the use of calculating travel time that apply to both the STT and the RTT. The first is in the use of standardized speed limits. In actuality, the speed limits of the roads may not match the standard; however determining the true speed limits of the roads in the study area exceeded the time and resource restraints of the project. The second limitation is that not everyone travels at the designated speed limit, so actual respondent travel times will likely vary. Additionally, the travel time calculations do not take into account variables such as traffic lights, one-way streets, or traffic congestion, all of which can affect the time a respondent needs to travel to the destination. Another major limitation of this approach is that it assumes the same mode of transportation is available for all routes; for individuals dependent on public transportation, this is likely to be a false assumption.
Data analysis
The analysis had three main objectives. (1) The first was to explore the similarities and differences between the measures using descriptive statistics; means and medians of area are presented along with factors thought to influence the size and shape of activity space. (2) Second, we compare the measures using indicators of accessibility constructed from the number of primary care opportunities located inside the activity space. (3) Finally, we assess how well each measure captures a respondent's actual health destination, and test the association between success at capturing a health destination and the area of the activity space, as well as the number of primary care opportunities. This is based on the assumption that the best measure of activity space for examining healthcare accessibility will be one that also successfully models where people actually go for health services. Statistical analysis was performed using Stata 8 [45].
Three main variables were constructed from each of the five activity space measures. First, area or the size of the activity space was calculated in square kilometers. The distribution is highly skewed towards the higher end of the range (e.g,. right-tailed). For use in analyses that require a linear measure, the measure is transformed from the quadratic scale to the linear scale by taking the square root of the area and truncated at the 2nd and 98th percentiles to minimize the effect of outliers.
The second variable is a count of the number of primary care opportunities (PCOs) located inside each respondent's activity space; ArcView's point-in-polygon procedure was used to calculate this variable. Like the area measure, the range of the number of primary care opportunities was highly skewed, and this measure also was transformed using the square root transformation.
The third variable was a dichotomous (no/yes) variable indicating whether an actual health destination was inside the activity space for each measure. Seventy-seven percent (n = 807) of respondents reported visiting an allopathic provider in the year preceding the interview; although many reported multiple providers and locations, we used only the location of the first reported provider. No significant differences in area or number of PCOs were found between those who reported a health destination and those who did not, for any of the activity space measures.
In order to better understand the differences between the measures, we examined the associations between the area of the activity space and a number of potential influences: number of routine activity destinations, maximum travel time to a routine destination (the threshold used to determine the extent of the RTT), travel time to nearest PCO, and proximity to primary roads. Both the number of destinations and the maximum travel time are components of the SDE1, SDE2, RNB and RTT measures, but not the STT measure. Since these were not used in the construction of the STT measure, we hypothesized that the road network characteristics were the primary determinant of the area of the STT. Respondents who lived near primary roads – roads that are assigned high average speeds – were likely to have the largest STT areas, and indeed, all of the network measures might be influenced by proximity to primary roads. To test this, we created a dichotomous (no/yes) variable, to indicate whether respondents lived within 5 km of a primary road. A 5 km buffer was constructed around the primary roads in the region, and the point-in-polygon procedure was used to determine which households were located inside the buffer.
Modeling individual accessibility using activity space
We compared three accessibility measures across the five representations of activity space: the proportion of respondents with at least one PCO inside the activity space; the mean/median number of primary care opportunities inside the activity space; and the proportion of respondents whose actual health destination was inside the activity space. Finally, to further explore the relationship between the area of activity space and healthcare accessibility, a logistic regression model was used to test the relationship between the area of the activity space and the presence of an actual health destination inside the activity space. Because the difference in range of area values between the measures, areas were normalized using z-scores and centered around the mean in order to place the five measures on a comparable scale. Results were obtained using Stata's logistic procedure, incorporating the cluster option to control for design effects.
Results
As expected, the area of the activity space varied dramatically by measure (see Table 1). The RNB activity space had the smallest mean area (107 km2), while the SDE1 had the smallest median area (71.5 km2). The area of the SDE2 measure was 4 times the SDE1 measure. The relative travel time polygon (RTT) had the highest mean value, while the 30-minute standard travel time polygon (STT) had the largest median area. For a majority of respondents (60%), the area of the STT was larger than the RTT, indicating that the maximum travel time of 30 minutes to a primary care provider could potentially require a large number of residents in this area to travel further than they routinely travel to obtain care.
Table 1 Area of activity space and proximity to primary roads for the five measures. Means and medians reported for untransformed measures. SDE1 = standard deviational ellipse at one standard deviation; SDE2 = standard deviational ellipse at 2 standard deviations; RNB = road network buffer; STT = standard travel time polygon; RTT = relative travel time polygon.
SDE1 SDE2 RNB STT RTT
Mean Median Mean Median Mean Median Mean Median Mean Median
Area (km2) (n = 1047) 182.2 71.5 728.7 286.0 107.0 95.7 2301 2160 2845 1425
Area and residential proximity to primary roads:
More than 5 km away from primary road 233.2 120.6 932.8 482.5 130.5 115.8 1333.2 1219.9 3256.0 2141.1
Within 5 km of a primary road 175.5 64.9 702.1 259.6 103.8 92.4 2432.6 2226.5 2789.7 1344.2
Correlations among area measures
Table 2 shows the correlation matrix of the areas for all five measures. The areas of the SDE measures were perfectly correlated; latter analyses will illustrate their differences. The RNB and RTT measures were strongly correlated (0.924) despite the drastically different size and shape of the measures. The RNB had the strongest correlation with the other measures. The STT was weakly, negatively correlated with the other four measures. The ranges of values suggest that each measure maps a different activity space and will capture different, perhaps complementary, information. The scatterplot matrix in Figure 5 provides a visual representation of the correlation matrix.
Table 2 Correlations between areas of the five measures, and between the measures and potential determinants. Spearman's rank correlation. SDE1 = standard deviational ellipse at 1 standard deviation; SDE2 = standard deviational ellipse at 2 standard deviations; RNB = road network buffer; STT = standard travel time polygon; RTT = relative travel time polygon.
SDE1 SDE2 RNB STT RTT
Correlations between the areas of the activity space measures
SDE1 1.000
SDE2 1.000 1.000
RNB 0.782 0.782 1.000
STT -0.113 -0.113 -0.153 1.000
RTT 0.709 0.709 0.924 -0.104 1.000
Correlation between area and the number of reported activity destinations 0.355 0.355 0.479 0.068 0.396
Correlation between area and maximum travel time to a routine destination 0.686 0.686 0.900 -0.362 0.954
Correlation between area and travel time to nearest primary care opportunity 0.277 0.277 0.188 -0.536 0.099
Correlation between area and residence located within 5 km of a primary road -0.154 -0.154 -0.121 0.385 -0.085
Figure 5 Scatterplot matrix of rank-transformed areas of the five activity space measures. Visual representation of the correlation matrix showing the Spearman's rank correlations between area measures. The matrix shows the perfect correlation between the SDE1 and SDE2, the strong correlation between RNB and RTT, and the lack of correlation between the STT and the other measures.
Determinants of area
Respondents reported a mean and median of 9 routine activity destinations, with a minimum of 1 and a maximum of 22; these values were approximately normally distributed. The area of the RNB activity space had the strongest correlation (.479) with the number of activity destinations (Table 2). Since the STT was dependent only on the household location, it had the weakest correlation (.068).
The mean and median travel times to the farthest activity destination were 28.6 and 24.7 minutes respectively, with a minimum of .39 and a maximum of 175 minutes. Because the threshold travel time of the RTT is the same as this time, the two measures were near perfectly correlated. Three other measures also had strong positive correlations with the activity travel time maximum: SDE1 and SDE2 (.686), and the RNB (.900). Again the exception, the STT had a moderate negative correlation (-.362).
None of the area measures were even moderately correlated with travel time to the nearest PCO, except for the STT, which had a moderate negative correlation (-.536). Because the STT measure behaved so differently from the other measures, we hypothesized that location within the road network was the primary determinant of the area of the STT polygon. We then tested the association between area and residential proximity to primary road (defined as within 5 km of a primary road). Only the STT area measure was positively correlated (.385); the other four measures had weak negative correlations. For the STT measure, the median area for those living near primary roads (2227 km2) was larger than for those living more than 5 km from primary roads (1220 km2), indicating that respondents who live near primary roads have the largest potential activity space at a given travel time (see Table 1). The reverse was true for the other measures, indicating a tendency for longer actual routine travel distances among those who live further away from primary roads. However, as 88% respondents in this sample lived within 5 km of a primary road, future work might explore the role of proximity to primary roads with shorter distances.
Access to primary care opportunities
We defined accessible primary healthcare opportunities (PCOs) as those located within a respondent's activity space. To examine accessibility, we looked at three sets of descriptive statistics for each representation of activity space: the mean and median number of primary care opportunities within an individual's activity space, the percent of respondents with at least one primary care opportunity inside their activity space, and correlation between the area of activity space and the number of primary care opportunities (Table 3).
Table 3 Access to primary care opportunities for the five activity space measures. Spearman's rank correlation. SDE1 = standard deviational ellipse at 1 standard deviation; SDE2 = standard deviational ellipse at 2 standard deviations; RNB = road network buffer; STT = standard travel time polygon; RTT = relative travel time polygon.
Measure of activity space
SDE1 SDE2 RNB STT RTT
Mean (S.D.) Median (Range) Mean (S.D.) Median (Range) Mean (S.D.) Median (Range) Mean (S.D.) Median (Range) Mean (S.D.) Median (Range)
Number of primary care opportunities 6.04 (9.47) 3 (0–104) 13.71 (18.68) 8 (0–155) 13.06 (12.90) 9 (0–86) 39.59 (37.55) 21 (0–142) 39.33 (44.55) 20 (0–232)
Percent with at least one primary care opportunity 77.6 92.8 97.5 99.9 99.0
Correlation between area and number of primary care opportunities 0.426 0.506 0.441 0.787 0.790
The distribution of the number of PCOs was skewed to the right, like the area measures. The SDE1 – the activity space measure with the smallest median area – also had the smallest median number of PCOs (3). The median number of PCOs was highest for the largest measures, STT (21) and RTT (20). The RNB (9) and SDE2 (8) were similar in spite of the larger area of the SDE2; because the SDE2 captures 95% of all activity destinations, the similarity in number of PCOs with the RNB measure (which captures 100% of activity destinations) is perhaps not surprising.
All but one respondent had at least one primary care opportunity inside the STT activity space, thus one respondent lacked access according to the federal maximum travel time of 30 minutes to a primary care provider. It is worth noting that this respondent's measured travel time to the nearest primary care opportunity was 30.52 minutes. Although at the limit of the 30-minute travel time standard, this respondent's threshold travel time for the RTT measure was 92 minutes, and the same respondent's RTT polygon had 134 primary care opportunities inside.
The range of opportunities was greatest for the RTT (0–232). Only 11 respondents lacked access using the RTT measure; of these, all had very small RTT areas, ranging from 0.04–136 square kilometers (all in the lowest quartile for area). However, all of these respondents did have primary care opportunities within their STT activity space (range of 6–52). The RNB and SDE2 also had a surprisingly high percentage with at least one primary care opportunity (97.5% and 92.8% respectively). Only the SDE1 measure indicated a substantial percentage of respondents without access to a primary care opportunity (22.4%).
Spearman's rank correlation was performed to test the strength of association between the area of activity space and the number of primary care opportunities for each activity space model. While each activity space model demonstrated a positive correlation between the area and the number of opportunities, the association was strongest for the STT (.787) and RTT (.790). This was not necessarily expected, given the weak negative correlation between the two area measures.
Correspondence between actual healthcare destinations and activity space
In order to assess which of the activity space measures best captures the health travel behavior of the respondents, we compared the proportion of respondents whose actual health destinations are captured by each measure (Table 4). The two largest area measures, STT and RTT, capture the health destinations of the highest percentage of respondents (86% and 82%, respectively), while the SDE1 captured the fewest. The SDE2 captured slightly more actual health destinations than the RNB. The RNB measure captured 59% of actual health destinations, indicating that in this sample, the majority did choose providers located near their routine activity destinations or along the assumed path of their travel to these destinations.
Table 4 Correspondence of actual healthcare destination with area and number of primary care opportunities. Among respondents reporting at least one visit to a health care provider in the past year and providing locational data for a provider (n = 807). Spearman's rank correlation. SDE1 = standard deviational ellipse at 1 standard deviation; SDE2 = standard deviational ellipse at 2 standard deviations; RNB = road network buffer; STT = standard travel time polygon; RTT = relative travel time polygon.
Activity space measures
SDE1 SDE2 RNB STT RTT
Percent of respondents whose reported health destination is inside the activity space
Total 40.2 62.3 58.6 86.1 81.7
By Area of Activity Space (quartiles)
Smallest – 1st 22.3 47.7 46.2 77.9 58.3
2nd 37.3 57.9 55.8 86.6 83.6
3rd 44.2 69.5 70.4 87.8 88.5
4th 57.1 75.7 61.9 93.0 95.6
By Number of Primary Care Opportunities (quartiles)
Fewest – 1st 13.5 39.0 29.9 58.7 44.4
2nd 36.3 58.5 46.4 64.5 63.2
3rd 51.3 71.9 50.8 68.5 70.1
4th 67.0 80.9 54.7 74.6 75.1
Correlations
between health destination inside and area of activity space .267 .242 .152 .166 .343
between heath destination inside and number of primary care opportunities .496 .360 .241 .219 .362
The percent of respondents with a health destination inside increased significantly by quartile of area for all measures. For example, the percentage with a health destination inside increased from 58% to 96% from the lowest to highest quartile for the RTT measure. Area was mildly correlated with having a health destination inside the activity space, with the RTT measure having the strongest association (.343) and the RNB the weakest (.152).
Positive correlations were found between having a health destination inside the activity space and the number of PCOs inside the activity space (Table 4). The strongest correlation was with the SDE1 measure (.496) and the weakest with the STT measure (.219). Again, the percentage increased significantly by quartile of number of PCOs. For the SDE1 measure, 14% had a health destination inside at lowest quartile of number of PCOs; this increased to 67% for the highest quartile.
Table 5 shows the results of logistic regression testing the association between area and the success of the activity space in capturing the actual health destination. Unadjusted odds ratios show the strongest effect of area of the RTT (OR = 3.37) and the weakest for the RNB (OR = 1.35). Adjusted odds ratios show the association with area after controlling for the number of primary care opportunities; although the effect of area is reduced, the area of the RTT (OR = 2.33) again has the strongest effect of the five area measures.
Table 5 Predicted odds of the health destination being located inside the activity space by area, adjusted for number of primary care opportunities. Area and Number of SDP measures truncated above 98th percentile to the 98th percentile. Area measures square root transformed. Area measures, counts of primary care opportunities normalized using Z scores and centered around the mean. SDE1 = standard deviational ellipse at 1 standard deviation; SDE2 = standard deviational ellipse at 2 standard deviations; RNB = road network buffer; STT = standard travel time polygon; RTT = relative travel time polygon.
Unadjusted odds ratio CI Adjusted odds ratio CI
Area
SDE1 1.75 1.50, 2.05 1.41 1.17, 1.68
SDE2 1.75 1.47, 2.09 1.40 1.14, 1.72
RNB 1.35 1.17, 1.56 1.18 1.01, 1.39
STT 1.68 1.37, 2.07 1.21 0.92, 1.59
RTT 3.37 2.51, 4.52 2.33 1.57, 3.46
Discussion
Activity space is a measure of where people go on a routine basis. Mapping a person is different from mapping a stationary object – people are not fixed to a single location, and so cannot adequately be represented with a fixed-location, one-dimensional point. By obtaining locations of routine destinations, a two-dimensional measure of activity space can be developed to represent the space a person occupies as they perform the routines of daily life. The increasing use of GIS in accessibility studies has not only made activity space a viable alternative to distance as a measure of geographic accessibility, but has increased the methodological options available to researchers to represent activity space and measure accessibility. This paper compares five of these options; a summary of each measure is provided in Table 6.
Table 6 Summary of activity space measures. SDE1 = standard deviational ellipse at 1 standard deviation; SDE2 = standard deviational ellipse at 2 standard deviations; RNB = road network buffer; STT = standard travel time polygon; RTT = relative travel time polygon.
Representation of accessibility Type and shape of measure Data sources Advantages Disadvantages Specific applications
SDE1
SDE2 Statistical approximation; abstract space Euclidean measure; ellipsoid shape Household + activity locations (multiple points), frequency-weighted Captures spread and orientation of points; can be weighted by frequency Capture 67% and 95% of points, respectively; Euclidean measures do not fully capture surface effects; poor representation of actual activity space; captures opportunities not in activity space; requires a minimum of three unique points to generate ellipse SDE1 has statistical/predictive power (in this study area)
RNB Descriptive, Actual access Network-based measure; shape is buffered network ("worm") Household + activity locations (multiple points), road network Captures 100% of activity destinations – best representation of actual space determined by nodes and routes May be too restricted for predictive purposes Bypassing; accessibility in actual activity space
STT Normative, potential access, single norm (30-minute travel time threshold) Network-based measure; shape is network-derived polygon Household (single point), road network, travel time standard Results are fundamentally different from other measures; captures the highest number of SDPs; area is indicator of relative location within road network Not based on activity destinations; "arbitrary" travel time limit? Most strongly conditioned by location relative to road network Evaluate accessibility according to standards/guidelines.
Area is indicator of proximity to primary roads, services
RTT Normative, potential access, relative norm (individually determined) Network-based measure; shape is network-derived polygon Household (single point), road network individual travel time threshold (maximum travel time to routine activity destination) Captures 100% of activity destinations; highly correlated with RNB (can be used in combination) May overlap with STT; also conditioned by location within road network hierarchy Relative accessibility; area has strongest correlation with health destination
The results support conclusions made by others that there is no one "best" measure for accessibility, that different measures capture or emphasize different dimensions of accessibility, and that the question being asked should determine the appropriate selection of measure [10,46]. The different measures provided different results when asking the same question. A primary care opportunity could appear "accessible" using the SDE1 but not when using the RNB, or vice-versa. The different activity space measures captured different numbers of primary care opportunities, providing a different assessment of accessibility. To the extent that a single measure can be considered the "best" measure to study accessibility, the question or problem to be solved largely determines the most appropriate measure. For example, the RNB measure might be the best approach for studying bypassing behavior, as the measure is the best representation of actual activity space. Different distance thresholds or other accessibility parameters can be established to measure potential access, as we did with the 30-minute travel time standard STT polygon. And RTT combines the descriptive with the normative by using observed distance tolerances to describe individual potential access; results demonstrate the value of this perspective. And the SDE remained valuable – the SDE2 often produced similar results as the other measures; yet given the high level of accessibility in this region, the more discriminating SDE1 may have more statistical power.
The STT measure was distinct from the other measures. The results presented here correspond with Kwan's finding of weak correlations between space-time measures and integral measures and her conclusion that the two types of measures are distinctive and "capture different dimensions of the accessibility experience of individuals" [10]. One interesting consequence of using the 30-minute travel time polygon was how the road network structured the results. Proximity to primary roads influenced the area of the STT activity space and its association with healthcare accessibility in two ways. First, visual analysis of the data confirmed that the primary care opportunities tend to be located on primary roads and clustered around the intersection of two or more primary roads, so that in general, respondents who lived nearer to primary roads also lived nearer to more service delivery points. Second, with higher speed limits assigned to primary roads, respondents who lived nearer to primary roads can travel longer distances in a 30-minute time span and have larger potential activity spaces, which then contain a higher number of primary care opportunities. Users of this type of measure should recognize how it represents this specific dimension of accessibility – proximity to primary roads, or more specifically, how it is conditioned by (data on) road type and speed limits. The contrasting trends between the STT and the other measures suggest that while those who live near primary roads can travel greater distances, they also live closer to services and other opportunities and thus tend to have smaller activity spaces. Conversely, those who live farther away from primary roads cannot travel as far within a given a given travel time limit, however they routinely travel farther to routine activities, and tend to have larger activity spaces. While it may seem obvious, this dynamic has not typically been recognized in discussions of activity space or accessibility.
Assumptions about the mode of transportation become very important, as not all respondents have the same access to a private vehicle. As reported elsewhere [47], 13% of respondents did not have a driver's license and approximately 5% of respondents had used public transportation to attend a healthcare visit in the year preceding the interview. In the U.K., GIS has been used to model travel times by various modes of public and private transportation [48], however, the public transportation system is much more well developed in the U.K. than in the U.S., let alone rural America. Developing alternative distance thresholds for persons who do not have access to a car or are unable to drive – situations with specific relevance for regions with high levels of poverty or high proportions of retired and elderly populations – may be useful to better approximate the activity space for those without access to a private automobile.
Although the goal of the paper was primarily methodological, the healthcare context of this region nonetheless influenced our results. The study site was selected for its relative isolation and sparse population density, as well as hypothesized of gaps in healthcare accessibility for some segments of the population; we instead found that the study area had both a well-developed transportation corridors and supply of primary care services [38]. Given the small numbers of respondents who lived more than 20 minutes from a primary care service delivery point, we were unable to compare those with poor access to those with good access (as defined by the 30-minute travel time standard).
While we believe that activity space as a conceptual model has the potential to become a more widely utilized tool in studies of spatial access, we do not wish to overstate the ease of incorporating these methods into many research projects. One of the unique strengths of this work is the rich source of activity data. Having the respondents' home locations as well as the locations and frequency of their activity destinations enabled us to create a reasonably complete activity space. The data and methods section details the data collection effort required for constructing activity spaces, and this level of effort may remain beyond the resources of many researchers. The activity data was only one part of a very long, in-depth survey that took up to two hours to complete and involved the payment of respondents for their time. Moreover, the project was able to contract with a professional research firm to implement the survey. In addition to the survey data, the project was fortunate to acquire a database of healthcare providers for the region that was compiled contemporaneously and had been georeferenced. This conjunction of detailed datasets, as well as access to some higher-end network analysis tools, afforded us the unique opportunity to create and analyze these models of activity space. Unfortunately, given the cost of datasets such as StreetMap 2000, the paucity of georeferenced health services databases, and the data collection requirements, others may not be able to fully duplicate the methods described.
On the other hand, these measures are relatively simple to construct when compared with the models described by Golledge and Stimson [12], or the algorithms that Kwan and colleagues employ to construct space-time measures [9,10,49]. There are tradeoffs between the specificity detailed in 2-day travel diaries and the more spatially- and temporally-generalized measures we constructed. We believe that for the purpose of studying healthcare accessibility, the measures we propose are appropriate and easier to implement; of course this may not be the case for all questions and applications. As desktop applications are developed that simplify the generation of activity spaces [50], the increasing ease of applying these measures will expand the range of potential applications.
Conclusion
The availability of geospatial technologies and data create multiple options for representing and operationalizing the construct of activity space. Echoing Kwan and colleagues [13], our aim is to advance accessibility research by examining the interrelated issues of method, representation, and application. Each of the five measures represent a methodological variation on a single theoretical construct of activity space. If we define "accessibility" as the availability of healthcare opportunities within that individual's activity space, access is influenced by the shape and area of the activity space, the spatial distribution of opportunities, and by the spatial structures that constrain and direct movement through space. The shape and area of the activity space is influenced by individual factors, spatial structures, and the locations of other opportunities; it is also influenced by how activity space is conceptualized and measured.
The paper demonstrates how activity space can be used to (a) examine the correspondence between location of health services and individual activity spaces, (b) assess the extent of bypassing in accessibility studies, and (c) test a travel time standard from the individual perspective, and (d) compare that standard to actual travel patterns. The analysis also shows that, in four of the five representations of activity space, the majority of respondents did in fact use a healthcare provider within their activity space, demonstrating the relevance of activity space to healthcare accessibility.
The five measures presented, SDE1, SDE2, RNB, STT and RTT together provide a multi-faceted picture of a respondent's activity space. The first approach, the SDE, provides a theoretical or abstract representation using Euclidean space. It quantifies the size and general orientation of the spread of destinations to which a respondent travels. It uses abstract space to represent material spatial arrangements and is not well suited for more sophisticated analysis looking at issues of bypassing or modeling routes taken by the respondent. The RNB measure is better at these types of analysis than SDE. By creating buffers around the road network its possible to analyze bypassing and know the network distance necessary to reach a destination, something that is important in an area that has physical barriers to straight, Euclidean travel. The two travel time measures, STT and RTT provide a glimpse of another important aspect of a respondent's activity space, namely travel time. The STT approach uses the federal standard of 30 minutes to determine access to primary care opportunities for each respondent. This can provide a picture of the degree of underservice for the respondent, or more generally, how they fare against the federal guidelines. The other travel time measure, RTT, is linked to the actual destinations a respondent travels to and provides a picture of the burden of travel to healthcare destinations relative to the burden of other routine activity destinations.
We have also shown that each measure of activity space can help inform the interpretation of the others, and conclude that the use of multiple measures is better than relying on a single measure. Triangulation in the social science context refers to examining a phenomenon from multiple perspectives to gain a more complete and nuanced interpretation of data. The most significant burden in any GIS-based analysis involves the collection of spatial data and the development of the spatial database; the additional burden of constructing and comparing multiple measures is relatively small once the database is established, and the benefits of using multiple measures outweigh the additional costs.
List of abbreviations
CFCC Census Feature Class Codes
GIS Geographic Information Systems
GISc Geographic Information Science
GPS Geographic Positioning System
I-26 Interstate 26
I-40 Interstate 40
I-85 Interstate 85
MAP Mountain Accessibility Project
MSA Metropolitan Statistical Area
PC Primary Care
PCO Primary Care Opportunity
RTT Relative Travel Time
RNB Road Network Buffer
SD Standard Deviation
SDE Standard Deviational Ellipse
SDE1 Standard Deviational Ellipse at 1 Standard Deviation
SDE2 Standard Deviational Ellipse at 2 Standard Deviations
STT Standard Travel Time
Authors' contributions
JES contributed to the study design, performed the statistical analysis, and drafted the manuscript. JS contributed to the study design, constructed the spatial measures and performed the spatial analysis, and helped to draft the manuscript. JSP contributed to the overall project design, oversaw data management and statistical analysis, and critically contributed to the manuscript. WMG conceived of the overall project of which this study is a part, assisted in the management of the data collection, and critically contributed to the manuscript. TAA assisted in the design of the overall project, managed data collection, and critically contributed to the manuscript.
Acknowledgements
This research was supported by Grant R01 HS09624 from the Agency for Health Care Quality and Research.
==== Refs
Fryer GE Drisko J Krugman RD Vojir CP Prochazka A Miller ME Multi-method assessment of access to primary medical care in rural Colorado J Rural Health 1999 15 113 121 10437338
Fortney J Rost K Warren J Comparing alternative methods of measuring geographic access to health services Health Services & Outcomes Research Methodology 2000 1 173 184
Parker EB Campbell JL Measuring access to primary medical care: some examples of the use of geographical information systems Health Place 1998 4 183 193 10671022 10.1016/S1353-8292(98)00010-0
Gesler WM Albert DP Albert DP, Gesler WM, Levergood B How spatial analysis can be used in medical geography Spatial Analysis, GIS and Remote Sensing Applications in the Health Sciences 2000 Chelsea, MI , Ann Arbor Press 11 38
Yuill RS The standard deviational ellipse: An updated tool for spatial description Geogr Ann 1971 53B 28 39
Shannon GW Spurlock CW Urban ecological containers, environmental risk cells, and the use of medical services Econ Geogr 1976 52 171 180
Gesler WM Meade MS Locational and population factors in health care-seeking behavior in Savannah, Georgia Health Serv Res 1988 23 443 462 3403277
Mey MG Ter Heide H Towards spatiotemporal planning: practicable analysis of day-to-day paths through space and time Environ Plan B 1997 24 709 723
Kwan MP Gender and individual access to urban opportunities: A study using space–time measures Prof Geogr 1999 51 211 227 10.1111/0033-0124.00158
Kwan MP Space-time and integral measures of individual accessibility: A comparative analysis using a point-based framework Geogr Anal 1998 30 191 217
Hoggart K Lees L Davies A Researching Human Geography 2001 London , Arnold 359
Golledge RG Stimson RJ Spatial Behavior: A Geographic Perspective 1997 New York and London , The Guilford Press
Kwan MP Murray AT O'Kelly ME Tiefelsdorf M Recent advances in accessibility research: Representation, methodology and applications Journal of Geographical Systems 2003 5 129 138 10.1007/s101090300107
Golledge RG Stimson RJ Analytical Behavioural Geography 1987 New York , Croom Helm 345
Nemet GF Bailey AJ Distance and health care utilization among the rural elderly Soc Sci Med 2000 50 1197 1208 10728841 10.1016/S0277-9536(99)00365-2
Rindfuss R Stern PC Liverman D, Moran EF, Rindfuss RR, Stern PC Linking remote sensing and social science: The need and the challenges People and Pixels: Linking Remote Sensing and Social Science 1998 Washington DC , National Academy Press 1 27
Johnston RJ Gregory D Smith DM The Dictionary of Human Geography 1994 Third Malden, MA , Blackwell Publisher
Girt JL Distance to general medical practice and its effect on revealed ill-health in a rural environment Can Geogr 1973 XVII 154 166
Shannon G Lovett J Bashshur R Travel for primary care: expectation and performance in a rural setting J Community Health 1979 5 113 125 318056 10.1007/BF01324013
Mayer JD The distance behavior of hospital patients: A disaggregated analysis Soc Sci Med 1983 17 819 827 6879240 10.1016/0277-9536(83)90032-1
Bronstein JM Morrisey MA Determinants of rural travel distance for obstetrics care Med Care 1990 28 853 866 2205769
Cloke P Rural life-styles: Material opportunity, cultural experience, and how theory can undermine policy Econ Geogr 1996 72 433 449
Thurston WE Meadows LM Rurality and health: perspectives of mid-life women Rural Remote Health 2003 3 (online):219
Lefever DW Measuring geographic concentration by means of the standard deviational ellipse Am J Sociol 1926 32 88 94 10.1086/214027
Joseph AE Phillips DR Accessibility and Utilization: Geographical Perspectives on Health Care Delivery 1984 New York , Harper & Row 214
Ricketts TC Savitz LA Gesler WM Osborne D Geographic Methods for Health Services Research: A Focus on the Rural-Urban Continuum 1994 Lanham, MD , University Press of America
Buczko W Bypassing of local hospitals by rural Medicare beneficiaries J Rural Health 1994 10 237 246 10139173
Elliot-Schmidt R Strong J The concept of well-being in a rural setting: understanding health and illness Aust N Z J Public Health 1997 5 59 63
Bronstein JM Morrisey MA Bypassing rural hospitals for obstetrics care J Health Polit Policy Law 1991 16 87 118 2066540
Akin JS Hutchinson P Health-care facility choice and the phenomenon of bypassing Health Policy Plan 1999 14 135 151 10538717 10.1093/heapol/14.2.135
Holoweiko M Bypassing primary-care physicians Med Econ 1997 74 208 216 (8) 10166515
Mackian S Bedri N Lovel H Up the garden path and over the edge: where might health-seeking behaviour take us? Health Policy Plan 2004 19 137 146 15070862 10.1093/heapol/czh017
Jones RE Fly JM Talley J Cordell HK Green migration into rural America: The new frontier of environmentalism? Soc Nat Resour 2003 16 221 238 10.1080/08941920309159
McGranahan DA Natural amenities drive rural population change Agricultural Economic Report No 781 1999 Washington, DC , Food and Rural Economics Division, Economic Research Service, U.S. Department of Agriculture
Cromartie J Migrants in the rural South choose urban and natural amenities Rural America 2001 15 7 18
Gesler W Arcury TA Preisser J Trevor J Sherman JE Spencer J Access to care issues for health professionals in the mountain region of North Carolina Int Q Community Health Educ 2000 20 82 102
Arcury TA Gesler WM Preisser JS Sherman J Spencer J Perin J The effects of geography and spatial behavior on health care utilization among the residents of a rural region Health Serv Res 2005 40 135 156 15663706 10.1111/j.1475-6773.2005.00346.x
Gesler WM Sherman JE Spencer J Arcury TA Preisser JS Perin J Exploring inequalities in health care coverage by degree of rurality in Western North Carolina Southeast Geogr
ESRI ArcView StreetMap 2000 Extension 1992 Redlands, CA , Environmental Systems Research International
McMillan and Moss Research Inc. Health Access GIS Database 2000 Mebane, NC
Levine N CrimeStat II: A Spatial Statistics Program for the Analysis of Crime Incident Locations 2002 Houston, TX and Washington, DC , Ned Levine & Associates and the National Institute of Justice
ESRI ArcView 3.3 Network Analyst Extension 1992 Redlands, CA , Environmental Systems Research International
Bosanac EM Parkinson RC Hall DS Geographic access to hospital care: a 30-minute travel time standard Med Care 1976 14 616 624 940405
U.S. Department of Health and Human Services Health Professional Shortage Area Primary Medical Care Designation Criteria 1993 Washington, DC , Bureau of Health Professions, Health Resources and Services Administration, U.S. Department of Health and Human Services
StataCorp Stata 8.2 1984 College Station, TX
Handy SL Niemeier DA Measuring accessibility: An exploration of issues and alternatives Environ Plan A 1997 29 1175 1194
Arcury TA Preisser JS Gesler WM Powers JM Access to transportation and health care utilization in a rural region J Rural Health 2005 21 31 38 15667007
Martin D Wrigley H Barnett S Roderick P Increasing the sophistication of access measurement in a rural healthcare study Health Place 2002 8 3 13 11852259 10.1016/S1353-8292(01)00031-4
Kim HM Kwan MP Space-time accessibility measures: A geocomputational algorithm with a focus on the feasible opportunity set and possible activity duration Journal of Geographical Systems 2003 5 71 91 10.1007/s101090300104
Buliung RN Kanaroglou PS A GIS toolkit for exploring geographies of household activity/travel behavior Journal of Transport Geography
|
16236174
|
PMC1283149
|
CC BY
|
2021-01-04 16:39:06
|
no
|
Int J Health Geogr. 2005 Oct 19; 4:24
|
utf-8
|
Int J Health Geogr
| 2,005 |
10.1186/1476-072X-4-24
|
oa_comm
|
==== Front
J Inflamm (Lond)Journal of Inflammation (London, England)1476-9255BioMed Central London 1476-9255-2-121624625810.1186/1476-9255-2-12ResearchHeat shock protein and heat shock factor 1 expression and localization in vaccinia virus infected human monocyte derived macrophages Kowalczyk Aleksandra [email protected] Krzysztof [email protected] Kinga [email protected] Jakub [email protected] Hanna [email protected] Jagiellonian University, Faculty of Biotechnology; 7, Gronostajowa St., 30-387 Krakow, Poland2005 24 10 2005 2 12 12 29 4 2005 24 10 2005 Copyright © 2005 Kowalczyk et al; licensee BioMed Central Ltd.2005Kowalczyk et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
Viruses remain one of the inducers of the stress response in the infected cells. Heat shock response induced by vaccinia virus (VV) infection was studied in vitro in human blood monocyte derived macrophages (MDMs) as blood cells usually constitute the primary site of the infection.
Methods
Human blood monocytes were cultured for 12 – 14 days. The transcripts of heat shock factor 1 (HSF1), heat shock protein 70 (HSP70), heat shock protein 90 (HSP90) and two viral genes (E3L and F17R) were assayed by reverse transcriptase-polymerase chain reaction (RT-PCR), and the corresponding proteins measured by Western blot. Heat shock factor 1 DNA binding activities were estimated by electrophoretic mobility shift assay (EMSA) and its subcellular localization analyzed by immunocytofluorescence.
Results
It appeared that infection with vaccinia virus leads to activation of the heat shock factor 1. Activation of HSF1 causes increased synthesis of an inducible form of the HSP70 both at the mRNA and the protein level. Although HSP90 mRNA was enhanced in vaccinia virus infected cells, the HSP90 protein content remained unchanged. At the time of maximum vaccinia virus gene expression, an inhibitory effect of the infection on the heat shock protein and the heat shock factor 1 was most pronounced. Moreover, at the early phase of the infection translocation of HSP70 and HSP90 from the cytoplasm to the nucleus of the infected cells was observed.
Conclusion
Preferential nuclear accumulation of HSP70, the major stress-inducible chaperone protein, suggests that VV employs this particular mechanism of cytoprotection to protect the infected cell rather than to help viral replication. The results taken together with our previuos data on monocytes or MDMs infected with VV or S. aureus strongly argue that VV employs multiple cellular antiapoptotic/cytoprotective mechanisms to prolong viability and proinflammatory activity of the cells of monocytic-macrophage lineage.
==== Body
Background
Manipulation of the immune system, especially interference with specific components of the apoptotic response of the infected cells is essential for a virus to replicate and to disseminate in a host.
Vaccinia virus belongs to the poxviruses super-family, a group of large DNA viruses known from their exclusive propagation outside the nucleus in the cytoplasm of the infected cell [1]. Vaccinia virus infections are commonly associated with a generalized host cell protein and nucleic acids synthesis inhibition, depending on time and an infectious dose. Despite the observed shutdown of host transcriptional and translational mechanisms and selective expression of many viral genes, several eukaryotic proteins are transiently induced or activated by poxviruses, e.g. transcription factors [2], cytokines [3,4], heat shock proteins [5] and antioxidant enzymes [6]. Moreover, although mainly necrotic, vaccinia virus is opposing the apoptosis due to several anti-apototic genes present and expressed from its genome [7,8].
Stress conditions like heat shock, infections, radiation, and exposure to chemicals induce increased levels of heat shock proteins in many cell lines [9]. The heat shock proteins can be induced in vitro following infection by a variety of viruses [10] such as Ad5 and HSV-1 which have been shown to induce synthesis of one of the main heat shock proteins, HSP70. Vaccinia virus was already found to be a potent inducer of HSP70 in mice [11,12]. The role of HSP70 in vaccinia virus infection has not been elucidated so far, however the results of the earlier studies in vaccinia virus infected U937 cells and primary macrophages suggest its role in viral protein folding and virus assembly [5]. Moreover, in vivo studies in mice reveal lack of the influence of infection on viral life cycle [12]. Obviously, HSPs constitute specific chaperons for the viral proteins necessary to secure proper folding, translocation and formation of multi-component complexes of the viral proteins. Recent investigations indicate that the heat shock proteins exert suppression of the apoptosis [13-15] and therefore might support vaccinia virus infection.
Induction of the heat shock protein synthesis requires earlier activation of heat shock factors. HSF1 is assumed to be the main mediator of the cellular stress response, which binds to the heat shock promoter element (HSE) [16,17]. It is believed that in normal conditions monomers of HSF1 exist inactive in the cytoplasm in large complexes with other heat shock proteins, e.g. HSP90, and HSP70. Upon stress, when the heat shock proteins are needed, HSF1 undergoes trimerization, subsequent translocation into the nucleus and binding to the heat shock elements within the regulatory sequences of the heat shock protein genes [15,18].
To understand further the role of vaccinia virus in the course of the infection, the heat shock response was studied in human blood monocyte derived macrophages infected with the vaccinia virus Western Reserve strain.
Methods
Cell culture
Peripheral blood leukocytes (PBL) were isolated by standard Ficoll-Paque (Pharmacia, Uppsala, Sweden) gradient centrifugation from the blood of healthy donors. The cells were cultured at the concentration of 2 × 107 of PBL cells per 5.5 cm dish for protein harvesting or at the concentration of 8 × 106 of PBL cells per 3.5 cm dish for immunocytochemical analyses. The cells were cultured for 10–14 days in RPMI medium (Gibco) with 10% human serum (AB serotype); medium was changed every 48 hours until the monocytes reached adherence. The adherent monocytes constitute 10 % of the total PBLs placed on the dish. Human hepatoma HepG2 cells were cultured in DMEM with 10% FCS in 60 mm-diameter culture dishes for 48 hours before infection or heat shock.
Virus propagation
The vaccinia virus Western Reserve strain was propagated in VERO-B4 cells (DSMZ, Germany) infected at multiplicity of infection (MOI) of 1 (one plaque forming unit, pfu, per cell) maintained in MEM supplemented with 4% heat-inactivated FCS. Infected cells were harvested when the maximum cytopathic effect was observed and infectivity was estimated by a quantal infectivity assay on VERO-B4 cells [19] and a standard plaque assay. Human macrophages were infected with the virus at multiplicity of infection 1 or 5. Infected cells were washed after 1 h of virus adsorption and fresh medium added.
Heat shock
The heat shock was performed in 42°C in the water bath for one hour, followed by 2–4 hours recovery at 37°C.
Protein isolation
The cells were washed with 1 ml cold PBS and harvested to the Eppendorf 2 ml tubes in 1–2 ml of PBS. The harvested cells were centrifuged at 250 × g for 5 min. The cell pellet was suspended either in 400 μl of the resuspension buffer for isolation of the nuclear and cytoplasmic fractions of proteins according to the Suzuki method [20], or in 150 μl of the extraction buffer (50 mM Tris pH 8.0, 10 mM CHAPS, 2 mM EDTA, 1 mM Na3VO4, 5 mM DTT, 1 mM PMSF, 10% glycerol) for whole cell extracts [21]. According to the Suzuki method, the nuclear fraction contains all nuclei, and the remaining supernatant is termed "cytoplasmic fraction". Contamination of the nuclei with cytoplasm was excluded based on lactate dehydrogenase activity measurements using a LDH detection kit from Boehringer Mannheim.
Protein concentrations were measured using the BCA assay (Sigma) based on bicinchoninic acid [22]. The absorbance was measured at 562 nm in SpectraMax 250 microplate reader (Molecular Devices).
Western blot
Equal amounts of protein extracts (10 μg/lane) were separated by SDS-PAGE according to the protocol described by Laemmli [23]. The protein transfer was performed in a semi-dry blotting system (Fastblot B31, Biometra) in the transfer buffer (25 mM Tris pH 8.3, 0.2 M glycine, 20% methanol, v/v) at 35 V for 30 min. Equal loading of samples, and even transfer, were confirmed by staining the membranes with Ponceau S. The membrane (Hybond, Amersham Pharmacia) was blocked with 5% powdered milk in TST buffer (10 mM Tris HCl pH 7.5, 0.9 % NaCl, 0.05 % Tween 20) for 1.5 h, followed by a 20-min wash in the TST buffer. The membrane was incubated either with the primary anti-HSF1 (H-311, sc-9144), or anti-HSP70 (K-20, sc-1060), or anti HSP90 antibodies (H-114, sc-7947) from Santa Cruz Biotechnology, in 1:1000 dilution in the TST buffer with 2 % BSA for 1 h. The membrane was then washed four times in TST buffer for 15 min. Secondary anti-rabbit IgG antibodies coupled to horseradish peroxidase (Amersham Pharmacia) were diluted 1:5000 in TST buffer with 2 % BSA. The membrane was incubated with the secondary antibodies for 1 hour, followed by four washes of 15 min. in TST buffer. The ECL-plus kit (Amersham Pharmacia) was used to visualize the protein. The membranes were exposed into X-ray films for 10 minutes to 1 hour, and the films were developed.
Electrophoretic mobility shift assay
A DNA mobility shift assay was carried out as described by Duyao [24]. The double-stranded oligonucleotides containing the HSF binding site (5'-CTAGAAGCTTCTAGAAGCTTCTAGAA-3') were an "optimal" heat shock element (HSE) containing five perfect inverted nGAAn repeats from the human hsp70 [25]. DNA fragments were labeled using Klenow polymerase and [α-32P]dCTP by filling 5'-overhangs of four bases at both ends after annealing. Equal amounts of protein (5 μg) in 10% glycerol were incubated at room temperature for 30 min. with 0.5 ng of the labeled dsDNA oligonucleotide in the presence of 2 μg of poly(dI-dC) in 10 mM Tris pH 7.5, 50 mM NaCl, 1 mM EDTA and 0.1 mM DTT in a total volume of 20 μl. For supershift analysis, the rabbit polyclonal antibodies against human HSF1 (H-311, sc-9144X) from Santa Cruz Biotechnology were also preincubated with protein extracts in 1:20 dilutions. Incubation mixtures were electrophoresed on 4.5% nondenaturing polyacrylamide gel in 0.5 × TBE. The dried gels were analyzed by autoradiography.
RNA extraction and RT-PCR
Total RNA was extracted from cultured cells using Trizol reagent (Gibco). RNA samples (2 μg) were used for cDNA synthesis reactions in a total volume of 20 μl containing 10 μl of each RNA sample, 0.5 μg oligo (dT)12–18 primer (Gibco) and 200 U of SuperScript II RNAse H-Reverse Transcriptase (Gibco) according to the protocol provided with the enzyme. Although some RNA samples were treated with RNase-free DNase to remove all genomic DNA prior to the RT reaction, similar results were received using DNase untreated RNA preparations. The PCR reactions were done using F105S Taq polymerase (Polygen) in the mixes containing: 5 μl 10 × PCR buffer, 1 μl 10 mM dNTPs, 2 μl of each primer, 50 mM KCl, 1.5 mM MgCl2, 2.5 U of (1 μl) Taq polymerase, 2 μl cDNA and 37 μl sterile water. Reactions were carried out at the following conditions: 94°C for 1 min, 60°C for 1 min, and 72°C for 1.5 min for 30 cycles (hsp70 and hsf1) or 95°C for 1 min, 50°C for 1 min, and 72°C for 1.5 min for 30 cycles (hsp90α) or 94°C for 1 min, 55°C for 1 min, and 72°C for 1.5 min for 30 cycles (E3L, F17R, and β-actin). Each thermal profile was ended with the final extension at 72°C for 15 min. The reaction products were then resolved on nondenaturing 2% agarose gel and visualized by staining with ethidium bromide. The primer sequences are listed in Table 1. The primers were designed to match sequences in separate exons (except for the hsp70 encoded by a single exon) to avoid the contribution of genome-templated product in the signal analysis.
Table 1 Sequences of the primers used in the RT-PCR reaction and the size of the amplified products of human and viral genes
Gene of interest Size of the product Primer orientation Primer sequence
β-actin 307 bp Forward
Reverse 5' AGCGGGAAATCGTGCGTG 3'
5' GGGTACATGGTGGTGCCG 3'
hsf1 577 bp Forward
Reverse 5' ATGGCCAGCTTCGTGCG 3'
5' ACAGCATCAGGGGCGTA 3'
hsp70 590 bp Forward
Reverse 5' TTTGACAACAGGCTGGTGAACC 3'
5' GTGAAGGATCTGCGTCTGCTTGG 3'
hsp90 1498 bp Forward
Reverse 5' GCTGTGCCGTTGGTCCTGTGC 3'
5' GGTTCTCCTTCATTCTGGTGC 3'
E3L 360 bp Forward
Reverse 5' TATATTGACGAGAGTTCTGAC 3'
5' ACTCATTAATAATGGTGACAGG 3'
F17R 283 bp Forward
Reverse 5' ATTCTCATTTTGCATCTGCTC 3'
5' AGCTACATTATCGCGATTAGC 3'
Immunofluorescence cell staining
The cells were cultured on sterile glass cover slips mounted in 3.5 cm culture dishes. Cells were fixed with 3% paraformaldehyde in PBS at 37°C for 15 min and permeabilized with 0.1% Triton-X-100 in PBS for 5 min at room temperature. Nonspecific binding sites were blocked with 3% bovine serum albumin solution in PBS and cells were stained with the anti-human HSF1 (H-311, sc-9144) rabbit polyclonal antibodies (Santa Cruz Biotechnology) in 1:200 dilution. The secondary sheep anti-rabbit Cy3 conjugated IgGs (Sigma, C2306) were used in 1:200 dilution. Nuclear DNA was additionally stained with Hoechst 33258 (Molecular Probes) at a concentration 0.5 μg/ml for 10 min at room temperature and then washed three times with PBS. Cover slips were mounted on microscopic glass slides using Vectashield (Vector Laboratories) to prevent fading of the fluorescent dye. Microphotographs were taken using a Leitz Orthoplan microscope with an epifluorescence and phase-contrast optics equipped with the Nikon FX-35DX camera on high sensitivity Kodak TMAX 3200 films. From one spot both, phase-contrast pictures as well as fluorescence pictures were taken.
Results
Changes in the heat shock factor 1 and the heat shock protein mRNAs content during vaccinia virus infection of human blood macrophages
The levels of HSF1, HSP70 and HSP90 mRNAs were determined by RT-PCR in control and virus-infected macrophages. In most unstressed cells, neither HSP70, nor HSP90 mRNA were detected (Fig. 1A), although in some cultures a basal level of both transcripts was visible (Fig. 1B). This heterogeneity was probably due to individual features of blood or serum donors. In contrary, HSF1 mRNA was constitutively accumulated in all examined cultures. The analysis showed no increase of the HSF1 transcripts up to 48 h p.i. (with a low infectious dose, 1 pfu/cell), similarly to the heat shock response when HSF1 mRNA is not induced (Fig. 1A). However, biphasic kinetics of HSF1 mRNA is observed after high dose infection (5 pfu/cell) with the minimum at 6 to 24 h p.i. corresponding to the maximal viral gene transcription (Fig. 1B) [5]. Subsequent decline in viral transcription was followed by increased HSF1 mRNA content.
Figure 1 RT-PCR analysis of the heat shock factor 1 and heat shock proteins in vaccinia virus infected human blood macrophages. HSF1, HSP70, HSP90α and β-actin mRNAs were measured at 3, 6, 16, 24, 48, 72 and 96 hours p.i. PCR data come from a single representative experiment being one of three separate experiments using the cells from healthy donors. Vaccinia virus (V) infection was carried out at MOI 1 (A) and 5 (B), control (C). HS – heat shock at 42°C for 1 h plus recovery at 37°C for 2 h. β-actin gene product was used as a control.
HSP70 mRNA increased early upon infection with a high vaccinia virus dose, and clear decrease was found fairly late, at 96 hours p.i. The transcript increase after low infectious dose was slowly reaching the maximum at 48 h p.i. The kinetics of HSP90 mRNA increase upon high vaccinia virus dose was similar to that of HSP70 mRNA, however its level decreased earlier than the levels of HSP70 transcript, as this was observed already at 72 hours p.i. At MOI 1, HSP90 mRNA increase was slow, similar to the increase of HSP70 mRNA. HSP70, HSP90 and HSF1 transcripts estimated after the heat shock of the macrophages are also included for comparison and β-actin transcript is shown as a control.
Viral gene expression in the macrophages
In order to check the viral infection itself, two viral genes were chosen: early gene E3L, responsible for the vaccinia virus antiapoptotic defence on the interferon pathway [26], and late viral gene, F17R, the product of which takes part in the mature virion assembly [27]. The RT-PCR of the viral genes showed an increased amount of E3L mRNA at 4 h p.i., which was maintained up to 96 h p.i. F17R mRNA was detected also at 4 h p.i. but increased at 14 h and maintained elevated up to 96 h p.i. (Fig. 2). The results evidenced, that late viral DNA replication had not been stopped in the macrophages. Moreover, increased and persistent levels of E3L mRNA support our conclusion on resistance to apoptosis elicited in the infected cells.
Figure 2 Viral early (E3L) and late (F17R) genes expression in the infected human adherent monocytes. Two vaccinia virus transcripts of E3L and F17R genes and of the cellular gene, β-actin, as a control, were estimated by RT-PCR. Representative results of three independent experiments are shown.
HSF 1 protein activity and localization in the vaccinia virus infected macrophages
Heat shock factor 1 DNA binding activity was analyzed in the nuclear and whole cell extracts of macrophages by an electrophoretic mobility shift assay. EMSA showed protein binding to the heat shock element in the control, at 16, and 24 hours of vaccinia virus infection (Fig. 3A). Supershift analysis of the whole cell extracts from vaccinia virus infected cells and the extracts from uninfected cells confirmed that heat shock factor 1 was present in equal quantities (Fig. 3A). However, nuclear proteins isolated at 16 h p.i. (Fig. 3A, lane 5) did not form the clear shifted antibody-HSE complex, suggesting that the epitopes recognized by the polyclonal antibodies are obscured by other proteins. Similar DNA-protein complex, without the antibodies against HSF1 added, was observed only in vaccinia virus-infected and heat shock treated human hepatoma HepG2 cell line (Fig. 3B).
Figure 3 Vaccinia virus-induced HSE binding activity in macrophages and a human hepatoma cell line. (A) Macrophages were infected with vaccinia virus (V) with 5 pfu/cell and cultured for 16 or 24 hours (supershift assay). Lane 1 – NE from infected macrophages at 16 h p.i., lane 2 – WCE from control macrophages, lane 3 – WCE from infected macrophages at 16 h p.i., lane 4 – WCE from infected macrophages at 24 h p.i.; lanes 5–8 – as lanes 1–4 plus preincubation with 1:20 dilution of antibodies against HSF1 (aHSF1). (B) HepG2 cells (3 × 106) were infected with vaccinia virus at MOI 1 for 24 h or heat shock treated (44°C, 20 min) or heat shock treated and recovered for 2 or 4 h at 37°C (shift assay). Lane 1 – control cells, lane 2 – heat shock treated, lane 3 – heat shock treated and recovered for 2 h, lane 4 – heat shock treated and recovered for 4 h, lane 5 – vaccinia virus infected for 24 h. O – ds oligoDNA (free HSE) incubated without proteins. Exposure time: 6 days (A) and 18 h (B). A single representative experiment being one of four separate experiments is shown.
Although HSF1 protein content did not change in the whole cell extracts after vaccinia virus infection and during the heat shock response (Fig. 4A), more HSF1 accumulated in the nuclei of infected cells than in the nuclei of control cells, especially at 48 h p.i. (Fig. 4B and 4C) as Western blot analysis revealed. The analysis with a polyclonal anti-HSF1 serum shows two HSF1 bands with mobilities of approximately 70 and 80 kDa, which differ in phosphorylation state (Fig. 4C) [28]. The hyperphosphorylation of HSF1 and translocation of the factor into the nucleus of vaccinia virus-infected macrophages was clearly seen at 24 h p.i. (Fig. 4C).
Figure 4 Western blot analysis of subcellular localization of HSF1 in vaccinia virus infected macrophages. Whole cell extracts (WCE) (A), nuclear extracts (NE) (B, C) and cytoplasmatic fraction(CYT)(B, C) of vaccinia virus infected macrophages were analysed by Western blot. C – control, HS – heat shock. Vaccinia virus (V) infection was carried out at MOI 5.
Indirect immunocytochemical staining of macrophages with anti-HSF1 antibodies (the secondary antibodies conjugated with Cy3) showed prevalent nuclear and weak cytoplasmic localization of the factor in the control cells (Fig. 5). Even more protein was observed in the nuclei and cytoplasm of vaccinia virus infected cells at 24 h p.i. The percentage of the cells containing HSF1 exclusively in their nuclei was calculated based on the immunocytochemical staining of the cells and mean values of at least 100 cells randomly selected on each sample were 24% and 46% for infected and control cells respectively.
Figure 5 Vaccinia virus-induced HSF1 redistribution. Cells uninfected (control) or infected for 24 h (MOI 5) were fixed and allowed to react with anti-human HSF1 antibodies (A) or stained with Hoechst 33258 (B). Panels A, B – epifluorescence, C – phase-contrast picture of the same cells. Representative images of three independent experiments are shown. The inserted bar – 20 μm.
HSP70 protein increases during infection and transiently accumulates in the nucleus
HSP70 protein content increased during the first 14 hours of infection (Fig. 6A) but no as much as it was shown for the heat shock treated macrophages. The increase reflected earlier changes in HSP70 mRNA content shown in Fig. 1. Prevalent nuclear accumulation of the protein was observed fairly late at 24 and 48 hours p.i. (Fig. 6B), while no change was found at 4 h p.i. (Fig. 6C). Data from the heat shock treated and the heat shock recovered cells are also included (Fig. 6C).
Figure 6 Changes in HSP70 content in vaccinia virus infected macrophages. Whole cell extracts (WCE) (A), nuclear extracts (NE) (B, C) and cytoplasmatic fraction (CYT) (B, C) of vaccinia virus infected macrophages were analysed by Western blot. C – control, HS – heat shock (proteins extracted after 4 hours recovery from heat shock). Vaccinia virus (V) infection was at MOI 5.
HSP90 protein does not increase during infection but transiently locates in the nucleus
HSP90 protein content did not change during in vitro infection as estimated by Western blot in the whole cell extracts (Fig. 7A). However, early (at 14 hours p.i.) increase in the nuclear content of the protein was found similarly to the results obtained for HSP70 protein (Fig. 7B). Additional analysis of HSP90 content in the heat shock treated and the heat shock recovered macrophages revealed that the heat shock similarly to the vaccinia virus infection, cause HSP90 protein translocation into the nucleus and the effect was clearly seen 4 hours after the heat shock (Fig. 7C).
Figure 7 HSP90 content in vaccinia virus infected macrophages. Whole cell extracts (WCE) (A), nuclear extracts (NE) (B, C) and cytoplasmatic fraction (CYT) (B, C) of vaccinia virus infected macrophages were analysed by Western blot. C – control, HS – heat shock, HS4 – heat shock and 4 h recovery. Vaccinia virus (V) infection was at MOI 5.
Discussion
Several cellular proteins are used by poxviruses and one of the examples is HSP70, which aggregates with viral proteins in the cytoplasm [5]. Although vaccinia virus life cycle does not appear to depend on HSP70 expression [12], the HSP70 transcripts as well as the protein increase significantly in human macrophages at 4 to 24 h p.i., as shown by us and others [5]. It has been well documented that protection against stress-induced apoptosis depends on the chaperone function of HSP70 [14]. Therefore, the results presented in this study indicate that HSP70 might be one of the factors responsible for the survival of VV-infected macrophages. In contrast to the results presented by others we also suggest a more important role of the nuclear pool of HSP70 [5]. Our data showing the predominant nuclear localization of HSP70 do not support the hypothesis on the possible role of HSP70 in folding of viral proteins [5], but speak rather in favor of its protective role in biogenesis of ribosomes within the nucleoli of the infected cells [29].
The role of HSP90 in viral infection, especially its nuclear accumulation (Fig. 7), remains unclear. Our earlier studies [4] have already revealed the stimulatory effect of the vaccinia virus infection on IL-10 gene expression in human blood elutriated monocytes. The finding stays in agreement with the data on the enhancement of the hsp90 gene expression by IL-10 in a human hepatoma HepG2 cell line and peripheral blood mononuclear cells [30]. The lack of HSP90 protein induction in the vaccinia virus infected cells was already found by others [10,31]. However, these authors [10] failed to detect an increased induction of HSP70, and this observation stays in contrast to our results (Fig. 6). Moreover, differential kinetics of HSP70 and HSP90 mRNA levels following exposure to a heat shock in human blood adherent monocytes was also found [32], therefore the heat shock response seems to be similar in this aspect to the vaccinia virus infection.
HSF1 is not a stress-inducible protein, neither is its expression level coupled to the rate of expression of the heat shock genes [33]. Although the vaccinia virus infection causes transient increase of HSF1 mRNA, no increase in HSF1 protein content is found, probably due to the instability of its mRNA. On the other hand, the decrease in HSF1 mRNA observed at the beginning of the infection does not severely affect the protein content because of a fairly long half life time of HSF1 protein [15]. The small decrease in HSF1 content found by us on the third day p.i. (Fig. 4A), might result from limited cellular protein synthesis observed during the prolonged viral infection.
In resting cells, HSF1 is predominantly found in a diffuse cytoplasmic and nuclear distribution, and after the heat shock it relocates rapidly to form large and irregularly shaped nuclear granules [34]. These nuclear structures, referred to as the HSF1 stress granules, can be induced by various stresses, and are detected in different cell types [35]. In resting human cells the predominant nuclear localization of HSF1 before and after the heat shock has been reported [36], and our analysis suggests that HSF1 partially remains in the cytoplasm of the infected macrophages (Fig. 5). Active translocation of several proteins from the nucleus to serve as transcription factors was already found for some viruses, which conduce their life cycle in the cytoplasm. Recent findings provide evidence that YY1 translocates into the cytoplasm of the vaccinia virus infected cells to serve as an activator of one of vaccinia late genes [37]. The factor is recruited to the cytoplasm of the vaccinia virus-infected macrophages through an exportin-1 system, sensitive to leptomycin B [38].
The vaccinia virus infection resulted in massive recruitment of the HSE-binding activity in the investigated cells (Fig. 3). Surprisingly, only a small fraction of this activity was recognised by the anti-HSF-1-specific antibody, and the 'supershifted' fraction was constitutive (Fig. 3A). We speculate that the observed HSE-binding activity contains HSF1, but the most of its epitopes were obscured by the virus-induced chaperones, which accumulated in the nuclei of the infected macrophages in abundant amounts. The speculation is supported by the results presented in Fig. 6B, which demonstrate the preferential nuclear accumulation of HSP70 and the lack of cytoplasmic accumulation of HSP70 (Fig. 6B) after the vaccinia virus infection. Consequently, the observed HSE-binding activity was much stronger in the nuclear extracts (lane 1) than in the whole cell extracts (lanes 2–4) (Fig. 3A). The similar HSE-binding activity was observed in the extracts from the vaccinia-infected or the heat shocked HepG2 cells (Fig. 3B). It is possible that mostly HSP70 and HSP90 recognise and strongly bind the preformed HSF1-HSE complexes in vitro [39]. It seems that the massive nuclear accumulation of stress chaperones is characteristic for the vaccinia virus-infected cells.
MDMs used in our study, survived the VV infection although the virus-induced stress reaction developed accordingly to the infecting dose (Fig. 1). The cytoprotective role of the stress seems evident, since the infected macrophages effectively accumulated different mRNA species for at least 4 days post infection. Routine fluorescent microscopic examination revealed no propidium iodide permeability of the infected cells (not shown). We have previously described that human peripheral blood monocytes retain viability following the infection with low doses of VV [4]. Moreover, in the same experimental conditions HSP70 protected the monocytes against Staphylococcus aureus-induced apoptosis [40]. Apparently, the challenge by S. aureus might be less tolerable for monocytes/macrophages than the one caused by vaccinia virus. It is due to the staphylococcal α-toxin, which is known to initiate this type of monocyte apoptosis [41]. The poxvirus-induced cytoprotection seems to be much more effective than the other types of stress reaction. Such conclusion can be drawn from the predominant nuclear localization of HSP70 induced in human cells by the vaccinia virus. The predominant nuclear localization of HSP70 was also observed in the respiratory syncytial virus infected cells [42]. The nuclear stress reaction has been found essential for protection also against hypoxia and oxidative stress [43]. Viral antiapoptotic proteins like the recently discovered F1L [44] certainly act in concert with Bcl-2 [45] and stress-induced chaperones to prolong lifespan of the infected cells. Little is known about the impact of pathogen-induced monocyte/macrophage apoptosis in immune system. Persistence of professional immune cells harboring intracellular pathogen in a circulation or a lymph tissue seems detrimental for immunity for at least two reasons: firstly, the immune response is deregulated by cytokines and impaired antigen presentation; secondly, the cells were proposed to serve as virus incubators [46].
Cells of monocytic lineage have recently been recognised as a crucial model to study virus-host interactions due to unique capability of these cells to cross-present endocytosed antigens, especially in the context of chaperone proteins [47]. Further understanding of the heat shock response during the vaccinia virus infection may improve strategies of application of vaccinia genome in recombinant gene expression, vaccination and gene therapy.
Declaration of competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
AK carried out monocyte isolation and culture and participated in the immunoassays, KG participated in the design of the study and carried out RT-PCR analysis, KS participated in the gel shift analysis and carried out immunocytochemical analyses, JD participated in the immunoassays, HR conceived the study, participated in its design and coordination, participated in the gel shift analysis and drafted the manuscript. All authors read and approved the final manuscript.
Acknowledgements
This work was supported by grant 6P04A 02116 from the Committee of Scientific Research (Warsaw, Poland).
==== Refs
Moss B Knipe DM, Howley PM The viruses and their replication Fields Virology 2001 2 Philadelphia: Lippincott Williams and Wilkins 2849 2883
Muller JM Ziegler-Heitbrock HWL Baeuerle PA Nuclear factor kappa B, a mediator of lipopolysaccharide effects Immunobiol 1993 187 233 256
Rokita H Kupiec T Guzik K Koj A Vaccinia virus-regulated acute phase cytokine production in human fibroblasts, U937 cells and endothelium Mediat Inflammation 1998 7 73 78 10.1080/09629359891207
Slezak K Guzik K Rokita H Regulation of interleukin 12 and interleukin 10 expression in vaccinia virus-infected human monocytes and U-937 cell line Cytokine 2000 12 900 908 10880234 10.1006/cyto.1999.0646
Jindal S Young RA Vaccinia virus infection induces a stress response that leads to association of Hsp70 with viral proteins J Virol 1992 66 5357 5362 1501279
Dobierzewska A Wator G Rokita H Edeas MA Changes in manganese superoxide dismutase mRNA content and enzymatic activity in vaccinia virus infected human astrocytoma U-373 cell line Superoxide Dismutases: Recent Advances and Clinical Applications 1999 Paris-Tokyo: Editions Mel Paris 168 173
McFadden G Barry M How Poxviruses oppose apoptosis Seminars in Virol 1998 8 429 442 10.1006/smvy.1997.0141
Nash P Barret J Cao J-X Hota-Mitchell S Lalani AS Everett H Xu X-M Robichaud J Hnatiuk S Ainslie C Seet BT McFadden G Immunomodulation by viruses: the myxoma story Immunol Rev 1999 168 103 120 10399068
Morimoto RI Regulation of the heat shock transcritonal response: cross talk between a family of heat shock factors, molecular chaperones, and negative regulators Genes Dev 1998 12 3788 3796 9869631
Phillips B Abravaya K Morimoto RI Analysis of the specificity and mechanism of transcriptional activation of the human hsp70 gene during infection by DNA viruses J Virol 1991 65 5680 5692 1656064
Sedger L Ruby J Heat shock response to vaccinia virus infection J Virol 1994 68 4685 4689 8207845
Sedger L Ramshaw I Condie A Medveczky J Braithwaite A Vaccinia virus replication is independent of cellular HSP72 expression which is induced during virus infection Virology 1996 225 423 427 8918931 10.1006/viro.1996.0619
Jaattela M Heat shock proteins as cellular lifeguards Ann Med 1999 31 261 271 10480757
Mosser DD Caron AW Bourget L Meriin AB Sherman MY Morimoto RI Massie B The chaperone function of hsp70 is required for protection against stress-induced apoptosis Mol Cell Biol 2000 20 7146 7159 10982831 10.1128/MCB.20.19.7146-7159.2000
Verbeke P Fonager J Clark BFC Rattan SIS Review. Heat shock response and ageing: mechanisms and applications Cell Biol Internat 2001 25 845 857 10.1006/cbir.2001.0789
Morimoto RI Cells in stress: transcriptional activation of heat shock genes Science 1993 259 1409 1410 8451637
Wu C Heat shock transcription factors: structure and regulation Annu Rev Cell Dev Biol 1995 11 441 469 8689565 10.1146/annurev.cb.11.110195.002301
Liu PCC Thiele DJ Modulation of human heat shock factor trimerization by the linker domain J Biol Chem 1999 274 17219 17225 10358080 10.1074/jbc.274.24.17219
Burleson FG Chambers TM Wiedbrauk DL Virology A laboratory manual 1991 San Diego: Academic Press
Suzuki YJ Mizuno M Packer L Signal transduction for nuclear factor-κB activation. Proposed location of antioxidant – inhibitable step J Immunol 1994 153 5008 5015 7963561
Sadowski HB Shuai K Darnell JE JrGilman MZ A common nuclear signal transduction pathway activated by growth factor and cytokine receptors Science 1993 261 1739 1744 8397445
Smith PK Krohn RI Hermanson GT Mallo AK Gartner FH Provenzano MD Fujimoto EK Goeke NM Olson BJ Klenk DC Measurement of protein using bicinchoninic acid Anal Biochem 1985 150 76 85 3843705 10.1016/0003-2697(85)90442-7
Laemmli UK Cleavage of structural proteins during the assembly of the head of bacteriophage T4 Nature 1970 227 680 685 5432063 10.1038/227680a0
Duyao MP Buckler AJ Sonenshein GE Interaction of an NF-κB-like factor with a site upstream of the c-myc promoter Proc Natl Acad Sci USA 1990 87 4727 4731 2191300
Mosser DD Theodorakis NG Morimoto RI Coordinate changes in heat shock element-binding activity and HSP70 gene transcription rates in human cell Mol Cell Biol 1988 8 4736 4744 3211126
Kibler KV Shors T Perkins KB Zeman CC Banaszak MP Biesterfeldt J Langland JO Jacobs BL Double-stranded RNA is a trigger for apoptosis in vaccinia virus-infected cells J Virol 1997 71 1992 2003 9032331
Moss B Conaway RC, Conaway JW Vaccinia virus transcription Transcription: Mechanisms and Regulation 1994 New York: Raven Press 185 205
Kline MP Morimoto RI Repression of the heat shock factor 1 transcriptional activation domain is modulated by constitutitve phosphorylation Mol Cell Biol 1997 17 2107 2115 9121459
Welch WJ How cells respond to stress Sci Am 1993 268 56 64 8097593
Ripley B Stephanou D Isenberg D Latchman D Interleukin-10 activates heat-shock protein 90β gene expression Immunol 1999 97 226 231 10.1046/j.1365-2567.1999.00773.x
Hung J-J Chung C-S Chang W Molecular chaperone Hsp90 is important for vaccinia virus growth in cells J Virol 2002 76 1379 1390 11773412 10.1128/JVI.76.6.2676-2682.2002
Jacquier-Sarlin MR Jornot L Polla BS Differential expression and regulation of hsp70 and hsp90 by phorbol esters and heat shock J Biol Chem 1995 270 14094 14099 7775471 10.1074/jbc.270.23.14094
Victor M Benecke BJ Expression levels of heat shock factors are not functionally coupled to the rate of expression of heat shock genes Mol Biol Rep 1998 25 135 141 9700049 10.1023/A:1006801205904
Jolly C Usson Y Morimoto RI Rapid and reversible relocalization of heat shock factor 1 within seconds to nuclear stress granules Proc Natl Acad Sci USA 1999 96 6769 6774 10359787 10.1073/pnas.96.12.6769
Cotto J Fox S Morimoto R HSF1 granules: a novel stress-induced nuclear compartment of human cells J Cell Sci 1997 110 2925 2934 9359875
Mercier PA Winegarden NA Westwood JT Heat shock factor 1 is predominantly a nuclear protein before and after heat stress J Cell Sci 1999 112 2765 2774 10413683
Broyless SS Liu X Zhu M Kremer M Transcription factor YY1 is a vaccinia virus late promoter activator J Biol Chem 1999 274 35662 35667 10585445 10.1074/jbc.274.50.35662
Slezak K Michalik M Kowalczyk A Rokita H YY1 is recruited to the cytoplasm of vaccinia virus-infected human macrophages by the Crm1 system Virus Res 2004 102 177 184 15084399 10.1016/j.virusres.2004.01.028
Nunes SL Calderwood SK Heat shock factor-1 and the heat shock cognate 70 protein associate in high molecular weight complexes in the cytoplasm of NIH-3T3 cells Biochem Biophys Res Commun 1995 213 1 6 7639722 10.1006/bbrc.1995.2090
Guzik K Bzowska M Dobrucki J Pryjma J Heat-shocked monocytes are resistant to Staphylococcus aureus-induced apoptotic DNA fragmentation due to expression of HSP72 Infect Immun 1999 67 4216 4222 10417194
Bantel H Sinha B Domschke W Peters G Schulze-Osthoff K Janicke RU alpha-Toxin is a mediator of Staphylococcus aureus-induced cell death and activates caspases via the intrinsic death pathway independently of death receptor signaling J Cell Biol 2001 155 637 648 11696559 10.1083/jcb.200105081
Brasier AR Spratt H Wu Z Boldogh I Zhang Y Garofalo RP Casola A Pashmi J Haag A Luxon B Kurosky A Nuclear heat shock response and novel nuclear domain 10 reorganization in respiratory syncytial virus-infected a549 cells identified by high-resolution two-dimensional gel electrophoresis J Virol 2004 78 11461 11476 15479789 10.1128/JVI.78.21.11461-11476.2004
Tanonaka K Toga W Takahashi M Kawana K Miyamoto Y Yoshida H Takeo S Hsp70 attenuates hypoxia/reoxygenation-induced activation of poly(ADP-ribose)synthetase in the nucleus of adult rat cardiomyocytes Mol Cell Biochem 2003 248 149 155 12870667 10.1023/A:1024192403758
Wasilenko ST Stewart TL Meyers AF Barry M Vaccinia virus encodes a previously uncharacterized mitochondrial-associated inhibitor of apoptosis Proc Natl Acad Sci USA 2003 100 14345 14350 14610284 10.1073/pnas.2235583100
Pirog KA Kowalczyk AK Rokita HB Changes in Bcl-2 expression in vaccinia virus-infected human peripheral blood monocytes Viral Immunol 2005 18 225 232 10.1089/vim.2005.18.224
Humlova Z Vokurka M Esteban M Melkowa Z Vaccinia virus induces apoptosis of infected macrophages J Gen Virol 2002 83 2821 2832 12388819
Srivastava P Roles of heat-shock proteins in innate and adaptive immunity Nat Rev Immunol 2002 2 185 94 11913069 10.1038/nri749
|
16246258
|
PMC1283150
|
CC BY
|
2021-01-04 16:36:23
|
no
|
J Inflamm (Lond). 2005 Oct 24; 2:12
|
utf-8
|
J Inflamm (Lond)
| 2,005 |
10.1186/1476-9255-2-12
|
oa_comm
|
==== Front
Nutr JNutrition Journal1475-2891BioMed Central London 1475-2891-4-291625313810.1186/1475-2891-4-29ResearchDietary inulin intake and age can significantly affect intestinal absorption of calcium and magnesium in rats: a stable isotope approach Coudray Charles [email protected] Mathieu [email protected] Christine [email protected] Jean Claude [email protected] Christian [email protected] Elyett [email protected] Andrzej [email protected] Yves [email protected] Centre de Recherche en Nutrition Humaine d'Auvergne, Unité Maladies Métaboliques et Micro-nutriments, INRA, Theix, 63122 St-Genès-Champanelle, France2005 27 10 2005 4 29 29 31 8 2005 27 10 2005 Copyright © 2005 Coudray et al; licensee BioMed Central Ltd.2005Coudray et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
previous studies have shown that non-digestible inulin-type fructan intake can increase intestinal mineral absorption in both humans and animals. However, this stimulatory effect on intestinal absorption may depend on experimental conditions such as duration of fermentable fiber intake, mineral diet levels and animals' physiological status, in particular their age.
Objectives
the aim of this study was to determine the effect of inulin intake on Ca and Mg absorption in rats at different age stages.
Methods
eighty male Wistar rats of four different ages (2, 5, 10 and 20 months) were randomized into either a control group or a group receiving 3.75% inulin in their diet for 4 days and then 7.5% inulin for three weeks. The animals were fed fresh food and water ad libitum for the duration of the experiment. Intestinal absorption of Ca and Mg was determined by fecal monitoring using stable isotopic tracers. Ca and Mg status was also assessed.
Results
absorption of Ca and Mg was significantly lower in the aged rats (10 and 20 mo) than in the young and adult rat groups. As expected, inulin intake increased Ca and Mg absorption in all four rat groups. However, inulin had a numerically greater effect on Ca absorption in aged rats than in younger rats whereas its effect on Mg absorption remained similar across all four rat age groups.
Conclusion
the extent of the stimulatory effect of inulin on absorption of Ca may differ according to animal ages. Further studies are required to explore this effect over longer inulin intake periods, and to confirm these results in humans.
Inulinintestinal absorptionstatuscalciummagnesiumfermentationstable isotopeagerat
==== Body
Introduction
When non-digestible inulin-type fructans reach the large intestine, they are fermented by the local microflora and stimulate the growth of bifidobacteria and lactobacilli, which may have health-promoting functions [1-3]. Several studies have demonstrated that rats fed with prebiotic fructans absorbed more Ca and Mg than control rats, despite an increase in total fecal mass [4-6]. Indeed, the products of fructan fermentation can influence the intestinal absorption of Ca and Mg in many ways. Short-chain fatty acids (SCFA) are fermentation products that are responsible for lowering the pH of cecal content, which in turn increases mineral solubility, leading to improved mineral absorption [7]. SCFA can directly influence mineral absorption by forming complexes with the minerals, thereby increase their uptake by the intestinal cells [8,9]. It is thought that the bacterial metabolites (e.g. butyrate) can stimulate the intestinal epithelium and increase its absorptive capacity [10]. These various factors are closely linked to the nature of the prebiotic carbohydrates and to experimental conditions [7,11,12]. Inulin has been shown to have generally high and consistent effects on intestinal Mg absorption in both animals and humans [13], but the effects of inulin on calcium (Ca) absorption seem to be dependent on experimental conditions (dose of inulin, dietary Ca content, experiment duration, animal age and mineral requirements). In this study, we investigated the relationship between animal age and the stimulatory effect of inulin on intestinal absorption and retention of Ca and Mg using a stable isotope approach following short-term administration of inulin in rats aged from 2 to 20 months. This is the first time that the effect of inulin is studied in rats using a stable isotope approach
Materials and methods
Materials and reagents
The enriched Ca isotope (44Ca) as CaCO3 and the enriched Mg isotope (25Mg) as MgO were obtained from Chemgas, (Boulogne, France). The atomic abundances of these enriched isotopes were as follows: 40Ca = 3.41%, 42Ca = 0.09%, 43Ca = 0.03%, 44Ca = 96.45% 46Ca =< 0.01% 48Ca = 0.02% and 24Mg = 1.6%, 25Mg = 97.8%, 26Mg = 0.6%. HNO3 (ultrapure), Mg and beryllium standard solutions (1 g/L) were obtained from Merck (Darmstadt, Germany). All other chemicals were of the highest quality available. Distilled water was used throughout. A Perkin-Elmer 6100DRC system (Perkin-Elmer Instruments, Courteboeuf, France) equipped with a Meinhard nebulizer was used for isotopic measurement, and a Perkin Elmer AA800 (Perkin Elmer Instruments, Courteboeuf, France) was used for total Mg measurement.
Animals and diets
Eighty male Wistar rats aged 2, 5, 10 or 20 months were purchased from Janvier (Le Genest Saint Ile, France). They were fed a commercial pellet diet (Ssniff R/S-breeding – until 3 mo, then Ssniff R/S maintenance from 3 to 24 mo age). Two groups were formed for each age bracket to receive either a control diet or a semi-purified diet containing inulin until the end of the experiment. The composition of these two diets is given in Table 1. Tested inulin was purchased from Orafti, Tienen, Belgium (Raftaline®). The target Ca and Mg levels in these diets were 5000 mg Ca/kg and 500 mg Mg/Kg diet. Powder diet (100 g) was made up with 100 ml of distilled water to form a kind of semi-liquid food prepared on-site each day. Chemical analysis of the diets offered confirmed the expected Ca and Mg contents in the experimental diets: 5107 mg Ca/kg and 5050 mg Ca/kg, and 495 mg Mg/kg and 514 mg Mg/kg in the control and inulin diets, respectively. Chemical analysis showed that the inulin contained approximately 40 mg Ca/kg and less than 1 mg Mg/kg. Dietary inulin level was maintained at 3.75% during the first 4 days and then 7.5% from day 5 until the end of the experiment. The 8 rat groups were given fresh food and water daily, made available ad libitum. Food consumption and body weight were recorded weekly. Throughout the experiment, the rats were housed two per cage (wire-bottomed to limit coprophagy) in a temperature-controlled room (22°C) with dark period from 08:00 pm to 08:00 am. Total experiment duration was 30 days. All procedures complied with the Institute's ethical guidelines on the care and use of laboratory animals.
Table 1 Diet composition (g/kg) during the experiment
Control diet Inulin diets
3.75% 7.5%
Wheat starch 650 612.5 575
Casein 200 200 200
Corn oil 50 50 50
Cellulose 50 50 50
Mineral mix (AIN 1993)a 35 35 35
Vitamin mix (AIN 1993)b 10 10 10
DL-Methionine 3 3 3
Choline bi-tartrate 2 2 2
Inulin 0 37.5 75
a: Mineral mix AIN 1993 ensures the following mineral levels in the diets (mg/kg): Na, 1020; K, 3600; P, 4000; Ca, 5000; Mg, 500; Zn, 30; Fe, 35; Cu, 6; Mn, 54; Se, 0.1; I, 0.2; Cr, 2.
b: Vitamin mix AIN 1993 ensures the following mineral levels in the diets (mg/kg): thiamine, 6; riboflavine, 6; pyridoxine, 7; nicotinic acid, 30; calcium pantothenate, 16; folic acid, 2; D-biotin: 0.2; and (μg/kg) cyanocobalamine (vitamin B12), 10; vitamin K, 50; and (IU/kg) vitamin A, 4000; vitamin E, 50; vitamin D, 1000.
Preparation of stable isotope solution
215 mg of the 44Ca (in carbonate form = 508 mg) and 255 mg of the 25Mg (in oxide form = 412 mg) were first individually moistened with 2 ml of distilled water. One ml of HCl 12 N (ultrapure) was added to the 44Ca suspension and two ml of HCl 12 N was added to the 25Mg suspension to transform the carbonate and the oxide into soluble chlorides of Ca and Mg, respectively. Each solution was then diluted with 50 ml of distilled water, both solutions were then mixed, and pH was adjusted to between 3 and 6 with 1 N sodium hydroxide solution. The resulting study solution was then completed to 150 ml with distilled water and maintained for several days at +4°C until utilization. Total and isotopic Ca and Mg contents were checked before use.
The rats were transferred to metabolic cages and housed individually three days before the beginning of the isotopic balance study to allow them to adapt to their new environment. Animals received by gavage about 1.7 ml of isotopic solution. The urine and faeces of each rat were quantitatively collected for four consecutive days, and excreted isotopes in these two media and in the gavage solution were quantitatively determined by ICP/MS, as described below.
Sampling procedures
The rats were sacrificed just after the dark period (between 08:00 am and 10:00 am), i.e. at a time when cecal fermentation was still very active. After anesthesia (40 mg sodium pentobarbital/kg body weight), blood was withdrawn from the abdominal aorta, placed into tubes containing sodium heparin and centrifuged at 1,000 g for 10 minutes. Plasma samples were stored at 4°C for mineral analysis. The cecum, complete with contents, was removed and weighed (total cecal weight). The cecal wall was flushed clean with ice-cold saline, blotted on filter paper, and weighed (cecal wall weight). For each rat, duplicate samples of cecal contents were collected into 2 ml microfuge tubes and immediately frozen at -20°C until analysis. The pH of cecal content was determined on site using a Sentron pH-system 1001 portable pH-meter (Sentron Europe B.V. Ac Roden, The Netherlands). Supernatants of the digestive contents were obtained by centrifuging one of the two microfuge tubes at 20,000 g for 10 minutes at 4°C, and then frozen until analysis. One tibia was also sampled for Ca and Mg analysis.
Analytical procedures
Ca and Mg concentrations were determined in the plasma and urine after adequate dilution into 0.1% (w/v) lanthanum chloride. Diet aliquots, fecal materials and tibia were dry-ashed (10 hours at 500°C) and dissolved with concentrated HNO3 and H2O2 on a heating plate until complete decoloration. The resulting mineral solutions were set at 10 ml with water and adequately diluted in 0.1% lanthanum chloride. Mineral concentrations were measured by atomic absorption spectrophotometry (on a Perkin-Elmer AA800) at wavelengths of 422 nm for Ca and 285 nm for Mg.
For isotopic 44Ca and 25Mg determination, samples were appropriately diluted before analysis using 1% HNO3. Ca and Mg concentration and isotope ratios were determined by ICP-MS using Ca and Mg as external standard and beryllium as internal standard. The instrument operating conditions were set as follows after optimization with a solution of 1 μg indium/l: RF Power = 1050 W, Nebulizer Ar flow rate = 0.79 L/min, Auxiliary Ar flow rate = 1.2 L/min, Outer Ar flow rate = 15 L/min. Data acquisition parameters were set as follows: Sweeps/reading = 50, Readings/replicate = 1, Number of replicates = 3, Dwell time = 50 ms for 24Mg, 75 ms for 9Be, 25Mg, 26Mg, and 44Ca, 150 ms for 42Ca and 300 ms for 43Ca, Scanning mode = peak hopping. DRC operating conditions (for 42Ca, 43Ca and 44Ca) were as follows: Cell Gas A Flow Rate = 0.5 L ammonia/min, RPa = 0, and RPq = 0.45.
Cecal SCFA concentrations, including acetic, propionic and butyric acid, were determined by gas-liquid chromatography on portions of supernatant fractions of cecal contents as previously described [14].
Calculations
Ca and Mg each have different stable isotopes with the following natural abundances: 40Ca = 96.941%, 42Ca = 0.647%, 43Ca = 0.135%, 44Ca = 2.086% 46Ca = 0.004% 48Ca = 0.187% and 24Mg = 78.99%, 25Mg = 10.00% and 26Mg = 11.01% [15]. 44Ca and 25Mg isotopic enrichments were obtained, respectively, from the following equations: (44Ca/43Ca measured ratio - 44Ca/43Ca baseline ratio)/(44Ca/43Ca baseline ratio) and (25Mg/26Mg measured ratio - 25Mg/26Mg baseline ratio)/(25Mg/26Mg baseline ratio).
Non-absorbed 44Ca and 25Mg isotopes in the fecal or urine samples (coming only from the 44Ca or 25Mg isotope labels) were calculated as follows:
for 44Ca (mg) = (total fecal or urine Ca (mg) × (natural abundance 44Ca × enriched 44Ca)/(1 + (natural abundance 44Ca × enriched 44Ca);
for 25Mg (mg) = (total fecal or urine Mg (mg) × (natural abundance 25Mg × enriched 25Mg)/(1 + (natural abundance 25Mg × enriched 25Mg)).
Calculations were also made directly from ICP-MS data. The two modes of calculation give the same results when the ICP-MS quantitative procedure is used [16].
Intestinal absorption of 44Ca and 25Mg was then calculated as administered 44Ca or 25Mg - 44Ca or 25Mg excreted in the feces, and retention of 44Ca and 25Mg was calculated as administered 44Ca or 25Mg - 44Ca or 25Mg excreted in the feces and in the urine.
Total cecal SCFA content (μmol/cecum) was calculated as the supernatant SCFA concentration (μmol/ml) × cecal water (ml/cecum).
Soluble Ca and Mg levels in the cecal contents were determined on the supernatant concentration (μg/ml), and soluble Ca and Mg contents per cecum were calculated as (μg Ca/ml or μg Mg/ml) × cecal water (ml).
Data analysis
Values are given as means ± SD, and data were tested by 2-way ANOVA using the General Linear Models procedure of the Super ANOVA package (Abacus, Berkeley, CA). Post-hoc comparisons were performed using Fisher's least significant difference procedures. Differences of p < 0.05 were considered statistically significant. Simple linear correlation analysis was used to assess the relationships between intestinal absorption of Ca and Mg and other relevant parameters. Values of p < 0.05 were considered statistically significant.
Results
Food intake and growth rate
Inulin intake at the dose of 7.5% showed only a tendency to decrease animal food intake in this study. The slight decrease in food intake in inulin-fed rats led to a non-significantly lower growth rate (p < 0.10) towards the end of the experiment in inulin-fed rats compared to controls. The lower calorific value of the inulin diets (-4%) compared to the control diets may also be responsible for this reduced weight gain. In addition, food intake decreased significantly with increasing age, as expected (data not shown).
Cecal fermentation parameters and total and cecal soluble Ca and Mg levels (table 2)
Table 2 Effect of age and inulin intake and their interaction on cecum fermentation parameters and cecal Ca and Mg levels in rats
Cont 3 Mo Cont 6 Mo Cont 11 Mo Cont 21 Mo Inulin 3 Mo Inulin 6 Mo Inulin 11 Mo Inulin 21 Mo inulin age interaction
Cecal content pH 6.92 ± 0.24 6.87 ± 0.17 6.72 ± 0.58 6.62 ± 0.31 5.71 ± 0.58 5.41 ± 0.22 5.64 ± 0.37 5.57 ± 0.22 <0.0001 NS NS
Cecal content, g 2.20 ± 0.35 2.34 ± 0.68 2.53 ± 0.97 2.86 ± 0.74 6.18 ± 1.68 6.46 ± 1.57 7.09 ± 2.31 7.10 ± 1.91 <0.0001 NS NS
Cecal wall, g 0.87 ± 0.07 1.11 ± 0.22 1.32 ± 0.29 1.25 ± 0.15 1.80 ± 0.37 2.32 ± 0.29 2.51 ± 0.39 2.46 ± 0.30 <0.0001 NS NS
Acetate, μmol/cecum 22.4 ± 6.3 24.5 ± 9.4 24.4 ± 11.8 28.0 ± 8.5 49.2 ± 22.3 70.4 ± 16.7 54.3 ± 17.3 63.5 ± 11.4 <0.0001 0.033 NS
Propionate, μmol/cecum 5.49 ± 1.48 6.31 ± 2.20 5.11 ± 2.10 6.11 ± 1.72 15.56 ± 12.29 12.84 ± 5.04 9.96 ± 4.86 11.26 ± 3.92 <0.0001 NS NS
Butyrate, μmol/cecum 9.90 ± 2.70 8.63 ± 3.10 6.37 ± 2.78 8.01 ± 3.38 48.13 ± 25.12 69.51 ± 32.50 39.98 ± 18.42 43.15 ± 8.63 <0.0001 0.018 0.037
Total SCFA, μmol/cecum 37.8 ± 9.1 39.5 ± 13.5 35.9 ± 15.9 42.2 ± 12.6 112.9 ± 50.8 152.8 ± 40.0 104.2 ± 27.7 117.9 ± 17.3 <0.0001 0.024 0.049
Values are mean ± SD, n = 10 animals.
The rats were received inulin (7.5%) for 4 weeks before cecal parameters assessment.
As expected, inulin intake significantly increased cecal wall weight and cecal content and significantly decreased the pH of cecal content. These variables did not change with rat age. In addition, inulin intake considerably increased the individual and total pools of SCFA in the cecal contents (p < 0.0001). The effect of age on these SCFA pools was less clear. No significant age-related difference was observed amongst the control group rats, whereas in the inulin-fed group, the intestinal bacteria produced higher acetate, butyrate and total SCFA in the rats aged 10 mo than in the three other groups (p < 0.05).
Intestinal absorption and retention of calcium (table 3)
Table 3 Effect of age and inulin intake and their interaction on intestinal absorption and retention of Ca in rats
Cont 3 Mo Cont 6 Mo Cont 11 Mo Cont 21 Mo Inulin 3 Mo Inulin 6 Mo Inulin 11 Mo Inulin 21 Mo inulin age interaction
Administered 44Ca, μg 1637 ± 46 1610 ± 14 1602 ± 17 1605 ± 24 1593 ± 19 1614 ± 19 1621 ± 25 1626 ± 22 NS NS <0.0005
Fecal 44Ca enrichment, % 12.5 ± 3.3 17.3 ± 2.4 20.2 ± 4.3 18.7 ± 2.8 10.9 ± 4.2 17.8 ± 6.2 17.4 ± 2.1 21.2 ± 3.1 NS <0.0001 NS
Fecal 44Ca level, μg/g 112 ± 30 182 ± 35 218 ± 45 204 ± 38 76 ± 30 152 ± 49 163 ± 26 188 ± 26 <0.0001 <0.0001 NS
Fecal 44Ca excretion, μg 856 ± 224 1139 ± 153 1389 ± 96 1366 ± 115 541 ± 223 926 ± 142 1207 ± 195 1192 ± 142 <0.0001 <0.0001 NS
Intestinal 44Ca absorption, μg 781 ± 206 471 ± 153 213 ± 90 239 ± 117 1052 ± 222 689 ± 142 413 ± 202 434 ± 143 <0.0001 <0.0001 NS
Intestinal 44Ca absorption, % 47.8 ± 12.9 29.3 ± 9.4 13.3 ± 5.6 14.9 ± 7.3 66.1 ± 13.9 42.7 ± 8.8 25.4 ± 12.4 26.7 ± 8.7 <0.0001 <0.0001 NS
Urinary 44Ca enrichment, % 17.4 ± 6.5 20.6 ± 6.7 14.8 ± 3.1 16.9 ± 3.9 13.7 ± 3.3 17.5 ± 6.7 18.6 ± 4.2 18.5 ± 4.1 NS NS 0.0569
Urinary 44Ca excretion, μg 15.3 ± 5.9 14.8 ± 6.0 25.8 ± 10.7 28.1 ± 7.9 23.5 ± 7.2 21.8 ± 9.9 49.0 ± 11.8 40.3 ± 12.5 <0.0001 <0.0001 0.036
44Ca retention, μg 765 ± 204 456 ± 153 188 ± 92 212 ± 113 1029 ± 219 667 ± 135 364 ± 203 394 ± 145 <0.0001 <0.0001 NS
44Ca retention, % 46.9 ± 12.8 28.3 ± 9.5 11.7 ± 5.7 13.2 ± 7.1 64.6 ± 13.8 41.3 ± 8.3 22.4 ± 12.4 24.2 ± 8.9 <0.0001 <0.0001 NS
Values are mean ± SD, n = 10 animals.
The rats were given 44Ca after 14 days of inulin intake (7.5%), and fecal non-absorbed 44Ca isotope was determined in a 4d feces and urine pools.
The amount of gavaged 44Ca was about 1.60 mg/rat, which led to a fecal 44Ca enrichment of 10% to 20% in the 4-day feces pool. Fecal 44Ca excretion expressed as mg/g of feces or as mg/day increased significantly with age. Consequently, net (mg) and relative (%) 44Ca absorption were significantly lower in the aged rats than in the young adult or adult rats. In addition, urinary 44Ca excretion (mg) increased significantly with age. Consequently, net (mg) and relative (%) 44Ca retention were considerably lower in the aged rats than in the young adult or adult rats. Inulin intake significantly decreased fecal 44Ca excretion, expressed as μg/g of feces or as μg, in all groups. Consequently, inulin intake significantly increased net (mg) and relative (%) 44Ca absorption. Moreover, inulin intake increased urinary 44Ca excretion (mg). Lastly, inulin intake significantly increased net (mg) and relative (%) 44Ca retention in the four age-related groups compared to the control diet groups.
Intestinal absorption and retention of magnesium (table 4)
Table 4 Effect of age and inulin intake and their interaction on intestinal absorption and retention of Mg in rats
Cont 3 Mo Cont 6 Mo Cont 11 Mo Cont 21 Mo Inulin 3 Mo Inulin 6 Mo Inulin 11 Mo Inulin 21 Mo inulin age interaction
Administered 25Mg, μg 2553 ± 71 2511 ± 22 2499 ± 26 2504 ± 38 2485 ± 30 2518 ± 30 2527 ± 39 2536 ± 35 NS NS <0.0005
Fecal 25Mg enrichment, % 47.9 ± 7.1 51.9 ± 7.4 54.6 ± 12.3 54.1 ± 8.1 33.9 ± 20.3 41.0 ± 19.9 47.4 ± 13.9 65.7 ± 17.6 NS 0.0007 0.027
Fecal 25Mg level, μg/g 149 ± 26 184 ± 34 205 ± 44 204 ± 38 51 ± 33 70 ± 35 91 ± 31 120 ± 35 <0.0001 <0.0001 NS
Fecal 25Mg excretion, μg 1138 ± 201 1157 ± 229 1311 ± 200 1366 ± 177 358 ± 224 430 ± 206 673 ± 205 757 ± 184 <0.0001 <0.0001 NS
Intestinal 25Mg absorption, μg 1415 ± 187 1354 ± 225 1188 ± 199 1137 ± 175 2127 ± 224 2087 ± 208 1855 ± 232 1780 ± 186 <0.0001 <0.0001 NS
Intestinal 25Mg absorption % 55.5 ± 7.5 54.0 ± 9.0 47.5 ± 8.0 45.4 ± 7.0 85.6 ± 8.9 82.9 ± 8.2 73.3 ± 8.4 70.2 ± 7.2 <0.0001 <0.0001 NS
Urinary 25Mg enrichment, % 29.3 ± 2.7 29.1 ± 3.5 28.0 ± 3.7 28.2 ± 2.32 34.2 ± 3.8 35.5 ± 2.4 33.5 ± 3.7 36.7 ± 6.5 <0.0001 NS NS
Urinary 25Mg excretion, μg 398 ± 64 323 ± 36 298 ± 80 292 ± 88 699 ± 91 792 ± 167 633 ± 146 551 ± 164 <0.0001 0.003 NS
25Mg retention, μg 1017 ± 189 1031 ± 225 890 ± 179 845 ± 142 1428 ± 216 1295 ± 192 1221 ± 242 1228 ± 165 <0.0001 0.011 NS
25Mg retention, % 39.8 ± 7.3 41.1 ± 9.0 35.6 ± 7.1 33.7 ± 5.7 57.8 ± 8.9 51.5 ± 7.8 48.3 ± 9.2 48.4 ± 6.5 <0.0001 0.008 NS
Values are mean ± SD, n = 10 animals.
The rats were given 25Mg after 14 days of inulin intake (7.5%), and fecal non-absorbed 25Mg isotope was determined in a 4d feces and urine pools.
The amount of gavaged 25Mg was about 2.50 mg/rat, which led to a fecal 25Mg enrichment of 35% to 65% in the 4-day feces pool. Fecal 25Mg excretion expressed as mg/g of feces or as mg increased significantly with age. Consequently, net (mg) and relative (%) 25Mg absorption were significantly lower in the aged rats than in the young adult or adult rats. In addition, urinary 25Mg excretion (mg) increased significantly with age. Consequently, net (mg) and relative (%) 25Mg retention were significantly lower in the aged rats than in the young adult or adult rats. As expected, inulin intake significantly decreased fecal 25Mg excretion, expressed as μg/g of feces or as μg, in all groups. Consequently, inulin intake significantly increased net (mg) and relative (%) 25Mg absorption. Similarly, inulin intake increased urinary 25Mg excretion (mg). However, inulin intake led to significantly higher net (mg) and relative (%) 25Mg retention in all four groups compared to the control diet.
Calcium and magnesium status (table 5)
Table 5 Effect of age and inulin intake and their interaction on status biomarkers of Ca and Mg in rats
Cont 3 M Cont 6 M Cont 11 M Cont 21 M Inulin 3 M Inulin 6 M Inulin 11 M Inulin 21 M inulin age interaction
Plasma Ca, mg/L 98 ± 4 98 ± 5 95 ± 6 100 ± 5 102 ± 5 99 ± 4 98 ± 3 100 ± 4 0.0601 0.0619 NS
Tibia weight, mg dw 480 ± 42 630 ± 80 717 ± 89 630 ± 93 489 ± 66 617 ± 66 841 ± 92 648 ± 47 0.042 <0.0001 0.023
Bone Ca, mg/g dw 207 ± 21 214 ± 15 216 ± 15 215 ± 21 205 ± 13 215 ± 18 202 ± 14 228 ± 7 NS 0.018 0.0639
Plasma Mg, mg/L 17.9 ± 1.1 17.7 ± 1.1 17.2 ± 1.0 16.9 ± 1.3 17.6 ± 1.1 17.9 ± 1.3 18.2 ± 1.5 18.2 ± 1.7 0.0570 NS NS
Erythrocyte Mg, mg/L 45.4 ± 3.8 44.2 ± 4.7 42.5 ± 3.4 43.4 ± 3.0 44.9 ± 4.9 43.8 ± 2.5 44.3 ± 3.6 43.8 ± 3.9 NS NS NS
Bone Mg, mg/g dw 3.92 ± 0.10 3.79 ± 0.08 3.72 ± 0.08 3.76 ± 0.08 3.91 ± 0.10 3.72 ± 0.07 3.73 ± 0.09 3.72 ± 0.09 NS <0.0001 NS
Values are mean ± SD, n = 10 animals.
Mean plasma Ca varied from 95 to 102 mg/L, showing a tendency to increase with inulin intake (+2%, p = 0.0601) and to decrease with increasing age (-1%, p = 0.0619). Mean bone Ca varied from 202 to 228 mg/g dry weight, and was unaffected by inulin intake. However, mean bone Ca increased significantly with increasing age. Mean plasma Mg varied from 16.9 to 18.2 mg/L, showing a tendency to increase with inulin intake (+3%, p = 0.0570). However, mean plasma Mg was not modified by age. Plasma Mg increased in the inulin-fed aged rats (+6.7%), whereas there was no increase in the young and adult rats (-0.3%). Mean red blood cell Mg levels varied from 42.5 to 45.4 mg/L and remained unchanged when age increases or under inulin intake. Mean bone Mg levels varied from 3.72 to 3.92 mg/g dry weight, decreasing significantly with aging (p < 0.0001). However, mean bone Mg was unaffected by inulin intake.
Discussion
Previous studies have repeatedly shown that intake of different inulin-type fructans can variably increase mineral intestinal absorption in humans and animals [4,5,17-19]. Indeed, inulin-type fructans strongly and consistently increase intestinal Mg absorption [12], whereas their effect on Ca absorption seems to be dependent on experimental conditions such as inulin type, dietary Ca levels, duration of fructan intake [20-22] and the animals' physiological state, particularly age. It is well known that the absorption mechanisms of Ca and Mg differ considerably [23,24], which may explain the observed differences between these two minerals in terms of inulin effect. In this study, we investigated the enhancing effect of fructan intake on Ca and Mg intestinal absorption and balance in rats of different ages.
1 – Effect of animal age and inulin intake on Ca absorption
Our results clearly showed that aged rats exhibited less efficient intestinal absorption and retention of Ca. 44Ca absorption ranged from 48% without inulin to 66% under inulin intake in the young and adult rats and from 15% without inulin to 27% under inulin intake in the old and very old rats. This decline in Ca absorption with age is not new, and has already been reported in animal and human studies [25-27] and is largely confirmed in this study. This decline is primarily due to an energy- and vitamin D-dependent Ca transport component in the elderly [28]. Our results clearly showed that inulin intake increased the efficiency of Ca intestinal absorption and retention. The mean 44Ca absorption in the four rat control groups was 26.3% compared to 40.2% in the four inulin-fed groups, with an overall increase in 44Ca absorption of 53%. These results are in agreement with literature data showing that the effect of inulin on Ca absorption seems to be optimal in the early weeks, then decreasing gradually with experiment duration [20,29,30]. One possible explanation for this phenomenon is a down-regulation of the active pathway of intestinal Ca absorption after several weeks of feeding inulin, as previously reported [31,32].
2 – Effect of animal age and inulin intake on Mg absorption
Our results showed that aged rats exhibited less efficient intestinal absorption and retention of Mg. 25Mg absorption ranged from 56% without inulin to 86% under inulin intake in the young and adult rats and from 45% without inulin to 70% under inulin intake in the old and very old rats. This decline in Mg absorption with age is not well documented in the literature in either animal or human studies. Few, if any, incomplete studies have reported an age effect on Mg absorption [33-35], and the results are inconsistent. Hence, to our knowledge, this is the first robust report to clearly show that Mg absorption decreases with age in the rat. Although Mg absorption is generally described as a passive phenomenon, one component of this absorption remains under hormonal control [36,37], which may explain the observed results. Our results clearly showed that inulin intake considerably increased Mg intestinal absorption and retention efficiency. Mean 25Mg absorption in the four rat control groups was 50.6%, compared to 78.0% in the four corresponding inulin-fed rat groups, with an overall increase in 25Mg absorption of 54%. These results are in agreement with literature data showing that inulin intake considerably increases Mg absorption in animals and humans (see recent review [13]).
3 – Modulation of the stimulatory inulin effect on Ca and Mg absorption with rat age
Since Ca absorption is generally well controlled, the observed absorption increase under inulin intake may be down-regulated (known as a feed-back phenomenon) in adult rats. Thus, given that Ca absorption is low and the adaptative phenomenon less well controlled in aged rats, we hypothesized that inulin intake would lead to a much greater increase in Ca absorption in aged rats than in the young or adult rats. Conversely, since Mg absorption is only weakly controlled with a generally consistent increase under inulin intake in adult rats, we hypothesized that inulin intake would increase Mg absorption in both aged rats and young or adult rats to the same extent.
The relative increase in 44Ca absorption under inulin intake was 41.5% and 84.5% in the younger (3 and 6 mo old) and older rats (11 and 21 mo old), respectively (figure 1). Although these increase percents are numerically more important in the older rats than in the younger rats, there was no statistically significant interaction between age and inuline. It is highly possible that the number of animals used in this experiment was not enough to reach significant level. Furthermore, the relative increase in 25Mg absorption under inulin intake was 53.5% and 54.5% in the younger and older rats, respectively (figure 1). This indicates that the stimulatory effect of inulin on 25Mg absorption was not age-dependent. It is possible that inulin intake may lead to a higher increase in 44Ca absorption in the older rats than in the younger rats, whereas inulin intake leads to a similar increase in 25Mg absorption in young, adult and aged rats, thus confirming the hypothesis we formulated for this study.
In conclusion, as expected, our results confirmed that short-term inulin intake stimulates the absorption of both Ca and Mg. Furthermore, not only these results confirmed that Ca absorption declines considerably with age but also showed for the first time that Mg absorption also declines with age in the rat. Moreover, these results confirmed our hypothesis of a greater stimulatory effect of inulin on Ca absorption in aged rats than in the young or adult rats, and a similar stimulatory effect of inulin on Mg absorption in aged rats and young and adult rats. Further studies are required to explore this effect on longer inulin intake periods and to validate these results on the stimulatory effect of inulin on Ca and Mg absorption in the elderly.
Abbreviations
Ca: calcium; Mg: Magnesium; ICP/MS: Inductively coupled plasma/mass spectrometry, OS: oligosaccharides; SCFA: Short-chain fatty acids;
Figure 1 Stimulatory effect of dietary inulin intake on intestinal absorption of 44Ca and 25Mg in rats of different ages. Intestinal 44Ca absorption (A) and ntestinal 25Mg absorption (B) in the control groups was normalized to 100% for each age group. The stimulatory effect of inulin (%) for a given age group was calculated as follows: 100* (intestinal absorption in the inulin-fed age group/intestinal absorption in the same age group without inulin). The rats were given 44Ca and 25Mg after 14 days of inulin intake (7.5%), and fecal non-absorbed isotopes were determined in a 4d feces pool.
Acknowledgements
The authors are grateful to ORAFTI (Tienen, Belgium) for providing the inulin for this study. The authors thank Séverine Thien, Lydia Jaffrelo, Claudine Lab and Pierre Lamby for their technical assistance.
==== Refs
Gibson GR Beatty ER Wang X Cummings JH Selective stimulation of bifidobacteria in the human colon by oligofructose and inulin Gastroenterology 1995 108 975 982 7698613 10.1016/0016-5085(95)90192-2
Jackson KG Taylor GR Clohessy AM Williams CM The effect of the daily intake of inulin on fasting lipid, insulin and glucose concentrations in middle-aged men and women Br J Nutr 1999 82 23 30 10655953
Kaur N Gupta AK Applications of inulin and oligofructose in health and nutrition J Biosci 2002 27 703 714 12571376
Delzenne N Aertssens J Verplaetse H Roccaro M Roberfroid M Effect of fermentable fructo-oligosaccharides on mineral, nitrogen and energy digestive balance in the rat Life Sci 1995 57 1579 1587 7564905 10.1016/0024-3205(95)02133-4
Ohta A Ohtsuki M Baba S Adachi T Sakata T Sakaguchi E Calcium and magnesium absorption from the colon and rectum are increased in rats fed fructooligosaccharides J Nutr 1995 125 2417 2424 7666261
Younes H Demigne C Remesy C Acidic fermentation in the caecum increases absorption of calcium and magnesium in the large intestine of the rat Br J Nutr 1996 75 301 314 8785206 10.1079/BJN19960132
Levrat MA Remesy C Demigne C High propionic acid fermentations and mineral accumulation in the cecum of rats adapted to different levels of inulin J Nutr 1991 121 1730 1737 1941180
Lutz T Scharrer E Effect of short-chain fatty acids on calcium absorption by the rat colon Exp Physiol 1991 76 615 618 1910768
Trinidad TP Wolever TM Thompson LU Effects of calcium concentration, acetate, and propionate on calcium absorption in the human distal colon Nutrition 1999 15 529 533 10422081 10.1016/S0899-9007(99)00110-0
Topping DL Clifton PM Short-chain fatty acids and human colonic function: roles of resistant starch and nonstarch polysaccharides Physiol Rev 2001 81 1031 1064 11427691
Remesy C Levrat MA Gamet L Demigne C Cecal fermentations in rats fed oligosaccharides (inulin) are modulated by dietary calcium level Am J Physiol 1993 264 G855 62 8388642
Coudray C Tressol JC Gueux E Rayssiguier Y Effects of inulin-type fructans of different chain length and type of branching on intestinal absorption and balance of calcium and magnesium in rats Eur J Nutr 2003 42 91 98 12638030 10.1007/s00394-003-0390-x
Coudray C Demigne C Rayssiguier Y Effects of dietary fibers on magnesium absorption in animals and humans J Nutr 2003 133 1 4 12514257
Demigne C Remesy C Rayssiguier Y Effect of fermentable carbohydrates on volatile fatty acids, ammonia and mineral absorption in the rat caecum Reprod Nutr Dev 1980 20 1351 1359 7349487
De Bievre P Taylor PDP Table of isotopic composition of the elements International Journal of Mass Spectrometry Ion Process 123 1993 123 149 166 10.1016/0168-1176(93)87009-H
Coudray C Pepin D Tressol JC Bellanger J Rayssiguier Y Study of magnesium bioavailability using stable isotopes and the inductively-coupled plasma mass spectrometry technique in the rat: single and double labelling approaches Br J Nutr 1997 77 957 970 9227192 10.1079/BJN19970093
Coudray C Bellanger J Castiglia-Delavaud C Remesy C Vermorel M Rayssignuier Y Effect of soluble or partly soluble dietary fibres supplementation on absorption and balance of calcium, magnesium, iron and zinc in healthy young men Eur J Clin Nutr 1997 51 375 380 9192195 10.1038/sj.ejcn.1600417
Lopez HW Coudray C Levrat-Verny MA Feillet-Coudray C Demigne C Remesy C Fructooligosaccharides enhance mineral apparent absorption and counteract the deleterious effects of phytic acid on mineral homeostasis in rats J Nutr Biochem 2000 11 500 508 11120448 10.1016/S0955-2863(00)00109-1
Younes H Coudray C Bellanger J Demigne C Rayssiguier Y Remesy C Effects of two fermentable carbohydrates (inulin and resistant starch) and their combination on calcium and magnesium balance in rats Br J Nutr 2001 86 479 485 11591235
Ohta A Ohtsuki M Baba S Hirayama M Adachi T Comparison of the nutritional effects of fructo-oligosaccharides of different sugar chain length in rats Nutr Res 1998 18 109 120 10.1016/S0271-5317(97)00205-4
Scholz-Ahrens KE Schaafsma G van den Heuvel EG Schrezenmeir J Effects of prebiotics on mineral metabolism Am J Clin Nutr 2001 73 459S 464S 11157358
Cashman K Prebiotics and calcium bioavailability Curr Issues Intest Microbiol 2003 4 21 32 12691259
Kayne LH Lee DB Intestinal magnesium absorption Miner Electrolyte Metab 1993 19 210 217 8264506
Schaafsma G Bioavailability of calcium and magnesium Eur J Clin Nutr 1997 51 Suppl 1 S13 6 9023473
Armbrecht HJ Effect of age on calcium and phosphate absorption. Role of 1,25-dihydroxyvitamin D Miner Electrolyte Metab 1990 16 159 166 2250622
Chan EL Lau E Shek CC MacDonald D Woo J Leung PC Swaminathan R Age-related changes in bone density, serum parathyroid hormone, calcium absorption and other indices of bone metabolism in Chinese women Clin Endocrinol (Oxf) 1992 36 375 381 1424171
Kinyamu HK Gallagher JC Prahl JM DeLuca HF Petranick KM Lanspa SJ Association between intestinal vitamin D receptor, calcium absorption, and serum 1,25 dihydroxyvitamin D in normal young and elderly women J Bone Miner Res 1997 12 922 928 9169351
Weaver CM Age related calcium requirements due to changes in absorption and utilization J Nutr 1994 124 1418S 1425S 8064395
Chonan O Watanuki M The effect of 6'-galactooligosaccharides on bone mineralization of rats adapted to different levels of dietary calcium Int J Vitam Nutr Res 1996 66 244 249 8899459
Coudray C Feillet-Coudray C Tressol JC Gueux E Thien S Jaffrelo L Mazur A Rayssiguier Y Stimulatory effect of inulin on intestinal absorption of calcium and magnesium in rats is modulated by dietary calcium intakesShort- and long-term balance studies Eur J Nutr 2004
Ohta A Motohashi Y Ohtsuki M Hirayama M Adachi T Sakuma K Dietary fructooligosaccharides change the concentration of calbindin-D9k differently in the mucosa of the small and large intestine of rats J Nutr 1998 128 934 939 9614150
Takasaki M Inaba H Ohta A Motohashi Y Sakai K Morris H Sakuma K Dietary short-chain fructooligosaccharides increase calbindin-D9k levels only in the large intestine in rats independent of dietary calcium deficiency or serum 1,25 dihydroxy vitamin D levels Int J Vitam Nutr Res 2000 70 206 213 11068700
Durlach J Bac P Durlach V Rayssiguier Y Bara M Guiet-Bara A Magnesium status and ageing: an update Magnes Res 1998 11 25 42 9595547
Coudray C Gaumet N Bellanger J Coxam V Barlet JP Rayssiguier Y Influence of age and hormonal treatment on intestinal absorption of magnesium in ovariectomised rats Magnes Res 1999 12 109 114 10423705
Vaquero MP Magnesium and trace elements in the elderly: intake, status and recommendations J Nutr Health Aging 2002 6 147 153 12166371
Ferment O Touitou Y Magnesium: metabolism and hormonal regulation in different species Comp Biochem Physiol A 1985 82 753 758 14570081 10.1016/0300-9629(85)90478-5
Hardwick LL Jones MR Brautbar N Lee DB Site and mechanism of intestinal magnesium absorption Miner Electrolyte Metab 1990 16 174 180 2250624
|
16253138
|
PMC1283151
|
CC BY
|
2021-01-04 16:39:31
|
no
|
Nutr J. 2005 Oct 27; 4:29
|
utf-8
|
Nutr J
| 2,005 |
10.1186/1475-2891-4-29
|
oa_comm
|
==== Front
Nutr JNutrition Journal1475-2891BioMed Central London 1475-2891-4-291625313810.1186/1475-2891-4-29ResearchDietary inulin intake and age can significantly affect intestinal absorption of calcium and magnesium in rats: a stable isotope approach Coudray Charles [email protected] Mathieu [email protected] Christine [email protected] Jean Claude [email protected] Christian [email protected] Elyett [email protected] Andrzej [email protected] Yves [email protected] Centre de Recherche en Nutrition Humaine d'Auvergne, Unité Maladies Métaboliques et Micro-nutriments, INRA, Theix, 63122 St-Genès-Champanelle, France2005 27 10 2005 4 29 29 31 8 2005 27 10 2005 Copyright © 2005 Coudray et al; licensee BioMed Central Ltd.2005Coudray et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
previous studies have shown that non-digestible inulin-type fructan intake can increase intestinal mineral absorption in both humans and animals. However, this stimulatory effect on intestinal absorption may depend on experimental conditions such as duration of fermentable fiber intake, mineral diet levels and animals' physiological status, in particular their age.
Objectives
the aim of this study was to determine the effect of inulin intake on Ca and Mg absorption in rats at different age stages.
Methods
eighty male Wistar rats of four different ages (2, 5, 10 and 20 months) were randomized into either a control group or a group receiving 3.75% inulin in their diet for 4 days and then 7.5% inulin for three weeks. The animals were fed fresh food and water ad libitum for the duration of the experiment. Intestinal absorption of Ca and Mg was determined by fecal monitoring using stable isotopic tracers. Ca and Mg status was also assessed.
Results
absorption of Ca and Mg was significantly lower in the aged rats (10 and 20 mo) than in the young and adult rat groups. As expected, inulin intake increased Ca and Mg absorption in all four rat groups. However, inulin had a numerically greater effect on Ca absorption in aged rats than in younger rats whereas its effect on Mg absorption remained similar across all four rat age groups.
Conclusion
the extent of the stimulatory effect of inulin on absorption of Ca may differ according to animal ages. Further studies are required to explore this effect over longer inulin intake periods, and to confirm these results in humans.
Inulinintestinal absorptionstatuscalciummagnesiumfermentationstable isotopeagerat
==== Body
Introduction
When non-digestible inulin-type fructans reach the large intestine, they are fermented by the local microflora and stimulate the growth of bifidobacteria and lactobacilli, which may have health-promoting functions [1-3]. Several studies have demonstrated that rats fed with prebiotic fructans absorbed more Ca and Mg than control rats, despite an increase in total fecal mass [4-6]. Indeed, the products of fructan fermentation can influence the intestinal absorption of Ca and Mg in many ways. Short-chain fatty acids (SCFA) are fermentation products that are responsible for lowering the pH of cecal content, which in turn increases mineral solubility, leading to improved mineral absorption [7]. SCFA can directly influence mineral absorption by forming complexes with the minerals, thereby increase their uptake by the intestinal cells [8,9]. It is thought that the bacterial metabolites (e.g. butyrate) can stimulate the intestinal epithelium and increase its absorptive capacity [10]. These various factors are closely linked to the nature of the prebiotic carbohydrates and to experimental conditions [7,11,12]. Inulin has been shown to have generally high and consistent effects on intestinal Mg absorption in both animals and humans [13], but the effects of inulin on calcium (Ca) absorption seem to be dependent on experimental conditions (dose of inulin, dietary Ca content, experiment duration, animal age and mineral requirements). In this study, we investigated the relationship between animal age and the stimulatory effect of inulin on intestinal absorption and retention of Ca and Mg using a stable isotope approach following short-term administration of inulin in rats aged from 2 to 20 months. This is the first time that the effect of inulin is studied in rats using a stable isotope approach
Materials and methods
Materials and reagents
The enriched Ca isotope (44Ca) as CaCO3 and the enriched Mg isotope (25Mg) as MgO were obtained from Chemgas, (Boulogne, France). The atomic abundances of these enriched isotopes were as follows: 40Ca = 3.41%, 42Ca = 0.09%, 43Ca = 0.03%, 44Ca = 96.45% 46Ca =< 0.01% 48Ca = 0.02% and 24Mg = 1.6%, 25Mg = 97.8%, 26Mg = 0.6%. HNO3 (ultrapure), Mg and beryllium standard solutions (1 g/L) were obtained from Merck (Darmstadt, Germany). All other chemicals were of the highest quality available. Distilled water was used throughout. A Perkin-Elmer 6100DRC system (Perkin-Elmer Instruments, Courteboeuf, France) equipped with a Meinhard nebulizer was used for isotopic measurement, and a Perkin Elmer AA800 (Perkin Elmer Instruments, Courteboeuf, France) was used for total Mg measurement.
Animals and diets
Eighty male Wistar rats aged 2, 5, 10 or 20 months were purchased from Janvier (Le Genest Saint Ile, France). They were fed a commercial pellet diet (Ssniff R/S-breeding – until 3 mo, then Ssniff R/S maintenance from 3 to 24 mo age). Two groups were formed for each age bracket to receive either a control diet or a semi-purified diet containing inulin until the end of the experiment. The composition of these two diets is given in Table 1. Tested inulin was purchased from Orafti, Tienen, Belgium (Raftaline®). The target Ca and Mg levels in these diets were 5000 mg Ca/kg and 500 mg Mg/Kg diet. Powder diet (100 g) was made up with 100 ml of distilled water to form a kind of semi-liquid food prepared on-site each day. Chemical analysis of the diets offered confirmed the expected Ca and Mg contents in the experimental diets: 5107 mg Ca/kg and 5050 mg Ca/kg, and 495 mg Mg/kg and 514 mg Mg/kg in the control and inulin diets, respectively. Chemical analysis showed that the inulin contained approximately 40 mg Ca/kg and less than 1 mg Mg/kg. Dietary inulin level was maintained at 3.75% during the first 4 days and then 7.5% from day 5 until the end of the experiment. The 8 rat groups were given fresh food and water daily, made available ad libitum. Food consumption and body weight were recorded weekly. Throughout the experiment, the rats were housed two per cage (wire-bottomed to limit coprophagy) in a temperature-controlled room (22°C) with dark period from 08:00 pm to 08:00 am. Total experiment duration was 30 days. All procedures complied with the Institute's ethical guidelines on the care and use of laboratory animals.
Table 1 Diet composition (g/kg) during the experiment
Control diet Inulin diets
3.75% 7.5%
Wheat starch 650 612.5 575
Casein 200 200 200
Corn oil 50 50 50
Cellulose 50 50 50
Mineral mix (AIN 1993)a 35 35 35
Vitamin mix (AIN 1993)b 10 10 10
DL-Methionine 3 3 3
Choline bi-tartrate 2 2 2
Inulin 0 37.5 75
a: Mineral mix AIN 1993 ensures the following mineral levels in the diets (mg/kg): Na, 1020; K, 3600; P, 4000; Ca, 5000; Mg, 500; Zn, 30; Fe, 35; Cu, 6; Mn, 54; Se, 0.1; I, 0.2; Cr, 2.
b: Vitamin mix AIN 1993 ensures the following mineral levels in the diets (mg/kg): thiamine, 6; riboflavine, 6; pyridoxine, 7; nicotinic acid, 30; calcium pantothenate, 16; folic acid, 2; D-biotin: 0.2; and (μg/kg) cyanocobalamine (vitamin B12), 10; vitamin K, 50; and (IU/kg) vitamin A, 4000; vitamin E, 50; vitamin D, 1000.
Preparation of stable isotope solution
215 mg of the 44Ca (in carbonate form = 508 mg) and 255 mg of the 25Mg (in oxide form = 412 mg) were first individually moistened with 2 ml of distilled water. One ml of HCl 12 N (ultrapure) was added to the 44Ca suspension and two ml of HCl 12 N was added to the 25Mg suspension to transform the carbonate and the oxide into soluble chlorides of Ca and Mg, respectively. Each solution was then diluted with 50 ml of distilled water, both solutions were then mixed, and pH was adjusted to between 3 and 6 with 1 N sodium hydroxide solution. The resulting study solution was then completed to 150 ml with distilled water and maintained for several days at +4°C until utilization. Total and isotopic Ca and Mg contents were checked before use.
The rats were transferred to metabolic cages and housed individually three days before the beginning of the isotopic balance study to allow them to adapt to their new environment. Animals received by gavage about 1.7 ml of isotopic solution. The urine and faeces of each rat were quantitatively collected for four consecutive days, and excreted isotopes in these two media and in the gavage solution were quantitatively determined by ICP/MS, as described below.
Sampling procedures
The rats were sacrificed just after the dark period (between 08:00 am and 10:00 am), i.e. at a time when cecal fermentation was still very active. After anesthesia (40 mg sodium pentobarbital/kg body weight), blood was withdrawn from the abdominal aorta, placed into tubes containing sodium heparin and centrifuged at 1,000 g for 10 minutes. Plasma samples were stored at 4°C for mineral analysis. The cecum, complete with contents, was removed and weighed (total cecal weight). The cecal wall was flushed clean with ice-cold saline, blotted on filter paper, and weighed (cecal wall weight). For each rat, duplicate samples of cecal contents were collected into 2 ml microfuge tubes and immediately frozen at -20°C until analysis. The pH of cecal content was determined on site using a Sentron pH-system 1001 portable pH-meter (Sentron Europe B.V. Ac Roden, The Netherlands). Supernatants of the digestive contents were obtained by centrifuging one of the two microfuge tubes at 20,000 g for 10 minutes at 4°C, and then frozen until analysis. One tibia was also sampled for Ca and Mg analysis.
Analytical procedures
Ca and Mg concentrations were determined in the plasma and urine after adequate dilution into 0.1% (w/v) lanthanum chloride. Diet aliquots, fecal materials and tibia were dry-ashed (10 hours at 500°C) and dissolved with concentrated HNO3 and H2O2 on a heating plate until complete decoloration. The resulting mineral solutions were set at 10 ml with water and adequately diluted in 0.1% lanthanum chloride. Mineral concentrations were measured by atomic absorption spectrophotometry (on a Perkin-Elmer AA800) at wavelengths of 422 nm for Ca and 285 nm for Mg.
For isotopic 44Ca and 25Mg determination, samples were appropriately diluted before analysis using 1% HNO3. Ca and Mg concentration and isotope ratios were determined by ICP-MS using Ca and Mg as external standard and beryllium as internal standard. The instrument operating conditions were set as follows after optimization with a solution of 1 μg indium/l: RF Power = 1050 W, Nebulizer Ar flow rate = 0.79 L/min, Auxiliary Ar flow rate = 1.2 L/min, Outer Ar flow rate = 15 L/min. Data acquisition parameters were set as follows: Sweeps/reading = 50, Readings/replicate = 1, Number of replicates = 3, Dwell time = 50 ms for 24Mg, 75 ms for 9Be, 25Mg, 26Mg, and 44Ca, 150 ms for 42Ca and 300 ms for 43Ca, Scanning mode = peak hopping. DRC operating conditions (for 42Ca, 43Ca and 44Ca) were as follows: Cell Gas A Flow Rate = 0.5 L ammonia/min, RPa = 0, and RPq = 0.45.
Cecal SCFA concentrations, including acetic, propionic and butyric acid, were determined by gas-liquid chromatography on portions of supernatant fractions of cecal contents as previously described [14].
Calculations
Ca and Mg each have different stable isotopes with the following natural abundances: 40Ca = 96.941%, 42Ca = 0.647%, 43Ca = 0.135%, 44Ca = 2.086% 46Ca = 0.004% 48Ca = 0.187% and 24Mg = 78.99%, 25Mg = 10.00% and 26Mg = 11.01% [15]. 44Ca and 25Mg isotopic enrichments were obtained, respectively, from the following equations: (44Ca/43Ca measured ratio - 44Ca/43Ca baseline ratio)/(44Ca/43Ca baseline ratio) and (25Mg/26Mg measured ratio - 25Mg/26Mg baseline ratio)/(25Mg/26Mg baseline ratio).
Non-absorbed 44Ca and 25Mg isotopes in the fecal or urine samples (coming only from the 44Ca or 25Mg isotope labels) were calculated as follows:
for 44Ca (mg) = (total fecal or urine Ca (mg) × (natural abundance 44Ca × enriched 44Ca)/(1 + (natural abundance 44Ca × enriched 44Ca);
for 25Mg (mg) = (total fecal or urine Mg (mg) × (natural abundance 25Mg × enriched 25Mg)/(1 + (natural abundance 25Mg × enriched 25Mg)).
Calculations were also made directly from ICP-MS data. The two modes of calculation give the same results when the ICP-MS quantitative procedure is used [16].
Intestinal absorption of 44Ca and 25Mg was then calculated as administered 44Ca or 25Mg - 44Ca or 25Mg excreted in the feces, and retention of 44Ca and 25Mg was calculated as administered 44Ca or 25Mg - 44Ca or 25Mg excreted in the feces and in the urine.
Total cecal SCFA content (μmol/cecum) was calculated as the supernatant SCFA concentration (μmol/ml) × cecal water (ml/cecum).
Soluble Ca and Mg levels in the cecal contents were determined on the supernatant concentration (μg/ml), and soluble Ca and Mg contents per cecum were calculated as (μg Ca/ml or μg Mg/ml) × cecal water (ml).
Data analysis
Values are given as means ± SD, and data were tested by 2-way ANOVA using the General Linear Models procedure of the Super ANOVA package (Abacus, Berkeley, CA). Post-hoc comparisons were performed using Fisher's least significant difference procedures. Differences of p < 0.05 were considered statistically significant. Simple linear correlation analysis was used to assess the relationships between intestinal absorption of Ca and Mg and other relevant parameters. Values of p < 0.05 were considered statistically significant.
Results
Food intake and growth rate
Inulin intake at the dose of 7.5% showed only a tendency to decrease animal food intake in this study. The slight decrease in food intake in inulin-fed rats led to a non-significantly lower growth rate (p < 0.10) towards the end of the experiment in inulin-fed rats compared to controls. The lower calorific value of the inulin diets (-4%) compared to the control diets may also be responsible for this reduced weight gain. In addition, food intake decreased significantly with increasing age, as expected (data not shown).
Cecal fermentation parameters and total and cecal soluble Ca and Mg levels (table 2)
Table 2 Effect of age and inulin intake and their interaction on cecum fermentation parameters and cecal Ca and Mg levels in rats
Cont 3 Mo Cont 6 Mo Cont 11 Mo Cont 21 Mo Inulin 3 Mo Inulin 6 Mo Inulin 11 Mo Inulin 21 Mo inulin age interaction
Cecal content pH 6.92 ± 0.24 6.87 ± 0.17 6.72 ± 0.58 6.62 ± 0.31 5.71 ± 0.58 5.41 ± 0.22 5.64 ± 0.37 5.57 ± 0.22 <0.0001 NS NS
Cecal content, g 2.20 ± 0.35 2.34 ± 0.68 2.53 ± 0.97 2.86 ± 0.74 6.18 ± 1.68 6.46 ± 1.57 7.09 ± 2.31 7.10 ± 1.91 <0.0001 NS NS
Cecal wall, g 0.87 ± 0.07 1.11 ± 0.22 1.32 ± 0.29 1.25 ± 0.15 1.80 ± 0.37 2.32 ± 0.29 2.51 ± 0.39 2.46 ± 0.30 <0.0001 NS NS
Acetate, μmol/cecum 22.4 ± 6.3 24.5 ± 9.4 24.4 ± 11.8 28.0 ± 8.5 49.2 ± 22.3 70.4 ± 16.7 54.3 ± 17.3 63.5 ± 11.4 <0.0001 0.033 NS
Propionate, μmol/cecum 5.49 ± 1.48 6.31 ± 2.20 5.11 ± 2.10 6.11 ± 1.72 15.56 ± 12.29 12.84 ± 5.04 9.96 ± 4.86 11.26 ± 3.92 <0.0001 NS NS
Butyrate, μmol/cecum 9.90 ± 2.70 8.63 ± 3.10 6.37 ± 2.78 8.01 ± 3.38 48.13 ± 25.12 69.51 ± 32.50 39.98 ± 18.42 43.15 ± 8.63 <0.0001 0.018 0.037
Total SCFA, μmol/cecum 37.8 ± 9.1 39.5 ± 13.5 35.9 ± 15.9 42.2 ± 12.6 112.9 ± 50.8 152.8 ± 40.0 104.2 ± 27.7 117.9 ± 17.3 <0.0001 0.024 0.049
Values are mean ± SD, n = 10 animals.
The rats were received inulin (7.5%) for 4 weeks before cecal parameters assessment.
As expected, inulin intake significantly increased cecal wall weight and cecal content and significantly decreased the pH of cecal content. These variables did not change with rat age. In addition, inulin intake considerably increased the individual and total pools of SCFA in the cecal contents (p < 0.0001). The effect of age on these SCFA pools was less clear. No significant age-related difference was observed amongst the control group rats, whereas in the inulin-fed group, the intestinal bacteria produced higher acetate, butyrate and total SCFA in the rats aged 10 mo than in the three other groups (p < 0.05).
Intestinal absorption and retention of calcium (table 3)
Table 3 Effect of age and inulin intake and their interaction on intestinal absorption and retention of Ca in rats
Cont 3 Mo Cont 6 Mo Cont 11 Mo Cont 21 Mo Inulin 3 Mo Inulin 6 Mo Inulin 11 Mo Inulin 21 Mo inulin age interaction
Administered 44Ca, μg 1637 ± 46 1610 ± 14 1602 ± 17 1605 ± 24 1593 ± 19 1614 ± 19 1621 ± 25 1626 ± 22 NS NS <0.0005
Fecal 44Ca enrichment, % 12.5 ± 3.3 17.3 ± 2.4 20.2 ± 4.3 18.7 ± 2.8 10.9 ± 4.2 17.8 ± 6.2 17.4 ± 2.1 21.2 ± 3.1 NS <0.0001 NS
Fecal 44Ca level, μg/g 112 ± 30 182 ± 35 218 ± 45 204 ± 38 76 ± 30 152 ± 49 163 ± 26 188 ± 26 <0.0001 <0.0001 NS
Fecal 44Ca excretion, μg 856 ± 224 1139 ± 153 1389 ± 96 1366 ± 115 541 ± 223 926 ± 142 1207 ± 195 1192 ± 142 <0.0001 <0.0001 NS
Intestinal 44Ca absorption, μg 781 ± 206 471 ± 153 213 ± 90 239 ± 117 1052 ± 222 689 ± 142 413 ± 202 434 ± 143 <0.0001 <0.0001 NS
Intestinal 44Ca absorption, % 47.8 ± 12.9 29.3 ± 9.4 13.3 ± 5.6 14.9 ± 7.3 66.1 ± 13.9 42.7 ± 8.8 25.4 ± 12.4 26.7 ± 8.7 <0.0001 <0.0001 NS
Urinary 44Ca enrichment, % 17.4 ± 6.5 20.6 ± 6.7 14.8 ± 3.1 16.9 ± 3.9 13.7 ± 3.3 17.5 ± 6.7 18.6 ± 4.2 18.5 ± 4.1 NS NS 0.0569
Urinary 44Ca excretion, μg 15.3 ± 5.9 14.8 ± 6.0 25.8 ± 10.7 28.1 ± 7.9 23.5 ± 7.2 21.8 ± 9.9 49.0 ± 11.8 40.3 ± 12.5 <0.0001 <0.0001 0.036
44Ca retention, μg 765 ± 204 456 ± 153 188 ± 92 212 ± 113 1029 ± 219 667 ± 135 364 ± 203 394 ± 145 <0.0001 <0.0001 NS
44Ca retention, % 46.9 ± 12.8 28.3 ± 9.5 11.7 ± 5.7 13.2 ± 7.1 64.6 ± 13.8 41.3 ± 8.3 22.4 ± 12.4 24.2 ± 8.9 <0.0001 <0.0001 NS
Values are mean ± SD, n = 10 animals.
The rats were given 44Ca after 14 days of inulin intake (7.5%), and fecal non-absorbed 44Ca isotope was determined in a 4d feces and urine pools.
The amount of gavaged 44Ca was about 1.60 mg/rat, which led to a fecal 44Ca enrichment of 10% to 20% in the 4-day feces pool. Fecal 44Ca excretion expressed as mg/g of feces or as mg/day increased significantly with age. Consequently, net (mg) and relative (%) 44Ca absorption were significantly lower in the aged rats than in the young adult or adult rats. In addition, urinary 44Ca excretion (mg) increased significantly with age. Consequently, net (mg) and relative (%) 44Ca retention were considerably lower in the aged rats than in the young adult or adult rats. Inulin intake significantly decreased fecal 44Ca excretion, expressed as μg/g of feces or as μg, in all groups. Consequently, inulin intake significantly increased net (mg) and relative (%) 44Ca absorption. Moreover, inulin intake increased urinary 44Ca excretion (mg). Lastly, inulin intake significantly increased net (mg) and relative (%) 44Ca retention in the four age-related groups compared to the control diet groups.
Intestinal absorption and retention of magnesium (table 4)
Table 4 Effect of age and inulin intake and their interaction on intestinal absorption and retention of Mg in rats
Cont 3 Mo Cont 6 Mo Cont 11 Mo Cont 21 Mo Inulin 3 Mo Inulin 6 Mo Inulin 11 Mo Inulin 21 Mo inulin age interaction
Administered 25Mg, μg 2553 ± 71 2511 ± 22 2499 ± 26 2504 ± 38 2485 ± 30 2518 ± 30 2527 ± 39 2536 ± 35 NS NS <0.0005
Fecal 25Mg enrichment, % 47.9 ± 7.1 51.9 ± 7.4 54.6 ± 12.3 54.1 ± 8.1 33.9 ± 20.3 41.0 ± 19.9 47.4 ± 13.9 65.7 ± 17.6 NS 0.0007 0.027
Fecal 25Mg level, μg/g 149 ± 26 184 ± 34 205 ± 44 204 ± 38 51 ± 33 70 ± 35 91 ± 31 120 ± 35 <0.0001 <0.0001 NS
Fecal 25Mg excretion, μg 1138 ± 201 1157 ± 229 1311 ± 200 1366 ± 177 358 ± 224 430 ± 206 673 ± 205 757 ± 184 <0.0001 <0.0001 NS
Intestinal 25Mg absorption, μg 1415 ± 187 1354 ± 225 1188 ± 199 1137 ± 175 2127 ± 224 2087 ± 208 1855 ± 232 1780 ± 186 <0.0001 <0.0001 NS
Intestinal 25Mg absorption % 55.5 ± 7.5 54.0 ± 9.0 47.5 ± 8.0 45.4 ± 7.0 85.6 ± 8.9 82.9 ± 8.2 73.3 ± 8.4 70.2 ± 7.2 <0.0001 <0.0001 NS
Urinary 25Mg enrichment, % 29.3 ± 2.7 29.1 ± 3.5 28.0 ± 3.7 28.2 ± 2.32 34.2 ± 3.8 35.5 ± 2.4 33.5 ± 3.7 36.7 ± 6.5 <0.0001 NS NS
Urinary 25Mg excretion, μg 398 ± 64 323 ± 36 298 ± 80 292 ± 88 699 ± 91 792 ± 167 633 ± 146 551 ± 164 <0.0001 0.003 NS
25Mg retention, μg 1017 ± 189 1031 ± 225 890 ± 179 845 ± 142 1428 ± 216 1295 ± 192 1221 ± 242 1228 ± 165 <0.0001 0.011 NS
25Mg retention, % 39.8 ± 7.3 41.1 ± 9.0 35.6 ± 7.1 33.7 ± 5.7 57.8 ± 8.9 51.5 ± 7.8 48.3 ± 9.2 48.4 ± 6.5 <0.0001 0.008 NS
Values are mean ± SD, n = 10 animals.
The rats were given 25Mg after 14 days of inulin intake (7.5%), and fecal non-absorbed 25Mg isotope was determined in a 4d feces and urine pools.
The amount of gavaged 25Mg was about 2.50 mg/rat, which led to a fecal 25Mg enrichment of 35% to 65% in the 4-day feces pool. Fecal 25Mg excretion expressed as mg/g of feces or as mg increased significantly with age. Consequently, net (mg) and relative (%) 25Mg absorption were significantly lower in the aged rats than in the young adult or adult rats. In addition, urinary 25Mg excretion (mg) increased significantly with age. Consequently, net (mg) and relative (%) 25Mg retention were significantly lower in the aged rats than in the young adult or adult rats. As expected, inulin intake significantly decreased fecal 25Mg excretion, expressed as μg/g of feces or as μg, in all groups. Consequently, inulin intake significantly increased net (mg) and relative (%) 25Mg absorption. Similarly, inulin intake increased urinary 25Mg excretion (mg). However, inulin intake led to significantly higher net (mg) and relative (%) 25Mg retention in all four groups compared to the control diet.
Calcium and magnesium status (table 5)
Table 5 Effect of age and inulin intake and their interaction on status biomarkers of Ca and Mg in rats
Cont 3 M Cont 6 M Cont 11 M Cont 21 M Inulin 3 M Inulin 6 M Inulin 11 M Inulin 21 M inulin age interaction
Plasma Ca, mg/L 98 ± 4 98 ± 5 95 ± 6 100 ± 5 102 ± 5 99 ± 4 98 ± 3 100 ± 4 0.0601 0.0619 NS
Tibia weight, mg dw 480 ± 42 630 ± 80 717 ± 89 630 ± 93 489 ± 66 617 ± 66 841 ± 92 648 ± 47 0.042 <0.0001 0.023
Bone Ca, mg/g dw 207 ± 21 214 ± 15 216 ± 15 215 ± 21 205 ± 13 215 ± 18 202 ± 14 228 ± 7 NS 0.018 0.0639
Plasma Mg, mg/L 17.9 ± 1.1 17.7 ± 1.1 17.2 ± 1.0 16.9 ± 1.3 17.6 ± 1.1 17.9 ± 1.3 18.2 ± 1.5 18.2 ± 1.7 0.0570 NS NS
Erythrocyte Mg, mg/L 45.4 ± 3.8 44.2 ± 4.7 42.5 ± 3.4 43.4 ± 3.0 44.9 ± 4.9 43.8 ± 2.5 44.3 ± 3.6 43.8 ± 3.9 NS NS NS
Bone Mg, mg/g dw 3.92 ± 0.10 3.79 ± 0.08 3.72 ± 0.08 3.76 ± 0.08 3.91 ± 0.10 3.72 ± 0.07 3.73 ± 0.09 3.72 ± 0.09 NS <0.0001 NS
Values are mean ± SD, n = 10 animals.
Mean plasma Ca varied from 95 to 102 mg/L, showing a tendency to increase with inulin intake (+2%, p = 0.0601) and to decrease with increasing age (-1%, p = 0.0619). Mean bone Ca varied from 202 to 228 mg/g dry weight, and was unaffected by inulin intake. However, mean bone Ca increased significantly with increasing age. Mean plasma Mg varied from 16.9 to 18.2 mg/L, showing a tendency to increase with inulin intake (+3%, p = 0.0570). However, mean plasma Mg was not modified by age. Plasma Mg increased in the inulin-fed aged rats (+6.7%), whereas there was no increase in the young and adult rats (-0.3%). Mean red blood cell Mg levels varied from 42.5 to 45.4 mg/L and remained unchanged when age increases or under inulin intake. Mean bone Mg levels varied from 3.72 to 3.92 mg/g dry weight, decreasing significantly with aging (p < 0.0001). However, mean bone Mg was unaffected by inulin intake.
Discussion
Previous studies have repeatedly shown that intake of different inulin-type fructans can variably increase mineral intestinal absorption in humans and animals [4,5,17-19]. Indeed, inulin-type fructans strongly and consistently increase intestinal Mg absorption [12], whereas their effect on Ca absorption seems to be dependent on experimental conditions such as inulin type, dietary Ca levels, duration of fructan intake [20-22] and the animals' physiological state, particularly age. It is well known that the absorption mechanisms of Ca and Mg differ considerably [23,24], which may explain the observed differences between these two minerals in terms of inulin effect. In this study, we investigated the enhancing effect of fructan intake on Ca and Mg intestinal absorption and balance in rats of different ages.
1 – Effect of animal age and inulin intake on Ca absorption
Our results clearly showed that aged rats exhibited less efficient intestinal absorption and retention of Ca. 44Ca absorption ranged from 48% without inulin to 66% under inulin intake in the young and adult rats and from 15% without inulin to 27% under inulin intake in the old and very old rats. This decline in Ca absorption with age is not new, and has already been reported in animal and human studies [25-27] and is largely confirmed in this study. This decline is primarily due to an energy- and vitamin D-dependent Ca transport component in the elderly [28]. Our results clearly showed that inulin intake increased the efficiency of Ca intestinal absorption and retention. The mean 44Ca absorption in the four rat control groups was 26.3% compared to 40.2% in the four inulin-fed groups, with an overall increase in 44Ca absorption of 53%. These results are in agreement with literature data showing that the effect of inulin on Ca absorption seems to be optimal in the early weeks, then decreasing gradually with experiment duration [20,29,30]. One possible explanation for this phenomenon is a down-regulation of the active pathway of intestinal Ca absorption after several weeks of feeding inulin, as previously reported [31,32].
2 – Effect of animal age and inulin intake on Mg absorption
Our results showed that aged rats exhibited less efficient intestinal absorption and retention of Mg. 25Mg absorption ranged from 56% without inulin to 86% under inulin intake in the young and adult rats and from 45% without inulin to 70% under inulin intake in the old and very old rats. This decline in Mg absorption with age is not well documented in the literature in either animal or human studies. Few, if any, incomplete studies have reported an age effect on Mg absorption [33-35], and the results are inconsistent. Hence, to our knowledge, this is the first robust report to clearly show that Mg absorption decreases with age in the rat. Although Mg absorption is generally described as a passive phenomenon, one component of this absorption remains under hormonal control [36,37], which may explain the observed results. Our results clearly showed that inulin intake considerably increased Mg intestinal absorption and retention efficiency. Mean 25Mg absorption in the four rat control groups was 50.6%, compared to 78.0% in the four corresponding inulin-fed rat groups, with an overall increase in 25Mg absorption of 54%. These results are in agreement with literature data showing that inulin intake considerably increases Mg absorption in animals and humans (see recent review [13]).
3 – Modulation of the stimulatory inulin effect on Ca and Mg absorption with rat age
Since Ca absorption is generally well controlled, the observed absorption increase under inulin intake may be down-regulated (known as a feed-back phenomenon) in adult rats. Thus, given that Ca absorption is low and the adaptative phenomenon less well controlled in aged rats, we hypothesized that inulin intake would lead to a much greater increase in Ca absorption in aged rats than in the young or adult rats. Conversely, since Mg absorption is only weakly controlled with a generally consistent increase under inulin intake in adult rats, we hypothesized that inulin intake would increase Mg absorption in both aged rats and young or adult rats to the same extent.
The relative increase in 44Ca absorption under inulin intake was 41.5% and 84.5% in the younger (3 and 6 mo old) and older rats (11 and 21 mo old), respectively (figure 1). Although these increase percents are numerically more important in the older rats than in the younger rats, there was no statistically significant interaction between age and inuline. It is highly possible that the number of animals used in this experiment was not enough to reach significant level. Furthermore, the relative increase in 25Mg absorption under inulin intake was 53.5% and 54.5% in the younger and older rats, respectively (figure 1). This indicates that the stimulatory effect of inulin on 25Mg absorption was not age-dependent. It is possible that inulin intake may lead to a higher increase in 44Ca absorption in the older rats than in the younger rats, whereas inulin intake leads to a similar increase in 25Mg absorption in young, adult and aged rats, thus confirming the hypothesis we formulated for this study.
In conclusion, as expected, our results confirmed that short-term inulin intake stimulates the absorption of both Ca and Mg. Furthermore, not only these results confirmed that Ca absorption declines considerably with age but also showed for the first time that Mg absorption also declines with age in the rat. Moreover, these results confirmed our hypothesis of a greater stimulatory effect of inulin on Ca absorption in aged rats than in the young or adult rats, and a similar stimulatory effect of inulin on Mg absorption in aged rats and young and adult rats. Further studies are required to explore this effect on longer inulin intake periods and to validate these results on the stimulatory effect of inulin on Ca and Mg absorption in the elderly.
Abbreviations
Ca: calcium; Mg: Magnesium; ICP/MS: Inductively coupled plasma/mass spectrometry, OS: oligosaccharides; SCFA: Short-chain fatty acids;
Figure 1 Stimulatory effect of dietary inulin intake on intestinal absorption of 44Ca and 25Mg in rats of different ages. Intestinal 44Ca absorption (A) and ntestinal 25Mg absorption (B) in the control groups was normalized to 100% for each age group. The stimulatory effect of inulin (%) for a given age group was calculated as follows: 100* (intestinal absorption in the inulin-fed age group/intestinal absorption in the same age group without inulin). The rats were given 44Ca and 25Mg after 14 days of inulin intake (7.5%), and fecal non-absorbed isotopes were determined in a 4d feces pool.
Acknowledgements
The authors are grateful to ORAFTI (Tienen, Belgium) for providing the inulin for this study. The authors thank Séverine Thien, Lydia Jaffrelo, Claudine Lab and Pierre Lamby for their technical assistance.
==== Refs
Gibson GR Beatty ER Wang X Cummings JH Selective stimulation of bifidobacteria in the human colon by oligofructose and inulin Gastroenterology 1995 108 975 982 7698613 10.1016/0016-5085(95)90192-2
Jackson KG Taylor GR Clohessy AM Williams CM The effect of the daily intake of inulin on fasting lipid, insulin and glucose concentrations in middle-aged men and women Br J Nutr 1999 82 23 30 10655953
Kaur N Gupta AK Applications of inulin and oligofructose in health and nutrition J Biosci 2002 27 703 714 12571376
Delzenne N Aertssens J Verplaetse H Roccaro M Roberfroid M Effect of fermentable fructo-oligosaccharides on mineral, nitrogen and energy digestive balance in the rat Life Sci 1995 57 1579 1587 7564905 10.1016/0024-3205(95)02133-4
Ohta A Ohtsuki M Baba S Adachi T Sakata T Sakaguchi E Calcium and magnesium absorption from the colon and rectum are increased in rats fed fructooligosaccharides J Nutr 1995 125 2417 2424 7666261
Younes H Demigne C Remesy C Acidic fermentation in the caecum increases absorption of calcium and magnesium in the large intestine of the rat Br J Nutr 1996 75 301 314 8785206 10.1079/BJN19960132
Levrat MA Remesy C Demigne C High propionic acid fermentations and mineral accumulation in the cecum of rats adapted to different levels of inulin J Nutr 1991 121 1730 1737 1941180
Lutz T Scharrer E Effect of short-chain fatty acids on calcium absorption by the rat colon Exp Physiol 1991 76 615 618 1910768
Trinidad TP Wolever TM Thompson LU Effects of calcium concentration, acetate, and propionate on calcium absorption in the human distal colon Nutrition 1999 15 529 533 10422081 10.1016/S0899-9007(99)00110-0
Topping DL Clifton PM Short-chain fatty acids and human colonic function: roles of resistant starch and nonstarch polysaccharides Physiol Rev 2001 81 1031 1064 11427691
Remesy C Levrat MA Gamet L Demigne C Cecal fermentations in rats fed oligosaccharides (inulin) are modulated by dietary calcium level Am J Physiol 1993 264 G855 62 8388642
Coudray C Tressol JC Gueux E Rayssiguier Y Effects of inulin-type fructans of different chain length and type of branching on intestinal absorption and balance of calcium and magnesium in rats Eur J Nutr 2003 42 91 98 12638030 10.1007/s00394-003-0390-x
Coudray C Demigne C Rayssiguier Y Effects of dietary fibers on magnesium absorption in animals and humans J Nutr 2003 133 1 4 12514257
Demigne C Remesy C Rayssiguier Y Effect of fermentable carbohydrates on volatile fatty acids, ammonia and mineral absorption in the rat caecum Reprod Nutr Dev 1980 20 1351 1359 7349487
De Bievre P Taylor PDP Table of isotopic composition of the elements International Journal of Mass Spectrometry Ion Process 123 1993 123 149 166 10.1016/0168-1176(93)87009-H
Coudray C Pepin D Tressol JC Bellanger J Rayssiguier Y Study of magnesium bioavailability using stable isotopes and the inductively-coupled plasma mass spectrometry technique in the rat: single and double labelling approaches Br J Nutr 1997 77 957 970 9227192 10.1079/BJN19970093
Coudray C Bellanger J Castiglia-Delavaud C Remesy C Vermorel M Rayssignuier Y Effect of soluble or partly soluble dietary fibres supplementation on absorption and balance of calcium, magnesium, iron and zinc in healthy young men Eur J Clin Nutr 1997 51 375 380 9192195 10.1038/sj.ejcn.1600417
Lopez HW Coudray C Levrat-Verny MA Feillet-Coudray C Demigne C Remesy C Fructooligosaccharides enhance mineral apparent absorption and counteract the deleterious effects of phytic acid on mineral homeostasis in rats J Nutr Biochem 2000 11 500 508 11120448 10.1016/S0955-2863(00)00109-1
Younes H Coudray C Bellanger J Demigne C Rayssiguier Y Remesy C Effects of two fermentable carbohydrates (inulin and resistant starch) and their combination on calcium and magnesium balance in rats Br J Nutr 2001 86 479 485 11591235
Ohta A Ohtsuki M Baba S Hirayama M Adachi T Comparison of the nutritional effects of fructo-oligosaccharides of different sugar chain length in rats Nutr Res 1998 18 109 120 10.1016/S0271-5317(97)00205-4
Scholz-Ahrens KE Schaafsma G van den Heuvel EG Schrezenmeir J Effects of prebiotics on mineral metabolism Am J Clin Nutr 2001 73 459S 464S 11157358
Cashman K Prebiotics and calcium bioavailability Curr Issues Intest Microbiol 2003 4 21 32 12691259
Kayne LH Lee DB Intestinal magnesium absorption Miner Electrolyte Metab 1993 19 210 217 8264506
Schaafsma G Bioavailability of calcium and magnesium Eur J Clin Nutr 1997 51 Suppl 1 S13 6 9023473
Armbrecht HJ Effect of age on calcium and phosphate absorption. Role of 1,25-dihydroxyvitamin D Miner Electrolyte Metab 1990 16 159 166 2250622
Chan EL Lau E Shek CC MacDonald D Woo J Leung PC Swaminathan R Age-related changes in bone density, serum parathyroid hormone, calcium absorption and other indices of bone metabolism in Chinese women Clin Endocrinol (Oxf) 1992 36 375 381 1424171
Kinyamu HK Gallagher JC Prahl JM DeLuca HF Petranick KM Lanspa SJ Association between intestinal vitamin D receptor, calcium absorption, and serum 1,25 dihydroxyvitamin D in normal young and elderly women J Bone Miner Res 1997 12 922 928 9169351
Weaver CM Age related calcium requirements due to changes in absorption and utilization J Nutr 1994 124 1418S 1425S 8064395
Chonan O Watanuki M The effect of 6'-galactooligosaccharides on bone mineralization of rats adapted to different levels of dietary calcium Int J Vitam Nutr Res 1996 66 244 249 8899459
Coudray C Feillet-Coudray C Tressol JC Gueux E Thien S Jaffrelo L Mazur A Rayssiguier Y Stimulatory effect of inulin on intestinal absorption of calcium and magnesium in rats is modulated by dietary calcium intakesShort- and long-term balance studies Eur J Nutr 2004
Ohta A Motohashi Y Ohtsuki M Hirayama M Adachi T Sakuma K Dietary fructooligosaccharides change the concentration of calbindin-D9k differently in the mucosa of the small and large intestine of rats J Nutr 1998 128 934 939 9614150
Takasaki M Inaba H Ohta A Motohashi Y Sakai K Morris H Sakuma K Dietary short-chain fructooligosaccharides increase calbindin-D9k levels only in the large intestine in rats independent of dietary calcium deficiency or serum 1,25 dihydroxy vitamin D levels Int J Vitam Nutr Res 2000 70 206 213 11068700
Durlach J Bac P Durlach V Rayssiguier Y Bara M Guiet-Bara A Magnesium status and ageing: an update Magnes Res 1998 11 25 42 9595547
Coudray C Gaumet N Bellanger J Coxam V Barlet JP Rayssiguier Y Influence of age and hormonal treatment on intestinal absorption of magnesium in ovariectomised rats Magnes Res 1999 12 109 114 10423705
Vaquero MP Magnesium and trace elements in the elderly: intake, status and recommendations J Nutr Health Aging 2002 6 147 153 12166371
Ferment O Touitou Y Magnesium: metabolism and hormonal regulation in different species Comp Biochem Physiol A 1985 82 753 758 14570081 10.1016/0300-9629(85)90478-5
Hardwick LL Jones MR Brautbar N Lee DB Site and mechanism of intestinal magnesium absorption Miner Electrolyte Metab 1990 16 174 180 2250624
|
16248894
|
PMC1283152
|
CC BY
|
2021-01-04 16:37:15
|
no
|
Proteome Sci. 2005 Oct 28; 3:8
|
latin-1
|
Proteome Sci
| 2,005 |
10.1186/1477-5956-3-8
|
oa_comm
|
==== Front
Respir ResRespiratory Research1465-99211465-993XBioMed Central London 1465-9921-6-1191623616910.1186/1465-9921-6-119ResearchLiquid chromatography/mass spectrometry analysis of exhaled leukotriene B4 in asthmatic children Montuschi Paolo [email protected] Simona [email protected] Marialinda [email protected] Chiara [email protected] Peter J [email protected] Marcello [email protected] Department of Pharmacology, Faculty of Medicine, Catholic University of the Sacred Heart, Rome, Italy2 Department of Forensic Medicine, Faculty of Medicine, Catholic University of the Sacred Heart, Rome, Italy3 Department of Immunodermatology, Istituto Dermopatico dell'Immacolata, IDI, IRCCS, Rome, Italy4 Department of Thoracic Medicine, Imperial College, School of Medicine at the National Heart and Lung Institute, London, UK2005 19 10 2005 6 1 119 119 16 9 2005 19 10 2005 Copyright © 2005 Montuschi et al; licensee BioMed Central Ltd.2005Montuschi et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
The role of leukotriene (LT) B4, a potent inflammatory mediator, in atopic asthmatic and atopic nonasthmatic children is largely unknown. The lack of a gold standard technique for measuring LTB4 in exhaled breath condensate (EBC) has hampered its quantitative assessment in this biological fluid. We sought to measure LTB4 in EBC in atopic asthmatic children and atopic nonasthmatic children. Exhaled nitric oxide (NO) was measured as an independent marker of airway inflammation.
Methods
Fifteen healthy children, 20 atopic nonasthmatic children, 25 steroid-naïve atopic asthmatic children, and 22 atopic asthmatic children receiving inhaled corticosteroids were studied. The study design was of cross-sectional type. Exhaled LTB4 concentrations were measured using liquid chromatography/mass spectrometry-mass spectrometry (LC/MS/MS) with a triple quadrupole mass spectrometer. Exhaled NO was measured by chemiluminescence with a single breath on-line method. LTB4 values were expressed as the total amount (in pg) of eicosanoid expired in the 15-minute breath test. Kruskal-Wallis test was used to compare groups.
Results
Compared with healthy children [87.5 (82.5–102.5) pg, median and interquartile range], exhaled LTB4 was increased in steroid-naïve atopic asthmatic [255.1 (175.0–314.7) pg, p < 0.001], but not in atopic nonasthmatic children [96.5 (87.3–102.5) pg, p = 0.59)]. Asthmatic children who were receiving inhaled corticosteroids had lower concentrations of exhaled LTB4 than steroid-naïve asthmatics [125.0 (25.0–245.0) pg vs 255.1 (175.0–314.7) pg, p < 0.01, respectively]. Exhaled NO was higher in atopic nonasthmatic children [16.2 (13.5–22.4) ppb, p < 0.05] and, to a greater extent, in atopic steroid-naïve asthmatic children [37.0 (31.7–57.6) ppb, p < 0.001] than in healthy children [8.3 (6.1–9.9) ppb]. Compared with steroid-naïve asthmatic children, exhaled NO levels were reduced in asthmatic children who were receiving inhaled corticosteroids [15.9 (11.5–31.7) ppb, p < 0.01].
Conclusion
In contrast to exhaled NO concentrations, exhaled LTB4 values are selectively elevated in steroid-naïve atopic asthmatic children, but not in atopic nonasthmatic children. Although placebo control studies are warranted, inhaled corticosteroids seem to reduce exhaled LTB4 in asthmatic children. LC/MS/MS analysis of exhaled LTB4 might provide a non-invasive, sensitive, and quantitative method for airway inflammation assessment in asthmatic children.
==== Body
Background
Chronic airway inflammation, the primary pathophysiological feature of asthma, underlies asthma symptoms and might have a role in airway and lung remodelling [1]. Elevated airway inflammation often precedes the onset of symptoms or airway limitation [1]. Quantification of airway inflammation is difficult as it requires invasive techniques or measurement of biomarkers in plasma or urine which reflect systemic rather than lung inflammation. Bronchoscopy with bronchoalveolar lavage remains the gold standard to assess airway inflammation, but invasiveness makes it unethical as a routine method particularly in children [2]. Sputum induction is less invasive [3], although this technique is particularly difficult to apply in young children and itself induces airway inflammation [4].
This has led to a search for non-invasive ways to measure airway and lung inflammation in order to aid diagnosis, to assess response to anti-inflammatory treatments, to predict loss of disease control and to assess the response to novel treatments [5]. There is increasing evidence that measurement of biomarkers in the breath may reflect pulmonary disease [6]. Direct sampling form the lung has major advantages compared with sampling form the blood or urine, when dilution and metabolism of inflammatory markers arising in the lungs makes interpretation very difficult. Exhaled nitric oxide (NO) measurement is currently the only completely noninvasive method for quantifying airway inflammation in patients with asthma who are not treated with corticosteroids [2,7]. However, as several inflammatory mediators are involved in asthma, identification of other exhaled markers might contribute to a better understanding of its complex pathophysiology. Exhaled breath condensate (EBC) is a completely noninvasive method for collecting airway secretions [8,9]. Cysteynyl-leukotrienes play an important role in the pathophysiology of asthma and their regulation is relevant for the management of this disease [10]. However, the role of leukotriene (LT) B4, a potent chemoattractant for neutrophils, in asthma is largely unknown. Using immunoassays, LTB4 has been detected in EBC in healthy subjects and found increased in asthmatic adults [11,12] and children [13,14]. However, the presence of LTB4 in EBC has not been confirmed by more specific analytical techniques. Moreover, a large variation in exhaled LTB4 concentrations in patients with similar clinical and functional features has been reported in different studies [13-15]. This may be partially explained by the lack of validation of commercially available immunoassay kits which were used to measure exhaled LTB4 in these studies [16]. Mass spectrometry, the reference analytical technique, is required for an accurate quantitative assessment of exhaled LTB4. In the present study, we aimed to measure LTB4 concentrations in EBC with liquid chromatography/mass spectrometry-mass spectrometry (LC/MS/MS) using deuterated (d4) LTB4 as internal standard in healthy children, atopic nonasthmatic children, steroid-naïve atopic asthmatic children, and atopic asthmatic children who were treated with inhaled corticosteroids. We also measured exhaled NO as an independent marker of airway inflammation.
Methods
Study subjects
Four groups of children were studied: 15 healthy children, 20 atopic nonasthmatic children, 25 steroid-naïve atopic asthmatic children, and 22 atopic asthmatic children who were receiving inhaled corticosteroids (Table 1). Atopic nonasthmatic and atopic asthmatic children were recruited from the Asthma and Allergy outpatient clinic, Istituto Dermopatico dell'Immacolata, Rome, Italy. The diagnosis and classification of asthma was based on clinical history and examination and pulmonary function parameters according to the Guidelines for the Diagnosis and Management of Asthma issued by the National Heart, Lung, and Blood Institute of the National Institutes of Health [17]. Steroid-naïve atopic asthmatic children had mild intermittent asthma with symptoms less often than twice a week, forced expiratory volume in one second (FEV1) of 80% or greater of predicted value and reversibility of 12% or greater to salbutamol, or positive provocation test result with methacholine or exercise. They were not taking any regular medication but used inhaled short-acting β2-agonists as needed for symptom relief. They were excluded from the study if they had used inhaled corticosteroids or LT receptor antagonists in the previous 4 weeks. Atopy was confirmed by skin prick testing. Steroid-treated atopic asthmatic children had persistent mild-to-moderate asthma and were receiving maintenance therapy with either 100 μg/day (10 children) or 200 μg/day (12 children) of inhaled fluticasone propionate administered by means of diskus (GSK, Uxbridge, United Kingdom) at a constant dose for at least 8 weeks. They had not been taking LT receptor antagonists for at least one month. Atopy was confirmed by skin prick tests.
Table 1 Subject characteristics§
Healthy children Atopic nonasthmatic children Steroid-naïve atopic asthmatic children Steroid-treated atopic asthmatic children
n 15 20 25 22
Sex, F/M 8/7 9/11 12/13 10/12
Age, yr 10 ± 1 9 ± 1 10 ± 1 10 ± 1
FEV1, % pred 101.5 ± 1.8 100.8 ± 2.3 96.1 ± 1.4 97.5 ± 2.3
FVC, % pred 102.7 ± 2.8 103.2 ± 2.5 100.4 ± 2.3 101.5 ± 2.7
FEV1/FVC, % 99.8 ± 1.5 97.5 ± 2.6 87.1 ± 2.3* 85.8 ± 2.5*
FEF25–75% 104.0 ± 3.8 101.8 ± 4.1 68.2 ± 3.9** 83.7 ± 4.5*
§ Data are expressed as numbers or mean ± SEM.
* P < 0.05 and ** P < 0.01 compared with healthy children. Steroid-treated asthmatic children were receiving either 100 μg/day or 200 μg/day of inhaled fluticasone at a constant dose for at least 8 weeks.
Atopic nonasthmatic children had a clinical history of atopy and positive skin test results. All had allergic rhinitis and were either sensitized to perennial allergens or were recently exposed to a relevant allergen.
Healthy control children had no history of asthma and atopic disease and negative skin prick test results. They were recruited from children of staff.
Study group children had no upper respiratory tract infections in the previous 3 weeks. Children were excluded from the study if they had used oral corticosteroids in the previous 4 weeks or non-steroidal anti-inflammatory drugs in the last 2 weeks.
Study design
The study was of cross-sectional design. Children attended the Department of Pharmacology, Catholic University of the Sacred Heart, Rome, Italy, for clinical examination, EBC collection, spirometry and skin prick testing. Informed consent was obtained by parents and the study was approved by the Ethics Committee of the Catholic University of the Sacred Heart, Rome, Italy.
Day-to-day repeatability for LTB4 measurements was assessed in 20 children with asthma in a randomized design by collecting three EBC samples within 7 days of the first.
Pulmonary function
Spirometry was performed by means of a 10 L bell spirometer (Biomedin, Padova, Italy) and the best of three maneuvers, expressed as a percentage of predicted values, was chosen.
Exhaled breath condensate collection
Exhaled breath condensate was collected using a condensing chamber (Ecoscreen, Jaeger, Hoechberg, Germany) as described previously [11]. Briefly, exhaled air entered and left the chamber through one-way valves at the inlet and outlet, thus keeping the chamber closed. Children were instructed to breath tidally through a mouthpiece connected to the condenser for 15 min. An average of 1.5 ml EBC per child was collected and stored a -80°C. LTB4 measurements were performed within 2 weeks from sample collection. α-Amylase concentrations in all EBC samples were measured by an in vitro colorimetric method to check for possible salivary contamination (Roche Diagnostics, Basel, Switzerland).
Measurement of exhaled LTB4
Exhaled LTB4 was measured with a Finnigan Surveyor®LC System pump coupled with a TSQ Quantum Ultra™ triple quadrupole mass spectrometer using electrospray ionization with negative ion polarity mode with electrospray source (Thermo Electron Corporation, San José, USA). The LC/MS/MS experimental conditions were similar to those described previously [18]. However, the use of a Finnigan Surveyor®LC System pump coupled with a TSQ Quantum Ultra™ triple quadrupole mass spectrometer made it possible to increase 10 fold the analytical sensitivity (10 pg/ml vs 100 pg/ml). The lower limit of quantification (LOQ) was 50 pg/ml. The optimized source parameters were as follows: spray voltage, 4500 V; sheath gas pressure, 40 arbitrary units; aux gas pressure, 10 arbitrary units; capillary temperature, 300°C; capillary offset, 30 V; tube lens offset, 110 V. Chromatography was performed on a BetaBasic C18 column (15 cm × 2, 1 mm internal diameter, Thermo Hypersil-Keyston, Bellefonte, PA, USA) using a linear gradient with acetonitrile-water-acetic acid (30:70:0.05, v/v, at pH 5–6) which was changed to 100% acetonitrile over 4 minutes. Flow rate was 0.25 ml/min. LTB4-d4 (Cayman Chemicals Co., Ann Arbor, MI) was used as internal standard and was added to each sample to reach a final concentration of 1000 pg/ml. The peak area ratios for LTB4/LTB4-d4 were plotted versus LTB4 concentrations. The calibration curve in the range 10–500 pg/ml had a good linearity with r2 = 0.9965. 10 μl of sample were directly injected into the liquid chromatograph without any pretreatment.
All solvents were high performance liquid chromatography grade obtained from Merck (Darmstadt, Germany); acetic acid extra pure was purchased by Riedel-de Haen, Sigma-Aldrich, Seelze, Germany. LTB4 and LTB4-d4 were purchased from Cayman Chemicals Co (Ann Arbor, MI, USA) and stored at -20°C until use. The working solutions were prepared daily using mobile phase as solvent.
LTB4 values in EBC were expressed as total amount (in picograms) of leukotriene expired in the 15-minute breath test (LT concentrations × volume of EBC). Concentration of LTB4 in undetectable samples was considered 25 pg/ml corresponding to 50% of the lower limit of quantification (50 pg/ml).
Exhaled nitric oxide measurement
Exhaled NO was measured with the NIOX system (Aerocrine, Stockholm, Sweden) with a single breath on-line method according to American Thoracic Society guidelines [7]. Children inhaled NO-free air and exhaled through a dynamic flow restrictor. Children were asked to exhale to residual volume, insert the mouthpiece, inhale to total lung capacity through a NO filter, and then to exhale into the device through the same mouthpiece for 10 seconds at constant flow of 50 ml/sec with the aid of a visual feedback on a computer screen. A negative pressure tracing on the screen was used to confirm that children were inhaling NO-free air. Exhalations were repeated after 1-minute relaxation period until the performance of three exhaled NO values varied less than 10%. Exhaled NO measurements were obtained before spirometry.
Statistical analysis
Linear regression analysis was used to assess the relationship between LC peak area ratios for LTB4/LTB4-d4 and LTB4 concentrations. Exhaled LTB4 and NO values were expressed as medians throughout with interquartile ranges (25th to 75th percentiles) shown in parentheses. Kruskal-Wallis test followed by pairwise Mann-Whitney U tests were used to compare groups. Correlations between variables were evaluated by Spearman test. Significance was defined as a value of p < 0.05.
Results
No α-amylase concentrations (<22 mU/ml) were detected in any study sample, excluding significant salivary contamination.
LTB4 in EBC
The mass spectra of LTB4 and LTB4-d4 revealed a base peak at m/z 335 for endogenous LTB4 and m/z 339 for internal standard LTB4-d4, corresponding to molecular ions. To increase specificity, LTB4 was detected in MS/MS mode. Parent ions m/z 335 and m/z 339 were fragmented with a collision energy of 17% and 19%, respectively. The transition 335 → 195 m/z and 339 → 197 m/z for internal standard was monitored (Figure 1). Exhaled LTB4 concentrations were detected in all steroid-naïve atopic asthmatics, atopic nonasthmatics, and healthy children and were undetectable in 7 steroid-treated atopic asthmatic children (Figure 2). Compared with healthy children [87.5 (82.5–102.5) pg], exhaled LTB4 was increased in steroid-naïve atopic asthmatic children [255.1 (175.0–314.7) pg, p < 0.001], but not in atopic nonasthmatic children [96.5 (87.3–102.5) pg, p = 0.59) (Figure 2). Steroid-naïve children with atopic asthma had higher exhaled LTB4 concentrations than atopic nonasthmatic children (p < 0.001) and healthy children (p < 0.001) (Figure 2).
Figure 1 Liquid chromatography/mass spectrometry-mass spectrometry chromatograms of exhaled breath condensate (EBC). Samples from an asthmatic child (A and B) and a healthy child (C and D) are shown. The ions m/z 195 (A and C) and m/z 197 (B and D) were used to monitor endogenous leukotriene (LT) B4 and LTB4-d4 (internal standard), respectively. LTB4 concentrations in EBC, calculated using peak area ratios for LTB4/LTB4-d4, were 87 pg/15 min (healthy child) and 450 pg/15 min (asthmatic child).
Figure 2 Leukotriene (LT) B4 concentrations in exhaled breath condensate in children with allergic disease. Exhaled LTB4 concentrations in healthy children (open squares), atopic non-asthmatic children (NA) (filled squares), steroid-naïve atopic asthmatic children (open triangles), and steroid-treated (ST) children with atopic asthma (filled triangles). LTB4 values are expressed as picograms produced during 15 minutes of breathing. Median values are shown with horizontal bars.
Asthmatic children who were receiving inhaled corticosteroids had lower concentrations of exhaled LTB4 than steroid-naïve asthmatics [125.0 (25.0–245.0) pg vs 255.1 (175.0–314.7) pg, p < 0.01, respectively] and similar to those in atopic nonasthmatic children (p = 0.41) and healthy controls (p = 0.43) (Figure 2). In steroid-treated atopic asthmatic children, subgroup analysis shows that children who were receiving 100 μg/day of inhaled fluticasone at a constant dose for at least 8 weeks had higher exhaled LTB4 values than those who were receiving with 200 μg/day of inhaled fluticasone at a constant dose for at least 8 weeks [245.0 (235.0–282.5) pg vs 25.0 (25.0–102.5) pg, p < 0.002, respectively) (Figure 3). In children receiving 200 μg/day of fluticasone, median exhaled LTB4 values were similar to those in atopic nonasthmatic children (p = 0.15) and healthy controls (p = 0.10). In contrast, exhaled LTB4 was elevated in asthmatic children who were receiving 100 μg/day of inhaled fluticasone (atopic nonasthmatic children: p < 0.005; healthy children: p < 0.001).
Figure 3 Leukotriene (LT) B4 concentrations in exhaled breath condensate in steroid-treated children with atopic asthma. Exhaled LTB4 concentrations in atopic asthmatic children who were receiving either 100 μg/day (open squares) or 200 μg/day (filled squares) of inhaled fluticasone at a constant dose for at least 8 weeks. LTB4 values are expressed as picograms produced during 15 minutes of breathing. Median values are shown with horizontal bars.
Seven asthmatic children who were treated with the higher dose of fluticasone had undetectable levels of exhaled LTB4 (Figure 3). MS/MS spectra of steroid-naïve asthmatic children (Figure 1A), healthy children (Figure 1C), and atopic nonasthmatic children (not shown) showed a major unknown peak in all samples. This peak is not always present in steroid-treated asthmatic children (not shown) and deserves further characterization. There was no correlation between exhaled LTB4 and exhaled NO in any study group. There was no correlation between exhaled LTB4 or exhaled NO and age, sex, or lung function in any study group. The intraclass correlation coefficient for LTB4 was 0.89.
Exhaled NO
Exhaled NO was higher in atopic nonasthmatic children [16.2 (13.5–22.4) ppb, p < 0.05] and, to a greater extent, in steroid-naïve atopic asthmatic children [37.0 (31.7–57.6) ppb, p < 0.001] than in healthy children [8.3 (6.1–9.9) ppb] (Figure 4). Compared with steroid-naïve asthmatic children, exhaled NO levels were reduced in asthmatic children who were receiving inhaled corticosteroids [15.9 (11.5–31.7) ppb, p < 0.01] (Figure 4). In these children, exhaled NO concentrations were higher than those in healthy controls (p < 0.01), but not than those in atopic nonasthmatic children (p = 0.98) (Figure 4). There was no difference in exhaled NO levels between asthmatic children receiving 200 μg/day of inhaled fluticasone and children receiving a daily dose of 100 μg [14.1 (11.3–22.9) ppb vs 19.8 (13.5–44.5) ppb, p = 0.27, respectively)] (Figure 5). Both subgroups of steroid-treated asthmatic children had higher exhaled NO levels than healthy children (children treated with fluticasone at a dose of 100 μg/day: p < 0.001; children treated with fluticasone at a dose of 200 μg/day: p < 0.001) but not than atopic nonasthmatic children (children treated with fluticasone at dose of 100 μg/day: p = 0.42; children treated with fluticasone at a dose of 200 μg/day: p = 0.46).
Figure 4 Exhaled nitric oxide (NO) concentrations in children with allergic disease. Exhaled NO concentrations in healthy children (open squares), atopic non-asthmatic children (NA) (filled squares), steroid-naïve atopic asthmatic children (open triangles), and steroid-treated (ST) children with atopic asthma (filled triangles). Median values are shown with horizontal bars.
Figure 5 Exhaled nitric oxide (NO) concentrations in exhaled breath condensate in steroid-treated children with atopic asthma. Exhaled NO concentrations in atopic asthmatic children who were receiving either 100 μg/day (open squares) or 200 μg/day (filled squares) of inhaled fluticasone at a constant dose for at least 8 weeks. Median values are shown with horizontal bars.
Discussion
EBC is a completely non-invasive method for sampling secretions from the airways which is suitable for repeated measures of lung inflammation even in young children [19]. Using LTB4-d4 as internal standard, we have definitively demonstrated that LTB4 is present in the EBC, providing an accurate quantitative assessment of its concentrations in this biological fluid. In our study, absolute LTB4 values in EBC are significantly higher than those previously reported in studies using enzyme immunoassays [14,15], but similar to those found in studies in which exhaled LTB4 was measured by MS analysis [18,20]. The reasons for this discrepancy are unknown but could be partially related to matrix effects as differences in protein composition of the immunoassay buffer used for LTB4 standard dilution versus EBC samples might influence LTB4 measurement. We avoided possible matrix effects as EBC samples were directly injected into the chromatograph without any pre-treatment. Our study population of atopic asthmatic children had well maintained lung function as indicated by FEV1 values greater than 80% of predicted values, but evidence for small airway obstruction given that values of FEV1/FVC ratio and FEF25–75%, which are more sensitive for detecting reduced lung function in children [1], were decreased. Compared with healthy control children, exhaled LTB4 concentrations were selectively elevated in steroid-naïve children with atopic asthma, but not in atopic nonasthmatic children and in steroid-treated atopic asthmatic children. Enhanced airway inflammation in steroid-naïve children with asthma was confirmed by elevated exhaled NO levels in these children. These findings indicate that exhaled LTB4 can be used as a new in vivo non-invasive marker of inflammation in steroid-naïve asthmatic children.
We did not study the effects of allergen challenge on exhaled LTB4 in atopic nonasthmatic children and atopic asthmatic children. Similar exhaled LTB4 values in healthy children and atopic nonasthmatic children who had either perennial rhinitis or were recently exposed to a relevant allergen indicate that allergen exposure has little, if any, effect on exhaled LTB4 in atopic nonasthmatic children. However, allergen challenge studies to formally address this issue are warranted. The effect of allergen challenge on exhaled LTB4 levels in atopic asthmatic children is unknown and needs to be clarified.
In contrast to exhaled LTB4, exhaled NO levels were elevated in atopic nonasthmatic children although this increase was less pronounced than in steroid-naïve atopic asthmatic children. These findings and the lack of correlation between exhaled LTB4 and exhaled NO levels suggest a different biological significance for these two biomarkers as exhaled LTB4 might be considered as a marker of asthma, whereas exhaled NO might reflect the different degree of inflammation within the respiratory tract in atopic children with and without asthma. This may have important implications for diagnosis and treatment of children with allergic disease.
Measurement of exhaled LTB4 might help identify those children with no current respiratory symptoms, well-maintained lung function and elevated exhaled NO who are not only atopic but have ongoing asthmatic inflammation which may require earlier pharmacological therapy. Moreover, elevated exhaled NO levels in atopic nonasthmatic children with rhinitis reflect the presence of inflammation within the airways as the restricted exhalation method ensures velum closure excluding contamination of the exhaled air with nasal NO [7]. This is consistent with a common origin for allergic rhinitis and asthma [1]. However, larger studies are needed to assess the biological significance of these findings and their implications.
In the present study, median exhaled LTB4 concentrations in children who were receiving inhaled fluticasone were lower than those in steroid-naïve asthmatic children. A significant reduction in exhaled LTB4 levels was observed only in the group of children who were receiving 200 μg/day of inhaled fluticasone, whereas in children who were receiving a daily dose of inhaled fluticasone of 100 μg median exhaled LTB4 levels were similar to those in steroid-naïve asthmatic children. These findings might indicate that the lower dose of fluticasone is unable to reduce exhaled LTB4 levels in asthmatic children. However, there was a high individual variability in exhaled LTB4 concentrations in both subgroups of steroid-treated asthmatic children as well as in steroid-naïve asthmatic children which needs to be addressed in future studies. In a previous open-label uncontrolled study, treatment with inhaled fluticasone (100 μg twice a day for 4 weeks) had no effect on exhaled LTB4 concentrations in atopic asthmatic children with similar features [14]. This might be due to differences in the study design (open-label vs observational), duration of treatment (4 weeks vs at least 8 weeks), and the analytical technique used for measuring LTB4 in EBC (enzyme immunoassay vs LC/MS/MS) as concentrations of exhaled LTB4 measured by MS are 10-fold higher than those measured by immunoassay. However, the cross-sectional study design of the present study precludes definitive conclusions on the effect of inhaled corticosteroids on exhaled LTB4 in asthmatic children for which large controlled studies are required.
One limitation of the EBC analysis is that the cellular source of the inflammatory mediators within the airways cannot be ascertained. To this aim, invasive techniques such as bronchial biopsies are required.
To adjust for possible variations in the EBC sample volume, LTB4 values in EBC were expressed as total amount of the eicosanoid expired over a standard period of collection. However, as pointed out by Effros and coworkers, part of the variation in non-volatile compound concentrations in EBC may be related to differences in the dilution of respiratory droplets by water vapor [21]. The lack of correlation between structurally related compounds such as LTB4 and LTE4 reported in previous studies [13,14] does not seem to support this evidence. However, reference indicators such as measurement of conductivity as proposed by Effros and coworkers [22] should be used in future studies aiming at quantifying exhaled eicosanoids.
Conclusion
Exhaled LTB4 is elevated in steroid-naïve atopic asthmatic children, but not in children receiving inhaled corticosteroids and in atopic nonasthmatic children. LC/MS/MS analysis of exhaled LTB4 might provide a completely non-invasive, sensitive, and quantitative method for airway inflammation assessment in asthmatic children. This technique is potentially applicable to patients with other lung diseases and suitable for longitudinal studies and assessing pharmacological therapy. The high cost is currently an important limitation. Further research is warranted for the complete characterization of the exhaled substances including cysteinyl-leukotrienes and for establishing the effect anti-asthmatic drugs on exhaled LTB4.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
PM conceived the idea for the study, collected exhaled breath condensate samples and measured exhaled nitric oxide. PM and PJB planned the investigation. CM recruited children and did clinical investigation and lung function tests. SM, MF and MC performed LC/MS analysis of leukotriene B4 in exhaled breath condensate. PM, PJB and MC participated in analysis and interpretation of data. PM wrote the report which was revised by PJB and MC. All authors read and approved the final manuscript.
Acknowledgements
This work was supported by the Catholic University of the Sacred Heart, Rome, Italy, academic funds 2004-2005. The sources of funding had no role in writing the report and did not participate in the decision to publish the results.
==== Refs
Stone KD Atopic diseases of childhood Curr Opin Pediatr 2003 15 495 511 14508299 10.1097/00008480-200310000-00009
Bates CA Silkoff PE Exhaled nitric oxide in asthma: from bench to bedside J Allergy Clin Immunol 2003 111 256 262 12589342 10.1067/mai.2003.103
Parameswaran K Pizzichini E Pizzichini MM Hussack Efthimiadis A Hargreave FE Clinical judgement of airway inflammation versus sputum cell counts in patients with asthma Eur Respir J 2000 15 486 490 10759441 10.1034/j.1399-3003.2000.15.10.x
Nightingale JA Rogers DF Barnes PJ Effect of inhaled ozone on exhaled nitric oxide, pulmonary function, and induced sputum in normal and asthmatic subjects Thorax 1999 54 1061 1069 10567624
Montuschi P Indirect monitoring of lung inflammation Nat Rev Drug Discov 2002 1 238 242 12120509 10.1038/nrd751
Kharitonov SA Barnes PJ Exhaled markers of pulmonary disease Am J Respir Crit Care Med 2001 163 1693 1722 11401895
Recommendations for standardized procedures for the on-line and off-line measurement of exhaled lower respiratory nitric oxide and nasal nitric oxide in adults and children-1999. Official statement of the American Thoracic Society 1999 Am J Respir Crit Care Med 1999 160 2104 2117 10588636
Hunt J Exhaled breath condensate: an evolving tool for non-invasive evaluation of lung disease J Allergy Clin Immunol 2002 110 28 34 12110814 10.1067/mai.2002.124966
Mutlu GM Garey KW Robbins RA Rubinstein I Collection and analysis of exhaled breath condensate in humans Am J Respir Crit Care Med 2001 164 731 737 11549524
Leff AR Regulation of leukotrienes and development of antileukotriene agents Annu Rev Med 2001 52 1 14 11160764 10.1146/annurev.med.52.1.1
Montuschi P Barnes PJ Exhaled leukotrienes and prostaglandins in asthma J Allergy Clin Immunol 2002 109 615 620 11941309 10.1067/mai.2002.122461
Hanazawa T Kharitonov SA Barnes PJ Increased nitrotyrosine in exhaled breath condensate of patients with asthma Am J Respir Crit Care Med 2000 162 1273 1276 11029330
Csoma Z Kharitonov SA Balint B Bush A Wilson NM Barnes PJ Increased leukotrienes in exhaled breath condensate in childhood asthma Am J Respir Crit Care Med 2002 166 1345 1349 12406853 10.1164/rccm.200203-233OC
Mondino C Ciabattoni G Koch P Pistelli R Trové A Barnes PJ Montuschi P Effects of inhaled corticosteroids on exhaled leukotrienes and prostaglandins in asthmatic children J Allergy Clin Immunol 2004 114 761 767 15480313 10.1016/j.jaci.2004.06.054
Bodini A Peroni D Vicentini L Loiacono A Baraldi E Ghiro L Corradi M Alinovi R Boner AL Piacentini GL Exhaled breath condensate eicosanoids and sputum eosinophils in asthmatic children: a pilot study Pediatr Allergy Immunol 2004 15 26 31 14998379 10.1046/j.0905-6157.2003.00097.x
Montuschi P Montuschi P Analysis of exhaled breath condensate: methodological issues New perspectives in monitoring lung inflammation: analysis of exhaled breath condensate 2004 Boca Raton: CRC Press 11 30
National Asthma Education and Prevention Program Expert Panel Report: Guidelines for the Diagnosis and Management of Asthma Update on Selected Topics J Allergy Clin Immunol 2002 111 S141 S219
Montuschi P Martello S Felli M Mondino C Chiarotti M Ion trap liquid chromatography/mass spectrometry analysis of leukotriene B4 in exhaled breath condensate Rapid Commun Mass Spectrom 2004 18 2723 2729 15499663 10.1002/rcm.1682
Montuschi P Barnes PJ Analysis of exhaled breath condensate for monitoring airway inflammation Trends Pharmacol Sci 2002 23 232 237 12008001 10.1016/S0165-6147(02)02020-5
Cap P Chladek J Pehal F Maly M Petru V Barnes PJ Montuschi P Gas chromatography/mass spectrometry analysis of exhaled leukotrienes in asthmatic patients Thorax 2004 59 465 470 15170025 10.1136/thx.2003.011866
Effros RM Hoagland KW Bosbous M Castillo D Foss B Dunning M Gare M Lin W Sun F Dilution of respiratory solutes in exhaled condensates Am J Respir Crit Care Med 2002 165 663 669 11874811
Effros RM Biller J Foss B Hoagland K Dunning MB Castillo D Bosbous M Sun F Shaker R A simple method for estimating respiratory solute dilution in exhaled breath condensate Am J Respir Crit Care Med 2003 168 1500 1505 14512268 10.1164/rccm.200307-920OC
|
16236169
|
PMC1283153
|
CC BY
|
2021-01-04 16:36:26
|
no
|
Respir Res. 2005 Oct 19; 6(1):119
|
utf-8
|
Respir Res
| 2,005 |
10.1186/1465-9921-6-119
|
oa_comm
|
==== Front
PLoS GenetPLoS GenetpgenplgeplosgenPLoS Genetics1553-73901553-7404Public Library of Science San Francisco, USA 1629958510.1371/journal.pgen.001005705-PLGE-RA-0192R1plge-01-05-02Research ArticleDevelopmentEvolutionGenetics/Population GeneticsDrosophilaMaternal Expression Relaxes Constraint on Innovation of the Anterior Determinant, bicoid
Maternal Expression Relaxes ConstraintBarker Michael S Demuth Jeffery P *Wade Michael J Department of Biology, Indiana University, Bloomington, Indiana, United States of AmericaStern David EditorPrinceton University, United States of America* To whom correspondence should be addressed. E-mail: [email protected] 2005 11 11 2005 1 5 e5710 8 2005 30 9 2005 Copyright: © 2005 Barker et al.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.The origin of evolutionary novelty is believed to involve both positive selection and relaxed developmental constraint. In flies, the redesign of anterior patterning during embryogenesis is a major developmental innovation and the rapidly evolving Hox gene, bicoid (bcd), plays a critical role. We report evidence for relaxation of selective constraint acting on bicoid as a result of its maternal pattern of gene expression. Evolutionary theory predicts 2-fold greater sequence diversity for maternal effect genes than for zygotically expressed genes, because natural selection is only half as effective acting on autosomal genes expressed in one sex as it is on genes expressed in both sexes. We sample an individual from ten populations of Drosophila melanogaster and nine populations of D. simulans for polymorphism in the tandem gene duplicates bcd, which is maternally expressed, and zerknüllt (zen), which is zygotically expressed. In both species, we find the ratio of bcd to zen nucleotide diversity to be two or more in the coding regions but one in the noncoding regions, providing the first quantitative support for the theoretical prediction of relaxed selective constraint on maternal-effect genes resulting from sex-limited expression. Our results suggest that the accelerated rate of evolution observed for bcd is owing, at least partly, to variation generated by relaxed selective constraint.
Synopsis
How do novel structures and functions originate? This question has proven more difficult to answer than the question of how existing structures are refined to better suit the environment. Evolution by natural selection explains the latter. Ironically, it is the power of natural selection to maintain genes that are suitable for their current roles that also works against the evolution of entirely new traits. The conservation of genes controlling the early stages of embryo development, from worms, to flies, to humans, is a famous example of natural selection's power to constrain evolution and thwart the spread of evolutionary novelties. However, the gene determining which end of the fly embryo becomes the head has a relatively recent origin. How did this gene, bicoid, escape purifying selection and take on a novel function? The authors investigate the hypothesis that because bicoid is expressed only in females it experiences only half as much constraint as a gene expressed in both sexes. Comparing sequences of bicoid with its duplicate gene zerknüllt, which retains expression in both sexes, the authors show that, as expected, the variation in bicoid is twice that of zerknüllt. The findings suggest that relaxed constraint is an important step in the origin of novel function.
Citation:Barker MS, Demuth JP, Wade MJ (2005) Maternal expression relaxes constraint on innovation of the anterior determinant, bicoid. PLoS Genet 1(5): e57.
==== Body
Introduction
The origin of novel structures and functions has long been problematic for evolutionary biology. A century-long debate has centered on the relative contribution of deterministic processes, such as natural selection, and stochastic processes, such as mutation and random genetic drift [1,2] to biological innovation. In modern parlance, “There are problems specific for the origin of novelties that are not the same as in the case of adaptive modification of existing structures ... novelties apparently arise in spite of strong developmental constraints that generally canalize morphological evolution” [3]. How descendant lineages overcome ancestral constraint is a principal research objective in the study of evolutionary novelty.
One route to gaining novel structures and functions is by relaxing ancestral evolutionary constraints. Although the degree of evolutionary lability can be discerned from the phylogenetic distribution of a given trait, the inference of relaxed constraint is more difficult and potentially tautological [4,5]. We show how to test the hypothesis of relaxed constraint using population genetic theory and sequence data from the novel Hox gene, bicoid (bcd), a critical component of the redesign of anterior patterning during fly embryogenesis, a major developmental innovation [6,7].
The gene bcd is present only in the derived Cyclorrhaphan flies, and is the result of duplication and divergence of a more widely conserved Hox3 descendant zerknüllt (zen) [7,8]. After the origin of myriapods in the arthropod lineage leading to insects, Hox3 underwent a functional transformation to what is now called zen [9,10]. In taxa prior to the duplication giving rise to bcd, zen is expressed both maternally and zygotically and is important for early designation of both embryonic and extraembryonic tissues [11]. Following duplication, the ancestral maternal-zygotic zen expression pattern was partitioned into solely maternally expressed bcd and solely zygotic expressed zen [6,8]. Since the partitioning of gene expression, bcd has evolved faster than zen [12,13].
The preservation of both gene copies and the partitioning of the ancestral maternal-zygotic expression pattern between them suggest that functional constraints on both genes remain despite bcd evolving additional, unique functionality. Furthermore, because both are homeotic genes critical to early development, selection on the coding regions of these genes is likely to be predominantly purifying (i.e., the selection coefficient, s, has tended to be negative). Thus, we can reasonably assume that the effective population size, N
e, the regional chromosomal environment (affecting mutation and recombination), and the average magnitude and sign of s have been approximately the same for both genes. There are several factors that might contribute to violations of our assumption of equivalent s. However, it is our purpose to test whether the reduced efficacy of selection that attends sex-limited expression is sufficient to explain the pattern of bcd diversity without invoking changes to s.
From a population genetics perspective, evolutionary constraint can be relaxed by either reducing the strength of purifying selection or limiting the proportion of genes exposed to selection. Here we refer to the former as relaxed selection and the latter as relaxed constraint. Population genetic theory shows that deleterious alleles at autosomal loci with sex-limited effects have equilibrium frequencies twice as high as those expressed in both sexes with similar fitness effects [14,15]. This relaxed constraint is due to deleterious alleles being “hidden” from selection in half of the population (i.e., males, assuming 1:1 sex ratio) in much the same way that recessive alleles have higher expected equilibrium frequencies (i.e., relaxed constraint) because they are “masked” from selection in heterozygotes. If bcd is experiencing relaxed constraint because it has sex-limited expression, then we expect the coding regions of bcd to be twice as diverse as those of zen [13]. Here, we report the first, to our knowledge, test of the prediction that maternal-effect genes have a 2-fold higher heterozygosity at mutation-selection balance than zygotically expressed genes by comparing within species sequence diversity for bcd and zen from ten populations of Drosophila melanogaster and nine of D. simulans.
Results
At mutation-drift balance, the expected value of Tajima's D is near zero [16,17] with positive (balancing selection) or negative (purifying selection) values caused by selection. The largest deviation from zero among our estimates of Tajima's D is −1.01513 for D. melanogaster bcd, and the smallest is −0.42264 for D. melanogaster zen. None of the values are significantly different from zero (or from one another), suggesting both genes in both species are near equilibrium (Table 1).
Table 1 Tajima's D and Nucleotide Diversities (π) for bcd and zen
The observed bcd/zen ratios of total sequence diversity are 1.82 for D. melanogaster and 1.60 for D. simulans. For the noncoding regions, the ratios are 0.74 and 1.28, respectively, while coding diversity ratios are 1.67 and 1.90 (Table 1). Further partitioning the diversity in coding regions, the bcd/zen diversity ratios for synonymous sites are 1.28 (0.00279/0.00218) for D. melanogaster and 1.35 (0.0229/0.017) for D. simulans. For the nonsynonymous sites in coding regions, the ratios are 2.18 (0.00107/0.00049) and 2.68 (0.00918/0.00342), respectively. Combining the nucleotide diversity of noncoding and synonymous sites for each gene in each species we find the bcd/zen ratios at silent sites to be 1.08 for D. melanogaster and 0.89 for D. simulans, very close to the theoretical expectation of 1.0. In contrast, for nonsynonymous sites in the coding regions, the ratios are somewhat in excess of 2, the theoretically predicted value.
Discussion
The redesign of anterior patterning during embryogenesis is a major developmental innovation, involving both positive selection and the overcoming of developmental constraints. A previous study of sequence variation in bcd from a Zimbabwe population of D. melanogaster [18] found a significant excess of replacement polymorphisms, indicative of reduced purifying selection, as well as “extensive” haplotype structure consistent with positive selection and linkage disequilibrium. The authors concluded that the selective history of bcd is “Complex, involving both relaxation of purifying selection in some parts of the protein and the action of selection in other parts of the gene region” [18]. We do not find evidence of haplotype structure; however, this absence is likely the result of sampling single flies from geographically diverse populations instead of the more intensive sampling of a single population, as in the study of Zimbabwe D. melanogaster.
Our comparison of bcd and zen sequence diversity suggests that the variation necessary for bcd to overcome developmental constraint is not due solely to redundancy attending gene duplication [13], but is more consistent with the relaxed constraint that accompanies sex-limited expression. Furthermore, our results indicate that most of the difference in bcd and zen variation can be accounted for by relaxed constraint without invoking differences in s. Although we find no extensive haplotype structure in our analysis, its existence in prior studies may be due to occasional selective sweeps that fix the high standing diversity of bcd within populations.
In the origin of evolutionary novelty, the balance between positive selection and relaxed constraint may be different for maternally expressed genes than for paternally expressed genes [19,20]. Although similarly expressed in only half of the members of a population, sexual selection acting on males through paternally expressed genes can be much stronger than the viability selection acting on embryos as a result of maternal-effect genes [21]. Maternal-effect genes are more similar in their evolutionary genetics to aging genes expressed late in life. Alleles with detrimental effects late in life accumulate in a population because selection against them is weaker than it is for alleles with deleterious effects of similar magnitude expressed earlier [22,23]. However, it is difficult to make a quantitative prediction regarding the elevated heterozygosity of aging genes because the degree of relaxed selective constraint depends upon a host of demographic variables. For maternal effect genes, we expect a 2-fold increase in heterozygosity at mutation-selection balance, because they are “hidden” from natural selection in half the population, i.e., in males, where they are not expressed. The levels of bcd and zen sequence diversity we observed in D. melanogaster and D. simulans support this theoretical prediction.
Materials and Methods
We purchased ten D. melanogaster and nine D. simulans wild-type strains from the Tucson Drosophila Species Stock Center (http://stockcenter.arl.arizona.edu/) (Table S1). We selected strains from a broad geographic range to sample genetic variation across the distribution of each species. From each strain, a single individual was randomly selected for DNA extraction and analyses of bcd and zen sequence diversity.
Genomic DNA was extracted from single flies using the Puregene tissue kit (Gentra Systems, Minneapolis, Minnesota, United States). Oligonucleotides for PCR amplification and direct sequencing were designed using the FastPCR program [24] from previously published D. melanogaster and D. simulans bcd and zen sequences. To increase primer specificity and avoid amplifying other genes, candidate primers were searched against the D. melanogaster genome to remove primers that may amplify other genes. Further, in silico PCR simulations of candidate primer pairs against all members of the Antennapedia Complex were conducted in FastPCR [24] to remove primers that may amplify related sequences.
The bcd primer pairs amplified a ~4-kb region that included ~60 bp of the 5′ and ~1 kb of the 3′ flanking regions. The zen primer pairs amplified an ~1.3-kb region that included ~30 bp of the 5′ and ~100 bp of the 3′ flanking regions. PCR products were purified using the QIAquick PCR Purification Kit (Qiagen, Valencia, California, United States). Forward and reverse strands were sequenced on an ABI 3730 DNA Analyzer (Applied Biosystems, Foster City, California, United States) using Big Dye Terminator v 3.1 chemistry (Applied Biosystems). Oligonucleotides used for PCR amplifications were also used for sequencing the D. melanogaster and D. simulans bcd and zen genes. The resulting sequences are deposited in GenBank (http://www.ncbi.nlm.nih.gov/) with the following accession numbers: D. melanogaster bcd (DQ114428–DQ114437), D. melanogaster zen (DQ114438–DQ114447), D. simulans bcd (DQ114448–DQ114456), and D. simulans zen (DQ114457–DQ114465).
We constructed contigs for each gene using CodonCode Aligner v. 1.3.4 (CodonCode, Dedham, Massachusetts, United States). All base calls in each contig were based upon at least one forward and reverse sequence pair, with most bases covered by two forward and reverse pairs. Contigs of each gene were aligned using BioEdit v. 7.0.4.1 [25]. We verified each alignment and variable site manually. Coding regions were defined for bcd following Baines et al. [18]; i.e., only the major transcript was considered coding. For our bcd sequences, we defined the coding regions of D. melanogaster as positions 68–232, 792–867, 923–1,864, and 2,377–2,663. For our D. simulans bcd, the coding regions were defined as positions 68–232, 776–851, 907–1,848, and 2,235–2,521. Coding regions for the zen sequences were identified by comparison to the D. melanogaster reference sequence. In our D. melanogaster zen sequences, the coding region was defined as positions 27–113 and 178–1,152. For D. simulans zen, the coding region was defined as positions 27–113 and 178–1,146.
Nucleotide diversity, π, is the average number of nucleotide differences per site between two randomly chosen sequences. It was calculated according to Nei [26] using DnaSP 4.10 [27]. To assess neutrality, we estimated Tajima's D [17]. Then, using the observed Tajima's D and the average number of nucleotide differences per sequence, we computed 95% confidence intervals by coalescent simulation in DnaSP.
Supporting Information
Table S1 Stock Center Accession Numbers and Collection Locations of Drosophila Populations
(41 KB DOC)
Click here for additional data file.
Accession Numbers
The GenBank (http://www.ncbi.nlm.nih.gov/) accession numbers discussed in this paper are D. melanogaster bcd (AF466621–AF466645), D. simulans bcd (AF465792), and D. melanogaster zen reference sequence (NT_033777). D. melanogaster GeneIDs are bcd (40830) and zen (40828).
We thank J. Baines for providing initial bcd and zen primer sequences. Thanks are also given to C. Hughes, T. Linksvayer, and N. Johnson for insightful discussions. MSB and JPD were supported by the National Science Foundation under the Integrated Graduate Education and Research Traineeship in Evolution, Development, and Genomics grant 9972830. MJW and JPD were supported by the National Institutes of Health under grant GM065414-01A.
Competing interests. The authors have declared that no competing interests exist.
Author contributions. MSB, JPD, and MJW conceived and designed the experiments, analyzed the data, and wrote the paper. MSB performed the experiments. MJW contributed reagents/materials/analysis tools.
==== Refs
References
Provine WB 1985 The R. A. Fisher – Sewall Wright controversy and its influence upon modern evolutionary biology Dawkins R Ridley M Oxford Surveys in Evolutionary Biology Oxford Oxford University Press 197 219
Seward AC 1909 Darwin and modern science: Essays in the commemoration of the centenary of the birth of Charles Darwin and of the fiftieth anniversary of the publication of the Origin of Species Cambridge Cambridge University Press 595 p.
Muller GB Wagner GP 1991 Novelty in evolution—Restructuring the concept Annu Rev Ecol Syst 22 229 256
Wagner GP Muller GB 2002 Evolutionary innovations overcome ancestral constraints: A re-examination of character evolution in male sepsid flies (Diptera: Sepsidae) Evol Dev 4 1 6 11868655
Eberhard WG 2001 Multiple origins of a major novelty: Moveable abdominal lobes in male sepsid flies (Diptera: epsidae), and the question of developmental constraints Evol Dev 3 206 222 11440252
Stauber M Taubert H Schmidt-Ott U 2000 Function of bicoid and hunchback homologs in the basal cyclorrhaphan fly Megaselia (Phoridae) Proc Natl Acad Sci U S A 97 10844 10849 10995461
Stauber M Jackle H Schmidt-Ott U 1999 The anterior determinant bicoid of Drosophila is a derived Hox class 3 gene Proc Natl Acad Sci U S A 96 3786 3789 10097115
Stauber M Prell A Schmidt-Ott U 2002 A single Hox3 gene with composite bicoid and zerknüllt expression characteristics in non-Cyclorrhaphan flies Proc Natl Acad Sci U S A 99 274 279 11773616
Falciani F Hausdorf B Schroeder R Akam M Tautz D 1996 Class 3 Hox genes in insects and the origin of zen
Proc Natl Acad Sci U S A 93 8479 8484 8710895
Hughes CL Kaufman TC 2002 Exploring the myriapod body plan: Expression patterns of the ten Hox genes in a centipede Development 129 1225 1238 11874918
Dearden P Grbic M Falciani F Akam M 2000 Maternal expression and early zygotic regulation of the Hox3 /zen gene in the grasshopper Schistocerca gregaria
Evol Dev 2 261 270 11252555
Schröder R Sander K 1993 A comparison of transplantable bicoid activity and partial bicoid homeobox sequences in several Drosophila and blowfly species (Calliphoridae) Roux's Arch Dev Biol 203 34 43
Demuth JP, Wade M J 2005 Maternal expression increases the rate of bicoid evolution by relaxing selective constraint Genetica In press.
Whitlock MC Wade MJ 1995 Speciation–Founder events and their effects on X-linked and autosomal genes Am Nat 145 676 685
Wade MJ 2000 The evolutionary genetics of maternal effects Mousseau TA Fox CW Maternal effects as adaptations Oxford Oxford University Press 5 21
Tajima F 1983 Evolutionary relationship of DNA sequences in finite populations Genetics 105 437 460 6628982
Tajima F 1989 Statistical method for testing the neutral mutation hypothesis Genetics 123 585 595 2513255
Baines JF Chen Y Das A Stephan W 2002 DNA sequence variation at a duplicated gene: Excess of replacement polymorphism and extensive haplotype structure in the Drosophila melanogaster bicoid region Mol Biol Evol 19 989 998 12082119
Metz EC Robles-Sikisaka R Vacquier VD 1998 Nonsynonymous substitution in abalone sperm fertilization genes exceeds substitution in introns and mitochondrial DNA Proc Natl Acad Sci U S A 95 10676 10681 9724763
Wyckoff G Wang W Wu C-I 2000 Rapid evolution of male reproductive genes in the descent of man Nature 403 394 309
Shuster SM Wade MJ 2003 Mating systems and strategies Princeton Princeton University Press 469 p.
Charlesworth B 1994 Evolution in age-structured populations Cambridge Cambridge University Press 320 p.
Hughes KA Alipaz JA Drnevich JM Reynolds RM 2002 A test of evolutionary theories of aging Proc Natl Acad Sci U S A 99 14286 14291 12386342
Kalendar R 2005 FastPCR: A PCR primer design and repeat sequence searching software with additional tools for the manipulation and analysis of DNA and protein. Available at: http://www.biocenter.helsinki.fi/bi/programs/fastpcr.htm Accessed 04 November 2004
Hall TA 1999 BioEdit: A user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT Nucl Acids Symp Ser 41 95 98
Nei M 1987 Molecular evolutionary genetics New York Columbia University Press 512 p.
Rozas J Sánchez-DelBarrio JC Messeguer X Rozas R 2003 DnaSP, DNA polymorphism analyses by the coalescent and other methods Bioinformatics 19 2496 2497 14668244
|
16299585
|
PMC1283158
|
CC BY
|
2021-01-05 08:00:24
|
no
|
PLoS Genet. 2005 Nov 11; 1(5):e57
|
utf-8
|
PLoS Genet
| 2,005 |
10.1371/journal.pgen.0010057
|
oa_comm
|
==== Front
PLoS GenetPLoS GenetpgenplgeplosgenPLoS Genetics1553-73901553-7404Public Library of Science San Francisco, USA 1629235510.1371/journal.pgen.0010060plge-01-05-03Research ArticleGenome-Wide Association Mapping in Arabidopsis Identifies Previously Known Flowering Time and Pathogen Resistance Genes Association Mapping in Arabidopsis thalianaAranzana María José 1Kim Sung 1Zhao Keyan 1Bakker Erica 2Horton Matthew 2Jakob Katrin 2Lister Clare 3Molitor John 4Shindo Chikako 3Tang Chunlao 1Toomajian Christopher 1Traw Brian 2Zheng Honggang 1Bergelson Joy 2Dean Caroline 3Marjoram Paul 4Nordborg Magnus 1*1 Molecular and Computational Biology, University of Southern California, Los Angeles, California, United States of America
2 Department of Ecology and Evolution, University of Chicago, Chicago, Illinois, United States of America
3 Cell and Developmental Biology, John Innes Centre, Norwich, United Kingdom
4 Department of Preventive Medicine, Keck School of Medicine, University of Southern California, Los Angeles, California, United States of America
Doebley John EditorUniversity of Wisconsin, United States of America* To whom correspondence should be addressed. E-mail: [email protected] 2005 11 11 2005 10 10 2005 1 5 e6029 8 2005 10 10 2005 Copyright: © 2005 Aranzana et al.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.There is currently tremendous interest in the possibility of using genome-wide association mapping to identify genes responsible for natural variation, particularly for human disease susceptibility. The model plant Arabidopsis thaliana is in many ways an ideal candidate for such studies, because it is a highly selfing hermaphrodite. As a result, the species largely exists as a collection of naturally occurring inbred lines, or accessions, which can be genotyped once and phenotyped repeatedly. Furthermore, linkage disequilibrium in such a species will be much more extensive than in a comparable outcrossing species. We tested the feasibility of genome-wide association mapping in A. thaliana by searching for associations with flowering time and pathogen resistance in a sample of 95 accessions for which genome-wide polymorphism data were available. In spite of an extremely high rate of false positives due to population structure, we were able to identify known major genes for all phenotypes tested, thus demonstrating the potential of genome-wide association mapping in A. thaliana and other species with similar patterns of variation. The rate of false positives differed strongly between traits, with more clinal traits showing the highest rate. However, the false positive rates were always substantial regardless of the trait, highlighting the necessity of an appropriate genomic control in association studies.
Synopsis
There is currently tremendous interest in using association mapping to find the genes responsible for natural variation, particularly for human disease. In association mapping, researchers seek to identify regions of the genome where individuals that are phenotypically similar (for example, they all have the same disease) are also unusually closely related. A potentially serious problem is that spurious correlations may arise if the population is structured so that members of a subgroup tend to be much more closely related. Because few genome-wide association studies have been carried out, it is not yet known how important this problem will be in practice.
In one of the first genome-wide association studies to date, this paper considers the model plant Arabidopsis thaliana. A very large number of spurious genotype–phenotype correlations are found, especially for traits that vary geographically. For example, plants from northern latitudes flower later; however, in addition to sharing genetic variants that make them flower late, they also tend to share variants across the genome, making it difficult to determine which genes are responsible for flowering. This notwithstanding, several previously known genes were successfully identified in this study, and the researchers are optimistic about the prospects for association mapping in this species.
Citation:Aranzana MJ, Kim S, Zhao K, Bakker E, Horton M, et al. (2005) Genome-wide association mapping in Arabidopsis identifies previously known flowering time and pathogen resistance genes. PLoS Genet 1(5): e60.
==== Body
Introduction
One of the main challenges of modern biology is achieving a better understanding of the molecular genetic basis for naturally occurring phenotypic variation. Primarily because of rapidly decreasing genotyping costs, genome-wide association mapping (also known as linkage disequilibrium mapping) has emerged as a very promising tool for accomplishing this. The basic idea is simple: rather than looking for marker–trait associations in a population with known relationships (such as the members of a pedigree, or the offspring of an experimental cross), we look for associations in the general population of “unrelated” individuals [1]. Because unrelated individuals are, of course, always related at some distance, phenotypically similar individuals may be similar because they share alleles inherited identical by descent, alleles that will be surrounded by short ancestral marker haplotypes that can be identified in genome-wide scans. Association mapping has two main advantages over traditional linkage mapping methods. First, the fact that no pedigrees or crosses are required often makes it easier to collect data. Second, because the extent of haplotype sharing between unrelated individuals reflects the action of recombination over very large numbers of generations, association mapping has several orders of magnitude higher resolution than linkage mapping.
The drawbacks of association mapping stem from the fact that it is not a controlled experiment. Power is unpredictable, partly because the decay of linkage disequilibrium is noisy, and partly because the genetic architecture of the trait is unknown (the latter is always a problem in mapping complex traits, but it is likely to be worse in association mapping because genetic heterogeneity is not limited by a small number of founders) [1–3]. The false positive rate is similarly difficult to predict: it is well known that population structure can cause strong spurious correlations [4]. The severity of these problems is not known, because few (if any) genome-wide association studies have been carried out to date.
Highly selfing organisms, like Arabidopsis thaliana, are ideal candidates for association mapping. First, they largely exist as collections of naturally occurring inbred accessions, which can be genotyped once and phenotyped repeatedly, for the same phenotype (to reduce environmental noise) or different phenotypes (allowing “in silico mapping” [5]). Second, inbreeding results in a pattern of polymorphism characterized by extensive haplotype structure, which should be well suited for association mapping [6].
Preliminary studies indicated that linkage disequilibrium in A. thaliana decayed over 50–250 kb [7]. Based on these results, a genome-wide polymorphism survey in which short (500–600 bp) fragments were resequenced approximately every 100 kb in 95 individuals was carried out. Analysis of these data resulted in two findings of direct relevance to association mapping [8]. First, linkage disequilibrium appears to decay faster than predicted, within 50 kb. This means that the available polymorphism data are not dense enough for a genome-wide association study. Second, A. thaliana exhibits substantial population structure. This means that the sample is less ideal for association mapping for the reasons alluded to above.
In spite of these problems, we have used the data to investigate the feasibility of genome-wide association mapping in A. thaliana. We considered four phenotypes for which major loci are known (the vernalization response locus FRI [9] and the three pathogen resistance loci, Rpm1, Rps5, and Rps2 [10–12]), and asked whether these loci could have been identified using genome-wide association mapping given a small, heavily structured sample such as the one available to us. We found that, in spite of an extremely high false positive rate, we were able to identify all of them, thus demonstrating the potential of genome-wide association studies in A. thaliana, and other species with similar patterns of variation.
Results
Genome-Wide Associations and the False Positive Rate
The data used in this study are summarized in Figure 1, which shows genotype and associated phenotype for four genes, for each of the 95 accessions, plotted against a tree representing the genome-wide relationships among the accessions (from [8]). The tree illustrates that accessions whose origins are geographically close tend to be more closely related, and it is clear by inspection that the phenotypes are not randomly distributed with respect to this tree. Flowering time was particularly strongly correlated with geographic origins, as would be expected for a trait that is likely to be under clinal selection. It follows that the standard null hypothesis in association mapping, independence between marker genotypes and traits, is false in a genome-wide sense. In other words, we should expect an elevated false positive rate, and this is precisely what we found. As illustrated in Figure 2, the distribution of p-values across the genome was heavily skewed towards zero, with flowering time showing the strongest deviation from the null expectation. To give some idea of the magnitude of the deviation, a naive application of a Kruskal–Wallis nonparametric test of association between flowering time and each of the approximately 850 sequenced loci (treating haplotypes as alleles) yielded 7% significant tests at the (nominal) 0.1% level, 18% significant tests at the 1% level, and 33% at the 5% level. The (nominally) significantly associated loci were distributed throughout the genome (Figure 3) and are clearly not all true positives. Indeed, given that we expect our study to have low power (due to both insufficient marker density and genetic heterogeneity), it is possible that none, except the previously known loci, are true positives.
Figure 1 Summary of the Data Used in the Study
The columns on the left give the genotype and associated phenotype for four loci, for each of the 95 accessions. The four loci are the flowering time locus FRI (+, wild-type; 1, Ler null allele; 2, Col null allele [9]), for which the associated phenotype is flowering time in long-day conditions without vernalization (late flowering is indicated by height and color of bar), and the three pathogen resistance loci Rps5, Rpm1, and Rps2 (+, wild-type; −, null allele [10,11,12]), for which the associated phenotypes are hypersensitive response to the appropriate bacterial avr gene (red indicates resistance, black indicates susceptibility, and missing data are indicated by missing bar). The tree on the right illustrates the genetic relationships between the accessions [8]. It is clear that phenotypes and genotypes are correlated, genome-wide.
Figure 2 The Genome-Wide Distribution of p-Values under Different Scenarios
(A) Cumulative distribution of p-values for association tests across approximately 850 loci. The sequenced haplotypes at each locus were treated as alleles (after eliminating singleton polymorphisms), and the significance of genotype–phenotype associations was tested using Kruskal–Wallis tests in the case of flowering time (a continuous trait), and using χ2 tests in the case of resistance (a binary trait). Under the null hypothesis of no association, the cumulative distribution should be a straight line: the observed distributions are all heavily skewed towards zero.
(B) The cumulative distribution of p-values for association with pathogen resistance, with and without correction for population structure using the program STRAT [13]. The false positive rate is decreased for avrPph3, but is unaffected for the other two phenotypes.
(C) The cumulative distribution of p-values for association with flowering time, with and without correction for population structure. ANOVA was used instead of the nonparametric Kruskal–Wallis test to make it possible to use population structure as cofactor (cf. [14]). The distribution for ANOVA with accessions from Finland and northern Sweden removed is also shown (“ANOVA − northern”). The false positive rate is decreased using both approaches.
Figure 3 Genome-Wide Scans for Association with Flowering Time and Pathogen Resistance
For flowering time (A), four different statistical methods were used (described in Materials and Methods): Voronoi focusing on “late” alleles (magenta line), Voronoi focusing on “early” alleles (blue line), CLASS (green line), and fragment-based Kruskal–Wallis tests (red line; see also Figure 2). For pathogen resistance (avrRpm1 [B], avrRpt2 [C], and avrPph3 [D]), only the last two tests were used. Higher peaks indicate stronger association (the y-axes are proportional to the negative log p-values, but have been normalized to the highest value within each test). The dotted lines correspond to the 95% percentile and are mainly intended to facilitate comparison between figures. Yellow vertical lines indicate the positions of the appropriate candidate loci. Peaks occur at these loci for all methods, but are otherwise distributed throughout the genome.
We attempted to decrease the false positive rate by taking population structure into account using so-called structured association, in which one uses genome-wide markers to infer population structure, and then carries out association tests conditional on the inferred structure [13,14]. For the pathogen resistance phenotypes, structured association reduced, but did not eliminate, the elevated positive rate for the most biased of the phenotypes (response to avrPph3); it had no effect on the other two rates (see Figure 2B). Similarly, the false positive rate for flowering time was strongly reduced, but remained extremely elevated relative to null expectations (Figure 2C). It is clear from Figure 1 that, at least for flowering time, much of the elevated false positive rate is due to the Swedish and Finnish accessions, which are genetically distinct and phenotypically extreme. Indeed, removing these accessions from the analysis reduced the false positive rate as much as using structured association (Figure 2C).
Mapping of Known Loci for Flowering Time and Pathogen Resistance
In spite of the high false positive rate, the four known loci were detectable in genome-wide scans (Figure 3). For the three pathogen resistance phenotypes the strongest association was found inside the appropriate R gene regardless of association method used. For flowering time, strong associations were evident in multiple locations throughout the genome, but associations in the FRI region were invariably among the ten most significant. Furthermore, FRI could readily be distinguished as true positive by clustering associations on the basis of which accessions were part of each association. Our rationale was that false positives due to population structure are expected to reoccur across the genome. This is precisely what we saw. Our haplotype-based association statistics identified loci for which clusters of phenotypically similar accessions exhibited excessive haplotype sharing. Figure 4 shows the result of clustering these clusters based on similarity in membership. We found that the vast majority of all significant associations were due to haplotype sharing among accessions from Finland and northern Sweden, sometimes with North American accessions also included. This type of association is thus found across the genome, and while nominally significant, is not significant in a genomic sense. Note that this does not mean that all these associations are false positives, but it does mean that most of them are. The very late flowering phenotype of the Finnish and northern Swedish accessions does have a genetic basis: we have identified a list of candidates, but we have no way of telling which (if any) of them is true.
Figure 4 Haplotypes Significantly Associated with Flowering Time Clustered by Haplotype Membership
To help determine which associations were real and which were due to population structure, the most significantly associated haplotypes (based on fragment-wise Kruskal–Wallis; see Materials and Methods) were clustered based on similarity in the list of accessions that carry each haplotype.
(A) The tree shows the resulting cluster with tips colored according to average flowering time among the accessions that carry the haplotype corresponding to each tip (the scale is given on the right along with a histogram showing the distribution of flowering time across the 95 accessions).
(B) The matrix shows the membership list for each haplotype. Each column corresponds to the haplotype (tip) in the tree above it; accessions highlighted in red carry the haplotype significantly associated with flowering time. The tree thus illustrates the clustering of the columns of the matrix: clustering was done based on pairwise distance as measured by the absolute value of the correlation in membership between columns. Phenotypes of the accessions are given on the right, and the rows of the matrix (i.e., the accessions) have been clustered based on pairwise Hamming distance. It is evident that most of the significant haplotypes, regardless of position in the genome, share similar membership lists that include the accessions from Finland and northern Sweden. On the other hand, the clusters corresponding to the known major alleles of FRI are unique, indicating that these are indeed true positives.
Figure 4 also identifies clusters with the property of being unique across the genome. In a hierarchical clustering, these would represent the deepest nodes because they are dissimilar from other clusters. Among the small number of “unique” clusters we identified one that corresponds to haplotype sharing among accessions carrying the Ler loss-of-function allele at FRI, and one that corresponds to the Col loss-of-function allele at the same locus [9]. These associations thus have the property that, in addition to being (nominally) significant, they are not found repeatedly across the genome. They are therefore more likely to be true positives.
The above analyses were intended to demonstrate that the signal of genotype–phenotype association for these four major loci would have been sufficient for genome-wide association mapping even in the small, heavily structured sample used by Nordborg et al. [8]. We have not addressed the other main aspect of power in association mapping, namely, the extent of linkage disequilibrium and what it implies about the marker density required for genome-wide scans. As mentioned in the Introduction, the marker density in the data of Nordborg et al. [8]—one resequenced fragment every 100 kb—is insufficient to cover the genome. The results above were based on denser marker coverage around the four loci, including markers within each target gene. As it turns out, we would have detected FRI and Rpm1 without adding additional markers, the former because (as we shall see below) the original marker coverage was sufficient to detect FRI, the latter because of luck. However, the denser marker coverage around all four loci allowed us to determine the required marker density by thinning the markers and noting when the signal disappeared. Figure 5 shows the result of successively eliminating resequenced fragments so that no markers were within 10, 25, 50, and 100 kb of the target locus. The difference between FRI and the three R genes is striking: while the former was readily picked up with the lowest marker density (corresponding to the density in the genome-wide data), the latter were only picked up with 10-kb spacing. When markers within 25 kb were eliminated, the association signal for the R genes was typically lost.
Figure 5 The Strength of Association (Using CLASS) around the Four Candidate Loci for Various Marker Densities
For each locus (FRI [A], Rpm1 [B], Rps2 [C], and Rps5 [D]), the bottom panel shows the pattern of association using all available fragment markers around the locus (the position of which is given by a grey vertical line), and the panels above show the effect of successively reducing the marker density so that no markers are within 10, 25, 50, and 100 kb (FRI only) of the causative polymorphisms. The dotted grey line represents the 95th percentile of all associations across the genome. Because we used an association statistic that utilizes the pattern of haplotype sharing across multiple fragments, the relative significance of any particular fragment may change depending on the presence or absence of other fragments. The FRI region (A) remains strongly associated with flowering time even for the lowest marker density, while the signal of association around the R genes (B–D) disappears as one goes from 10- to 25-kb spacing.
Discussion
Genome-Wide Association Mapping and Population Structure
Our results present a striking demonstration of the potential effect of population structure in causing an elevated false positive rate in association mapping. As genome-wide association studies in humans are becoming increasingly feasible, the seriousness of this problem has been the subject of considerable debate [15–19]. In this context, our study is roughly equivalent to a genome-wide scan for association with skin color using a world-wide sample of humans. Most human association mapping studies are likely to be case–control studies, which, given a judiciously chosen control, should be less prone to false positives [17].
Nonetheless, more studies like ours are likely to be carried out, in humans as well as in other organisms, and it seems likely that population structure will then be a problem. The extent of the problem will of course depend on the extent to which the sample is structured, but it will also depend on the phenotype. Traits that are strongly correlated with population structure will display a more highly elevated rate of false positives. In the present case, flowering time, which is likely involved in local adaptation [20,21], shows a more highly elevated rate than pathogen resistance, variation for which appears to be maintained by frequency-dependent balancing selection [10–12]. It should be noted, however, that differences between the resistance phenotypes were also found: the false positive rate for avrPph3 is more highly elevated than for the other resistance-related rates (see Figure 2). Why this should be the case is not clear, but might tell us something about the ecology of the pathogens responsible for maintaining polymorphism at these loci.
Several methods for dealing with false positives due to population structure have been proposed. The best known are “genomic control” [22] and “structured association” [13]. We found that structured association based on the approach of Pritchard et al. [13] and Thornsberry et al. [14] did not successfully correct the elevated false positive rate in our sample. This should not be surprising. The model underlying the approach of Pritchard et al. [13] is one of admixture between a small number of homogeneous, randomly mating populations. While this may be a reasonable approximation for many human samples, it is clearly not valid for our sample of A. thaliana, which shows all signs of isolation by distance [8].
Genomic control [22] is an alternative approach in which genome-wide markers are used to estimate the effect of population structure on association statistics and correct these statistics to achieve valid significance levels. We did not try this approach for several reasons. First, it, too, is based on a simple model of population structure. Second, the approach has only been developed for relatively simple contingency table statistics, and it is not clear how it should be implemented for the haplotype-based methods used here. Third, it is clear from our FRI results that genomic control would lack power. Association with FRI is not necessarily stronger than the false positives due to structure, and any approach that eliminated the latter based on the strength of association would also eliminate the former. In contrast, Figure 4 suggests that methods that simultaneously infer the structure and the associations should be able to separate true from false positives.
It is clear that more work is needed in this area. Indeed, given the difficulty of modeling population histories, it may be fruitful to abandon the notion of “population structure” (with its implication that unstructured populations actually exist), and instead view all population samples as members of a gigantic, unknowable pedigree. Models appropriate for handling such data have been developed in the animal breeding community [23], and can be extended to genome-wide association mapping [24,25].
The Prospect for Genome-Wide Association Mapping in A. thaliana
We have demonstrated that FRI, Rpm1, Rps2, and Rps5 could have been detected using genome-wide association mapping even in the small and heavily structured sample used by Nordborg et al. [8]. It should be emphasized that these are genes of major effect: the two loss-of-function alleles at FRI account for 13% of the variation in flowering time in our study, and correlation between being susceptible and carrying the known susceptibility allele is 0.66, 0.77, and 0.62 for Rpm1, Rps2, and Rps5, respectively. To map genes of more subtle effect, a much larger sample is surely needed. Furthermore, since power in association mapping is determined both by the effects of alleles and by their frequencies [3,26], the structure of the sample matters greatly. In addition to elevating the false positive rate, the presence of population structure may increase genetic heterogeneity—avoiding this problem is one of the main arguments for the use of population isolates in human genetics [27]. Whether genetic heterogeneity is a problem or not depends on the genetic architecture of the trait, which is of course unknown a priori.
In addition to a different sample, it is clear that a denser marker map than the one generated by Nordborg et al. [8] is needed. Although we were able to map FRI using 100-kb marker spacing, it is now clear that linkage disequilibrium around this gene is unusually extensive, probably because of a combination of local adaptation and recent selective sweeps (as was suggested by earlier studies [7]). On the other hand, the extent of linkage disequilibrium surrounding the R genes is likely to be smaller than usual because variation at these loci is due to ancient polymorphism maintained by balancing selection [10,11]. The observation that we can map these genes using linkage disequilibrium with markers 10 kb away suggests that a marker spacing of roughly 20 kb (which guarantees at least one marker within 10 kb of a causative polymorphism) would provide reasonable power. This implies that on the order of 6,000 single nucleotide polymorphisms (SNPs) chosen to be maximally informative about the local haplotype structure (so-called tag-SNPs [28,29]) might be sufficient for genome-wide association mapping in A. thaliana. Needless to say, the marker spacing required will vary across the genome depending on the local haplotype structure, and also depends on the sample. Further studies to investigate the required density are underway.
Materials and Methods
Plant material.
The accessions used are described in [8].
Sequencing and genotyping.
We used the resequencing data of Nordborg et al. [8], plus additional fragments resequenced around the four loci. Genotyping for the loss-of-function deletion alleles at FRI, Rpm1, and Rps5 was done using PCR assays as previously described [10,11,21]. Genotyping at Rps2 (not a deletion polymorphism) was done by sequencing the entire leucine-rich repeat region and comparing the results with those of [12]. All data are available as Datasets S1 and S2.
Measuring flowering time.
Flowering time was measured in days using plants grown under long-day conditions (16 h light, 8 h dark) at a constant temperature of 18 °C. Measurements were generally taken for six plants per accession, and the average used in the analysis. The experiment was stopped at 200 d, and accessions that had not flowered at that point were assigned a value of 200. The flowering time data are available as Dataset S1.
Measuring pathogen resistance.
Seedlings of each accession were germinated in flats containing a 1:1 mixture of Premier Pro-Mix and MetroMix (Premier Horticulture, Red Hill, Pennsylvania, United States). Flats were first placed at 4 °C for 7 d to promote germination, then placed in a growth room at 20 °C with short-day lighting (12 h light, 12 h dark). On the 23rd day of growth, two leaves per plant were inoculated with 0.1 ml of 108 cfu/ml bacteria in 10 mM MgSO4 buffer using a blunt-tipped syringe [30]. Leaf collapse was scored at 20 h and again at 24 h after inoculation. A positive score at either time point was deemed a hypersensitive response. The four avr genes were tested using the following transformed strains of Pseudomonas syringae: Pst DC3000::avrPphB [31], Pst DC3000::avrRpm1 [32], Pst DC3000::avrB (from J. Greenberg, University of Chicago), and Pst DC3000::avrRpt2 [33]. As a negative control, P. syringae DC3000 without the avr genes was also tested [33]. Each of the five strains was tested in a separate experiment consisting of six replicates of each of the 95 accessions, planted two per cell, for a total of 576 plants and six flats in each test. Accessions were considered to exhibit a hypersensitive response if at least eight of the 12 replicate leaves exhibited collapse. Accessions were considered to lack the hypersensitive response if at least eight of the 12 replicate leaves exhibited no leaf collapse. Accessions that exhibited ambiguous responses to a strain were excluded from further analysis. The negative control strain, P. syringae DC3000 without the added avr genes, caused no hypersensitive response in any of the lines. Results for avrPphB were almost identical to those for avrRpm1, and are not shown. The resistance data are available as Dataset S1.
Association mapping methods.
There has been considerable debate over how much power is gained by using haplotype-based instead of single SNP methods. In organisms where linkage disequilibrium decays rapidly (e.g., Drosophila melanogaster [26]), or where haplotypes have to be inferred (e.g., humans [34,35]), this is indeed a relevant question. In the present case, the polymorphism data come in the form of short haplotypes within which linkage disequilibrium is nearly complete, and it is thus natural to utilize haplotype-based methods. Indeed, we have found that methods incorporating longer-range disequilibrium sometimes perform substantially better [40]. We utilized three different methods here.
Single-fragment haplotypes.
After removing singleton polymorphisms, each resequenced fragment was treated as a multi-allelic marker locus with haplotypes corresponding to alleles. Haplotypes with frequency lower than 5% were grouped. Phenotypic associations were then tested using either a Kruskal–Wallis test in the case of flowering time (a continuous trait), or χ2 tests in the case of resistance (a binary trait).
CLASS (cladistic association).
We developed a simple clustering method similar in spirit to what has been proposed by several other researchers [36–38]. For each resequenced fragment, we first generated a similarity matrix using the extent of pairwise haplotype sharing between all pairs of accessions. We then clustered the accessions using a standard hierarchical clustering algorithm (we used neighbor joining), and heuristically searched for clades of accessions that were strongly associated with the phenotype (using either Kruskal–Wallis or χ2 tests to evaluate the strength of association). Our algorithm found clades using the following steps. (1) Search all clades and choose the one that gives the lowest p-value in a test with one degree of freedom. (2) Search the tree obtained by removing this clade for the clade that gives the lowest p-value in a test with three factors (and two degrees of freedom): the target clade, the clade identified in the previous step, and the remaining individuals. We repeated step 2, increasing the degrees of freedom by one each step, until the p-values no longer decreased.
Voronoi.
We utilized a slightly modified version of the spatial clustering algorithm described elsewhere [39] and that has previously been used to fine-map FRI [40]. To summarize, each haplotype cluster searched by Voronoi contains a prototypic haplotype to which all observed haplotypes are compared, with respect to a starting location, or center. The simple similarity measure used to compare the two haplotypes is the calculated shared length identical by state originating from the center. Standard Markov chain Monte Carlo techniques were used to identify parameters such as haplotype risks for each cluster, which could then associate a haplotype cluster to an observed phenotype.
We deviated from the original version of this algorithm by assigning haplotypes to a specific cluster in a probabilistic way rather than a deterministic fashion. At any given step of the Markov chain Monte Carlo algorithm, a randomly observed haplotype was selected as the prototypic haplotype. We then assigned haplotype hi to cluster cn according to the following probability:
where ssin is the normalized shared length between the hi haplotype and
cluster center haplotype. ssin is the ratio of the observed and the mean shared length at xc, where xc is the putative functional mutation location in cluster c.
Furthermore, rather than using the Bayes factor as a summary statistic, we used the posterior likelihood as our final statistic. We constructed the 95% confidence interval of the likelihood for each haplotype and considered a haplotype to be significant if the confidence interval did not contain zero. This procedure also allowed the distinction between positive and negative effects. For those significant haplotypes, if the confidence interval was above zero, we concluded a positive association to late flowering; otherwise, the haplotype was negatively associated with early flowering. The posterior likelihood distribution of the functional mutation associated with the significant haplotypes gave likelihood for both positive and negative effects.
Significance thresholds.
To generate the clustering in Figure 4, the 75 most significant fragments were selected, and, from among these, all haplotype clusters with a Bonferroni-corrected p-value less than 0.005 were selected. Note that the p-value for a fragment reflects all haplotypes observed for that fragment (the number of categories in the Kruskal–Wallis tests equals the number of haplotypes), whereas the p-value for a particular haplotype reflects the contribution of that haplotype only (two categories). These thresholds were chosen to yield an interpretable figure.
Correcting for population structure.
We attempted to decrease the false positive rate due to population structure using structured association, in which one looks for associations conditional on inferred population structure [13]. We used the population structure estimate from the program STRUCTURE [41], with K = 8 clusters, generated as described in [8].
For the binary pathogen resistance phenotypes, association analysis was then carried out using the program STRAT [13]. However, since STRAT only works with binary data, it could not be used with the quantitative flowering time phenotype. Thornsberry et al. [14] extended the structured association approach to quantitative phenotypes, but their method is restricted to binary (SNP) genotypes, and cannot be used with the haplotype data available to us. Instead, we used a simple modification, in which the cluster assignment produced by STRUCTURE (the Q matrix) was used as a cofactor in a standard ANOVA. Basically, we carried out a likelihood ratio test of two models: H
0 was FT ~ Q and H
1 was FT ~ as.factor(marker genotype) + Q. The p-values were based on the χ2 distribution of the likelihood ratio test statistic.
Supporting Information
Dataset S1 Genomic Alignments
(1.3 MB ZIP)
Click here for additional data file.
Dataset S2 Genotypes and Phenotypes
(3 KB CSV)
Click here for additional data file.
This work was mainly supported by National Science Foundation 2010 grant DEB-0115062 (JB and MN), a grant from the W. H. Keck Foundation (MN), National Institutes of Health (NIH) grant GM57994 (JB), and NIH Center of Excellence in Genomic Science grant P50 HG002790 (M. Waterman, PI). In addition, CT was supported by an NIH postdoctoral grant, BT was supported by a Dropkin Fellowship, and HZ was supported by a grant from the Fletcher Jones Foundation (Simon Tavaré, PI).
Competing interests. The authors have declared that no competing interests exist.
Author contributions. JB, CD, PM, and MN designed and directed the overall project. MJA, CL, CS, CT, HZ, CD, and MN generated the flowering time data. EB, MH, KJ, BT, and JB generated the resistance data. MJA, CT, CT, BT, and HJ generated the candidate locus polymorphism data. MJA, SK, KZ, JM, PM, and MN analyzed the data. SK, KZ, and MN wrote the paper.
A previous version of this article appeared as an Early Online Release on October 10, 2005 (DOI: 10.1371/journal.pgen.0010060.eor).
Abbreviations
SNPsingle nucleotide polymorphism
==== Refs
References
Nordborg M Tavaré S 2002 Linkage disequilibrium: What history has to tell us Trends Genet 18 83 90 11818140
Weiss KM Terwilliger JD 2000 How many diseases does it take to map a gene with SNPs? Nature Genet 26 151 157 11017069
Zondervan KT Cardon LR 2004 The complex interplay among factors that influence allelic association Nature Rev Genet 5 89 100 14735120
Lander ES Schork NJ 1994 Genetic dissection of complex traits Science 265 2037 2048 8091226
Grupe A Germer S Usuka J Aud D Belknap JK 2001 In silico mapping of complex disease-related traits in mice Science 292 1915 1918 11397946
Nordborg M 2000 Linkage disequilibrium, gene trees, and selfing: An ancestral recombination graph with partial self-fertilization Genetics 154 923 929 10655241
Nordborg M Borevitz JO Bergelson J Berry CC Chory J 2002 The extent of linkage disequilibrium in Arabidopsis thaliana
Nature Genet 30 190 193 11780140
Nordborg M Hu TT Ishino Y Jhaveri J Toomajian C 2005 The pattern of polymorphism in Arabidopsis thaliana
PLoS Biol 3 e196 15907155
Johanson U West J Lister C Michaels S Amasino R 2000 Molecular analysis of FRIGIDA , a major determinant of natural variation in Arabidopsis flowering time Science 290 344 347 11030654
Stahl EA Dwyer G Mauricio R Kreitman M Bergelson J 1999 Dynamics of disease resistance at the Rpm1 locus of Arabidopsis
Nature 400 667 671 10458161
Tian D Araki H Stahl E Bergelson J Kreitman M 2002 Signature of balancing selection in Arabidopsis
Proc Natl Acad Sci USA 99 11525 11530 12172007
Mauricio R Stahl EA Korves T Tian D Kreitman M 2003 Natural selection for polymorphism in the disease resistance gene Rps2 of Arabidopsis thaliana
Genetics 163 735 746 12618410
Pritchard JK Stephens M Rosenberg NA Donnelly P 2000 Association mapping in structured populations Am J Hum Genet 67 170 181 10827107
Thornsberry JM Goodman MM Doebley J Kresovich S Nielsen D 2001
Dwarf8 polymorphisms associate with variation in flowering time Nature Genet 28 286 289 11431702
Freedman ML Penney KL McDonald GJ Mignault AA Patterson N 2004 Assessing the impact of population stratification on genetic association studies Nature Genet 36 388 393 15052270
Marchini J Cardon LR Phillips MS Donnelly P 2004 The effects of human population structure on large genetic association studies Nature Genet 36 512 517 15052271
Hinds DA Stokowski RP Patil N Konvicka K Kershenobich D 2004 Matching strategies for genetic association studies in structured populations Amer J Hum Genet 74 317 325 14740319
Helgason A Yngvad'ottir B Hrafnkelsson B Gulcher JR Stef'ansson K 2005 An Icelandic example of the impact of population structure on association studies Nature Genet 37 90 95 15608637
Reiner AP Ziv E Lind DL Nievergelt CM Schork NJ 2005 Population structure, admixture, and aging-related phenotypes in African-American adults: The cardiovascular health study Amer J Hum Genet 76 463 477 15660291
Stinchcombe JR Weinig C Ungerer M Olsen KM Mays C 2004 A latitudinal cline in flowering time in Arabidopsis thaliana modulated by the flowering time gene FRIGIDA
Proc Natl Acad Sci USA 101 4712 4717 15070783
Shindo C Aranzana MJ Lister C Baxter C Nicholls C 2005 Role of FRIGIDA and FLC in determining variation in flowering time of Arabidopsis thaliana
Plant Physiol 138 1163 1173 15908596
Devlin B Roeder K 1999 Genomic control for association studies Biometrics 55 997 1004 11315092
Kennedy BW Quinton M van Arendonk JAM 1992 Estimation of effects of single genes on quantitative traits J Anim Sci 70 2000 2012 1644672
Schork NJ 2001 Genome partitioning and whole-genome analysis Adv Genet 42 299 322 11037329
Yu J Pressoir G Briggs WH Bi IV Yamasaki M 2005 A unified mixed-model method for association mapping accounting for multiple levels of relatedness Nature Genet in press.
Long AD Langley CH 1999 The power of association studies to detect the contribution of candidate genetic loci to variation in complex traits Genome Research 9 720 731 10447507
Wright AF Carothers AD Pirastu M 1999 Population choice in mapping genes for complex diseases Nature Genet 23 397 404 10581024
Johnson GCL Esposito L Barratt BJ Smith AN Genova JHGD 2001 Haplotype tagging for the identification of common disease genes Nature Genet 29 233 237 11586306
Patil N Berno AJ Hinds DA Barrett WA Doshi JM 2001 Blocks of limited haplotype diversity revealed by high-resolution scanning of human chromosome 21 Science 294 1719 1723 11721056
Jakob K Goss EM Araki H Van T Kreitman M 2002
Pseudomonas viridiflava and P. syringae — natural pathogens of Arabidopsis thaliana
Mol Plant Microbe Interact 15 1195 1203 12481991
Simonich MT Innes RW 1995 A disease resistance gene in arabidopsis with specificity for the avrPph3 gene of Pseudomonas syringae pv. phaseolicola
Mol Plant Microbe Interact 8 637 640 8589418
Tian D Traw MB Chen JQ Kreitman M Bergelson J 2003 Fitness costs of R -gene-mediated resistance in Arabidopsis thaliana
Nature 423 74 77 12721627
Whalen MC Innes RW Bent AF Staskawicz BJ 1991 Identification of Pseudomonas syringae pathogens of Arabidopsis and a bacterial locus determining avirulence on both Arabidopsis and soybean Plant Cell 3 49 59 1824334
Chapman JM Cooper JD Todd JA Clayton DG 2003 Detecting disease associations due to linkage disequilibrium using haplotype tags: A class of tests and the determinants of statistical power Human Heredity 56 18 31 14614235
Morris AP Whittaker JC Xu CF Hosking LK Balding DJ 2003 Multipoint linkagedisequilibrium mapping narrows location interval and identifies mutation heterogeneity Proc Natl Acad Sci USA 100 13442 13446 14597696
Durrant C Zondervan KT Cardon LR Hunt S Deloukas P 2004 Linkage disequilibrium mapping via cladistic analysis of single-nucleotide polymorphism haplotypes Amer J Hum Genet 75 35 43 15148658
Templeton AR Maxwell T Posada D Stengard JH Boerwinkle E 2005 Tree scanning: A method for using haplotype trees in phenotype/genotype association studies Genetics 169 441 453 15371364
Tzeng JY 2005 Evolutionary-based grouping of haplotypes in association analysis Genetic Epidemiology 28 220 231 15726584
Molitor J Marjoram P Thomas D 2003 Fine-scale mapping of disease genes with multiple mutations via spatial clustering techniques Amer J Hum Genet 73 1368 1384 14631555
Hagenblad J Tang C Molitor J Werner J Zhao K 2004 Haplotype structure and phenotypic associations in the chromosomal regions surrounding two Arabidopsis thaliana flowering time loci Genetics 168 1627 1638 15579712
Falush D Stephens M Pritchard JK 2003 Inference of population structure using multilocus genotype data: Linked loci and correlated allele frequencies Genetics 164 1567 1587 12930761
|
16292355
|
PMC1283159
|
CC BY
|
2021-01-05 07:59:37
|
no
|
PLoS Genet. 2005 Nov 11; 1(5):e60
|
utf-8
|
PLoS Genet
| 2,005 |
10.1371/journal.pgen.0010060
|
oa_comm
|
==== Front
PLoS PathogPLoS PathogppatplpaplospathPLoS Pathogens1553-73661553-7374Public Library of Science San Francisco, USA 1629235610.1371/journal.ppat.001001805-PLPA-RA-0107R1plpa-01-02-07Research ArticleHIV - AIDSVirologyVirusesEukaryotesHomo (human)In VitroThe Cell Cycle Independence of HIV Infections Is Not Determined by Known Karyophilic Viral Elements Chimeric HIV and Non-Dividing CellsYamashita Masahiro Emerman Michael *Division of Human Biology and Basic Sciences, Fred Hutchinson Cancer Research Center, Seattle, Washington, United States of AmericaMalim Michael EditorKing′s College London, United Kingdom* To whom correspondence should be addressed. E-mail: [email protected] 2005 11 11 2005 1 3 e1825 7 2005 19 9 2005 Copyright: © 2005 Yamashita and Emerman.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.Human immunodeficiency virus and other lentiviruses infect cells independent of cell cycle progression, but gammaretroviruses, such as the murine leukemia virus (MLV) require passage of cells through mitosis. This property is thought to be important for the ability of HIV to infect resting CD4+ T cells and terminally differentiated macrophages. Multiple and independent redundant nuclear localization signals encoded by HIV have been hypothesized to facilitate migration of viral genomes into the nucleus. The integrase (IN) protein of HIV is one of the HIV elements that targets to the nucleus; however, its role in nuclear entry of virus genomes has been difficult to describe because mutations in IN are pleiotropic. To investigate the importance of the HIV IN protein for infection of non-dividing cells, and to investigate whether or not IN was redundant with other viral signals for cell cycle-independent nuclear entry, we constructed an HIV-based chimeric virus in which the entire IN protein of HIV was replaced by that of MLV. This chimeric virus with a heterologous IN was infectious at a low level, and was able to integrate in an IN-dependent manner. Furthermore, this virus infected non-dividing cells as well as it infected dividing cells. Moreover, we used the chimeric HIV with MLV IN to further eliminate all of the other described nuclear localization signals from an HIV genome—matrix, IN, Viral Protein R, and the central polypurine tract—and show that no combination of the virally encoded NLS is essential for the ability of HIV to infect non-dividing cells.
Synopsis
Human immunodeficiency virus can infect many cells irrespective of whether or not they are dividing, whereas some other retroviruses, such as the murine leukemia virus can only infect cells that are proliferating. This property is important for the ability of HIV to establish infections in critical cell types in infected people. Multiple and redundant signals encoded by HIV have been hypothesized to facilitate migration of viral genomes into the nucleus. However, here the authors eliminated all four described nuclear localizing signals from an HIV genome and show that no combination of these virally encoded signals is essential for the ability of HIV to infect non-dividing cells. They suggest that another step of the virus lifecycle, other than nuclear import, is the rate-limiting step that determines the cell cycle dependence/independence of retroviral infections.
Citation:Yamashita M, Emerman M (2005) The cell cycle independence of HIV infections is not determined by known karyophilic viral elements. PLoS Pathog 1(3): e18.
==== Body
Introduction
Human immunodeficiency virus and other lentiviruses have the ability to infect non-dividing cells [1–3]. This property allows HIV to integrate into two major types of virus reservoirs in vivo: resting CD4+ T cells and macrophages [4]. However, the ability to productively infect non-dividing cells is not shared by all retroviruses [5]. For example, the gamma retroviruses as exemplified by the murine leukemia virus (MLV) requires mitosis for integration [6,7]. Infection and transduction with foamy retroviruses also depends on cell cycle and requires mitosis [8–10]. An alpharetrovirus, the avian sarcoma virus, appears to be able to integrate viral genomes in non-dividing cells [11,12], but fails to produce virus particles, indicating that it requires mitosis for a later stage of the viral life-cycle [13,14].
After entry into the cytoplasm, retroviruses undergo an uncoating and reverse transcription process that yields a large nucleoprotein complex called the preintegration complex (PIC) [15]. Nuclear entry of viral DNA is an essential step in the retroviral life cycle since viral genomic DNA in the PIC must enter the nucleus to be integrated into host cell chromosomes. The prevailing model to explain the ability of lentiviruses to infect cells independent of the cell cycle is that lentiviruses can target their viral genomes into the nucleus of non-dividing cells via active nuclear transport, while gammaretroviruses that cannot infect non-dividing cells gain an access to the host chromosomes only when the nuclear membrane breaks down at mitosis [6,7]. Thus, it has been hypothesized that the PIC of lentiviruses contain virally encoded nuclear localization signals (NLS), which allow active nuclear transport independent of the cell cycle, whereas the PIC of gammaretroviruses do not contain virally encoded NLS, and thus can not enter the nucleus until mitosis (reviewed in [16]).
Several lentiviral elements that contain a potential NLS and are present in the PIC have been identified including the matrix (MA) [17], integrase (IN) [18], and Viral Protein R (Vpr) [19] proteins and a cis-acting element called the central polypurine tract (cPPT) [20]. However, the importance of each of these elements is controversial since subsequent studies have shown that HIV lacking one or several mutations in these NLS elements still retains a significant ability to infect non-dividing cells [16,21–27]. The IN protein is a particularly attractive candidate to mediate nuclear import of HIV genomes since it is part of the PIC through all steps of infection until viral integration, and IN is necessary for nuclear localization and transposition of the yeast elements Ty1 and Ty3 [28–30]. Moreover, HIV IN contains nuclear import activity [18,25,31–34], whereas MLV IN lacks such nuclear import activity [35,36]. However, the role of HIV IN within nuclear import of viral genomes has been difficult to definitively address because mutations or deletions within IN often show pleiotropic effects on virus replication, including assembly, and reverse transcription, in addition to integration (reviewed in [37]). Recent work has suggested that HIV IN itself does not contain an NLS, but rather traffics to the nucleus by virtue of binding the lens epithelium-derived growth factor (LEDGF)/p75 protein [36,38–39].
We recently reported that the capsid protein (CA) is a dominant determinant of retroviral infectivity in non-dividing cells since HIV containing MLV CA lost the ability to infect non-dividing cells, even though it still contains proteins with an NLS [40]. Because HIV CA is not nucleophilic [41,42] and is not stably associated with the HIV PIC [43–48], these data led us to propose that nuclear entry is not the rate-limiting step in the ability of HIV to infect non-dividing cells [40,49]. However, because we were unable to eliminate all of the proposed NLS in HIV, we could not rule out the possibility that the chimeric HIV containing MLV CA masked a pathway usually used by HIV for entry into the nucleus.
Here, we directly tested the involvement of IN within infection of non-dividing cells by constructing an HIV-MLV chimeric virus in which the HIV IN coding sequence was replaced with the IN coding sequence of MLV. Somewhat remarkably, this chimeric virus was infectious at a low level and was able to integrate in an IN-dependent manner. Furthermore, this virus infected non-dividing cells as well as it infected dividing cells.
While individual NLS-containing proteins, in some cases combinations, have been mutated or deleted from HIV in previous studies, it could be argued that the effect of the different NLS are redundant, and therefore HIV still retained some ability to infect non-dividing cells because of the presence of other NLS on other proteins. The ability to generate an HIV-based chimera with MLV IN allowed us to further eliminate all of the other described NLS (MA, Vpr, and the cPPT) from an HIV infectious clone. We report here that this chimeric virus without any of the previously described NLS is still able to infect non-dividing cells. We discuss the possibility that uncoating of the entering viral particle, rather than nuclear import, is the rate-limiting step that determines the cell cycle dependence/independence of retroviral infections.
Results
Generation of an Infectious Chimeric HIV-1 with MLV IN That Is Integration-Competent
HIV IN localizes to the nucleus when stably expressed in cells, whereas MLV IN does not [36]. Therefore, to determine if the karyophilic property of IN is essential for the infectivity of HIV in non-dividing cells, HIV-1 IN was replaced with MLV IN within an HIV-based provirus, generating the chimeric clone called MHIV-mIN (which encodes MLV IN instead of HIV IN while the rest of the provirus is HIV) (Figure 1). Transfection of this chimeric provirus showed that it produces virus particles as indicated by the presence of virus-specific proteins in culture supernatants of transfected cells (Figure 2). As expected, MLV IN, and not HIV-1 IN, was detected in virions (Figure 2). The amount of virions made by MHIV-mIN was between 3- to 30-fold lower than that made by wild-type HIV-1 as measured by p24gag ELISA (unpublished data). Nevertheless, processing of reverse transcriptase (RT) and IN appeared normal in virions produced by MHIV-mIN virus (Figure 2).
Figure 1 Schematic Representation of the Genomic Organization of Chimeric HIV/MLV Proviruses
Portions originated from the HIV genome are shown in white, while those from the MLV genome are in gray. The junction between HIV-1 RT and MLV IN within HIV-mIN was created by direct joining of DNA sequences encoding the C-terminus of HIV RT to the N-terminus of MLV IN. Part of the 3′ end of the HIV-1 IN encoding sequence is retained in the construct of MHIV-mIN to preserve the overlapping Vif sequence and cis-acting elements such as cPPT and splice acceptor(s). However, no part of HIV IN should be expressed in the chimeric virus because of the presence of two stop codons following the sequence encoding MLV IN (see Materials and Methods). The molecular clone encoding MHIV-mIN/matt has the MLV att sites in 5′ U5 and 3′ U3. After reverse transcription, both ends of U5 and U3 will have the MLV att sites. MHIV-mMA12CA has been previously described [40]. MHIV-mMA12CA/mIN is similar except it contains MLV IN in addition to the MLV Gag region. A molecular clone of MHIV-mMA12CA/mIN was created by putting the DNA sequence encompassing the MLV IN encoding sequence of the MHIV-mIN with a Vpr mutation into the infectious provirus pMHIV-mMA12CA. HIVΔNLS contains MLV MA instead of HIV MA, MLV IN instead of HIV IN and mutations in the cPPT and in Vpr.
Figure 2 Western Blot Analysis of Purified Virus Particles of MHIV-mIN Together with HIV-1 and MLV
Polyprotein processing was tested for HIV-1 RT, HIV-1 IN and MLV IN. Because of low protein production by MHIV-mIN, 10-times more virions were loaded for MHIV-mIN than for HIV-1 and MLV in this experiment.
We tested the infectivity of MHIV-mIN together with wild-type HIV-1 in a single-cycle replication assay [50]. While the titer of MHIV-mIN was about 3-logs lower than that of wild-type HIV-1 when normalized by the amount of p24gag (Figure 3A), it was still well above the background. Real-time PCR data indicated that MHIV-mIN produces 3- to 5-fold less cDNA than wild-type HIV-1 (Figure 3B). Thus, a decrease of reverse transcription of MHIV-mIN alone cannot explain the reduced infectivity of MHIV-mIN.
Figure 3 MHIV-mIN Is Infectious
(A) Single-cycle infectivity of MHIV-mIN. VSV-G-pseudotyped HIV and MHIV-mIN were made by transfection of 293T cells with plasmid DNA. Infectivity was measured with the MAGI assay by counting β-galactosidase positive cells 2 d post-infection. Virus titers were normalized by the amount of p24 (μg). Infections were performed in triplicate. Mean values are shown here with standard derivation. The background in the assay is about 10 blue cells.
(B) Copy numbers of late products of reverse transcription were measured by using real-time quantitative PCR. Viral cDNA numbers were normalized by p24. Infections were performed in triplicate. This is a representative experiment of three independent trials.
(C) Infectivity of MHIV-mIN and MHIV-mIN/matt was compared in the MAGI assay. VSV-G-pseudotyped viruses were prepared and concentrated at 100-fold by ultracentrifugation. Infections of MAGI cells were also performed in the presence or absence of reverse transcriptase inhibitors (50 μM 3TC and AZT). Formation of blue cells in this assay must result from retrovirus infection since addition of reverse transcription inhibitors (shown as black bars) eliminated most of positive cells.
(D) Comparison of infectivity between wild-type HIV-1 and MHIV-mIN as judged by the ability to make puromycin-resistant clones. Infectivity was measured by counting puromycin-resistant colonies 14 d after infection. To facilitate stable transduction of HIV genomes, a mutation was introduced into the vpr gene in these proviruses because expression of Vpr would preclude formation of colonies due to its cytotoxicity [69,77]
The integration reaction requires specific recognition by viral IN of short DNA sequences (~10 bp) at both ends of viral DNA, called the attachment (att) site. A previous report indicated that replacement of HIV att sites with MLV att sites at both ends of the long terminal repeat (LTR) reduced viral titer to 0.5% level of the wild-type level [51], while others have found that the att sequences other than the conserved CA dinucleotide motif are not very important in vivo [52,53]. To test this, we also made a chimeric clone that contains MLV att sequences in both ends of the LTR (Figure 1, called pMHIV-mIN/matt), and examined the infectivity of these chimeras (Figure 3C). We could obtain titers of up to 1 × 106 infectious units per ml after concentration of both viruses, but MHIV-mIN/matt did not show any significant increase in infectivity when compared with MHIV-mIN (Figure 3C). The infectivity of MHIV-mIN was also sensitive to reverse transcriptase inhibitors (Figure 3C and Figure S1), and thus depends on de novo genomic DNA synthesis. Therefore, we found that there is not a requirement for an MLV-specific att site in the context of a chimeric HIV with MLV IN.
IN mutants of HIV that are defective for integration support low levels of infectivity in the multinuclear activation of galactosidase indicate cell (MAGI) assay, probably due to weak expression of the tat gene products from unintegrated DNA [54–56]. However stable expression of transduced genes usually requires integration of viral DNA into host chromosome [57,58]. Thus, to genetically test for integration, we made use of a reporter virus system in which the puromycin-resistant gene was put in place of the nef gene, and infected cells were selected for puromycin resistance. Compared with HIV, MHIV-mIN exhibits ~4 log decrease of infectivity in the puromycin-based assay (Figure 3D), which is about one log lower than the virus titer difference in MAGI assay (Figure 3A). The difference between the MAGI titer and the puromycin-resistance titer is likely due to expression of Tat from unintegrated DNA [54–56]. Nonetheless, these data show that MHIV-mIN is capable of stable transduction.
To more directly address the question of whether or not the MHIV-mIN virus can carry out bona fide IN-mediated integration, we extracted genomic DNA from the puromycin-resistant colonies to amplify and sequence the junction between viral and host sequences. There are two characteristic features of retroviral integration of viral DNA into the host genome. First, two nucleotides are deleted from both ends of viral DNA. Indeed, we observed the deletion of two nucleotides of both ends of all sequenced clones (Figure 4). The second characteristic of IN-mediated integration is that the target sequence of the host DNA is duplicated after the integration event. The size of duplication differs among retroviruses [59]; for example, HIV integration yields 5-bp duplication of the target sequence, whereas MLV integration creates 4-bp duplication. In each case, the length of the duplicated sequence was 4-bp, which is consistent with integration of the HIV chimeric virus mediated by the MLV IN. Taken together these findings demonstrate that MHIV-mIN is competent for all of the early steps of virus replication including integration.
Figure 4 Integration Sites of MHIV-mIN
A schematic illustration of the unintegrated viral DNA is shown with the detailed structures of both ends of LTR (top). The two terminal base pairs at each end of the linear DNA precursor, which are removed in the integration process, as shown in highlight. Two other DNA sequences confirmed the removal of dinucleotides from one end of viral DNA (unpublished data). DNA sequences flanking integrated proviral DNA are shown (bottom). Junction sequences between the integrated MHIV-mIN genome and human genomic DNA at each end of the provirus were obtained by nested PCR based on the sequence of integration sites that were mapped with human genome sequences as described in the Materials and Methods. The 4-bp sequence duplications that flank the integrated provirus are shown in boxes.
HIV IN Is Not Essential for Infection of Non-Dividing Cells
HIV efficiently infects non-dividing cells, whereas MLV infection is restricted in non-dividing cells. To determine if IN plays an essential role in this difference, growth-arrested cells prepared by treatment of HeLa cells with aphidicolin were challenged with the chimeric virus MHIV-mIN along with control viruses, and infectivity was judged by measuring the output of the luciferase gene encoded by reporter virus constructs. As expected from previous studies, wild-type HIV was capable of infecting non-dividing cells as efficiently as dividing cells, while transduction of the luciferase gene by MLV was reduced in non-dividing cells compared with in dividing cells (Figure 5). The phenotype of MHIV-mIN was similar to that with HIV, but not with MLV, in that it was not decreased in non-dividing cells relative to dividing cells (Figure 5). In fact, we saw a slight increase of infectivity by MHIV-mIN on non-dividing cells relative to dividing cells (Figure 5). This increase may be due to expression of the reporter gene from unintegrated DNA in non-dividing cells, as described in the case of infection of non-dividing cells with feline immunodeficiency virus IN mutants [57]. Nonetheless, these results demonstrate that IN is not an essential determinant for the ability of HIV-1 to infect non-dividing cells relative to dividing cells.
Figure 5 IN Does Not Determine the Infectivity in Non-Dividing Cells
Aphidicolin-treated HeLa cells were infected with increasing amount of luciferase-encoding viruses described in Figure 1. Culture supernatants of transfected cells were used as the inoculum. Virus infectivity was judged by measuring luciferase titers of infected cell lysates 2 d after infection. RLU; relative light units. White circles indicate cells without aphidicolin. Filled circles indicate cells with aphidicolin (2 μg per ml). This is a representative experiment that was done at least three times for each virus.
We previously showed that replacement of part of the gag gene of HIV with that of MLV would convert HIV into a virus that had lost the ability to infect non-dividing cells [40]. Similarly, we found that we can change the phenotype of the MHIV-mIN by replacing the MA and CA proteins of HIV with the MA, p12, and CA proteins of MLV (MHIV-mMA12CA-mIN in Figure 1). Indeed, addition of Gag proteins of MLV into the HIV provirus that already contains MLV IN increased the infectivity in dividing cells, but specifically lost the ability to infect non-dividing cells (Figure 5: Compare MHIV-mMA12CA/mIN with MHIV-mIN). These data demonstrate that Gag, rather than IN, is the dominant determinant for the ability of HIV to infect cells independent of cell cycle progression.
Normal Levels of Nuclear Import by MHIV-mIN
A recent report showed that efficient nuclear entry of HIV can occur independently of mitotic nuclear disassembly in cycling cells [60]. Thus, one interpretation of our results is that elimination of an NLS from HIV would result in lack of infectivity both in dividing and non-dividing cell populations. Indeed, the new chimeric virus created in the present study, MHIV-mIN, infects dividing cells and non-dividing cells with an equal efficiency, but the overall infectivity by MHIV-mIN is severely reduced from that of wild-type HIV-1 (Figure 3A). Thus, to directly determine whether or not MHIV-mIN is restricted at nuclear import of viral DNA, infected cells were separated into cytoplasmic and nuclear fractions and real-time PCR was used to measure late reverse transcription products. The results indicate that there is little apparent difference of viral DNA associated with nuclear fractions between MHIV-mIN and HIV (Figure 6A). Although higher amounts of viral DNA were associated with the nuclear fractions of HIV (~75%) than MHIV-mIN (~50%), this level of difference cannot explain the decrease of infectivity by MHIV-mIN (3-log reduction compared with wild-type HIV). Control experiments using a cytoplasmic protein (LDH I) as a maker for the cytoplasmic fraction indicated that contamination of cytoplasm into the nuclear fraction is less than 1% (Figure 6B. compare the 125-fold dilution of the cytoplasmic fraction in lane 2 with the nuclear fraction in lane 6). These data indicate that nuclear entry of MHIV-mIN is essentially not inhibited and that reduced infectivity of the chimeric virus is due to a post-nuclear entry event (most likely integration). We also examined 2-LTR circles, which are often used as a surrogate marker for nuclear entry. We found that the ratio of 2-LTR circles to total viral DNA in cells infected with MHIV-mIN is roughly equivalent (or even slightly higher) to the ratio in cells infected with parental HIV-1 (Figure 6C). The slight increase in the average number of 2-LTR circles per total viral DNA for MHIV-mIN relative to wild-type virus likely reflects the fact that mutants that integrate inefficiently often accumulate 2-LTR circles [56]. Nonetheless, in sum, these data further support the idea that HIV IN is not essential for the nuclear transport of viral DNA and infectivity in non-dividing cells.
Figure 6 Nuclear Entry of MHIV-mIN
(A) Subcellular localization of viral DNA. Infected cells were fractionated to cytoplasmic (C) and nuclear (N) fractions. Viral DNA was extracted and subject to real-time PCR to measure late products of reverse transcription. These data represent one of two independent experiments. Control viruses, with the presence of reverse transcription inhibitors or without VSV-G protein, were also used in the experiments to monitor retrovirus-dependent DNA synthesis, and showed that contamination of plasmid DNA used for transfection is less than 1% of the total DNA.
(B) Western blot analysis of total cell lysates (To), cytoplasmic extract (Cy) and nuclear lysates (Nu). Contamination of cytoplasmic extract and the presence of intact cells in nuclear fractions were tested by checking for the presence of a cytoplasmic protein, LDH-I, in each fraction (upper lanes). Five-fold dilutions of the cytoplasmic extract (5-, 25-, and 125-fold dilutions, lanes 4, 3, and 2, respectively) were made to assess the degree of contamination of the nuclear fraction. Presence of proteins was confirmed by antibody against a nuclear pore complex protein (mAb414; lower lane).
(C) Nuclear import was monitored by measuring late reverse transcription products and 2-LTR circles. The ratio of total viral DNA and 2-LTR circles was obtained by dividing the copy number of late RT products by the copy number of 2-LTR circle. The parental wild-type strain of HIV-1 (shown here as wt) was compared with the chimeric virus MHIV-mIN (shown as mIN). Control infections with reverse transcription inhibitors (AZT and 3TC: 50 μM each) yielded viral copy numbers that are less than 10% of copy numbers of the samples without reverse transcription inhibitors, indicating that contamination of plasmid DNA used to produce virus stocks does not affect the final results. Two independent experiments gave substantially identical data.
HIV Lacking All of the Known Types of NLS Still Infects Non-Dividing Cells as Efficiently as Dividing Cells
As mentioned above, IN is not the sole candidate that potentially encodes a viral NLS. MA, Vpr, and the cPPT have all been described as elements that are important for entry of HIV-1 PIC into the nucleus. To formally address the argument that other described NLSs in HIV as well as the cPPT are redundant for nuclear import with the NLS in HIV IN, a mutant HIV-1 lacking all the NLS candidates was generated. This mutant (HIV-ΔNLS), carrying MLV MA and IN instead of HIV counterparts, lacks a functional vpr gene, and has a mutated cPPT (Figure 1). We found that HIV-ΔNLS had reduced infectivity relative to wild-type HIV (Figure 7), but the infectivity of HIV-ΔNLS is sensitive to reverse transcriptase inhibitors (Figure S1), and thus is not an artifact of the virus concentration. Importantly, the infectivity of HIV-ΔNLS is independent of cell cycle conditions (Figure 7). It should be noted that our luciferase system can detect reduction even when the activity is low (See reduction of MLV infectivity in 0.08 μl in Figure 7, for example). The phenotype of HIV-ΔNLS in non-dividing cells is in marked contrast to that of MLV which is dependent on the cell cycle, and in contrast to a previously described chimeric HIV virus containing MLV MA, p12, and CA (MHIV-mMA12CA: Figure 1) [40], which has specifically lost the ability to efficiently infect non-dividing cells (Figure 7). Therefore, these data demonstrate that HIV without any of the previously described NLS elements is fully capable of infecting non-dividing cells as well as it infects dividing cells, and suggest that the virally encoded NLS elements are not rate-limiting for this process.
Figure 7 HIV Lacking NLSs Infects Non-Dividing Cells
Single-cycle infectivity assay of HIV-ΔNLS together with HIV, MLV, and MHIV-mMA12CA. Both HIV-ΔNLS and MHIV-mMA12CA were concentrated by ultracentrifugation for infections. For details, see the legend to Figure 5. White circles indicate cells without aphidicolin. Filled circles indicate cells with aphidicolin (2 μg per ml). This is a representative experiment that was done at least 3 times for each virus.
Discussion
In the present study, we created a chimeric HIV-1 in which HIV IN is replaced by its counterpart from MLV and demonstrated that HIV can integrate with a heterologous IN protein. Thus, while inefficient, MLV IN can replace HIV IN within the context of an infectious virus. The infectivity of this chimeric virus having MLV IN was not dependent on the presence of the MLV att sites at the ends of the LTR, yet did result in duplications of genomic DNA that were consistent with MLV IN. We used this chimeric virus as a template to eliminate all previously described viral NLS elements and found that it was still able to infect non-dividing cells as well as it infected dividing cells.
A popular model for lentiviral infection of non-dividing cells is that the karyophilic activity of an HIV-encoded NLS such as IN is important to bring HIV PIC into the nucleus of non-dividing cells, thereby allowing efficient infection in the absence of mitosis. Our experiments using a chimeric HIV-1 with MLV IN showed that the exchange of IN does not cause any phenotypic change of infectivity that is specific to non-dividing cells. This finding indicates that IN is not an essential determinant that governs the infectious phenotype in non-dividing cells. These results are in agreement with some previous studies that examined individual putative NLS elements within HIV IN and argued against a role of an IN NLS in nuclear import of HIV [16,24,25,61]. Indeed, a recent report by Lu et al demonstrated that mutations in a putative NLS of HIV IN results in class II mutations, which are defective at a postnuclear entry step rather than at nuclear import [61]. In contrast, Ikeda et al claimed that nuclear import of viral DNA is affected by reduced binding of IN to viral cDNA [62]. However, the reduction of nuclear import of such IN mutants (with reduced binding ability to viral cDNA) was at most 40% of the wild-type level as judged by nuclear DNA and that amount of reduction does not seem to explain severely reduced infectivity of their mutants (less than 1% of the wild-type level) [62]. Moreover, other studies have shown that HIV IN localizes to the nucleus by virtue of binding LEDGF/p75 [36,38,39]. However, reduction of LEDGF levels by small interfering RNA (siRNA) affected the nuclear localization of HIV IN, but did not affect the ability of HIV to infect non-dividing cells [36]. Therefore, although we cannot completely rule out the possibility that HIV IN is involved in nuclear migration of viral DNA, we believe that its role is minor.
Although other studies have ruled out a role for individual and some combinations of putative karyophilic viral elements in the HIV PIC, it has not been possible up to now to eliminate all of the identified elements at once in order to test the hypothesis that infection of non-dividing cells is reliant on multiple redundant NLS. However, we were able to create an HIV mutant lacking all of the known NLS-encoding elements, and demonstrated that not only IN, but also none of the other NLS-encoding elements have any effect on the ability of HIV-1 to infect non-dividing cells. Thus, our data are not consistent with a previous suggestion that mutation of single (or double) NLS-encoding elements had little phenotypic change because of redundant NLS-encoding elements that are responsible for nuclear transport of HIV PIC and for infection in non-dividing cells. One possible interpretation of our results is that we have not yet found the most important NLS encoded by HIV. While this is still formally possible, the present results along with our previous results that found that CA is a dominant determinant for retrovirus infectivity in non-dividing cells [40], suggest that these virally encoded karyophilic elements are not the major determinants for the infectivity of HIV in non-dividing cells. Rather, we consider that our data lend support to the alternative hypothesis that nuclear entry is not the rate-limiting step for infection of non-dividing cells. Our hypothesis is also consistent with the findings that the addition of NLS encoding sequences to MLV does not render it infectious to non-dividing cells [35,63].
Instead, we propose that the difference in CA between HIV and MLV affects the progress of uncoating, thereby influencing downstream events such as nuclear import and integration. In this model, uncoating of HIV progresses normally in non-dividing cells and functional PIC enter the nucleus where they integrate viral DNA. In contrast, uncoating of MLV is impaired in non-dividing cells, which results in the failure of subsequent steps of the replication cycle. In this scenario, gammaretroviruses may need mitosis to complete uncoating. In fact, in the case of HIV, CA is dissociated from viral nucleoprotein complexes [43–48], while larger amounts of CA are associated with MLV PICs [15,64–65], suggesting that uncoating of MLV may not be as efficient as that of HIV. An optimal stability of the HIV core appears to be essential for infectivity [66], and complete uncoating may be a prerequisite for nuclear import of PIC. In this hypothesis, the tight association of MLV CA with PIC prevents cellular machinery from interacting with a putative NLS on MLV PICs, thereby retaining PICs within the cytoplasm of interphase cells. On the other hand, the HIV PIC can migrate into the nucleus of interphase cells by using cellular transport machinery. Thus, we are not arguing that nuclear import of the HIV PIC is not essential. Rather, that it is not the rate-limiting step and that cellular rather than viral components of the PIC might play the major role in viral nuclear import after uncoating.
Materials and Methods
Nomenclature and construction of proviruses.
Proviruses were named as follows: MHIV-mIN encodes the MLV IN instead of the HIV IN while the rest of the provirus is HIV. MHIV-mIN/matt has the MLV IN and MLV att sites. MHIV-MA12CA/mIN encodes the MLV IN as well as the MA, p12, and CA proteins of MLV, instead of the HIV MA and CA. HIV and MLV genes were taken from the infectious proviruses pLai [67] and pAMS [68], respectively. The AMS clone encodes a chimeric strain of amphotropic MLV of which 3′ part of the genome was obtained from the amphotropic virus clone 4070A and the 5′ end from MLV-K. The IN region derived from 4070A was cloned into the proviral DNA of HIV-1 in place of part of the HIV IN encoding sequence. The 5′ end of the MLV IN encoding sequence starts at the same position of the original 5′ end of the HIV IN encoding sequence. Therefore, the junction between the HIV RT encoding sequence and the MLV IN encoding sequence is aggaaagtactaATAGAAAACTCAA (HIV sequence is shown in small letters; MLV sequence is shown in capital letters). Part of the 3′ end of the HIV-1 IN encoding sequence was preserved in the construct of MHIV-mIN, since it contains several important cis-acting elements such as central polypurine tract (cPPT). The original TAA stop codon for the MLV IN is followed by additional stop codon (TGA) that prevents expression of a possible fusion protein containing MLV Env, since MLV IN encoding sequence also contains the initiation site of MLV Env. Thus, the junction between the MLV IN encoding sequence and HIV sequence is CGTGGAAGCCCTTAATAGTCTgaattc (MLV sequence is shown in capital letters; HIV sequence is shown in small letters; two stop codons are shown in underlined).
A molecular clone of HIV-matt was constructed by replacing the att sites of HIV with those of MLV in pLai. The 3′ U3 att site of HIV-1 (ctggaagggcta) was replaced with the U3 att site of MLV (TGAAAGACCCCAA). The 5′ U5 att site of HIV-1 (tctctagcag) was replaced with the U5 att site of MLV (ggtctttcat). Then, the clone HIV-matt was used to create the construct encoding MHIV-mIN/matt by swapping the DNA sequence encompassing the MLV IN encoding sequence in the MHIV-mIN. An additional chimeric HIV-1 having both portion of the gag gene and IN of MLV was made by replacing the DNA sequence encoding the HIV MA and CA proteins with the DNA sequence of the MLV MA, p12 and CA in the context of the proviral clone MHIV-mIN, resulting in MHIV-mMA12CA/mIN.
An HIV mutant lacking all of the putative NLS encoding genes was made by mutating the cPPT and the vpr gene and by replacing the HIV-1 MA and IN with the MLV MA and IN, respectively. The cPPT-D mutation [20] was introduced into pLai by PCR mutagenesis. Construction of a Vpr mutant (pLai-ΔVpr) [69] and MHIV having the MLV MA [40] was reported previously. All of these mutations and replacements were combined together with the env-deficient provirus clone pLai-ΔEnv to create the NLS-minus mutant HIV-ΔNLS.
The reporter virus constructs encoding the luciferase gene were made by introducing the luciferase gene from the wild-type Env-minus HIV-1 (pLai-ΔEnv-luc2) [40] into the new HIV-based constructs. The puromycin resistant gene is cloned into the nef gene of molecular clones of HIV-1, MHIV-mIN, and MHIV-Mma12CA/mIN. The HIV puromycin resistant constructs were created in the same way as the luciferase constructs as described above. The vpr mutant pLai-ΔVpr was used to introduce an insertional mutation in the vpr gene of the puromycin virus constructs.
Western blot analysis.
Western blots were probed with the following antibodies: rabbit anti-HIV-1 RT antibody (through the AIDS Research and Reference Reagent Program, Division of AIDS, NIAID); mouse monoclonal anti-HIV-1 IN antibody (Michael Malim, King's College, London); rabbit anti-MLV IN (Frederick Bushman, University of Pennsylvania, [65]); sheep antibody against LDH I (Cortex Biochem, San Leandro, CA); and mouse monoclonal antibody against nuclear pore complex proteins MAb414 ( Covance, Devner, Pennsylvania) [70]. The membranes were washed for 30 min in wash buffer (PBS containing 0.2% Tween 20) and then incubated with a 1:10,000 dilution of horseradish peroxidase-conjugated antibodies that match with the primary antibody for 60 min at room temperature. The membranes were washed three times for 30 min, and the bound antibody was detected with ECL Plus Western blotting detection reagents (Amersham Biosciences, Little Chalfont, United Kingdom). In some cases, membranes were stripped and reprobed with another primary antibody.
Infectivity assays.
Vesicular stomatitis virus G protein (VSV-G)-pseudotyped viruses were prepared by transient transfections of 293T cells performed with the FuGene 6 reagent. HIV and MHIV expression plasmids were co-transfected with a VSV-G-expression vector (pL-VSV-G [71]) in addition to pCMV-tat to express the VSV-G for pL-VSV-G. For the production of VSV-G-pseudotyped MLV, the MLV Gag-Pol expression vector (pCS2-mGP) [40] were used along with the murine retrovirus-based vectors [72] encoding the luciferase gene (pLNCluc) [40] as well as the VSV-G construct. To enhance the infectivity, MHIV-mIN and HIV-ΔNLS were concentrated by ultracentrifugation. Briefly, 25–35 ml of culture supernatant of transfected 293T cells were centrifuged at 500 × g for 5 min to remove cell debris and then filtered through a 0.22 μm filter. The supernatant were transferred into ultracentrifuge tubes and centrifuged at 64,000 × g for 90 min within a SW28 rotor (Beckman Instruments, Fullerton, California, United States). The supernatants were carefully removed and 250–350 μl of culture medium was added at 4 °C for 1 hr and freshly used for infection.
Single-cycle infectivity of HIV and MHIV was measured by challenging MAGI cells with serial dilution of virus and staining for β-galactosidase expression as basically described previously [50]. HeLa cells were used for infections with the luciferase reporter virus stocks. Luciferase titer was assayed with the luciferase assay kit (Promega, Madison, Wisconsin, United States) and read on a luminometer. Growth-arrested cells were prepared by treatment with 2 μg per ml of aphidicolin (Sigma, St. Louis, Missouri, United States). Virus binding was enhanced by spinoculation [73] and by addition of 20 μg per ml of DEAE/dextran.
Quantification of p24gag and viral cDNA.
The p24gag content of the viral supernatants was determined by an enzyme-linked immunosorbent assay (ELISA; Beckman Coulter, Hialeah, Florida, United States.). Late products of reverse transcription and 2-LTR circles of HIV-1 were measured by using real-time PCR based on a previously published protocol [74] as described previously [40].
Subcellular fractionation.
One day before infection, approximately 5 million HeLa cells were seeded onto four 75 cm2 flasks. The cells were challenged either by the VSV-G-pseudotyped HIV-1 or MHIV-mIN. Cells were infected with virus stocks that can synthesize equivalent amount of viral DNA in target cells. Virus stock of MHIV-mIN was concentrated by ultracentrifugation. Both virus stocks were treated with 50 units of Turbo DNase (Ambion, Austin, Texas, United States) per ml at 37 °C for an hour. Infections were performed with the presence of DEAE/dextran (20 μg per ml).
Subcellular fractionations were carried out based on the method described by Yuan et al [75] with minor modifications. One day after infection, cells were washed, tripsinized, and washed once again with phosphate-buffered saline. In order to extract cell lysates and DNA from intact cells, 20% of the infected cells were kept for further experiments. All the manipulations after this step were carried out at 4 °C. The remaining 80% cells were resuspended in 3 volumes of hypotonic buffer (10 mM HEPES, [pH 7.9]; 1.5 mM MgCl2; 10 mM KCl; 2 mM dithiothereitol ; 20 μg of aprotinin per ml). Resuspended cells were centrifuged at 2,300 × g for 5 min. The cell pellet was resuspended in 3 volumes of hypotonic buffer and kept on ice for 10 min. The cells were homogenized with 30 strokes in a Dounce homogenizer. Nuclei and cell debris were pelleted by centrifugation at 3,300 × g for 15 min. The supernatant of this centrifugation was directly used to extract viral DNA or additionally spun down at 13,400 × g for 20 min. The nuclear pellet were washed with 3 volumes of hypotonic buffer containing 0.005% digitonin once and then washed with hypotonic buffer twice. DNA was extracted from half of each fraction using the QIAamp DNA Mini Kit (Qiagen, Valencia, California, United States), and the other half was used for Western blotting to assess the integrity of the fractionation procedure using antibodies to a cytoplasmic protein or a nuclear pore antigen.
To assess the integrity of the fractionation procedure, we examined the contamination of cytoplasmic fraction into nuclear fraction by monitoring the presence of LDH I. Intact cells (10% of the infected cells) and nuclei (half of the purified nuclei) were first resuspended with 50 μl and 100 μl of NTE buffer (10 mM Tris-HCl, [pH 8.0]; 1 mM EDTA; 50 mM NaCl; 2 mM DTT; 20 μg of aprotinin per ml), respectively. After incubation on ice for 5 min, equivalent amount of NP40-doc buffer (1% NP40; 0.2% sodium deoxycholate; 0.12 M NaCl; 20 mM Tris-HCl, [pH 8.0]) were added to the samples and kept on ice for 10 min. The samples were mixed by vortex and spun down at 9,300 × g for 5 min. Twenty μg of protein samples was used in SDS-PAGE and western blotting analysis.
The most serious problem for our experiments was potential contamination of cytoplasmic viral DNA into purified nuclei. Viral DNA is present in a nucleoprotein complex or a free-DNA form, and those viral DNA may behave differently than cytoplasmic proteins such as LDH I in the process of fractionation. To address this possibility, we used a control to determine if quantity of contamination of viral DNA from the cytoplasmic fraction into the nuclear fraction during the washing steps. To this end, we mixed cytoplasmic extracts of infected cells with nuclei of uninfected cells. In these experiments, we used MLV instead of HIV because of the ease of manipulation. Virus stocks of MLV were prepared by harvesting culture supernatant of ecotropic MLV-producing NIH/3T3 cells. Cytoplasmic extracts of acutely infected cells and nuclei of uninfected NIH/3T3 cells were prepared, mixed on ice, and washed as described above. Nuclear-associated DNA was extracted and subject to real-time PCR to measure the copy number of late reverse transcription products as described above. We found that there was less than 1% introduction of cytoplasmic viral DNA into the nuclear fraction during the washing steps (unpublished data).
Sequencing of junctions between host DNA and integrated viral DNA.
Junction sequences between host DNA and viral DNA were determined by using an inverse PCR strategy as described before [76]. HeLa cells were infected with VSV-G-pseudotyped MHIV-mIN-ΔVpr-Puro. Puromycin-resistant cell clones (~130 colonies) were selected for 2–3 weeks with the presence of puromycin (0.7 μg per ml) and used to extract genomic DNA. Genomic DNA (2 μg) from infected cells was digested with 20 U of PstI at 37 °C for 12 h. After heat inactivation at 65 °C for 40 min, 200 ng of digested DNA were taken out for ligation reaction. The ligation reaction was carried out at 16 °C for 12 h. Ligase was then heat inactivated at 65 °C for 15 min. The region of the junction between cellular DNA and the 5′ end of the integrated proviral DNA was amplified by nested inverse PCR. The first PCR primers were U3RRG2, 5′-GGCAAGCTTTATTGAGGC-3′and Gag716, 5′-GGTCAGCCAAAATTACCCTATAGTG-3′. The second PCR primers were MH536, 5′-TCCACAGATCAAGGATATCTTGTC-3′, and Gag934, 5′-TGTTAAAAGAGACCATCAATGAGGAAG-3′. The PCR was carried out in 50 μl solution, which contains 1 × PCR buffer, dNTPs (0.2 mM), primers (1 μM), 10 units of Taq polymerase (Roche, Basel, Switzerland), and 200 ng of ligated DNA. PCR products were purified and used for cloning by using pGEM-T Vector System (Promega). Positive clones were sequenced by using the T7 primer.
Junction sequences between 3′ ends of viral DNA and host DNA were determined for three of the clones by nested PCR with 5′sense primers matching with 3′ LTR of proviral DNA and with 3′ anti-sense primers matching with host DNA downstream of viral DNA. The information obtained from 5′ junction sequences between host DNA and viral DNA allowed us to map integration sites of these three clones in the human genome sequence deposited in GenBank. Based on this information, 3′ primers were designed. Amplified products were cloned into T-vector and sequenced.
Supporting Information
Figure S1 Single-Cycle Infectivity Assay of MHIV-mIN and HIV-ΔNLS with Reverse Transcriptase Inhibitors
Infections with viruses that encode the luciferase gene in place of nef were performed with the presence (shown in black) or absence (shown in gray) of a reverse transcriptase inhibitor (RTI) (AZT and 3TC: 50 μM each). For details, see the legend to Figure 5. In both cases, the luciferase activity is decreased by RTI which indicates that expression of luciferase relies on de novo RT activity. Also, the presence of aphidicolin does not change the dependence on de novo reverse transcriptase activity for luciferase activity. This is a representative experiment done with two different virus stocks with virtually identical results.
(15 MB TIF)
Click here for additional data file.
We are grateful to Frederick Bushman, Yegor Voronin, and Tsai-Yu Lin for critical reading of the manuscript. We thank the National Institute of Health AIDS Reagent Program for Antiserum to HIV-1 RT; Frederick Bushman for antiserum to MLV IN; and Michael H. Malim for mouse monoclonal antibody against HIV-1 IN. This work was supported by NIH grants R01 AI 51153 and R37 AI 30927.
Competing interests. The authors have declared that no competing interests exist.
Author contributions. MY and ME conceived and designed the experiments. MY performed the experiments. MY and ME analyzed the data. MY contributed reagents, materials, and analysis tools. MY and ME wrote the paper.
Abbreviations
CAcapsid protein
cPPTcentral polypurine tract
INintegrase
MAmatrix
MLVmurine leukemia virus
NLSnuclear localization signal
PICpreintegration complex
RTreverse transcriptase
RTIreverse transcriptase inhibitor
VprViral Protein R
VSV-GVesicular stomatitis virus G protein
==== Refs
References
Weinberg JB Matthews TJ Cullen BR Malim MH 1991 Productive human immunodeficiency virus type 1 (HIV-1) infection of nonproliferating human monocytes J Exp Med 174 1477 1482 1720811
Lewis P Hensel M Emerman M 1992 Human immunodeficiency virus infection of cells arrested in the cell cycle Embo J 11 3053 3058 1322294
Desrosiers R 2001 Nonhuman lentiviruses In: Knipe D, Howley P, Griffin D, Lamb R, Martin M et al., editors. Fields Virology. 4 ed. Philadephia: Lippincott, Williams, and Wilkins. pp. 2095–2121.
Stevenson M 2003 HIV-1 pathogenesis Nat Med 9 853 860 12835705
Katz RA Greger JG Skalka AM 2005 Effects of cell cycle status on early events in retroviral replication J Cell Biochem 94 880 889 15669021
Roe T Reynolds TC Yu G Brown PO 1993 Integration of murine leukemia virus DNA depends on mitosis Embo J 12 2099 2108 8491198
Lewis PF Emerman M 1994 Passage through mitosis is required for oncoretroviruses but not for the human immunodeficiency virus J Virol 68 510 516 8254763
Bieniasz PD Weiss RA McClure MO 1995 Cell cycle dependence of foamy retrovirus infection J Virol 69 7295 7299 7474157
Trobridge G Russell DW 2004 Cell cycle requirements for transduction by foamy virus vectors compared to those of oncovirus and lentivirus vectors J Virol 78 2327 2335 14963129
Patton GS Erlwein O McClure MO 2005 Cell-cycle dependence of foamy virus vectors J Gen Virol 2925–2930
Hatziioannou T Goff SP 2001 Infection of non-dividing cells by Rous sarcoma virus J Virol 75 9526 9531 11533215
Katz RA Greger JG Darby K Boimel P Rall GF 2002 Transduction of interphase cells by avian sarcoma virus J Virol 76 5422 5434 11991971
Humphries EH Temin HM 1972 Cell cycle-dependent activation of rous sarcoma virus-infected stationary chicken cells: Avian leukosis virus group-specific antigens and ribonucleic acid J Virol 10 82 87 4339198
Humphries EH Temin HM 1974 Requirement for cell division for initiation of transcription of Rous sarcoma virus RNA J Virol 14 531 546 4369144
Bowerman B Brown PO Bishop JM Varmus HE 1989 A nucleoprotein complex mediates the integration of retroviral DNA Genes Dev 3 469 478 2721960
Dvorin JD Malim MH 2003 Intracellular trafficking of HIV-1 cores: Journey to the center of the cell Curr Top Microbiol Immunol 281 179 208 12932078
Bukrinsky MI Haggerty S Dempsey MP Sharova N Adzhubel A 1993 A nuclear localization signal within HIV-1 matrix protein that governs infection of non-dividing cells Nature 365 666 669 8105392
Gallay P Hope T Chin D Trono D 1997 HIV-1 infection of non-dividing cells through the recognition of integrase by the importin/karyopherin pathway Proc Natl Acad Sci U S A 94 9825 9830 9275210
Heinzinger NK Bukinsky MI Haggerty SA Ragland AM Kewalramani V 1994 The Vpr protein of human immunodeficiency virus type 1 influences nuclear localization of viral nucleic acids in non-dividing host cells Proc Natl Acad Sci U S A 91 7311 7315 8041786
Zennou V Petit C Guetard D Nerhbass U Montagnier L 2000 HIV-1 genome nuclear import is mediated by a central DNA flap Cell 101 173 185 10786833
Fouchier RA Meyer BE Simon JH Fischer U Malim MH 1997 HIV-1 infection of non-dividing cells: Evidence that the amino-terminal basic region of the viral matrix protein is important for Gag processing but not for post-entry nuclear import Embo J 16 4531 4539 9303297
Freed EO Englund G Maldarelli F Martin MA 1997 Phosphorylation of residue 131 of HIV-1 matrix is not required for macrophage infection Cell 88 171 173 discussion 173 174 9008157
Reil H Bukovsky AA Gelderblom HR Gottlinger HG 1998 Efficient HIV-1 replication can occur in the absence of the viral matrix protein Embo J 17 2699 2708 9564051
Petit C Schwartz O Mammano F 2000 The karyophilic properties of human immunodeficiency virus type 1 integrase are not required for nuclear import of proviral DNA J Virol 74 7119 7126 10888652
Limon A Devroe E Lu R Ghory HZ Silver PA 2002 Nuclear localization of human immunodeficiency virus type 1 preintegration complexes (PICs): V165A and R166A are pleiotropic integrase mutants primarily defective for integration, not PIC nuclear import J Virol 76 10598 10607 12368302
Limon A Nakajima N Lu R Ghory HZ Engelman A 2002 Wild-type levels of nuclear localization and human immunodeficiency virus type 1 replication in the absence of the central DNA flap J Virol 76 12078 12086 12414949
Kootstra NA Schuitemaker H 1999 Phenotype of HIV-1 lacking a functional nuclear localization signal in matrix protein of gag and Vpr is comparable to wild-type HIV-1 in primary macrophages Virology 253 170 180 9918876
Lin SS Nymark-McMahon MH Yieh L Sandmeyer SB 2001 Integrase mediates nuclear localization of Ty3 Mol Cell Biol 21 7826 7838 11604517
Kenna MA Brachmann CB Devine SE Boeke JD 1998 Invading the yeast nucleus: A nuclear localization signal at the C terminus of Ty1 integrase is required for transposition in vivo Mol Cell Biol 18 1115 1124 9448009
Moore SP Rinckel LA Garfinkel DJ 1998 A Ty1 integrase nuclear localization signal required for retrotransposition Mol Cell Biol 18 1105 1114 9448008
Pluymers W Cherepanov P Schols D De CE Debyser Z 1999 Nuclear localization of human immunodeficiency virus type 1 integrase expressed as a fusion protein with green fluorescent protein Virology 258 327 332 10366569
Petit C Schwartz O Mammano F 1999 Oligomerization within virions and subcellular localization of human immunodeficiency virus type 1 integrase J Virol 73 5079 5088 10233971
Tsurutani N Kubo M Maeda Y Ohashi T Yamamoto N 2000 Identification of critical amino acid residues in human immunodeficiency virus type 1 IN required for efficient proviral DNA formation at steps prior to integration in dividing and non-dividing cells J Virol 74 4795 4806 10775618
Depienne C Mousnier A Leh H Le RE Dormont D 2001 Characterization of the nuclear import pathway for HIV-1 integrase J Biol Chem 276 18102 18107 11278458
Seamon JA Jones KS Miller C Roth MJ 2002 Inserting a nuclear targeting signal into a replication-competent Moloney murine leukemia virus affects viral export and is not sufficient for cell cycle-independent infection J Virol 76 8475 8484 12134052
Llano M Vanegas M Fregoso O Saenz D Chung S 2004 LEDGF/p75 determines cellular trafficking of diverse lentiviral but not murine oncoretroviral integrase proteins and is a component of functional lentiviral preintegration complexes J Virol 78 9524 9537 15308744
Engelman A 1999 In vivo analysis of retroviral integrase structure and function Adv Virus Res 52 411 426 10384245
Maertens G Cherepanov P Debyser Z Engelborghs Y Engelman A 2004 Identification and characterization of a functional nuclear localization signal in the HIV-1 integrase interactor LEDGF/p75 J Biol Chem 279 33421 33429 15163664
Maertens G Cherepanov P Pluymers W Busschots K De Clercq E 2003 LEDGF/p75 is essential for nuclear and chromosomal targeting of HIV-1 integrase in human cells J Biol Chem 278 33528 33539 12796494
Yamashita M Emerman M 2004 Capsid is a dominant determinant of retrovirus infectivity in non-dividing cells J Virol 78 5670 5678 15140964
Bukrinskaya AG Ghorpade A Heinzinger NK Smithgall TE Lewis RE 1996 Phosphorylation-dependent human immunodeficiency virus type 1 infection and nuclear targeting of viral DNA Proc Natl Acad Sci U S A 93 367 371 8552640
Gallay P Swingler S Aiken C Trono D 1995 HIV-1 infection of non-dividing cells: C-terminal tyrosine phosphorylation of the viral matrix protein is a key regulator Cell 80 379 388 7859280
Bukrinsky MI Sharova N McDonald TL Pushkarskaya T Tarpley WG 1993 Association of integrase, matrix, and reverse transcriptase antigens of human immunodeficiency virus type 1 with viral nucleic acids following acute infection Proc Natl Acad Sci U S A 90 6125 6129 7687060
Farnet CM Haseltine WA 1991 Determination of viral proteins present in the human immunodeficiency virus type 1 preintegration complex J Virol 65 1910 1915 2002549
Fassati A Goff SP 2001 Characterization of intracellular reverse transcription complexes of human immunodeficiency virus type 1 J Virol 75 3626 3635 11264352
Karageorgos L Li P Burrell C 1993 Characterization of HIV replication complexes early after cell-to-cell infection Aids Res Hum Retroviruses 9 817 823 7504934
Miller MD Farnet CM Bushman FD 1997 Human immunodeficiency virus type 1 preintegration complexes: Studies of organization and composition J Virol 71 5382 5390 9188609
Khiytani DK Dimmock NJ 2002 Characterization of a human immunodeficiency virus type 1 pre-integration complex in which the majority of the cDNA is resistant to DNase I digestion J Gen Virol 2523–2532
Yamashita M Emerman M 2005 Retroviral infection of non-dividing cells: Old and new perspectives Virology 344: In press.
Kimpton J Emerman M 1992 Detection of replication-competent and pseudotyped human immunodeficiency virus with a sensitive cell line on the basis of activation of an integrated beta-galactosidase gene J Virol 66 2232 2239 1548759
Masuda T Kuroda MJ Harada S 1998 Specific and independent recognition of U3 and U5 att sites by human immunodeficiency virus type 1 integrase in vivo J Virol 72 8396 8402 9733892
Katzman M Katz RA 1999 Substrate recognition by retroviral integrases Adv Virus Res 52 371 395 10384243
Brown HE Chen H Engelman A 1999 Structure-based mutagenesis of the human immunodeficiency virus type 1 DNA attachment site: Effects on integration and cDNA synthesis J Virol 73 9011 9020 10516007
Wiskerchen M Muesing MA 1995 Human immunodeficiency virus type 1 integrase: Effects of mutations on viral ability to integrate, direct viral gene expression from unintegrated viral DNA templates, and sustain viral propagation in primary cells J Virol 69 376 386 7983732
Ansari-Lari MA Donehower LA Gibbs RA 1995 Analysis of human immunodeficiency virus type 1 integrase mutants Virology 211 332 335 7544046
Engelman A Englund G Orenstein JM Martin MA Craigie R 1995 Multiple effects of mutations in human immunodeficiency virus type 1 integrase on viral replication J Virol 69 2729 2736 7535863
Saenz DT Loewen N Peretz M Whitwam T Barraza R 2004 Unintegrated lentivirus DNA persistence and accessibility to expression in non-dividing cells: Analysis with class I integrase mutants J Virol 78 2906 2920 14990709
Nakajima N Lu R Engelman A 2001 Human immunodeficiency virus type 1 replication in the absence of integrase-mediated DNA recombination: Definition of permissive and nonpermissive T-cell lines J Virol 75 7944 7955 11483739
Brown PO 1997 Integration In: Coffin JM, Hughes SH, Varmus HE, editors. Retroviruses: Cold Spring Harbor Laboratory Press. pp. 161–203.
Katz RA Greger JG Boimel P Skalka AM 2003 Human immunodeficiency virus type 1 DNA nuclear import and integration are mitosis-independent in cycling cells J Virol 77 13412 13417 14645598
Lu R Limon A Devroe E Silver PA Cherepanov P 2004 Class II integrase mutants with changes in putative nuclear localization signals are primarily blocked at a postnuclear entry step of human immunodeficiency virus type 1 replication J Virol 78 12735 12746 15542626
Ikeda T Nishitsuji H Zhou X Nara N Ohashi T 2004 Evaluation of the functional involvement of human immunodeficiency virus type 1 integrase in nuclear import of viral cDNA during acute infection J Virol 78 11563 11573 15479797
Deminie CA Emerman M 1994 Functional exchange of an oncoretrovirus and a lentivirus matrix protein J Virol 68 4442 4449 8207817
Fassati A Goff SP 1999 Characterization of intracellular reverse transcription complexes of Moloney murine leukemia virus J Virol 73 8919 8925 10515996
Li L Olvera JM Yoder KE Mitchell RS Butler SL 2001 Role of the non-homologous DNA end joining pathway in the early steps of retroviral infection EMBO J 20 3272 3281 11406603
Forshey BM von Schwedler U Sundquist WI Aiken C 2002 Formation of a human immunodeficiency virus type 1 core of optimal stability is crucial for viral replication J Virol 76 5667 5677 11991995
Peden K Emerman M Montagnier L 1991 Changes in growth properties on passage in tissue culture of viruses derived from infectious molecular clones of HIV-1LAI, HIV-1MAL, and HIV-1ELI Virology 185 661 672 1683726
Miller AD Law MF Verma IM 1985 Generation of helper-free amphotropic retroviruses that transduce a dominant-acting, methotrexate-resistant dihydrofolate reductase gene Mol Cell Biol 5 431 437 2985952
Rogel ME Wu LI Emerman M 1995 The human immunodeficiency virus type 1 vpr gene prevents cell proliferation during chronic infection J Virol 69 882 888 7815556
Davis LI Blobel G 1986 Identification and characterization of a nuclear pore complex protein Cell 45 699 709 3518946
Bartz SR Vodicka MA 1997 Production of high-titer human immunodeficiency virus type 1 pseudotyped with vesicular stomatitis virus glycoprotein Methods 12 337 342 9245614
Miller AD Rosman GJ 1989 Improved retroviral vectors for gene transfer and expression Biotechniques 7 980 982 , 984 986 , 989 990 2631796
O′Doherty U Swiggard WJ Malim MH 2000 Human immunodeficiency virus type 1 spinoculation enhances infection through virus binding J Virol 74 10074 10080 11024136
Butler SL Hansen MS Bushman FD 2001 A quantitative assay for HIV DNA integration in vivo Nat Med 7 631 634 11329067
Yuan B Fassati A Yueh A Goff SP 2002 Characterization of Moloney murine leukemia virus p12 mutants blocked during early events of infection J Virol 76 10801 10810 12368323
Chun TW Finzi D Margolick J Chadwick K Schwartz D 1995 In vivo fate of HIV-1-infected T cells: Quantitative analysis of the transition to stable latency Nat Med 1 1284 1290 7489410
Planelles V Bachelerie F Jowett JB Haislip A Xie Y 1995 Fate of the human immunodeficiency virus type 1 provirus in infected cells: A role for vpr J Virol 69 5883 5889 7637036
|
16292356
|
PMC1283251
|
CC BY
|
2021-01-05 12:11:25
|
no
|
PLoS Pathog. 2005 Nov 11; 1(3):e18
|
utf-8
|
PLoS Pathog
| 2,005 |
10.1371/journal.ppat.0010018
|
oa_comm
|
==== Front
PLoS PathogPLoS PathogppatplpaplospathPLoS Pathogens1553-73661553-7374Public Library of Science San Francisco, USA 1629235710.1371/journal.ppat.001002305-PLPA-RA-0017R1plpa-01-03-02Research ArticleInfectious DiseasesMicrobiologyIn VitroEubacteriaHomo (human)Human Neutrophils Kill Bacillus anthracis
Neutrophils Kill
B. anthracisMayer-Scholl Anne 1Hurwitz Robert 2Brinkmann Volker 3Schmid Monika 4Jungblut Peter 4Weinrauch Yvette 5Zychlinsky Arturo 1*
1 Department of Cellular Microbiology, Max-Planck-Institute for Infection Biology, Berlin, Germany
2 Biochemistry Core Facility, Max-Planck-Institute for Infection Biology, Berlin, Germany
3 Microscopy Core Facility, Max-Planck-Institute for Infection Biology, Berlin, Germany
4 Protein Analysis Core Facilities, Max-Planck-Institute for Infection Biology, Berlin. Germany
5 Department of Microbiology, New York University School of Medicine, New York, New York, United States of America
Young John EditorThe Salk Institute for Biological Studies, United States of America* To whom correspondence should be addressed. E-mail: [email protected] 2005 11 11 2005 1 3 e234 4 2005 3 10 2005 Copyright: © 2005 Mayer-Scholl et al.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
Bacillus anthracis spores cause natural infections and are used as biological weapons. Inhalation infection with B. anthracis, the etiological agent of anthrax, is almost always lethal, yet cutaneous infections usually remain localized and resolve spontaneously. Neutrophils are typically recruited to cutaneous but seldom to other forms of anthrax infections, raising the possibility that neutrophils kill B. anthracis. In this study we infected human neutrophils with either spores or vegetative bacteria of a wild-type strain, or strains, expressing only one of the two major virulence factors. The human neutrophils engulfed B. anthracis spores, which germinated intracellularly and were then efficiently killed. Interestingly, neutrophil killing was independent of reactive oxygen species production. We fractionated a human neutrophil granule extract by high-performance liquid chromatography and identified α-defensins as the component responsible for B. anthracis killing. These data suggest that the timely recruitment of neutrophils can control cutaneous infections and possibly other forms of B. anthracis infections, and that α-defensins play an important role in the potent anti-B. anthracis activity of neutrophils.
Synopsis
Bacillus anthracis is the bacterium that causes anthrax, a disease that can occur through natural infections and also through intentional release. B. anthracis makes spores, which are in a dormant state, similar to seeds of a plant, and are extremely resistant to the environment. B. anthracis spores can infect through the skin or the lung. Lung infections disseminate through the body and are lethal. In contrast, skin infections often remain localized, and patients survive even without treatment. It is not well understood why these bacteria cause a localized infection through the skin and a lethal disease through the lung.
Little is known about how B. anthracis is controlled. Neutrophils are the first white blood cells recruited to a site of infection and are specialized in killing microbes. Previous studies show that neutrophils are abundant in the skin form, but not in the lung form of anthrax. The researchers report that human neutrophils can take up B. anthracis spores. Once inside, the spores germinate to form vegetative bacteria. The vegetative bacteria are extremely susceptible to neutrophil-killing mechanisms. The B. anthracis virulence factors (molecules that make bacteria cause diseases) manipulate other human cells but do not deter neutrophils. B. anthracis is indeed exquisitely sensitive to the neutrophil protein α-defensin. These data support a new model where B. anthracis skin, but not lung, infections are controlled by the antimicrobial activity of neutrophils.
Citation:Mayer-Scholl A, Hurwitz R, Brinkmann V, Schmid M, Jungblut P, et al. (2005) Human neutrophils kill Bacillus anthracis. PLoS Pathog 1(3): e23.
==== Body
Introduction
The Gram-positive bacterium Bacillus anthracis infects through intradermal inoculation, ingestion, or inhalation of spores. Spores are dormant forms of B. anthracis and are extremely resistant to environmental stress. In the current model of respiratory infections, the spores are first taken up by macrophages where they germinate and become vegetative bacteria [1]. Macrophages then transport the bacteria to the regional lymph nodes [2]. In these organs, the bacteria escape from the macrophage and spread through the lymphatics and blood stream causing massive septicemia. Vegetative B. anthracis express two essential virulence factors: the tripartite anthrax toxin and the poly-γ-D glutamic acid capsule [3]. The toxin-protective antigen binds to its specific receptor and translocates the edema and lethal factors into the cytosol [4]. Edema factor is an adenylate cyclase that causes tissue edema [5], whereas lethal factor is a metalloprotease that inactivates mitogen-activated protein kinase-kinase and provokes cell death [6].
Neutrophils are a vital component of the acute inflammatory response and play a key role in the resolution of microbial infections. They are terminally differentiated cells, incapable of cell division, and synthesize very low levels of RNA and protein. Neutrophils engulf microbes into a phagosome that fuses with intracellular granules to form a phagolysosome. In the phagolysosome the bacteria are killed through the interaction of reactive oxygen species (ROS) [7] and oxygen-independent mediators such as enzymes and antimicrobial peptides [8]. Antimicrobial peptides are predominantly cationic and are thought to permeabilize the bacterial membrane and lyse microbes [9].
Inhalation B. anthracis infections result in sepsis and death, while cutaneous anthrax almost always remains localized. Interestingly, in untreated cutaneous cases, neutrophils surround the necrotic, bacteria-containing tissue [10], whereas neutrophil infiltration is rarely seen in the lung during inhalation anthrax [11]. Although there are reports of pulmonary infiltration in B. anthracis infections [12], they were thought to be due to preexisting lesions [13,14]. Pigs and dogs also develop a cutaneous form when infected subcutaneously. This form is reminiscent of human cutaneous anthrax and is accompanied by massive neutrophil infiltrations [13]. Also, neutrophil recruitment is found in the lungs of pulmonary infected dogs and pigs that survive high infection doses [14]. Therefore, we hypothesized that neutrophils can kill B. anthracis.
Results/Discussion
Human Neutrophils Kill B.anthracis Spores Independent of Reactive Oxygen Species
We tested wild-type B. anthracis and strains expressing only the toxin or the capsule. We incubated neutrophils with a wild-type strain at a multiplicity of infection (MOI) of 5:1 (spores: neutrophils) and measured bacterial viability at different times. Activated neutrophils killed 60% of spores after 3 h, and at 4 h only 30% of the spores survived. Similar results were obtained from infections with the toxin- or capsule-producing strains (Figure 1A). Spores of wild-type (Figure 1B), as well as the other two strains (unpublished data), were phagocytosed by activated neutrophils. In the absence of serum, around 50% of the spores were found intracellularly 30 min post-infection as determined by counting spores in electron microscopy samples. These data are in agreement with Welkos et al. [15] who showed that murine neutrophils can kill toxin-positive and capsule-negative spores, albeit with lower efficiency, than we observed with human neutrophils. To ensure that spores and not vegetative bacteria were ingested, the killing efficiency was tested in serum-free media since some sera induce germination [16].
Figure 1 Human Neutrophils Kill B. anthracis Spores Independently of Reactive Oxygen Species
(A) Human neutrophils kill B. anthracis spores. Activated neutrophils were infected with spores generated either from the wild-type or toxin- or capsule-producing strains at an MOI (spore: neutrophil) of 5:1 (wild-type) or 1:1 (toxin- and capsule-producers). Colony forming units (CFUs) were counted at indicated time points.
(B) TEM of activated neutrophils infected with wild-type spores (arrow) after 30 min incubation. Bar indicates 2 μm.
(C) ROS are not required for killing of B. anthracis spores. Activated neutrophils were infected with spores (capsule producer) in presence (□) or absence (⋄) of NADPH oxidase inhibitor.
(D) hNGE does not kill B. anthracis spores. Spores from the toxin-producing strain were incubated in 50% hNGE or in buffer for the indicated time points. The CFU was determined by serial dilution. Survival of 100% refers to the number of bacteria present in the buffer control. Error bars indicate the SD from three experiments.
Neutrophils kill microbes through oxygen-dependent and independent mechanisms. ROS are generated by the nicotinamide adenosine dinucleotide phosphate (NADPH) oxidase and are important in some bacterial infections. To determine the role of ROS in the killing of B. anthracis, we inactivated neutrophil NADPH oxidase with diphenyleneiodonium (DPI) [17] (Figure S1), and found that the antimicrobial activity was not dependent on ROS (Figure 1C). We then tested whether antimicrobial proteins were effective against B. anthracis by incubating spores with a human neutrophil extract enriched in granule proteins (hNGE). Even at very high hNGE concentrations the spores remained viable throughout a 4-h incubation (Figure 1D) and did not germinate during this time (unpublished data). These findings prompted us to test whether spores germinate in neutrophils and whether the vegetative forms are then susceptible to neutrophil killing.
B. anthracis Spores Germinate within Human Neutrophils
Neutrophils were infected with spores, and the germination of B. anthracis in the cells was analyzed by transmission electron microscopy (TEM). At 90 min post-infection, newly germinated vegetative bacteria were observed for the first time (Figure 2A). In contrast, spore uptake was recorded after 30 min. By 180 min, neutrophils contained mostly germinated bacteria (Figure 2B). The spores and vegetative bacteria were found in membrane-bound vacuoles, most likely phagosomes. Intriguingly, the germination process in both neutrophils and macrophages [1] does not seem synchronized, as spores were found throughout the infection period. Whether the spores remain unresponsive to cell-specific germinants, remains to be examined.
Figure 2
B. anthracis Spores Germinate within Human Neutrophils
(A) TEM of neutrophil infected with spores of the toxin-producing strain for 90 min at an MOI of 1:1. At this time point both spores (arrow) and germinated bacteria (arrowhead) were present (see inset for amplification).
(B) Neutrophil infected for 180 min. At this time point neutrophils contained mostly germinated bacteria (arrowheads). Bars in (A) indicate 5 μm, in (B) 2 μm, and in insets 500 nm.
(C) Phagocytosis of B. anthracis spores is essential for neutrophil killing. Phagocytosis of activated neutrophils was inhibited with cytochalasin D (grey bars). Cells were infected with wild-type strain at an MOI of 5:1. Controls were PMA-activated cells not treated with cytochalasin D (black bars).
(D) B. anthracis spores do not germinate in the presence of neutrophils. PMA-activated neutrophils were pretreated with cytochalasin D in the presence (black bars) or absence (grey bars) of penicillin G (25μg/ml). The cells were infected with wild-type strain spores at an MOI of 5:1 and incubated for the given time points. The CFU was determined by serial dilution. Survival of 100% refers to the number of spores present in the controls lacking penicillin G. Error bars indicate the SD from three experiments.
To confirm that germination occurred intracellularly, we prevented neutrophil phagocytosis by incubating activated neutrophils with cytochalasin D, an inhibitor of actin polymerization. Inhibition of phagocytosis completely abrogated the sporicidal activity of the neutrophils (Figure 2C), supporting the electron microscopy data that spores germinate within the cells prior to killing. To further verify that spores did not germinate extracellularly, we incubated spores in the presence of cytochalasin D-treated neutrophils and added an antibiotic to kill any newly developed vegetative cells. There was no difference in the counts of spores incubated with cytochalsin D-treated neutrophils in the absence or presence of antibiotics (Figure 2D). These data indicate that degranulating neutrophils do not induce extracellular germination.
Human Neutrophils Efficiently Kill Vegetative B. anthracis
To determine whether vegetative B. anthracis are killed by neutrophils, we infected activated cells with the wild-type strain or with those expressing either the toxin or the capsule. Both the fully virulent and the toxin-expressing strains were phagocytosed despite the large size of the B. anthracis bacilli (~5 microns) (Figure 3A and 3B). The capsule of the wild-type strain can clearly be distinguished in the inset of Figure 3A. Three hours post-infection, activated neutrophils killed 70% of the wild-type strain, more than 90% of the toxin-producing bacteria, and 80% of the capsule producers (Figure 3C). These data show that the capsulated strains are slightly more resistant to neutrophil killing at an MOI of 10:1. In contrast, at an MOI of 1:1 and a 30-min infection period, neither the toxin nor the capsule significantly affects the killing capacity of the neutrophils (Figure 3D).
Figure 3 Human Neutrophils Efficiently Kill Vegetative B. anthracis Intra- and Extracellularly
(A) TEM of neutrophils infected with wild-type B. anthracis (arrow) at an MOI of 1:1. Bar = 1μm. Inset indicates the inoculum. Bar = 100 nm.
(B) Scanning electron microscopy of neutrophils infected with vegetative form of the toxin-producing strain for 30 min. Phagocytosed bacteria shown by arrow. Neutrophils efficiently engulf B. anthracis despite their size. Bar = 5μm.
(C) Neutrophils kill B. anthracis efficiently. Activated neutrophils were infected with wild-type, capsule-, or toxin-expressing vegetative bacteria for indicated time points at an MOI of 10:1.
(D) B. anthracis capsule and toxin are not antiphagocytic. Activated neutrophils were pretreated with the phagocytosis inhibiting cytochalsin D (10 μg/ml), infected with wild-type, capsule-, or toxin-expressing vegetative bacteria at an MOI of 1:1, and incubated for 30 min.
To determine if extracellular mechanisms are responsible for B. anthracis killing, we again inhibited neutrophil phagocytosis. In the presence of cytochalasin D, all three B. anthracis strains were killed equally well (Figure 3D). This implies that extracellular mechanisms play an important role in the neutrophil-mediated killing. Interestingly, the proportion of intracellular killing was comparable in all three strains, indicating that neither the toxin nor the capsule has a pronounced antiphagocytic effect. These data are in agreement with O'Brien et al. [18] who showed that when the neutrophil is exposed to the toxin and the bacteria at the same time, the anthrax toxin does not inhibit phagocytosis. Although the capsule is an absolute prerequisite for virulence [19], the role of the capsule in the interaction of B. anthracis with host cells is not well understood [20,21]. It has been inferred, however, from studies with other capsulated microorganisms, that a capsule can affect phagocytosis. The data presented here show that, at least in the conditions tested, the B. anthracis capsule does not inhibit phagocytosis.
Interestingly, opsonization did not significantly affect the rate of killing of either capsule- or toxin-expressing vegetative bacteria (Figure 4A) indicating that activated neutrophils might recognize B. anthracis directly. The independence of opsonization was seen in both phorbol 12-myristate 13-acetate (PMA) (Figure 4A) and IL-8-activated neutrophils (unpublished data). Interestingly, previous reports [20,22] showed that opsonization of vegetative bacilli had an effect on the generation of ROS, but bacterial killing was not directly tested.
Figure 4 Human Neutrophils Efficiently Kill Vegetative B. anthracis Independently of Oxygen Radicals
(A) Opsonization is not required for neutrophil anti-B. anthracis activity. PMA-activated neutrophils were infected with capsule (gray bar)- and toxin (black bar)-producing strains at an MOI of 1:1 in the presence or absence of serum for 30 min.
(B) Neutrophils kill both capsule (gray bar)- and toxin (black bar)-producing B. anthracis independently of NADPH oxidase activity. Activated neutrophils were incubated with DPI (10 μM) before infection. Bacterial counts were determined 30 min post-infection (MOI 1:1).
(C) hNGE kills B. anthracis at low concentrations. Wild-type, toxin-, and capsule-producing strains were incubated with 5%, 7%, and 10% hNGE. After 30 min, remaining CFUs were determined and referred to the number of bacteria in controls of bacteria incubated in buffer. Error bars show SD of three experiments.
Analogously to spore killing, vegetative B. anthracis are killed independent of ROS production (Figure 4B). These data imply that the described reduced ROS production in neutrophils of anthrax patients [22] might not play a significant role in vivo. Interestingly, these data are in contrast to the Gram-positive bacterium Staphylococcus aureus, which is killed in a ROS-dependent manner [23]. Therefore, we investigated whether B. anthracis is killed by granule antimicrobials. Indeed, all three B. anthracis strains were surprisingly susceptible to low concentrations of hNGE (Figure 4C). In comparison, 10% hNGE only killed 50% of a S. aureus culture (unpublished data). Corresponding to neutrophil killing, both capsulated strains were less susceptible to hNGE in comparison to the toxin-expressing strain.
Neutrophils Kill B. anthracis through α-Defensins
To identify the components of the hNGE responsible for B. anthracis killing, hNGE was fractionated by C4 reverse phase high-performance liquid chromatography (RP-HPLC). The peak fraction exhibiting antimicrobial activity in the first column (Figure 5A) was further purified in a C4 RP-HPLC column with a different mobile phase (see Materials and Methods) followed by a C18 RP-HPLC column. Tricine SDS-PAGE analysis of the antimicrobial fraction showed a peptide migrating at three to four kDa (unpublished data). The active component was identified by MALDI MS peptide mass fingerprinting (PMF) and MS/MS data of five peptides as human neutrophil α-defensin (Figure 5B). Only the 1-fold and the 2-fold charged molecule peaks were detected (Figure 5B, inset) in the MS spectrum of the non-trypsinized peptide, confirming purity of the sample. These results were verified by LC-ESI-MS analysis (unpublished data).
Figure 5 Neutrophils Kill B. anthracis through α-Defensins
(A) Purification of the anti-B.anthracis activity in hNGE; hNGE was resolved in a C4 RP- HPLC column (A) and the fractions were tested for their killing activity of B. anthracis (bars). The main chromatographic peak coincides with the B. anthracis killing activity (arrow).
(B) The active component was identified by MALDI MS PMF as human neutrophil α-defensins. Combined MASCOT search of PMF and MS/MS data (five MS/MS spectra) resulted in a Score value of 313. Peak heights of molecules not belonging to the identified protein were below 3%. Purity of sample was confirmed by the analysis of the uncleaved peptide (inset). As the protein was analyzed in non-reducing conditions it is present in its oxidized form reducing the Mr from 3,446 to 3,440.
(C) The purified fraction and synthetic α-defensin have comparable specific activity. Vegetative toxin-producing bacteria were incubated for varying time points with 8 μg/ml synthetic (□) or purified (○) peptide and referred to the number of bacteria in buffer only. Error bars indicate SD of three experiments.
α-defensins are a group of small antimicrobial peptides found in human neutrophil granules [24]. The three major human α-defensins differ by one amino acid, and all kill bacteria [25] effectively, although some Gram-positive pathogens such as S. aureus are resistant [26]. To confirm that α-defensins are responsible for neutrophil killing of B. anthracis, we compared the antimicrobial activity of the purified fraction to synthetic human α-defensin (Figure 5C). These two samples had a comparable specific activity confirming human α-defensins as an hNGE component responsible for killing wild-type, toxin-producing, and capsule-producing (unpublished data) B. anthracis strains.
The data presented here show that human neutrophils engulf B. anthracis spores and induce them to germinate. Neutrophils then efficiently kill the bacteria independent of toxin or capsule expression. The timely recruitment of neutrophils to the site of a cutaneous anthrax infection could explain why these infections resolve spontaneously. In inhalation anthrax, where there is little neutrophil infiltration [11], the infection progresses unchecked and leads to sepsis and death. Here we also show that α-defensins play an important role in the potent anti-B. anthracis activity of neutrophils.
Materials and Methods
Human neutrophil isolation.
Heparinized venous blood was obtained after informed consent of healthy volunteers. Neutrophils ( > 95% pure) were isolated using a dextran-sedimentation protocol followed by centrifugation in a Ficoll-Paque gradient [27]. Erythrocytes were lysed with ultra-pure water for 30 s and cells washed in HBSS- and diluted in RPMI/10 mM HEPES with or without 10% FCS.
Bacterial strains.
All experiments were performed using either the wild-type (Vollum), the capsule-producing (Pasteur), or toxin-producing (Sterne) B. anthracis strain and performed in a BSL-3 facility. The anthrax strains were a kind gift of W. Beyer, University of Hohenheim. For toxin expression, bacteria were grown on BHI agar plates with 0.8% sodium bicarbonate at 37 °C, 5% CO2. A single colony was subcultured overnight in the supplemented BHI. The wild-type strain was additionally incubated in 5% CO2 atmosphere. The capsule-producing strain was grown on BHI plates for 2 d at 37 °C and 20% CO2 and a single colony was subcultured overnight in RPMI supplemented with 10% BHI, 10 mM HEPES, and 0.8% sodium bicarbonate.
Toxin and capsule production.
Lethal and edema factors, as well as protective antigen expressions, were monitored by Western Blot analysis of TCA-precipitated bacterial culture supernatant using antibodies from Abcam (Abcam, Cambridge, United Kingdom); γ-D-glutamic acid capsule expression was examined by India ink stain.
Preparation of B. anthracis endospore suspensions.
Strains were grown in BHI overnight at 37 °C before plating on meat yeast agar and incubated for 10 d at 37 °C and 5% CO2. The spores were scraped off, washed five times with PBS, and kept in PBS at 4 °C until use. The wild-type strain spores were purified with Renografin as described by Ivins et al. [28]. They were washed four times with water and kept in water at 4 °C. Purity was confirmed by phase contrast microscopy and heat treatment (60 °C, 45 min).
Antimicrobial activity of neutrophils.
2 × 106/ml neutrophils were allowed to adhere to plastic plates for 15 min before activation with 25 nM PMA Sigma (Sigma, St. Louis, Missouri, United States) or 10 ng/ml Il-8 (BioCat, Heidelberg, Germany) for 30 min. Bacterial suspensions (vegetative cells or spores) were centrifuged onto the neutrophils (1,250 rpm, 10 min) and incubated for the indicated time points at 37 °C. For opsinization experiments vegetative bacteria were preincubated for 10 min with 5% non-inactivated human serum. Killing experiments were performed in the presence of a final concentration of 5% human serum. The cells were scraped off and serial dilutions plated on BHI agar plates. Colonies were counted after 12-h incubation at 37 °C, 5% CO2. Bacterial killing was measured as percentages of control values (bacteria incubated alone in media without neutrophils). To measure antimicrobial activity in the absence of ROS, neutrophils were preincubated with 10 μM of the NADPH oxidase inhibitor DPI [17] for 30 min. Inhibition of the respiratory burst was confirmed by enhanced chemiluminescence of PMA (25 nM)-activated neutrophils [29]. The experiment was read in a BD Pharmagen TM, Monolight 3096 microplate luminometer (BD Biosciences Pharmingen, San Diego, California, United States).
Inhibition of phagocytosis.
Neutrophils were isolated as described above and allowed to adhere to plastic plates for 15 min. Cells were activated with PMA (25 ng/ml) for 30 min, spun down at 1,250 rpm for 5 min. The supernatant was discarded and media containing 10 μg/ml cytochalsin D (Sigma) was added for 15 min at 37 °C. Cells were infected with spores or vegetative bacteria as described above. To confirm that spores do not germinate in the presence of neutrophils, 25 μg/ml PenicillinG (Sigma-Aldrich) was added to the cytochalsin D- treated cells prior to infection.
Chromatography.
hNGE was fractionated with a C4 RP-HPLC column (Vydac Protein C4 Column, 3.9 × 250 mm, 5 μm). Proteins were eluted with a gradient of increasing concentrations of acetonitril containing 0.1% (v/v) trifluoroacetic acid (flow rate 1 ml/min). Fractions were lyophilized, dissolved in 20 mM of sodium acetate buffer and tested for antimicrobial activity. The antimicrobial activity was further purified in a second C4 column, (mobile phase 100 mM of ammonium acetate [pH5]), followed by a C18 RP-HPLC column (X-terra RP18, 3.9 × 150 mm, 3.5 μm) with an acetonitril and 0.1% (v/v) trifluoroacetic acid gradient (flow rate 1 ml/min). Separation was performed in a Waters 626 LC System with a Water 996 Photodiode Array Detector (Waters, Milford, Massachusetts, United States).
Bactericidal activity.
hNGE was prepared as described in [30]. The bactericidal activity of hNGE, HPLC fractions, or synthetic human α-defensin 2 (American Peptide Company, Sunnyvale, California, United States) was determined by incubation with bacteria (107 bacteria/ml) in a casamino buffer (0.3% casamino acids, HBSS-, 10 mM HEPES [pH7.4]) at 37 °C, with the indicated doses, shaking for 30 min. Antimicrobial activity of each hNGE batch was tested with the toxin- expressing B. anthracis strain. Survivors were counted in serial dilutions. Bacterial killing was measured as percentages of control values (bacteria incubated in media alone).
Electron microscopy.
For TEM, cells were fixed with 2.5% glutaraldehyde, post-fixed with 1% osmium tetroxide, contrasted with uranylacetate and tannic acid, and dehydrated and embedded in Spurr's (Ted Pella, Redding, California, United States). After polymerization, specimens were cut at 60 nm and contrasted with lead citrate. Specimens were analyzed in a Leo 906E TEM. For scanning electron microscopy: Cells were fixed with 2.5% glutaraldehyde, post-fixed using repeated incubations with 1% osmium tetroxide/1% tannic acid, dehydrated with a graded ethanol series, critical-point dried, and coated with 2 nm platinum. After dehydration and critical-point drying, the specimens were coated with 5 nm Platinum/Carbon and analyzed in a Leo 1550 scanning electron microscopy.
Mass spectrometry.
The identity and purity of the antimicrobial component was analyzed by MALDI mass spectrometry (Proteomics 4700 workstation, Applied Biosystems, Foster City, California, United States), PMF, MS/MS analysis, and mass analysis of the uncleaved protein. The lyophilized sample was digested with 50 mM NH4HCO3, 5% acetonitrile, 2% (w/v) trypsin (Sequencing grade modified Trypsin, Promega, Madison, Wisconsin, United States), and 0.15M DTT for 4 h at 37 °C. The reaction was stopped with 0.2% TFA and mixed with matrix alpha-Cyano-4-hydroxycinnamic acid (CHCA) solubilized in 50% acetonitrile 0.3% TFA with a concentration of 5 mg/ml.
Analysis of PMFs was obtained with parameters: reflectron mode, 20 kV accelerating voltage, and a low mass gate of 800 Da. MS/MS spectra were obtained without collision gas. Parameters for database searches (MASCOT, http://www.matrixscience.com) were: 30 ppm peptide mass tolerance for PMF and 0.3 Da for MS/MS spectra. The uncleaved protein was analyzed in linear mode with CHCA as matrix with an internal marker (Mr 2465.21).
Supporting Information
Figure S1 Inhibition of Respiratory Burst by DPI
PMA-(25 nM) activated neutrophils were incubated in the presence (□) or absence (⋄) of DPI and chemiluminescence monitored in a BD Pharmagen TM, Monolight 3096 microplate luminometer in the presence of luminol.
(782 KB TIF)
Click here for additional data file.
We thank Wolfgang Beyer from the University of Hohenheim for providing us with the B. anthracis strains and many helpful comments. We thank Ulrike Reichard and Beatrix Fauler for help with electron microscopy, as well as Wolfgang Höhenwarter for the LC/ESI/MS analysis. We thank Constance Scharff and members of the Department of Cellular Microbiology, Max-Planck-Institute for Infection Biology for helpful comments. YW is supported by the National Institutes of Health IH grant 1 R21 AI057579-01A1.
Competing interests. The authors have declared that no competing interests exist.
Author contributions. AMS, YW, and AZ conceived and designed the experiments. AMS, RH, VB, and MS performed the experiments. AMS, RH, VB, MS, and AZ analyzed the data. PJ contributed reagents/materials/analysis tools. AMS and AZ wrote the paper.
Abbreviations
CFUcolony forming unit
DPIdiphenyleneiodonium
hNGEhuman neutrophil granule extract
MOImultiplicity of infection
NADPHnicotinamide adenosine dinucleotide phosphate
PMAphorbol 12-myristate 13-acetate
PMFpeptide mass fingerprinting
ROSreactive oxygen species production
RP-HPLCreverse phase high-performance liquid chromatography
TEMtransmission electron microscopy
==== Refs
References
Guidi-Rontani C Weber-Levy M Labruyere E Mock M 1999 Germination of Bacillus anthracis spores within alveolar macrophages Mol Microbiol 31 9 17 9987105
Guidi-Rontani C Mock M 2002 Macrophage interactions Curr Top Microbiol Immunol 271 115 141 12224520
Mock M Fouet A 2001 Anthrax Annu Rev Microbiol 55 647 671 11544370
Bradley KA Mogridge J Mourez M Collier RJ Young JA 2001 Identification of the cellular receptor for anthrax toxin Nature 414 225 229 11700562
Leppla SH 1982 Anthrax toxin edema factor: A bacterial adenylate cyclase that increases cyclic AMP concentrations of eukaryotic cells Proc Natl Acad Sci U S A 79 3162 3166 6285339
Duesbery NS Webb CP Leppla SH Gordon VM Klimpel KR 1998 Proteolytic inactivation of MAP-kinase-kinase by anthrax lethal factor Science 280 734 737 9563949
Hampton MB Kettle AJ Winterbourn CC 1998 Inside the neutrophil phagosome: Oxidants, myeloperoxidase, and bacterial killing Blood 92 3007 3017 9787133
Borregaard N Cowland JB 1997 Granules of the human neutrophilic polymorphonuclear leukocyte Blood 89 3503 3521 9160655
Lehrer RI Ganz T 1999 Antimicrobial peptides in mammalian and insect host defense Curr Opin Immunol 11 23 27 10047545
Lebowich RJ McKillip BG Conboy JR 1943 Cutaneous anthrax: A pathological study with clinical correlation Am J Clin Pathol 13 505 515
Albrink WS Brooks SM Biron RE Kopel M 1960 Human inhalation anthrax: A report of three fatal cases Am J Pathol 36 147 471
Abramova FA Grinberg LM Yampolskaya OV Walker DH 1993 Pathology of inhalational anthrax in 42 cases from the Sverdlovsk outbreak of 1979 Proc Natl Acad Sci U S A 90 2291 2294 8460135
Brachman PS 1980 Inhalation anthrax Ann N Y Acad Sci 353 83 93 7013615
Meselson M Guillemin J Hugh-Jones M Langmuir A Popova I 1994 The Sverdlovsk anthrax outbreak of 1979 Science 266 1202 1208 7973702
Cromartie WJ Bloom WL Watson DW 1947 Studies in infection with Bacillus anthracis
J Infect Dis 80 1 13
Gleiser CA Gochenour WS Jr. Ward MK 1968 Pulmonary lesions in dogs and pigs exposed to a cloud of anthrax spores J Comp Pathol 78 445 448 4971601
Welkos SL Trotter RW Becker DM Nelson GO 1989 Resistance to the Sterne strain of B. anthracis: Phagocytic cell responses of resistant and susceptible mice Microb Pathog 7 15 35 2509851
Ireland JA Hanna PC 2002 Macrophage-enhanced germination of Bacillus anthracis endospores requires gerS Infect Immun 70 5870 5872 12228320
Cross AR Jones OT 1986 The effect of the inhibitor diphenylene iodonium on the superoxide-generating system of neutrophils: Specific labelling of a component polypeptide of the oxidase Biochem J 237 111 116 3800872
O'Brien J Friedlander A Dreier T Ezzell J Leppla S 1985 Effects of anthrax toxin components on human neutrophils Infect Immun 47 306 310 3917427
Drysdale M Heninger S Hutt J Chen Y Lyons CR 2005 Capsule synthesis by Bacillus anthracis is required for dissemination in murine inhalation anthrax Embo J 24 221 227 15616593
Keppie J Harris-Smith PW Smith H 1963 The chemical basis of the virulence of Bacillus anthracis. Ix. Its aggressins and their mode of action Br J Exp Pathol 44 446 453 14079018
Makino S Watarai M Cheun HI Shirahata T Uchida I 2002 Effect of the lower molecular capsule released from the cell surface of Bacillus anthracis on the pathogenesis of anthrax J Infect Dis 186 227 233 12134259
Alexeyev OA Morozov VG Suzdaltseva TV Mishukov AS Steinberg LA 1994 Impaired neutrophil function in the cutaneous form of anthrax Infection 22 281 282 8002089
Kristian SA Durr M Van Strijp JA Neumeister B Peschel A 2003 MprF-mediated lysinylation of phospholipids in Staphylococcus aureus leads to protection against oxygen-independent neutrophil killing Infect Immun 71 546 549 12496209
Ganz T 2003 Defensins: Antimicrobial peptides of innate immunity Nat Rev Immunol 3 710 720 12949495
Ganz T Lehrer RI 1995 Defensins Pharmacol Ther 66 191 205 7667395
Peschel A Collins LV 2001 Staphylococcal resistance to antimicrobial peptides of mammalian and bacterial origin Peptides 22 1651 1659 11587793
Boyum A 1968 Isolation of mononuclear cells and granulocytes from human blood. Isolation of monuclear cells by one centrifugation, and of granulocytes by combining centrifugation and sedimentation at 1 g Scand J Clin Lab Invest Suppl 97 77 89 4179068
Dahlgren C Karlsson A 1999 Respiratory burst in human neutrophils J Immunol Methods 232 3 14 10618505
Weiss J Elsbach P Olsson I Odeberg H 1978 Purification and characterization of a potent bactericidal and membrane active protein from the granules of human polymorphonuclear leukocytes J Biol Chem 253 2664 2672 344320
|
16292357
|
PMC1283252
|
CC BY
|
2021-01-05 12:11:21
|
no
|
PLoS Pathog. 2005 Nov 11; 1(3):e23
|
utf-8
|
PLoS Pathog
| 2,005 |
10.1371/journal.ppat.0010023
|
oa_comm
|
==== Front
PLoS BiolPLoS BiolpbioplosbiolPLoS Biology1544-91731545-7885Public Library of Science San Francisco, USA 1627755210.1371/journal.pbio.0030401Research ArticleCancer BiologyMus (Mouse)Impaired DNA Replication within Progenitor Cell Pools Promotes Leukemogenesis Poor Competition Promotes TumorigenesisBilousova Ganna
1
Marusyk Andriy
2
Porter Christopher C
3
Cardiff Robert D
4
DeGregori James [email protected]
1
2
3
5
1Department of Biochemistry and Molecular Genetics, University of Colorado Health Sciences Center, Aurora, Colorado, United States of America,2Program in Molecular Biology, University of Colorado Health Sciences Center, Aurora, Colorado, United States of America,3Department of Pediatrics, University of Colorado Health Sciences Center, Aurora, Colorado, United States of America,4Center for Comparative Medicine and Department of Pathology, School of Medicine, University of California, Davis, California, United States of America,5Integrated Department of Immunology, University of Colorado Health Sciences Center, Aurora, Colorado, United States of AmericaKemp Christopher Academic EditorFred Hutchinson Cancer Research CenterUnited States of America12 2005 15 11 2005 15 11 2005 3 12 e40113 7 2005 23 9 2005 Copyright: © 2005 Bilousova et al.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
A Surprising New Path to Tumor Development
Impaired cell cycle progression can be paradoxically associated with increased rates of malignancies. Using retroviral transduction of bone marrow progenitors followed by transplantation into mice, we demonstrate that inhibition of hematopoietic progenitor cell proliferation impairs competition, promoting the expansion of progenitors that acquire oncogenic mutations which restore cell cycle progression. Conditions that impair DNA replication dramatically enhance the proliferative advantage provided by the expression of Bcr-Abl or mutant p53, which provide no apparent competitive advantage under conditions of healthy replication. Furthermore, for the Bcr-Abl oncogene the competitive advantage in contexts of impaired DNA replication dramatically increases leukemogenesis. Impaired replication within hematopoietic progenitor cell pools can select for oncogenic events and thereby promote leukemia, demonstrating the importance of replicative competence in the prevention of tumorigenesis. The demonstration that replication-impaired, poorly competitive progenitor cell pools can promote tumorigenesis provides a new rationale for links between tumorigenesis and common human conditions of impaired DNA replication such as dietary folate deficiency, chemotherapeutics targeting dNTP synthesis, and polymorphisms in genes important for DNA metabolism.
The authors show that the replicative competence of normal body cells can influence the rate of outgrowth of cells that carry an oncogene, helping explain a poorly understood aspect of carcinogenesis.
==== Body
Introduction
Despite significant advances in leukemia research, the factors that promote the selection and expansion of mutated cells leading to leukemia are not well known. Leukemias represent abnormal and poorly regulated hematopoiesis, with leukemia stem cells capable of self-renewal as well as the generation of multiple cell types [1]. Acquired mutations contributing to leukemogenesis, including chromosomal translocations that generate oncogenic fusion genes, may need to occur in hematopoietic stem cells (HSCs), as these cells have sufficient lifespan necessary for the acquisition of additional mutagenic hits. Alternatively, it is possible that oncogenic events happen in more committed progenitors, particularly if the mutation confers self-renewal on the cell or if the cell (such as a B cell progenitor) exhibits a sufficient life span [1,2].
Chronic myelogenous leukemia (CML) is a clonal myeloproliferative disease characterized by excessive proliferation of progenitor cells with massive accumulation of mature myeloid lineage cells. Untreated, the chronic phase is typically followed by an accelerated phase and blast crisis [3]. The t(9:22) Philadelphia chromosome translocation product generating p210 Bcr-Abl is the major cause of CML, and translocations generating either p210 or p190 Bcr-Abl fusions are present in childhood and adult acute lymphoblastic leukemias (ALLs). The CML-associated p210 Bcr-Abl gene is found in myeloid, erythroid, and lymphoid lineage cells in CML patients, suggesting that the translocation originally occurred in a pluripotent stem cell [3]. Recent evidence suggests that while the p210 Bcr-Abl fusion is found in HSC, the leukemia stem cells for both p210- and p190-associated ALLs possess committed B progenitor phenotypes [4]. The Bcr-Abl kinase promotes proliferation and growth factor independence, through the activation of a number of growth factor signaling pathways [5]. In addition, Bcr-Abl expression increases genomic instability, in part by stimulating homologous recombination repair, which can promote survival after DNA damage, while increasing recombination and loss of heterozygosity events [6].
A number of contexts associated with decreased cell cycle progression are, somewhat paradoxically, associated with high rates of tumors. The loss of the E2F1 and E2F2 transcription factors are generally associated with decreased proliferation in vitro and in vivo [7,8], consistent with their roles in promoting the transcription of a variety of genes required for cell cycle progression. Surprisingly, mice with mutations in E2f1 and/or E2f2 genes exhibit significantly increased cancer incidence [9,10]. Until now, the proposed tumor suppressive function of E2F1 and E2F2 has been attributed to decreased E2F-dependent apoptosis, DNA repair, or gene repression (reviewed in [11,12]). In addition, dietary folate deficiency, resulting in impaired DNA replication in cells in vivo, and polymorphisms that reduce dNTP synthesis are associated with increased colon cancer and leukemia rates [13,14]. Previous studies have substantiated that dNTP imbalances caused by folate deficiency or drugs that inhibit dNTP synthesis can increase DNA damage [14–16], and enhanced mutation accumulation is thought to underlie increased tumorigenesis [13,14]. Finally, patients treated with particular chemotherapeutic regimens, such as etoposide with methotrexate, are at increased risk of treatment-related acute myeloid leukemia [17]. Etoposide and methotrexate, which inhibit topoisomerase II and dihydrofolate reductase respectively, are effective inhibitors of DNA replication. Chemotherapeutics such as etoposide promote chromosomal translocations [17], and dNTP imbalances engendered by methotrexate may further promote mutation accumulation. Relevant to the studies presented here, genetic, dietary, and chemotherapeutic inhibitions of the dNTP and folate metabolic pathways are expected to promote replicative stress and DNA damage, leading to checkpoint activation and cell cycle inhibition.
While it is logical that conditions that negatively impact on a cell population could select for mutations conferring resistance, whether contexts of impaired proliferation contribute to tumorigenesis has not been previously demonstrated. We show that impaired DNA replication of hematopoietic progenitors in E2f1/E2f2 mutant mice or mice treated with hydroxyurea (HU) allows Bcr-Abl-expressing and p53-mutated progenitors to outcompete nonmutated cells in the same niche, promoting leukemogenesis.
Results
The Poorly Replicating E2f1
−/−
E2f2
−/− Progenitor Environment Favors the Competitive Expansion of Bcr-Abl-Expressing Progenitors
DNA replication in bone marrow (BM) progenitor cell populations is severely impaired in E2f1
−/−
E2f2
−/− (DKO) mice, but not in E2f1
+/−
E2f2
+/− mice (phenotypically wild type; henceforth called E2f1
+
2
+), leading to severe hematopoietic deficiencies [7]. We tested the effects of E2f1 and E2f2 mutation on the expansion and leukemogenicity of cells expressing the p190 Bcr-Abl fusion oncogene, which is preferentially associated with ALL in humans [5]. For most experiments, we limited transduction with mouse stem cell viruses (MSCVs) encoding Bcr-Abl and/or green fluorescent protein (GFP) to only a small percentage of progenitors to more accurately reflect the rare occurrence of oncogenic initiating events in a background of unmutated cells, allowing for competition between Bcr-Abl-positive and -negative progenitor cells. The transduced cells were then transplanted into wild-type irradiated recipient mice, and competitive reconstitution was analyzed (see Figure 1 for the experimental design).
Figure 1 Experimental Design
At low transduction efficiencies (less than 5% of transplanted cells expressing Bcr-Abl), we found that Bcr-Abl expression clearly provides a substantial advantage for DKO, but not E2F1
+
2
+ progenitors, as reflected by contributions to peripheral blood cells at 3 wk post-BM transplantation (BMT) (Figure 2A and 2B). The recipients of Bcr-Abl-transduced DKO cells had dramatically increased percentages of B cells and myeloid cells expressing Bcr-Abl. Increased contributions of Bcr-Abl-expressing cells in the DKO relative to E2F1
+
2
+ background were also observed at high infection efficiency, although a modest (but for many recipients, transient) advantage of Bcr-Abl expression was evident among E2F1
+
2
+ cells in terms of contributions to mature blood cells (Figure S1). As a control, we transduced BM progenitors with vector expressing GFP only, and in numerous experiments we found that E2f1/E2f2 loss did not noticeably affect the ability to transduce stem and progenitor populations, as similar short-term and long-term reconstitution of all lineages as determined by GFP expression was observed in recipients of vector-transduced E2f1
+
2
+ and DKO BM progenitors (Figures 2A, 2C, and unpublished data).
Figure 2 The Expression of Bcr-Abl Provides a Competitive Advantage to DKO Progenitors
Purified c-Kit+ cells from either E2f1
+
2
+ or DKO donor mice (pooled sets of littermates) were transduced with MSCV-expressing p190 Bcr-Abl and GFP or GFP only (vector), and transplanted into lethally irradiated BALB/c recipients. The same pool of Bcr-Abl transduced DKO progenitors was either mixed with untransduced E2f1
+
2
+ competitor progenitors (“DKO + E2F1+2+ competitors”) or with untransduced DKO progenitors (“DKO”). Bcr-Abl-transduced E2f1
+
2
+ progenitors similarly were mixed with untransduced E2f1
+
2
+ progenitors (“E2F1+2+”). Specifically, each recipient received 2.5 × 105 untransduced whole BM cells combined with 1.2 × 106 MSCV-transduced c-Kit+ cells. Initial infection efficiencies measured for the whole c-Kit-enriched population were 5.1% and 3.75% among E2f1
+
2
+, and 4.73% and 3.4% among DKO progenitors for Bcr-Abl and vector, respectively. Tail blood isolated at 3 wk post-BMT was analyzed for the expression of GFP and B220 (upper bar graph; B-cell lineage) or GR-1 (lower bar graph; myeloid lineage) by flow cytometry.
(A) Each bar represents data from five mice with standard error (SE) indicated, and statistical significance determined using Student's unpaired t-test (for all figures in this paper).
(B) Representative flow cytometric profiles of peripheral B cells are shown, with the percentages of peripheral blood cells expressing both GFP and B220 indicated.
(C) BM cells were analyzed by flow cytometry for the expression of Lin antigens, Sca1, Flk2 and GFP. Lin antigens included GR-1, B220, Ter119, CD3, and CD4. Cells analyzed in the profiles shown were gated as Lin− Flk2−. The upper right quadrant represents progenitor-enriched Lin− Flk2− Sca1+ cells expressing either Bcr-Abl or vector, and the percentages of Lin− Flk2− cells that express Sca1 and/or GFP are indicated. In multiple experiments, recipients of Bcr-Abl transduced E2f1
+
2
+ cells exhibited reduced percentages of GFP− Lin− Flk2− Sca1+ cells (the upper left quadrant) relative to vector controls, a phenomenon that we do not currently understand.
In order to directly determine contributions to hematopoietic progenitor populations, we analyzed BM from the same set of mice for the expression of GFP (vector or Bcr-Abl) in particular progenitor pools. The lineage negative (Lin−) population in the BM expresses low levels of lineage-specific markers, and is thus enriched for progenitor cells. The Lin− Flk2− Sca1+ population is enriched for HSC but will also include short-term progenitors [18]. Of note, while Sca1 is expressed on only about 25% of HSC in BALB/c mice, the Lin− Sca1+ population is substantially enriched for long-term HSC activity in these mice [19]. We observed a dramatic selective advantage for Bcr-Abl expression in the Lin− Flk2− Sca1+ population in the DKO, but not the E2f1
+
2
+, background (Figure 2C, upper right quadrants), correlating with contributions to peripheral blood cells. Of note, a transient advantage within more committed progenitors (such as Lin− Sca1− cells; Figure 2C, lower right quadrants) is often evident for Bcr-Abl expression in E2f1
+
2
+ cells. As we find that c-Kit is down-regulated on Lin− Sca1+ cells post-BMT (unpublished data), we have been unable to further examine Bcr-Abl expression among highly defined HSC. In summary, Bcr-Abl expression provides a dramatic competitive advantage to progenitor cells in the replication-impaired DKO background, but not in the replication-competent E2f1
+
2
+ background.
The Expression of Bcr-Abl Restores S Phase Progression in DKO Progenitor Cells
The competitive expansion of Bcr-Abl-expressing cells relative to untransduced cells in the DKO background could be the result of overcoming cell cycle defects observed in DKO progenitors. To test this possibility, we assayed the effect of Bcr-Abl expression on S-phase progression rates in DKO progenitor cells using a bromodeoxyuridine (BrdU) pulse-chase method [20]. Two weeks post-BMT, at which point no signs of leukemia development were evident, mice were injected with BrdU and then sacrificed after 2 h. BrdU injection essentially provides a short pulse of the label, since BrdU half-life in vivo is very short [21]. Following flow cytometric isolation of GFP+ Lin− or GFP− Lin− progenitors from the spleen, the average DNA content of BrdU-positive cells was assayed by propidium iodide (PI) intensity profiling. The rates of S phase progression were determined as the shift of the average PI intensity from the middle of S phase toward the G2 peak using formulas described by Begg et al. [20]. Although the calculated progression rates do not provide accurate absolute values for S phase length, since the presumption that the average PI intensity lies in the middle of S phase is not true for DKO cells (which accumulate in early S phase [7]), this method can still accurately analyze the changes of the rates caused by Bcr-Abl expression. Significantly, while S phase progression is substantially slowed in DKO progenitor cells, as previously reported [7], the expression of Bcr-Abl in these cells essentially restored S phase progression to control progression rates (Figure 3). In contrast, the expression of Bcr-Abl had no noticeable effects on S phase progression in wild-type progenitors. In addition, we previously showed that a fraction of DKO BM progenitors appear arrested in S phase, exhibiting S phase DNA content but no detectable BrdU incorporation [7]. As shown in Figure 3A, Bcr-Abl expression in DKO progenitors prevents the accumulation of this S phase-arrested population.
Figure 3 Bcr-Abl Restores S Phase Progression in DKO Progenitors
Wild-type or DKO donor mouse c-Kit+ cells were transduced with MSCV-Bcr-Abl or vector and transplanted as in Figure 2 but at higher efficiency.
(A) Recipients were injected 3 wk post-transplant with BrdU and spleen B cell and myeloid progenitors (combined) isolated by flow sorting, separating GFP+ from GFP−. Immunofluorescence for BrdU together with PI staining for DNA content are shown for GFP− cells (internal control) and GFP+ cells from E2f1
+
2
+ or DKO recipients of Bcr-Abl-transduced stem cells. The percentages of cells with S phase DNA content but without significant BrdU incorporation (in the small square) are indicated.
(B) The histograms shown in the insets in (A) plot the DNA content of cells gated as BrdU+, which are used to calculate the average lengths of S phase (from two experiments, ± SE).
Thus, the competitive advantage provided by Bcr-Abl expression to DKO progenitors appears to derive from restored cell cycle progression. Furthermore, as we find that the expression of Bcl2 does not provide an advantage to DKO progenitors relative to wild-type progenitors (Figure S2), the ability of Bcr-Abl to block apoptosis does not appear to underlie the selective advantage conferred by Bcr-Abl expression in the DKO background, although decreased apoptosis could be a contributing factor.
The DKO Background Promotes Bcr-Abl-Mediated Leukemogenesis
As leukemogenesis appears to require initiating mutations in stem or progenitor cell populations, we addressed whether the greater competitive advantage provided to Bcr-Abl-expressing DKO progenitors translates to increased tumor incidence in transplanted mice. Strikingly, all recipients of Bcr-Abl-transduced DKO progenitors developed rapid pre-B ALL-like disease using both low and high initial transduction efficiencies (Figures 4A and 4B), preceded by rapid increases of peripheral GFP+ B220+ cells (Figures 2A and S1). In contrast, only about half of the recipients of E2f1
+
2
+ cells transduced at high efficiency, and none of the recipients at low transduction efficiency, developed leukemia (also pre-B ALL; Figure 4). For leukemias arising in both genetic backgrounds, leukemic infiltrates were evident in spleen, liver, and lung, and peripheral blood GFP+ cells expressed low B220, exhibiting a large, immature morphology with pale-staining nuclei (Figure 4C and unpublished data). Leukemia cells in the spleen and BM were mostly negative for the Lin markers GR-1, Ter119, CD4, and CD3 (unpublished data), expressing from undetectable to low B220 levels depending on the leukemias (Figure 4C; examples shown express little to no B220). Interestingly, we consistently observed a greater fraction of Sca1+CD34+ cells in DKO relative to E2f1
+
2
+ leukemias (Figure 4C), consistent with the early advantage provided by Bcr-Abl expression in DKO progenitor-enriched populations (Figure 2C).
Figure 4 E2f1/E2f2 Loss Promotes Bcr-Abl-Mediated Leukemias
(A and B) E2f1
+
2
+ or DKO mouse c-Kit+ cells were transduced with MSCV-Bcr-Abl and transplanted as in Figure 2, except that in (B) a higher initial transduction efficiency was obtained (23.2% for E2f1
+
2
+ and 24.1% for DKO). Initial transduction efficiencies for (A) were the same as in Figure 2. Untransduced competitors were included as described in Figure 2, except that for the experiments shown in (B), competitors were c-Kit+-purified BM cells cultured in parallel to the transduced cultures but not infected, and each recipient mouse received 1.5 × 105 untransduced c-Kit+ competitors combined with 1.4 × 106
Bcr-Abl-transduced progenitors. Kaplan-Meier curves are shown for both experiments. Mice were sacrificed when moribund, all with splenomegaly and lymphadenopathy. No mice transplanted with vector-expressing cells developed leukemia (unpublished data). The Kaplan-Meier curves for Bcr-Abl/DKO are statistically different from the Bcr-Abl/ E2f1
+
2
+ (p < 0.001 for [A] and p < 0.0001 for [B]) and Bcr-Abl/DKO + E2f1
+
2
+ competitors (p < 0.001 for [A], p < 0.02 for [B]) curves.
(C) Peripheral blood, spleen, and BM from morbid leukemic Bcr-Abl BMT mice or from healthy vector BMT mice (from [B]) were analyzed for the expression of the indicated antigens and GFP by flow cytometry. Peripheral blood was analyzed 21 d post-BMT. The recipient of Bcr-Abl/DKO cells was sacrificed at 23 d and other recipients sacrificed at 29 d post-BMT. The mean fluorescence intensities (mfi) for peripheral B220+ cells are indicated.
E2f1
+
2
+ Competitors Effectively Suppress Leukemia Development by Bcr-Abl-Expressing DKO Progenitors
While leukemogenesis correlates with the ability of Bcr-Abl-expressing cells to competitively expand within progenitor pools, based on increased leukemia rates alone we cannot distinguish whether or not cell autonomous effects of E2f1/E2f2 loss, such as accelerated mutagenesis or decreased apoptosis, are the major factors underlying increased leukemia rates. To directly address this issue, we included E2f1
+
2
+ untransduced competitor cells with Bcr-Abl-transduced DKO progenitors; amazingly, these competitors were able to almost completely eliminate leukemogenesis (Figures 4A and 4B). E2f1
+
2
+ competitors also substantially reversed the selective advantage conferred by Bcr-Abl expression in the DKO background as reflected in peripheral blood cells (Figures 2A, 2B, and S1) and within stem cell-enriched populations (Figure 2C). Similarly, transplantation of Bcr-Abl-transduced DKO progenitors into sublethally irradiated recipients resulted in the rebound of host (wild-type) hematopoiesis and complete prevention of leukemias (unpublished data). Thus, enhanced leukemogenesis is not primarily due to cell autonomous effects of E2f1 and E2f2 loss, but most likely results from the poor ability of DKO progenitor cells to compete with Bcr-Abl-expressing DKO progenitors.
Importantly, these results suggest that the failure of many recipients of Bcr-Abl-transduced E2f1
+
2
+ (particularly at low transduction efficiency) to develop leukemia was not due to our inability to introduce Bcr-Abl into the appropriate progenitor cells, but to the inability of Bcr-Abl
+ progenitors to be maintained in the face of replication competent competitors. Finally, while it remains formally possible that DKO stem/progenitor cells home poorly to BM niches and that Bcr-Abl restores homing to DKO progenitors, the fact that a low percentage of cotransplanted E2f1
+
2
+ BM can reverse the selective expansion of Bcr-Abl
+ DKO progenitors and leukemogenesis (Figures 4 and S1) strongly argues against a role for differential homing in genotype-dependent leukemia rates.
In summary, a DKO progenitor pool promotes Bcr-Abl-dependent leukemogenesis due to the inability of untransduced DKO progenitors to provide effective competition. The ability of Bcr-Abl to restore S phase progression to replication impaired DKO progenitors most likely underlies the competitive advantage provided by Bcr-Abl specifically within the DKO background.
HU Promotes the Competitive Expansion of Bcr-Abl-Expressing Blood Progenitors Ex Vivo
In order to further explore the generality of the selection for Bcr-Abl-expressing cells under conditions of impaired DNA replication as well as to examine Bcr-Abl-mediated leukemogenesis in a more clinically relevant context, we asked how the suppression of proliferation within hematopoietic progenitor pools during treatment of mice with HU, an inhibitor of ribonucleotide reductase, affects Bcr-Abl-dependent competition within wild-type BM. HU has been used for over 40 y as a cytostatic drug [22]. HU-induced dNTP depletion results in replication stalling and the activation of checkpoint pathways [23]. We initially asked whether HU treatment creates a poorly competitive environment that favors the expansion of Bcr-Abl-expressing blood progenitors ex vivo. As shown in Figure 5A, HU differentially affected Bcr-Abl- and vector-expressing progenitors, allowing Bcr-Abl
+ cells to competitively expand. In fact, while the expansion of vector-expressing progenitors was almost completely inhibited by all concentrations of HU used, Bcr-Abl-expressing progenitors continued expanding at doses ranging from 5 to 25 μM (Figure 5A, lower panels). Higher doses of 50 and 100 μM HU did effectively inhibit the expansion of Bcr-Abl
+ progenitors, indicating that the advantage provided by Bcr-Abl expression during HU treatment is relative but not absolute. While Bcr-Abl expression alone provided a modest advantage to wild-type progenitors in culture, 5 to 25 μM HU treatment conferred an even greater advantage to Bcr-Abl-expressing cells (Figure 5A, upper panel). Furthermore, as determined by GFP expression levels, HU selects for cells expressing higher levels of Bcr-Abl (Figure 5B). Notably, increased Bcr-Abl expression is commonly observed in patients during CML progression from chronic phase to blast crisis [24].
Figure 5 HU Promotes the Competitive Expansion of Bcr-Abl
+ Blood Progenitors Ex Vivo
(A) BALB/c c-Kit+ BM cells were transduced with MSCV-Bcr-Abl or vector and cultured in stem cell medium with HU (0, 5, 10, 25, 50, or 100 μM) added starting at day 3. Flow cytometric analysis of GFP expression and total cell counting in cultures were performed every 3–4 d. The changes in GFP percentages in vector and Bcr-Abl cultures (upper graphs) and the cumulative expansion (loge scale) of Bcr-Abl
+ and vector cells (lower graphs) during HU treatment were plotted.
(B) The percentages of cells expressing Bcr-Abl (day 13) as well as the mean fluorescence intensities (mfi) for the GFP+ population, which has been shown to reflect Bcr-Abl protein expression [35], are indicated. Most Bcr-Abl
+ cells were c-Kit+CD34+, with or without HU treatment. HU treatment did not affect the intensity of GFP in vector cells (unpublished data).
HU Treatment Provides a Competitive Advantage to Bcr-Abl-Expressing Progenitors In Vivo
Mice were transplanted with MSCV-p190 Bcr-Abl-transduced BALB/c BM progenitor cells, and half of the recipients were switched to water containing HU (20 mg/kg/d; similar to doses used clinically for cytoreduction [22]) at 2 wk post-BMT. Strikingly, by 5 and 8 d after switching recipient mice to HU, a substantial competitive advantage for Bcr-Abl
+ cells within the progenitor-enriched Lin− Sca1+ population was evident, while a similar selection for Bcr-Abl expression was not observed without HU treatment (Figure 6A). Note that about two-thirds of the progenitor-enriched Lin− Sca1+ cells expressed Bcr-Abl in HU-treated mice, while very few Bcr-Abl
+ Lin− Sca1+ cells were detected in untreated mice, despite the fact that these mice were transplanted with the same pool of transduced BM progenitors. The rapid competitive advantage provided by Bcr-Abl specifically following HU treatment in vivo, together with the almost immediate selection for Bcr-Abl expression by HU ex vivo, argues that HU-induced mutagenesis does not account for the effects of HU on the competitive expansion of Bcr-Abl
+ progenitors.
Figure 6 HU Selects for Bcr-Abl Expression Within Hematopoietic Progenitor Pools In Vivo
BALB/c c-Kit+ BM cells were transduced with the indicated retroviruses, and transplanted into sublethally irradiated BALB/c recipients. At 2 wk post-BMT, half of each group was switched to water containing HU.
(A) BM cells were analyzed for the expression of Lin antigens, Sca1, and GFP as described in Figure 2C on days 5 and 8 of HU treatment. Cells analyzed in the profiles shown were gated as Lin−. The percentages of the progenitor enriched Lin− Sca1+ population that expressed vector or Bcr-Abl are indicated in the upper right corners (underlined).
(B) Four days after switching to HU water (or not), mice were injected with BrdU, and spleen GFP+ and GFP− Lin− progenitors isolated by flow sorting as in Figure 3. BrdU incorporation is detected in GFP− cells (internal control) and GFP+ cells, and the calculated average lengths of S phase from two experiments is shown in the graph.
To understand why Bcr-Abl provides such an advantage specifically following HU treatment, we analyzed cell cycle progression in Bcr-Abl-expressing progenitors in vivo using the same method used for Figure 3. While HU decreased the percentage of cells undergoing DNA synthesis and increased the length of S phase, Bcr-Abl expression resulted in more normal S phase progression in leukocytic progenitors in HU-treated recipients (Figure 6B). This result provides a mechanistic insight into why HU treatment improves the competitive expansion of Bcr-Abl-expressing progenitors in vivo and ex vivo: While HU impedes the proliferation of competing progenitors, Bcr-Abl overcomes this block. In contrast, Bcr-Abl expression does not appear to affect DNA replication in untreated progenitors, and thus Bcr-Abl expression does not provide an apparent competitive edge without HU.
HU Treatment Accelerates and Increases the Penetrance of Bcr-Abl-Mediated Leukemias
We addressed whether HU treatment of mice with Bcr-Abl
+ progenitors promotes leukemogenesis in vivo. For each experiment, aliquots from a single pool of MSCV-Bcr-Abl (or vector)-transduced BALB/c BM stem cells were transplanted into multiple syngeneic recipients. For all recipients, ∼3% of peripheral B cells were GFP+ at 2 wk post-BMT, at which point half of the mice were switched to water containing HU. By treating with HU starting 2 wk post-BMT, potential effects of HU on retroviral infection efficiency or homing to BM niches are avoided. Strikingly, all of the HU-treated mice developed a pre-B ALL-like disease within 2 mo (Figure 7A) preceded by increases in the percentages of Bcr-Abl
+ cells in the B220+ subset (unpublished data). In contrast, mock-treated mice that received Bcr-Abl-transduced stem cells developed leukemia with much lower penetrance (3 of 8) and longer latency periods. HU treatment also enhanced leukemogenesis induced by p210 Bcr-Abl, although interestingly, these mice all succumbed to CML-like disease (Figure 7B). Thus, in our model system, p190 Bcr-Abl causes B-ALL and p210 causes CML, thus closely modeling the human diseases caused by the respective translocation products. Given that HU treatment promotes the expansion of Bcr-Abl
+ progenitors within 5 d (Figure 6A), it is improbable that HU-promoted mutagenesis alone accounts for more rapid disease, although increased mutation accumulation may be a contributing factor. Furthermore, as HU similarly promotes leukemogenesis in analogous experiments using donor BM from Rag2
−/− mice, in which case mature T and B cells were nearly undetectable in recipients (Figure S3), HU-mediated increases in Bcr-Abl-dependent leukemogenesis does not appear to result from immunosuppression.
Figure 7 HU Treatment Enhances Bcr-Abl-Mediated Leukemogenesis
Bcr-Abl BMT recipients were generated and treated with HU as described in Figure 6.
(A–B) Kaplan-Meier curves for vector (A), p190 Bcr-Abl (A), or p210 Bcr-Abl (B) transduced stem cell-transplanted mice with or without HU treatment are shown. The curves for p190 and p210 Bcr-Abl with HU are statistically different from the respective Bcr-Abl-untreated curves (p < 0.001 and p < 0.05, respectively). For (A) and (B), the average percent GFP+ in peripheral blood cells was 3%, and mice with similar percentages were segregated into the HU-treated and untreated groups. Mice were sacrificed when moribund, all with splenomegaly and massive increases in B220+GFP+ (for p190) or GR-1+GFP+ (for p210) cells in the spleen and peripheral blood.
(C) Peripheral blood from morbid p190 Bcr-Abl-BMT mice (with or without HU, as in Figure 6) or from a healthy vector BMT mouse was analyzed for B220 and GR-1 expression by flow cytometry.
(D) BM cells from morbid p190 BMT mice were analyzed for the expression of c-Kit, CD34, CD43, B220, with side scatter and forward scatter (reflects cell size) as indicated. The dashed oval indicates the typical position of normal BM pre/pro-B-cells. All leukemic BM cells in untreated Bcr-Abl recipients expressed B220 and CD43 as shown in the upper panels. Leukemic BM cells from HU treated mice generally expressed less B220 (despite B220 expression in peripheral cells; see (A) and displayed more c-Kit+ and many more CD34+ cells. Percentages shown are averages (± SD) for three mice for each group. Two Bcr-Abl
+ mice treated with HU that developed leukemia late (7 and 10 wk post-BMT) displayed an immunophenotype more similar to the Bcr-Abl-untreated mice.
Leukemic p190 Bcr-Abl recipient mice exhibited massive increases in pre-B-like large B220+ CD43+ cells, high leukocyte blast counts in peripheral blood, and splenomegaly (Figures 7C and S4). Pathological characterization of spleens and other tissues determined that the lymphoproliferative diseases arising in all HU and the three non-HU-treated mice were large cell leukemias, with frequent involvement of liver, lung, and spleen (unpublished data). Spleens contained a mixture of leukemic cells, with most exhibiting lymphoid characteristics and a subset exhibiting a myeloid phenotype (unpublished data). For most HU (relative to non-HU)-treated mice, the predominant GFP+ (i.e., Bcr-Abl
+) population in the BM and spleen contained a more substantial subpopulation of cells expressing the progenitor markers c-Kit and CD34 (Figure 7D). We transplanted spleen or BM cells (103–105) from morbid leukemic HU- and non-HU-treated Bcr-Abl BMT recipients into BALB/c recipients, and most recipients for both groups developed leukemia, underscoring the malignant nature of the leukemias (Figure S5). Thus, by providing a competitive advantage to Bcr-Abl
+ progenitor cells (as shown ex vivo and in vivo), HU treatment promotes Bcr-Abl-mediated ALL and CML.
The Inhibition of p53 Provides a Competitive Advantage to DKO BM Progenitors
While our experiments provide strong evidence that backgrounds of impaired DNA replication enhance the competitive expansion of Bcr-Abl-expressing cells, we wanted to test if this phenomenon can be applicable to other oncogenic events. The p53 tumor suppressor gene is mutated in over half of human tumors, and mutation of p53 has been shown to disrupt checkpoints that prevent cell cycle progression in the face of various cellular stresses [25]. Mutation of p53 in human cancers frequently results in the expression of a dominant negative mutant p53 protein (DNp53) that forms inactive multimers with wild-type p53. Given the known ability of DNp53 to alleviate checkpoint-dependent cell cycle arrests, we asked whether the expression of DNp53 (175H) could provide a competitive advantage for E2f1/E2f2 mutant progenitor expansion. In addition, we engineered a retrovirus to express the oligomerization domain of p53 (DDp53), which also inhibits p53 activity [26]. Strikingly, MSCV-mediated expression of DNp53 or DDp53 in hematopoietic progenitors from DKO, but not wild-type, mice resulted in a dramatic selective advantage for these cells after only 2 wk post-BMT (Figure 8A). Again, the percentages of peripheral blood cells expressing vector were very similar for wild-type and DKO cells. Thus, the inhibition of p53 provides a competitive advantage during hematopoiesis in DKO, but not wild-type, BM.
Figure 8 The Inhibition of p53 Provides a Competitive Advantage for DKO, but Not Wild-Type, BM Progenitors
(A) Wild-type or DKO BM cells were transduced with MSCV-DNp53, MSCV-DDp53, or vector, and transplanted into lethally irradiated BALB/c recipients. At 2 wk post-BMT, the percentage of GFP+ cells among peripheral blood GR-1+ cells was determined by flow cytometry. The contribution of GFP+ cells after 2 wk relative to the initial infection efficiency is shown. Initial infection efficiencies for each virus were similar for the two BM genotypes.
(B) Genetic disruption of p53 provides a proliferative advantage in the DKO but not wild-type background. The indicated mixtures of freshly harvested BM were used to reconstitute the hematopoietic system of lethally irradiated recipient BALB/c mice. At 6 wk post-BMT, the percentage of GFP+ nucleated cells in peripheral blood was determined by flow cytometry and fold expansion was determined relative to the initial mixture.
We used genetic disruption of p53 [27] as an alternative to retroviral DNp53 expression. We bred the p53 mutation into the E2f1/E2f2 mutant and BALB/c backgrounds together with a ubiquitously expressed GFP transgene [28]. In competitive BMT experiments using wild-type BM in competition with p53
−/−GFP+ BM (93:7 ratio), p53
−/−GFP+ BM contributions to hematopoiesis were similar to its percentage within the original BM transplanted (Figure 8B), indicating that p53 mutation does not provide an immediate competitive advantage during hematopoiesis. In contrast, DKO/p53
−/−GFP+ BM progenitors exhibited significant competitive expansion relative to DKO progenitors, and this competitive expansion was completely abrogated by the inclusion of E2f1
+
2
+ competitors (Figure 8B; compare second and third columns). Thus, as in the Bcr-Abl experiments shown in Figure 2, the increased expansion of DKO/p53
−/− cells in a DKO background relative to p53
−/− cells in a wild-type background is not due to cell autonomous contributions of E2f1/E2f2 loss, but instead to the poor competition provided by DKO hematopoiesis.
Discussion
Given that deregulated cell cycling is considered a hallmark of tumorigenesis [29], it is natural to assume that contexts that promote proliferation would promote cancer. Studies in search of connections between genetic polymorphisms, dietary factors, and environmental exposures that predispose to cancer are often predicated on the assumption that increased cancer rates in part result from the enhancement of proliferation by these conditions [30–33]. We show that the opposite can be true. Our data indicate that the reduced ability of progenitor cell pools to proliferate actually promotes tumorigenesis, in that cells acquiring mutations that improve proliferation outcompete poorly replicating competitors. In contrast, a replication-competent pool is inherently tumor-suppressive, and otherwise oncogenic mutations may not be favored.
We have shown that within a wild-type, healthy progenitor cell pool, Bcr-Abl expression or p53 inhibition does not provide an apparent selective advantage. In contrast, in contexts of impaired DNA replication of blood progenitors, expression of Bcr-Abl or inhibition of p53 provides an immediate advantage within progenitor pools. The expression of Bcr-Abl bypasses the S phase blocks imposed by E2f mutation or HU treatment, consistent with the known ability of Bcr-Abl to abrogate cell cycle checkpoints. As Bcr-Abl expression provides an immediate competitive advantage following HU treatment ex vivo, the advantage appears to result from the better proliferation of Bcr-Abl
+ cells relative to untransduced peers, not effects of HU on mutagenesis or the immune system, or other potential complications found in vivo. Most importantly, replication-impaired contexts (both genetic and chemotherapeutic) can substantially promote Bcr-Abl-dependent leukemias. That Bcr-Abl-expressing cells with restored S phase progression outcompete untransduced cells with highly impaired replication provides a logical explanation for the promotion of leukemogenesis by HU treatment and E2f1/E2f2 loss. Indeed, the promotion of Bcr-Abl-dependent leukemias by E2F mutation is not due to cell autonomous contributions of E2f1/E2f2 loss, as E2f1
+
2
+ competitors virtually eliminate leukemias resulting from Bcr-Abl expression in DKO progenitors. Thus, healthy competitors are potently tumor-suppressive.
The presented results support a model whereby the reduced proliferation of blood progenitors provides a powerful selection for oncogenic mutations that improve cell cycle progression. In essence, impeded proliferation of progenitor cells provides for poor competition, promoting the expansion of progenitors that acquire oncogenic mutations which relieve cell cycle blocks (Figure 9). In addition, the ability of some oncogenic mutations to impede apoptosis resulting from checkpoint activation might further contribute to a net competitive advantage. It has been previously proposed that carcinogen-mediated inhibition of cell survival or proliferation can select for oncogenically mutated cells resistant to these conditions ([34] and references therein). Analogously, we show that conditions that impair DNA replication in the entire population of progenitor cells select for oncogenic mutations that initiate tumorigenesis. In contrast, oncogenic mutations prove less advantageous or even disadvantageous in a normal hematopoietic system, as cell cycle progression in the progenitor pool is already efficient, and checkpoints should not be activated. Finally, contexts that involve nucleotide deprivation and DNA damage, and thus impeded DNA replication, may enhance mutation accumulation, and mutator phenotypes should synergize with poor competition to promote tumorigenesis. Thus, these conditions provide not only an increased chance for the acquisition of initiating mutations, but also a poorly competitive environment that favors the expansion of these oncogenically mutated cells. In this regard, it is unlikely that all conditions that globally impair cell cycle progression in a stem cell pool will promote tumorigenesis, particularly if these impaired contexts cannot be easily overcome by oncogenic mutation or do not sufficiently promote mutation accumulation. Moreover, while impaired replication can clearly select for the expression of oncogenes such as Bcr-Abl and DNp53, other oncogenic events (like Bcl2 overexpression) may not be selected for and may even be selected against.
Figure 9 The Poor Competition Model
In a wild-type, healthy progenitor cell pool, potentially oncogenic mutations (represented as a red “X” in a cell) will often provide little or no advantage in the short term, and may even be disadvantageous (top images). In contrast, in a replication-impaired hematopoietic system (bottom images), acquisition of some oncogenic mutations should provide a selective advantage by partly or fully restoring cell cycle progression, even when the same mutation might be disadvantageous in a wild-type pool. The preferential expansion of the mutated population increases the target size and thus the chances for additional mutations and cancer. In this figure, the rectangle indicates the limited niche size, implying competition for contacts and growth factors. The thickness of curved arrows indicates proliferative ability.
The low leukemia incidence that we observed in the replication competent-backgrounds may seem at odds with previous reports of efficient induction of CML-like disease in recipients of MSCV-Bcr-Abl-transduced BM cells using high-efficiency infections [35–37]. So why would MSCV-Bcr-Abl transduction result in rapid and penetrant onset of leukemia with high- but not low-efficiency infection, given that normal hematopoietic stem and progenitor cells, and thus healthy competitors, are still present in both cases? One possibility is that the resulting high percentage of cells expressing Bcr-Abl could result in substantial autocrine stimulation, which may not reflect the normal context of a rare Bcr-Abl fusion event. Indeed, expression of Bcr-Abl in cytokine-dependent cell lines leads to growth factor-independent proliferation associated with autocrine production of interleukin-3 (IL-3) and granulocyte/monocyte-colony stimulating factor [38–40], and increased levels of IL-3 are detected in serum in murine models of CML [36,37]. Importantly, retroviral expression of IL-3 by BM cells is sufficient to induce a myeloproliferative syndrome in mice [41–43]. Furthermore, Bcr-Abl
+ cells secrete inhibitors of normal hematopoiesis that induce apoptosis in normal but not Bcr-Abl
+ cells [44]. Autocrine stimulation by Bcr-Abl-expressing cells and inhibition of normal hematopoiesis should have a far greater impact when a large percentage of cells express Bcr-Abl, as compared to the physiological context of an initiating mutation leading to the expression of the oncogene in a single clone. Nonetheless, these activities could still be very relevant to the progression of Bcr-Abl
+ leukemias, in that once a Bcr-Abl
+ clone has reached some critical mass, the clone may promote its own competitive expansion even in the face of normal hematopoietic competitors. Finally, high-efficiency transduction of progenitors should increase the chance of secondary mutations that promote Bcr-Abl-dependent leukemogenesis, including gene alterations mediated by retroviral integrations.
An important point is that most of the recipients of control (E2f1
+
2
+ or wild-type without HU) progenitors transduced with Bcr-Abl did not develop leukemia. Indeed, for surviving mice, GFP expression in peripheral blood became undetectable (unpublished data), indicating that Bcr-Abl expression is disfavored in stem cell pools in the long-term. The facts that (1) the same pool of Bcr-Abl-transduced BM progenitors was transferred into recipients that subsequently were mock- or HU-treated, and (2) all HU-treated Bcr-Abl recipients rapidly developed leukemia, indicate that most of the nonleukemic mice in the control group did receive Bcr-Abl
+ progenitors capable of causing leukemia, but that these cells were not maintained. In contrast, HU treatment clearly promoted the expansion of Bcr-Abl
+ cells, stimulating leukemogenesis. Analogously, Bcr-Abl, IgH-Bcl2, and LMO2-TCR translocations are frequently found in nonleukemic individuals who do not subsequently develop leukemias [45]. Based in part on the detection of Bcr-Abl in peripheral blood of healthy people for several months to years, an argument has been made that Bcr-Abl translocations must occur frequently in HSCs of people that never develop leukemia [46]. In light of the data presented here, we suggest that the relative inability of progenitor cells possessing Bcr-Abl translocations to competitively expand and to acquire additional mutations under most contexts limits the development of leukemia.
Our result that HU promotes Bcr-Abl-dependent leukemias may seem surprising, given that HU is used for cytoreduction of CML cells in patients. However, in human disease the leukemic cells are already established and dominant, while our experiments are more relevant to the context of leukemia initiation. Thus, our use of HU may be less pertinent to the use of this drug to treat CML, but can be viewed as a general means to suppress S phase progression in the hematopoietic system, which is relevant to the effects of clinically used drugs or other conditions that inhibit DNA replication in progenitor populations. Indeed, a number of contexts that have been presumed or demonstrated to reduce dNTP production are associated with increased cancer rates, including folate deficiency [14] and polymorphisms resulting in decreased thymidylate synthase expression [47]. These contexts of dNTP restriction are thought to promote cancer in part by increasing mutation rates, and our results suggest that poorly competitive stem cell pools also play an important role. While links between Bcr-Abl
+ CML or ALL incidence and dietary variables or polymorphisms affecting dNTP metabolism have not been sufficiently examined, based on our demonstration that the inhibition of dNTP synthesis by HU promotes Bcr-Abl-dependent leukemias, such studies are warranted.
HU induces replication-associated double-stranded DNA breaks in cultured cells [15,16], although whether HU induces DNA damage or mutagenesis in vivo is still unresolved. Long-term treatment of sickle cell anemia patients with HU does not result in increased mutation rates or cancer formation [48]. However, some studies have found an association between cytoreductive HU treatment of myeloproliferative disorders and the development of acute leukemias, including a high proportion with 17p deletion [49,50]. Still, while commonly used chemotherapeutic antimetabolites such as HU and methotrexate do not appear to be substantially leukemogenic when used as single agents [48,51], combinations of antimetabolites with etoposide are associated with higher incidences of treatment-related leukemias relative to etoposide alone [17]. While clinical experiences are insufficiently controlled to conclude that antimetabolite therapies contribute to secondary leukemias, our demonstration that a chemotherapeutic agent that inhibits dNTP synthesis can promote leukemia suggests that this issue needs to be further addressed in the clinic. The inhibition of DNA replication by antimetabolites may promote the competitive expansion of progenitor cells acquiring etoposide-induced oncogenic mutations, much as HU promotes the expansion of Bcr-Abl-expressing progenitors. Thus, conditions that generally impair DNA replication, such as drugs that inhibit dNTP synthesis, may function as tumor promoters for oncogenically initiated cells by creating a poorly competitive environment. This view contrasts with the generally described function of tumor promoters as enhancers of proliferation.
Finally, while numerous properties of E2F1 and E2F2, including gene repression and the promotion of apoptosis and DNA repair, have been suggested to underlie observed increases in tumors in E2f1 and E2f2 mutant mice (reviewed in [11,12]), we demonstrate that the promotion of Bcr-Abl-dependent leukemias and the expansion of p53 mutant cells by E2f1/E2f2 disruption result from non-cell autonomous consequences of E2f loss. According to our model, E2f1/E2f2 mutations at least in part predispose mice to tumors, since the E2f genes are disrupted in all cells in the mouse, thus providing a poorly competitive progenitor population. In contrast, loss-of-function mutations in E2f1 or E2f2 in an individual cell would not be expected to provide an advantage to the cell, and thus would not contribute to tumorigenesis, thus perhaps explaining the paucity of such mutations in human tumors. Poor competition should select for oncogenic mutations only when the entire population of competing cells are similarly debilitated.
In summary, we have demonstrated that impaired DNA replication in hematopoietic progenitor pools contributes to a powerful selection for mutations (such as Bcr-Abl translocation) that improve cell cycle progression, providing a selective advantage over the poor competition offered by replication-debilitated progenitors and thus promoting leukemogenesis. As cancer development has attributes clearly reminiscent of Darwinian evolution, it is logical that it should become easier for a mutant cell clone to be the “fittest” as the quality of the competition declines. Thus, as with the natural selection of species, poor fitness selects for adaptive mutations in stem cell pools.
Materials and Methods
Mice
The generation of the E2f1 and E2f2 mutant mice (backcrossed seven times into BALB/c) has been previously described [52,53]. BALB/c and p53
−/− mice were purchased from Jackson Labs (Bar Harbor, Maine, United States). GFP transgenic mice were the generous gift of Dr. Brian Schaefer, and were backcrossed with p53 mutation into the BALB/c background. The University of Colorado Health Sciences Center Animal Care and Use Committee approved all mouse experiments.
Retroviral constructs
The retroviral MSCV-iresGFP vectors expressing either p190 or p210 Bcr-Abl were provided by Drs. Zhonghan Dai and Warren Pear. MSCV-Bcl2-iresGFP was provided by Dr. Scott Lowe. To construct similar retroviral vectors expressing DNp53 and DDp53, the p53Arg175His cDNA fragment from plasmid pCMV-NEO pC53–175 (provided by Dr. Bert Vogelstein) and DDp53 cDNA from pLXSNp53DD (provided by Dr. Xiao-Fan Wang) were cloned into the MSCV-iresGFP vector.
Retroviral transduction and bone marrow transplantation
MSCV viruses were prepared by transient transfection of Phoenix-E cells together with pCL-Eco, and titered on fibroblasts using GFP as the marker, so that similar multiplicities of infection for oncogene and vector viruses were used. Donor DKO, E2f1
+
2
+, or wild-type mice were injected intraperitoneally with 100 mg/kg 5-fluorouracil (Sigma, St. Louis, Missouri, United States) 3 d prior to BM harvest. Harvested donor BM cells were enriched for c-Kit+ cells by MACS (Miltenyi Biotec, Bad Gladbach, Germany) and then cultured overnight in IMDM plus 15% defined fetal bovine serum (HyClone, South Logan, Utah, United States) and 0.1% bovine serum antigen (Sigma), with the cytokines 100 ng/ml human stem cell factor (gift from Dr. Chris Hogan), 100 ng/ml human IL-11 (PeproTech, Rocky Hill, New Jersey, United States),1.5 ng/ml mouse IL-3 (PeproTech), and 100 ng/ml hIL-6 (PeproTech). Cells were transduced with MSCV-containing supernatants (∼1/5 volume) supplemented with 6 μg/ml Polybrene (Sigma) for 6 h. Cells were then transplanted intravenously into lethally (1,200 Rads for E2F experiments) or sublethally (600 Rads for HU experiments) irradiated BALB/c mice. A subset of the transduced cells was further cultured for 2 d, and the percentage of GFP+ cells was determined by flow cytometry in order to determine the initial infection efficiency. Ex vivo culturing of Bcr-Abl-transduced cells for Figure 5 was performed under the same culture conditions.
Flow cytometry and cell cycle analysis
Single-cell suspensions were washed in PBS containing 5% FBS (FBS-PBS). About 106 cells were stained in 20 μl of antibody solution (1:100 dilution of each antibody) for 30 min at room temperature. Cells were washed once with 1 ml of FBS-PBS and resuspended in 400 μl of PBS for flow cytometric analysis. The following PharMingen (San Diego, California, United States) antibodies against mouse were used: phycoerythrin (PE)-linked anti-B220, PE-anti-Ter119, PE-anti-GR-1, PE-anti-CD3, PE-anti-CD4, PE-anti-Flk2/Flt3, and PE-anti-CD43; and allophycocyanin (APC)-linked anti-B220, APC-anti-Thy1.2, and APC-anti-CD117; and biotin-linked anti-GR-1, biotin-anti-CD34, biotin-anti-IgM, and PE-cyanine (PE-Cy7)-linked anti-biotin. APC-anti-Sca1 antibodies were purchased from eBioscience (San Diego, California, United States). Fluorescence was detected with a Cytomics FC 500 (Beckman Coulter, Allendale, New Jersey, United States) cytometer. For BrdU incorporation analyses, mice were injected intraperitoneally with BrdU (1 mg per 25 g of body weight; Roche, Indianapolis, Indiana). After 2 h, pooled spleen and BM cells were purified using a MoFlo (Cytomation, Fort Collins, Colorado, United States) cell sorter for GFP+ or GFP− subsets of progenitors: B220+ plus GR-1+ (B-cell and myeloid lineages) for Figure 3 or Lin− for Figure 6. Immunofluorescence, flow cytometric analyses, and calculations of the length of S phase were performed as described previously [7,20].
Pathology
Mice were monitored for disease development, as judged by increasing percentages of GFP+ cells with blast morphology in peripheral blood of transplanted animals, as well as symptoms, such as abnormal gait and labored breathing. Moribund animals were sacrificed and examined for tumors. Touch preps and tissue sections were obtained from spleen and liver.
Supporting Information
Figure S1 The Loss of E2f1 and E2f2 Promotes the Competitive Repopulation of the B Cell Lineage by Bcr-Abl-Expressing Progenitors
C-kit+ cells from either E2f1
+
2
+ or DKO donor mice were transduced at low- and high-efficiency and transplanted as in Figures 2 and 4 (same experiments). The graph shows the changes in the percentages of GFP+ cells in the peripheral B220+ lineage of Bcr-Abl transduced E2f1
+
2
+E2F1+ recipients (left), Bcr-Abl transduced DKO recipients (middle) and Bcr-Abl transduced DKO recipients supplemented with untransduced E2f1
+
2
+ competitor progenitors (right). Each curve follows the individual mouse used in experiments with low (upper) and high (lower) transduction efficiency. Curves end if the mouse develops leukemia and is sacrificed prior to the next tail blood analysis. Surviving mice from the Bcr-Abl→E2f1
+
2
+ and Bcr-Abl→DKO + E2f1
+
2
+ competitors groups were bled at 180 and 240 d post-BMT (for low- and high-transduction experiments, respectively), and no GFP+ cells were detected above background in any recipient (unpublished data).
(54 KB PPT).
Click here for additional data file.
Figure S2 \Inhibition of Apoptosis by Bcl2 Overexpression Does Not Provide a Competitive Advantage to DKO Relative to E2f1
+
2
+ Progenitors
E2f1
+
2
+ or DKO BM cells were transduced with MSCV viruses expressing Bcl2 (with iresGFP), and then transplanted into lethally irradiated BALB/c mice. The mice were sacrificed at 3 wk post-transplant, and flow cytometric analysis of GFP and GR-1 expression in BM was performed. Initial infection efficiencies for both genotypes were similar. MSCV-expressed Bcl2 is functional, given that it conferred increased survival to BM progenitor cultures from E2f1
+
2
+ and DKO mice (unpublished data).
(54 KB PPT).
Click here for additional data file.
Figure S3 HU Treatment Increases p190 Bcr-Abl-Mediated Leukemogenesis in the Rag2
−/− Background
(A) C-Kit+ cells from Rag2
−/− BM (BALB/c congenic) were transduced with MSCV-p190 Bcr-Abl, and transplanted into lethally irradiated BALB/c recipients as in Figure 2. At 2 wk post-BMT, half of each group was switched to water containing HU. The average percent GFP+ in peripheral blood cells was 10%, and mice with similar percentages were segregated into the HU treated and untreated groups. Kaplan-Meier curves for transplanted mice are shown. Mice were sacrificed when moribund; all with splenomegaly and massive increases in GFP+ cells in the spleen and peripheral blood.
(B) Representative flow cytometric plots are shown for recipients of Bcr-Abl-transduced Rag2
−/− BM. Peripheral blood cells of the transplanted mice were analyzed for the expression of CD3 and IgM 2 wk post-BMT (before HU treatment). Cells expressing these markers (T and B cells) were almost undetectable in transplanted animals (left). Spleen cells from morbid Bcr-Abl/Rag2
−/− (with and without HU) mice were analyzed for the expression of GFP, B220, Ter119, GR-1, and Thy1.2. Leukemias arising in untreated and HU-treated mice did not express B220 and Thy1.2 (unpublished data), but displayed many GR-1+ and Ter119+ cells (right; gated on GFP+ blast population). A similar blast population expressing both myeloid (GR-1+) and erythroid (Ter119+) markers was present in the one morbid non-HU treated recipient (unpublished data).
(114 KB PPT).
Click here for additional data file.
Figure S4 Recipients of Bcr-Abl-Transduced Progenitors That Developed Leukemias With or Without HU Treatment Exhibited Splenomegaly and High Numbers of Leukocytic Blasts in Peripheral Blood
Recipients of vector or p190 Bcr-Abl transduced progenitors were treated or not with HU as in Figure 7 and sacrificed when morbid. Peripheral blood from the tail vein was drawn before sacrifice and analyzed for a complete blood count using a Cell-Dyn 1700 System. HU treatment of vector recipients did result in modest but significant (p < 0.05) leukopenia. For Bcr-Abl recipients with or without HU treatment, high peripheral leukocyte counts were evident, most of which were blasts (also confirmed microscopically), but the difference between HU-treated and untreated was not statistically significant. Following sacrifice, spleens were weighed. The increased spleen weight in HU-treated relative to untreated recipients of Bcr-Abl transduced progenitors is significant (p < 0.05). The average weight of a spleen from a healthy mouse is about 0.1 g.
(40 KB PPT).
Click here for additional data file.
Figure S5 Leukemias That Developed in Untreated and HU-Treated Animals Were Transferable
BM cells (5 × 103, 5 × 104, and 5 × 105) from morbid leukemic HU and non-HU treated Bcr-Abl BMT recipients were transplanted into BALB/c recipients. The fraction of transplanted mice that developed secondary leukemias is shown, and the time post-transplant that mice became morbid indicated in parentheses. Similar results were obtained with spleen cells from a different set of leukemic donors.
(36 KB PPT).
Click here for additional data file.
Accession Numbers
The GenBank (http://www.ncbi.nlm.nih.gov/Genbank/) accession numbers of the genes discussed in this paper are Bcr-Abl (X06418), E2f1 (L21973), E2f2 (AK087452), and p53 (P04637).
JD is supported by grants from the National Institutes of Health (NIH RO1 CA77314) and the American Cancer Society (RSG LIB-101051), and by a Scholar Award from the Leukemia and Lymphoma Society. CP is supported by a National Research Service Award (T32 CA 82086). RDC is supported by NIH grants U42 RR14905 and R01 CA89140. We would like to thank Drs. David Johnson, Erik Knudsen, and Rachel Rempel for their critical review of this manuscript, and Dr. Zonghan Dai for advice. We also thank K. Helm, C. Childs and M. Ashton of the Cancer Center Flow Cytometry Core (supported by grant 2 P30 CA 46934–09). JD dedicates this paper in the memory of Monika Weber, Kay Sutherland, and Atilio Ramos, relatives who succumbed to their leukemias.
Competing interests. The authors have declared that no competing interests exist.
Author contributions. GB and JVD conceived and designed the experiments. GB, AM, and CCP performed the experiments. GB and JVD analyzed the data. RDC performed pathological analyses of leukemias. GB and JVD wrote the paper.
Citation: Bilousova G, Marusyk A, Porter CC, Cardiff RD, DeGregori J (2005) Impaired DNA replication within progenitor cell pools promotes leukemogenesis. PLoS Biol 3(12): e401.
Abbreviations
ALLacute lymphoblastic leukemia
BMbone marrow
BMTbone marrow transplantation
BrdUbromodeoxyuridine
CMLchronic myelogenous leukemia
DDp53oligomerization domain of p53
DKOE2f1−/−E2f2−/−
DNp53dominant negative p53 mutant
E2f1+2+E2f1+/−E2f2+/−
GFPgreen fluorescent protein
HSChematopoietic stem cell
HUhydroxyurea
ILinterleukin
Lin−lineage negative
MSCVmouse stem cell virus
PIpropidium iodide
SEstandard error
==== Refs
References
Passegue E Jamieson CH Ailles LE Weissman IL Normal and leukemic hematopoiesis: Are leukemias a stem cell disorder or a reacquisition of stem cell characteristics? Proc Natl Acad Sci U S A 2003 100 Suppl 1 11842 11849 14504387
Jamieson CH Ailles LE Dylla SJ Muijtjens M Jones C Granulocyte-macrophage progenitors as candidate leukemic stem cells in blast-crisis CML N Engl J Med 2004 351 657 667 15306667
Sawyers CL Chronic myeloid leukemia N Engl J Med 1999 340 1330 1340 10219069
Castor A Nilsson L Astrand-Grundstrom I Buitenhuis M Ramirez C Distinct patterns of hematopoietic stem cell involvement in acute lymphoblastic leukemia Nat Med 2005 11 630 637 15908956
Van Etten RA Studying the pathogenesis of BCR-ABL+ leukemia in mice Oncogene 2002 21 8643 8651 12476310
Skorski T BCR/ABL regulates response to DNA damage: The role in resistance to genotoxic treatment and in genomic instability Oncogene 2002 21 8591 8604 12476306
Li FX Zhu JW Hogan CJ DeGregori J Defective gene expression, S phase progression, and maturation during hematopoiesis in E2F1/E2F2 mutant mice Mol Cell Biol 2003 23 3607 3622 12724419
Wu L Timmers C Maiti B Saavedra HI Sang L The E2F1–3 transcription factors are essential for cellular proliferation Nature 2001 414 457 462 11719808
Yamasaki L Jacks T Bronson R Goillot E Harlow E Tumor induction and tissue atrophy in mice lacking E2F-1 Cell 1996 85 537 548 8653789
Zhu JW Field SJ Gore L Thompson M Yang H E2F1 and E2F2 determine thresholds for antigen-induced T-cell proliferation and suppress tumorigenesis Mol Cell Biol 2001 21 8547 8564 11713289
DeGregori J The genetics of the E2F family of transcription factors: Shared functions and unique roles Biochim Biophys Acta 2002 1602 131 150 12020800
Dimova DK Dyson NJ The E2F transcriptional network: Old acquaintances with new faces Oncogene 2005 24 2810 2826 15838517
Lightfoot TJ Roman E Causes of childhood leukaemia and lymphoma Toxicol Appl Pharmacol 2004 199 104 117 15313583
Ames BN DNA damage from micronutrient deficiencies is likely to be a major cause of cancer Mutat Res 2001 475 7 20 11295149
Lundin C Erixon K Arnaudeau C Schultz N Jenssen D Different roles for nonhomologous end joining and homologous recombination following replication arrest in mammalian cells Mol Cell Biol 2002 22 5869 5878 12138197
Saintigny Y Delacote F Vares G Petitot F Lambert S Characterization of homologous recombination induced by replication inhibition in mammalian cells EMBO J 2001 20 3861 3870 11447127
Pui CH Relling MV Topoisomerase II inhibitor-related acute myeloid leukaemia Br J Haematol 2000 109 13 23 10848777
Christensen JL Weissman IL Flk-2 is a marker in hematopoietic stem cell differentiation: A simple method to isolate long-term stem cells Proc Natl Acad Sci U S A 2001 98 14541 14546 11724967
Spangrude GJ Brooks DM Mouse strain variability in the expression of the hematopoietic stem cell antigen Ly-6A/E by bone marrow cells Blood 1993 82 3327 3332 8241503
Begg AC McNally NJ Shrieve DC Karcher H A method to measure the duration of DNA synthesis and the potential doubling time from a single sample Cytometry 1985 6 620 626 4064842
Penit C In vivo thymocyte maturation. BUdR labeling of cycling thymocytes and phenotypic analysis of their progeny support the single lineage model J Immunol 1986 137 2115 2121 3093565
Chabner BA Longo DL Cancer chemotherapy and biotherapy: Principles and practice 2001 Philadelphia Lippincott Williams and Wilkins 1,140 pp
Shechter D Costanzo V Gautier J Regulation of DNA replication by ATR: Signaling in response to DNA intermediates DNA Repair (Amst) 2004 3 901 908 15279775
Calabretta B Perrotti D The biology of CML blast crisis Blood 2004 103 4010 4022 14982876
Wallace-Brodeur RR Lowe SW Clinical implications of p53 mutations Cell Mol Life Sci 1999 55 64 75 10065152
Gottlieb E Haffner R von Ruden T Wagner EF Oren M Down-regulation of wild-type p53 activity interferes with apoptosis of IL-3-dependent hematopoietic cells following IL-3 withdrawal EMBO J 1994 13 1368 1374 8137820
Jacks T Remington L Williams BO Schmitt EM Halachmi S Tumor spectrum analysis in p53-mutant mice Curr Biol 1994 4 1 7 7922305
Schaefer BC Schaefer ML Kappler JW Marrack P Kedl RM Observation of antigen-dependent CD8(+) T-cell/ dendritic cell interactions in vivo Cell Immunol 2001 214 110 122 12088410
Hanahan D Weinberg RA The hallmarks of cancer Cell 2000 100 57 70 10647931
Luch A Nature and nurture—Lessons from chemical carcinogenesis Nat Rev Cancer 2005 5 113 125 15660110
Milner JA Incorporating basic nutrition science into health interventions for cancer prevention J Nutr 2003 133 3820S 3826S 14608120
Choi SW Mason JB Folate status: Effects on pathways of colorectal carcinogenesis J Nutr 2002 132 2413S 2418S 12163703
Ryan BM Weir DG Relevance of folate metabolism in the pathogenesis of colorectal cancer J Lab Clin Med 2001 138 164 176 11528369
Blagosklonny MV Carcinogenesis, cancer therapy and chemoprevention Cell Death Differ 2005 12 592 602 15818400
Pear WS Miller JP Xu L Pui JC Soffer B Efficient and rapid induction of a chronic myelogenous leukemia-like myeloproliferative disease in mice receiving P210 bcr/abl-transduced bone marrow Blood 1998 92 3780 3792 9808572
Zhang X Ren R Bcr-Abl efficiently induces a myeloproliferative disease and production of excess interleukin-3 and granulocyte-macrophage colony-stimulating factor in mice: a novel model for chronic myelogenous leukemia Blood 1998 92 3829 3840 9808576
Li S Ilaria RL Million RP Daley GQ Van Etten RA The P190, P210, and P230 forms of the BCR/ABL oncogene induce a similar chronic myeloid leukemia-like syndrome in mice but have different lymphoid leukemogenic activity J Exp Med 1999 189 1399 1412 10224280
Anderson SM Mladenovic J The BCR-ABL oncogene requires both kinase activity and src-homology 2 domain to induce cytokine secretion Blood 1996 87 238 244 8547647
Hariharan IK Adams JM Cory S Bcr-abl oncogene renders myeloid cell line factor independent: Potential autocrine mechanism in chronic myeloid leukemia Oncogene Res 1988 3 387 399 3147434
Sirard C Laneuville P Dick JE Expression of bcr-abl abrogates factor-dependent growth of human hematopoietic M07E cells by an autocrine mechanism Blood 1994 83 1575 1585 8123848
Chang JM Metcalf D Lang RA Gonda TJ Johnson GR Nonneoplastic hematopoietic myeloproliferative syndrome induced by dysregulated multi-CSF (IL-3) expression Blood 1989 73 1487 1497 2653457
Just U Katsuno M Stocking C Spooncer E Dexter M Targeted in vivo infection with a retroviral vector carrying the interleukin-3 (multi-CSF) gene leads to immortalization and leukemic transformation of primitive hematopoietic progenitor cells Growth Factors 1993 9 41 55 8102237
Wong PM Chung SW Dunbar CE Bodine DM Ruscetti S Retrovirus-mediated transfer and expression of the interleukin-3 gene in mouse hematopoietic cells result in a myeloproliferative disorder Mol Cell Biol 1989 9 798 808 2565534
Lin H Monaco G Sun T Ling X Stephens C Bcr-Abl-mediated suppression of normal hematopoiesis in leukemia Oncogene 2005 24 3246 3256 15735695
Greaves MF Wiemels J Origins of chromosome translocations in childhood leukemia Nat Rev Cancer 2003 3 1 10
Matioli GT BCR-ABL insufficiency for the transformation of human stem cells into CML Med Hypotheses 2002 59 588 589 12376083
Skibola CF Smith MT Hubbard A Shane B Roberts AC Polymorphisms in the thymidylate synthase and serine hydroxymethyltransferase genes and risk of adult acute lymphocytic leukemia Blood 2002 99 3786 3791 11986237
Zimmerman SA Schultz WH Davis JS Pickens CV Mortier NA Sustained long-term hematologic efficacy of hydroxyurea at maximum tolerated dose in children with sickle cell disease Blood 2004 103 2039 2045 14630791
Furgerson JL Vukelja SJ Baker WJ O'Rourke TJ Acute myeloid leukemia evolving from essential thrombocythemia in two patients treated with hydroxyurea Am J Hematol 1996 51 137 140 8579054
Sterkers Y Preudhomme C Lai JL Demory JL Caulier MT Acute myeloid leukemia and myelodysplastic syndromes following essential thrombocythemia treated with hydroxyurea: High proportion of cases with 17p deletion Blood 1998 91 616 622 9427717
Kinder AJ Hassell AB Brand J Brownfield A Grove M The treatment of inflammatory arthritis with methotrexate in clinical practice: Treatment duration and incidence of adverse drug reactions Rheumatology (Oxford) 2005 44 61 66 15611303
Field SJ Tsai F-Y Kuo F Zubiaga AM Kaelin WG E2F-1 functions in mice to promote apoptosis and suppress proliferation Cell 1996 85 549 561 8653790
Murga M Fernandez-Capetillo O Field SJ Moreno B Borlado LR Mutation of E2F2 in mice causes enhanced T lymphocyte proliferation, leading to the development of autoimmunity Immunity 2001 15 959 970 11754817
|
16277552
|
PMC1283331
|
CC BY
|
2021-01-05 08:21:45
|
no
|
PLoS Biol. 2005 Dec 15; 3(12):e401
|
utf-8
|
PLoS Biol
| 2,005 |
10.1371/journal.pbio.0030401
|
oa_comm
|
==== Front
PLoS BiolPLoS BiolpbioplosbiolPLoS Biology1544-91731545-7885Public Library of Science San Francisco, USA 1627755310.1371/journal.pbio.0030404Research ArticleMicrobiologyIn VitroVirusesEpstein-Barr Virus Provides a New Paradigm: A Requirement for the Immediate Inhibition of Apoptosis Viral Bcl-2 Homologs in Latent InfectionAltmann Markus
1
Hammerschmidt Wolfgang [email protected]
1
1Department of Gene Vectors, GSF-National Research Center for Environment and Health, München, GermanySugden Bill Academic EditorUniversity of Wisconsin at MadisonUnited States of America12 2005 15 11 2005 15 11 2005 3 12 e4044 7 2005 27 9 2005 Copyright: © 2005 Altmann and Hammerschmidt.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
Virus Proteins Prevent Cell Suicide Long Enough to Establish Latent Infection
DNA viruses such as herpesviruses are known to encode homologs of cellular antiapoptotic viral Bcl-2 proteins (vBcl-2s), which protect the virus from apoptosis in its host cell during virus synthesis. Epstein-Barr virus (EBV), a human tumor virus and a prominent member of γ-herpesviruses, infects primary resting B lymphocytes to establish a latent infection and yield proliferating, growth-transformed B cells in vitro. In these cells, 11 viral genes that contribute to cellular transformation are consistently expressed. EBV also encodes two vBcl-2 genes whose roles are unclear. Here we show that the genetic inactivation of both vBcl-2 genes disabled EBV's ability to transform primary resting B lymphocytes. Primary B cells infected with a vBcl-2-negative virus did not enter the cell cycle and died of immediate apoptosis. Apoptosis was abrogated in infected cells in which vBcl-2 genes were maximally expressed within the first 24 h postinfection. During latent infection, however, the expression of vBcl-2 genes became undetectable. Thus, both vBcl-2 homologs are essential for initial cellular transformation but become dispensable once a latent infection is established. Because long-lived, latently infected memory B cells and EBV-associated B-cell lymphomas are derived from EBV-infected proapoptotic germinal center B cells, we conclude that vBcl-2 genes are essential for the initial evasion of apoptosis in cells in vivo in which the virus establishes a latent infection or causes cellular transformation or both.
The transformation of resting B-lymphocytes by a human tumor virus is shown to require vBcl-2 genes, which abrogate host cell apoptosis. These genes are not required once latent infection is established.
==== Body
Introduction
Apoptosis is a mechanism used by the infected cell as part of its antiviral response; cells can commit suicide as a direct response to viral infection prior to viral maturation. Therefore, regulating apoptosis during infection is a high priority for many viruses that frequently encode Bcl-2-like mechanisms to do so [1]. Cellular Bcl-2 was originally discovered as an oncogenic protein in follicular lymphomas and up to now many Bcl-2 family members have been identified that share distinct domains of high sequence homology. Bcl-2 family members include apoptosis-inhibitory proteins, such as Bcl-2 and Bcl-XL, or those that mediate proapoptotic functions exemplified by Bax. All viral Bcl-2 (vBcl-2) homologs identified so far block apoptosis. The major role of vBcl-2 proteins is therefore to prevent premature death of the host cell during virus production, which would otherwise reduce the amount of progeny virus [2,3]. Adenovirus E1B 19K, for example, is a minimal version of the cellular Bcl-2 prototype, which prevents premature cell death during productive infection. Bcl-2 and E1B 19K can functionally substitute for each other in the suppression of apoptosis during virion synthesis and oncogenic transformation [1,3]. While members of the γ-herpesvirus family carry one vBcl-2 gene in general, Epstein-Barr virus (EBV) encodes two Bcl-2 homologs, BALF1 and BHRF1. The BHRF1 protein clearly resembles Bcl-2 in its antiapoptotic function during in vitro assays [4], while the role of BALF1 is controversial [5,6]; however, no function has been assigned to either BALF1 or BHRF1 in the context of viral infection [1,7]. Interestingly, vBcl-2 homologs have not been recognized to contribute to viral latent infection directly [2,8,9], probably because viral latency can only be studied with very few viruses. In addition, functions involved in viral “fitness” often overlap with specific functions, which contribute directly to initiation and maintenance of viral latency in vivo [10–12].
EBV provides a unique in vitro model, which permits the dissection of viral contributions to latent infection. EBV can infect all cells of the B-cell lineage, but its main targets in vivo are naïve B cells and B cells that undergo affinity maturation during a germinal center reaction to establish long-term latent infections in memory B cells [13,14]. In vivo and in vitro EBV's latent state is characterized by the absence of virus synthesis and maintenance of the viral genome as plasmids in the infected cell. In vitro, EBV infects resting human B lymphocytes and transforms them into lymphoblastoid cell lines (LCLs), a process that is termed growth transformation and a hallmark of this virus. In LCLs, 11 so-called latent genes are consistently expressed. These are the EBV nuclear antigens EBNA1, EBNA2, EBNA-LP, EBNA3A, -B, -C, and the latent membrane proteins LMP1, LMP2A, and -B, and two noncoding RNAs [15]. Genetic and biochemical experiments demonstrate that several of these viral proteins directly contribute to growth transformation and latency of infected cells in vitro and in vivo, although only a subset of these 11 genes are expressed in the human host [16]. Typically, latent genes of EBV mimic cellular functions. For example, the viral proteins LMP1 and LMP2A target physiological signaling pathways, which are engaged by the B-cell and CD40 receptors upon contact with antigen and T helper cells, respectively. LMP1 and LMP2A are also expressed in EBV-positive B-cell lymphomas such as Hodgkin's disease and posttransplant lymphomas [13], which are two of several malignancies with which EBV has been associated [16,17]. LMP1 indirectly inhibits apoptosis by upregulating several cellular antiapoptotic genes [18], presumably through the induction of the NF-κB pathway [19–21]. LMP1 promotes cell proliferation [18,22] and scores as an oncogene in transgenic animals [23]. In LCLs, but not in EBV-associated tumors, LMP1 and LMP2A are transcriptionally upregulated by EBNA2, which is the first latent viral gene expressed after infection of B cells in vitro. EBNA2 is the key viral factor, which is essential for in vitro growth transformation of resting primary B cells [24,25], along with LMP1 [26], EBNA1 [15], EBNA3A, and -C [27]. This minimal set of five latent EBV genes has been regarded as sufficient to yield growth-transformed LCLs in vitro [15], although the requirements for EBNA3A, EBNA1, and LMP1 appear to be less stringent [28–30]. None of the known latent gene products of EBV directly regulate apoptosis, and little is known about viral genes expressed very early in primary B cells upon viral infection.
We analyzed the two vBcl-2 genes of EBV—BHRF1 and BALF1—and demonstrate that their gene products in addition to the known latent EBV genes are essential for the process of B-cell transformation. The vBcl-2 homologs are maximally expressed initially after infection but are neither expressed nor required once a latent infection is established. The very early expression of vBcl-2 genes prevents EBV-infected B cells from undergoing spontaneous programmed cell death and is mandatory to establish a latent infection and cause cellular transformation. We conclude that EBV employs a biphasic control—characterized by viral genes expressed transiently and those that are expressed stably—to initiate and maintain a latent infection.
Results
vBcl-2 Genes Are Essential for Growth Transformation of Resting B Cells but Dispensable for Proliferating B Blasts
In order to gain insight into the function of BALF1 and BHRF1, we constructed several viral mutants, which carry singly or dually inactivated alleles of BALF1 and BHRF1 (Figure 1A and 1B). Mutant virus stocks were generated as described [31]. Surprisingly, virus production was not impaired even in the case of the dually inactivated vBcl-2 mutants, and transcomplementation of either BALF1 or BHRF1 in virus-producing cells did not yield higher titers (data not shown). Virus stocks were quantified by infecting Raji cells, an established B-cell line [29]. Because our recombinant EBVs encode green fluorescence protein (GFP), we could measure the concentration of GFP-transducing virions as “green Raji units” (GRUs), which were in the range of 104–105/ml GRUs similar to wild-type 2089 EBV stocks [32]. Primary B cells were infected with serial virus dilutions to determine the number of GRUs required statistically to give rise to clonal LCLs [33]. Two GRUs were sufficient in the case of 2089 wild-type EBV, and the single BALF1– or BHRF1– mutants were equally capable of yielding clonal LCLs but at a slightly higher virus dose (Figure 1B–1D). Two independent viral mutants with inactivated vBcl-2 genes failed to generate LCLs, even at a dose of 104 GRUs per well (Figure 1C and 1D). Reconstruction of the BALF1 allele in the revertant BALF1+/BHRF1− virus reconstituted EBV's capacity to growth transform human B lymphocytes (Figure 1C and 1D).
Figure 1 Construction and Genealogy of EBV Mutants and Their Efficiency of B-Cell Growth Transformation
(A) Construction of the loss of function mutation in BALF1 and BHRF1. (B) Wild-type (WT) 2089 EBV is based on the genome of the EBV strain B95.8. The BALF1 and BHRF1 genes were inactivated through insertional mutagenesis with antibiotic resistance genes. Two vBcl-2 mutants 2636 and 2765 were constructed independently on the basis of the singly inactivated viruses, and the BALF1−/BHRF1− mutant 2636 was reverted to BALF1+/BHRF1−. (C) The numbers indicate the number of GRUs of the virus stocks, which were required to yield one clonal LCL for each virus noted in (B). (D) Efficiency of B-cell growth transformation. The different singly and dually vBcl-2− EBV virus mutants were compared with wild-type 2089 EBV for their efficiencies to growth transform primary B cells in limiting dilution assays. Forty-eight wells with 105 target B cells per well were infected with each virus dilution, and wells with proliferating cells were recorded 6 wk p.i. The data are graphed to identify the number of GRUs of the different virus isolates required to yield one LCL. The horizontal line at 30 wells positive for 48 wells plated identifies for a Poisson distribution the required number of GRUs shown on the x-axis. Infection of primary B cells with the vBcl-2− mutants 2636 and 2765 did not yield stable LCLs even with up to 104 GRUs per 105 cells per well. (E) Comparison of the efficiency of wild-type 2089 EBV and the BALF1−/BHRF1− mutant 2636 to yield stable, clonal LCLs in limiting dilution assays with primary B cells or activated B blasts as target cells. Activated B blasts readily gave rise to LCL clones with the BALF1−/BHRF1− mutant 2636 in contrast to the situation in (D), although about 200 GRUs were statistically required to establish clonal lines, which all could be further expanded and characterized (data not shown).
We determined whether infection of preactivated human B cells might overcome the failure of the vBcl-2− mutants to establish growth transformed LCLs from resting primary B cells. B blasts, which can be generated from resting primary B cells by cocultivation on CD40 ligand-expressing feeder cells in conjunction with IL-4, proliferate in vitro and express cellular antiapoptotic Bcl-2 family members, which are upregulated through the CD40 and JAK/Stat-induced signaling pathways [34,35]. The continued proliferation of B blasts is totally dependent on both activating stimuli. Upon removal of either CD40L or IL-4 signals, the B blasts cease to proliferate immediately and undergo apoptosis within 10 d (data not shown). Limiting dilution assays with activated B blasts as target cells and serially diluted BALF1−/BHRF1− 2636 virus (Figure 1B) resulted in clonal LCLs, which proliferated in the absence of either CD40L or IL-4 signals, indicating that preactivated B cells compensate for the functional defect of a vBcl-2− virus (Figure 1E). Once established, these LCLs infected with vBcl2− EBV proliferated normally and did not show an apparently different phenotype when compared with wild-type EBV-infected LCLs (data not shown).
vBcl-2 Genes Are Maximally Expressed Very Early After EBV Infection of Primary B Cells
Eleven EBV genes have previously been identified as being constitutively expressed in established LCLs. Neither BALF1 nor BHRF1 is among this group, however [15,36]. Latent EBV genes such as EBNA2 and LMP1 encode key regulators of cellular proliferation, which are required to initiate and maintain proliferation of lymphoblastoid cells in vitro [22,24–26,37]. Therefore, we set out to analyze the expression kinetics of BALF1 and BHRF1 in infected primary B cells. As shown in Figure 2, upon infection with the B95.8 prototype EBV strain, BHRF1 and BALF1 transcripts were readily detected within 24 h postinfection (p.i.) but not in cells infected at 4 °C or in uninfected cells (Figure 2A and 2E). Whereas EBNA2 is constitutively expressed in established latently infected LCLs (data not shown [15]), the expression of both BALF1 and BHRF1 rapidly decreased, and their transcripts were weakly or not detected by RT-PCR 3 wk p.i. (Figure 2A and 2E). EBNA2 transcripts could be detected 1 d p.i. in wild-type 2089 EBV or 2636 BALF1−/BHRF1− mutant infected primary B cells (Figure 2B). Similarly, vBcl-2 transcripts are readily detectable in cells infected with an EBNA2– 2491 mutant EBV (Figure 2C), indicating that EBNA2 and vBcl-2 genes are independently expressed very early after infection.
Figure 2 Detection of EBNA2, BALF1, and BHRF1 Transcripts by RT-PCR Analysis
(A). Primary B cells were infected with B95.8 EBV, and RNA was prepared at various time points. The expression of BALF1, BHRF1, and HPRT (as control) was analyzed by RT-PCR. (B). Primary B cells were infected with wild-type 2089 EBV, the BZFL1− mutant 2809, or the BALF1−/BHRF1− mutant 2636 (MOI 0.1), and the expression of EBNA2 was determined by RT-PCR. EBNA2 mRNA was detectable in comparable amounts in cells infected with either virus stock on day 1 and on day 5 p.i. (C). Primary B cells were infected with wild-type 2089 EBV, the BZFL1− mutant 2809, or the EBNA2− mutant 2491. Similar to (A), vBcl-2 transcripts were detectable in comparable amounts in cells infected with either virus stock on day 1 and on day 5 p.i. but at a reduced level on day 5. (D). BALF1 and BHRF1 transcript levels in latently infected and lytically induced cell lines. Neither BALF1 nor BHRF1 are detectably expressed in the latently infected, EBV-positive Akata cell line [62] or in 293HEK cells stably transfected with wild-type 2089 EBV. Both mRNAs are easily detectable in B95.8 cells, which spontaneously support productive infection, or in lytically induced 293HEK cells carrying the maxi-EBV 2089). (E) Schematic overview of the PCR primers and their location used to detect EBNA2-, BALF1-, or BHRF1-specific cDNAs. The PCR product indicative of correctly spliced EBNA2 transcripts is 426 bp in length; the BALF1 PCR product is 443 bp in length. BHRF1-specific PCR products are expected to be 551 bp and 991 bp in length, representing the spliced and unspliced transcripts, respectively.
As expected, BALF1 and BHRF1 were not expressed in other latently EBV-infected cells (Figure 2D) but readily detectable in cells that spontaneously support the lytic phase of EBV's life cycle (B95.8 cells in Figure 2A and 2D) and maximally expressed in cells in which the lytic phase was induced (Figure 2D). Because both vBcl-2 genes were induced during EBV's lytic phase, and several reports indicate that lytic expression of BHRF1 is strictly dependent on the immediate-early transcriptional activator BZFL1 ([15,38] and references therein), we wanted to learn whether initial expression of BALF1 or BHRF1 in newly infected primary B cells is also regulated by BZLF1. BZLF1 encodes a molecular switch protein, which is instrumental in inducing the lytic phase in latently EBV-infected cells ([15] and references therein). The EBV mutant 2809, which is BZLF1− [39], was used to infect primary B cells, and the expression of BALF1 and BHRF1 was assessed by RT-PCR. Again, vBcl-2 transcripts were readily detectable in cells infected with the BZFL1− 2809 EBV mutant, indicating that BZLF1 does not regulate BALF1 and BHRF1 expression (Figure 2C). As expected, EBNA2 expression was also BZLF1 independent (Figure 2B).
Cell Cycle Activation Is Abrogated in vBcl-2– EBV-Infected Cells
We questioned whether vBcl-2 gene products might be critical for cell cycle entry similar to EBNA2 [15,40]. Stocks of the BALF1−/BHRF1− 2636 virus were used to infect primary B cells at a multiplicity of infection (MOI) of 0.5 GRU with the EBNA2− mutant 2491 or wild-type EBV 2089 serving as controls. For direct comparison, a fraction of the cells was left uninfected. The cell cycle status of the cells was analyzed by 5-bromo-2′-deoxyuridine (BrdU)-incorporation and fluorescence-activated cell-sorting (FACS) analysis, which revealed cycling cells in wild-type EBV-infected cells as early as day 3 p.i. (Figure 3). Uninfected cells did not enter the cell cycle nor did the cells infected with the EBNA2− or the BALF1−/BHRF1− 2636 mutants. These latter infected cells all showed a rapid increase in the fraction of cells with a subG1 DNA content indicating that the inactivation of both vBcl-2 genes prevented cell cycle entry similar to EBNA2− EBV, which lacks the key regulator of EBV's latent genes (Figure 3).
Figure 3 Cell Cycle Analysis of Uninfected Primary B Cells (PBls) or PBls Infected with EBNA2− Virus, BALF1−/BHRF1− Mutant 2636, or Wild-Type 2089 EBV
Primary B cells (105) per time point were infected with a viral dose such that about 50% of the cells were infected or left uninfected, and their cell cycle profiles were determined by BrdU incorporation and FACS analysis. Only wild-type 2089 EBV induced cell cycle progression, whereas uninfected cells as well as cells infected with the two mutants became apoptotic as indicated by their subG1 DNA content. (A) shows the compiled data of the primary results, which are illustrated in (B). One representative experiment out of three is shown.
B-Cell Apoptosis Is Delayed in vBcl-2+ EBV-Infected Primary B Cells
Given that EBNA2 is a pivotal mediator of EBV's known latent genes, it is important to know if EBV's vBcl-2 genes, which are required to establish latency (see Figure 1), depend on any of EBNA2′s functions. We tested whether primary B cells infected with the EBNA2– mutant, which is capable of expressing both vBcl-2 genes (see Figure 2C), showed a phenotype different from cells infected with a mutant null for vBcl-2s. Because both viral mutants are deficient in inducing cellular proliferation (see Figure 3), we could concentrate on effects related to apoptotic markers independent of global cellular activation (Figure 4). Toward this end, we infected primary B cells with the EBNA2− 2491 or the BALF1−/BHRF1− 2636 viruses (or wild-type 2089 EBV as a positive control) and determined the occurrence of GFP+ EBV-infected cells by FACS (Figure 4C). To monitor the numbers of GFP+ cells over a period of 8 d, calibration beads were used as an internal reference (Figure 4A). By three-parameter FACS analysis, GFP+ B cells were also analyzed for binding of Annexin-V and propidium iodine (PI) uptake as indicators of early apoptosis and loss of membrane integrity, respectively (Figure 4D). In parallel, uninfected cells with typical lymphocytic characteristics by forward and sideward scatter criteria (Figure 4B) were compared for their Annexin-V and PI staining. As shown in Figure 4D and 4E, uninfected primary B cells as well as those infected with BALF1−/BHRF1− 2636 mutant rapidly lost their viability. As early as day 3 p.i., about 70% of the BALF1−/BHRF1− virus-infected cells were Annexin-V+/PI+ comparable to uninfected cells. By day 5, few lymphocytes had survived and none were found on day 8 (Figure 4E). In stark contrast, all primary B cells infected with the EBNA2− mutant 2491, which expressed BALF1 and BHRF1 transcripts immediately after infection (see Figure 2C), were alive on day 3 p.i., about 15% survived until day 5, and a few percent of the cells were GFP+/Annexin-V−/PI− 8 d p.i. (Figure 4D and 4E). Thus, EBV's Bcl-2 homologs support initial B-cell survival and rescue EBV-infected B cells, which would otherwise succumb rapidly to spontaneous apoptosis.
Figure 4 Apoptosis of Primary B Cells
Primary B cells were infected with the EBNA2− mutant 2491, the BALF1−/BHRF1− mutant 2636, or wild-type 2089 EBV, at an MOI of 0.1, or the cells were left uninfected. Cells were analyzed by FACS at days 1, 3, 5, and 8 p.i. (A) Total cellular events were collected until 3 × 103 BD CaliBRITE beads as a volume standard were counted. These beads, which are indicated by red circles in the FACS diagrams in (A) and (C), were used as an internal volume reference corresponding to 3 × 104 cells plated initially. (B) Uninfected cells that fulfilled the criteria of lymphocytes according to their forward (FSC) and sideward (SSC) scatter characteristics were gated as indicated. One day after B-cell preparation, lymphocytes were present in the lymphocyte gate as expected, but only a few lymphocytes were still present in this gate 8 d after cell preparation when the cells had been left uninfected. Similarly, EBV-infected lymphocytes were selected according to the same scatter criteria (data not shown). (C) EBV-infected GFP+ cells were gated as indicated. The example shows an uninfected and a 2491 EBNA2− EBV-infected B-cell sample with 3 × 103 BD CaliBRITE beads added 1 d p.i. (D) Uninfected primary B cells within the lymphocyte gate or EBV-infected GFP+ lymphocytes were analyzed by FACS for Annexin-V binding and PI staining at different time points p.i. The absolute numbers of Annexin-V−/PI−/GFP+ cells allowed the calculation of surviving cells at each time point p.i. Uninfected cells indicate the rate and kinetics of spontaneous apoptosis of primary B cells ex vivo. Primary B cells infected with the BALF1−/BHRF1− mutant 2636 died as rapidly as uninfected cells. Only B cells infected with the EBNA2− mutant 2491, which is vBcl-2+, survived considerably longer. Primary B cells infected with wild-type 2089 EBV as a positive control rapidly increased in the number of Annexin-V−/PI−/GFP+ cells. (E) Summary of the primary data shown in (D). Annexin-V−/PI− cells (uninfected control to determine spontaneous programmed cell death), or Annexin-V−/PI−/GFP+ (infected) cells were set to 100% at day 1, and the percentile of Annexin-V−/PI− cells at each time point was calculated. One representative experiment out of three is shown.
Discussion
vBcl-2 Genes Are Essential for Growth Transformation of Primary B Lymphocytes
Some viruses encode proteins that interfere with the host's apoptotic machinery to ensure short-term cell survival and to yield progeny before the cell lyses and dies [3,41,42]. Similarly, γ-herpesviruses encode vBcl-2 genes, whose gene products are antiapoptotic Bcl-2 family members and act during productive infection or contribute to viral virulence [10,12]. Only EBV is known to encode two Bcl-2 homologs, BALF1 and BHRF1. BHRF1 was shown to be dispensable for B-cell growth transformation [43,44], but its product does have antiapoptotic functions [4,45]. The function of the BALF1 gene product is controversial [5,6]. Our experiments indicate that (i) both BALF1 and BHRF1 belong to the same genetic complementation group and are functionally redundant (see Figure 1). Either BALF1 or BHRF1 is (ii) essential to establish latently infected LCLs (see Figures 1 and 3) but (iii) dispensable for their continued proliferation (see Figure 1E) and (iv) generation of virus progeny (data not shown). In newly infected primary B cells, (v) both vBcl-2 genes are expressed immediately but transiently after infection (see Figure 2), in contrast to the group of 11 latent EBV genes, which are constitutively expressed [15,46]. Surprisingly, BALF1 and BHRF1 proteins (vi) prevent spontaneous programmed cell death in newly infected resting primary B cells (see Figure 4). This newly identified function of vBcl-2 ensures successful infection of primary B cells, gives rise to latent infection, and supports growth transformation of resting B cells in vitro.
This finding seems to be without precedent because vBcl-2 homologs have not been recognized to contribute to viral latent infection directly [2,8,9], probably because herpesviral latency can only be studied with a few viruses. Similar to EBV, long-term latency of the murine γ-herpesvirus MHV-68 is narrowly confined to cells expressing a B-cell phenotype [47,48]. Infection of mice with MHV-68 provides a tractable and authentic animal model to study herpesviral latency in the context of its natural host. A viral mutant deficient for the MHV-68 vBcl-2 gene M11 was found compromised in reactivating from latency [12], whereas a more recent paper identified a vBcl-2 associated deficit in establishment of latency [10]. Although the molecular basis for these observations remains unclear, both seem to be consistent with vBcl-2 protecting latently infected B cells from apoptotic death. Similar to our findings, no role for vBcl-2 of MHV-68 was apparent during acute viral replication in vivo or viral virulence [10–12].
How Are vBcl-2 Genes Regulated in Newly Infected Primary B Cells?
Experimental data clearly demonstrate that BALF1 and BHRF1 are maximally expressed in primary B cells within 24 h p.i. (see Figure 2A). Because the expression of BALF1 and BHRF1 is also prominent during EBV's lytic phase (see Figure 2D), we wondered whether their expression might be regulated by the viral immediate-early gene BZLF1. BZLF1 is a transcription factor and acts like a molecular switch to induce the lytic phase in latently EBV-infected cells ([15] and references therein). Unexpectedly, the expression of both vBcl-2 genes was found to be independent of BZLF1 (see Figure 2C). Thus, we do not know the mechanisms regulating the expression of both BALF1 and BHRF1, but we would like to speculate that both genes are directly and spontaneously expressed from the transducing EBV genome. The herpesviral DNA delivered to the nucleus of the infected cell is unmethylated, coated with polyamines [49], and not in a chromatin-like state [50], which should permit universal access of the transcriptional machinery to many viral promoter elements. Modifications of the EBV genomic DNA over time will alter its accessibility, which might cause a rapid decline in the expression of BALF1 and BHRF1 when the virus eventually establishes a genuine latent infection (see Figure 2A).
vBcl-2 Homologs Are Likely to Be Involved in In Vivo Latency and Oncogenesis of EBV-Associated Tumors
EBV-associated B-cell lymphomas originate from germinal center B cells, which are inherently prone to apoptosis as they undergo affinity maturation and somatic hypermutation in their B-cell receptor genes, events characteristic of this stage of B-cell differentiation. It is unclear whether EBV infects naïve B cells prior to their germinal center passage [51,52] or germinal center B cells directly [53]. Our data indicate in either case that BALF1 and BHRF1 are involved in the in vivo survival of EBV-infected CD77+ germinal center B cells. These cells normally are highly sensitized to apoptotic signals, and most will be eliminated via programmed cell death [54]. EBV is expected to provide an important signal or signals to prevent apoptosis common to germinal center B cells because a fraction of EBV-infected and malignant tumor cells contain only nonfunctional B-cell receptor genes and would have been expected to undergo programmed cell death. EBV's vBcl-2 proteins likely constitute this survival signal, which would be superseded by the eventual expression of LMP2A [55,56] and/or other “classical” latent EBV genes. Thus, EBV's Bcl-2 homologs are likely to counteract the physiological elimination of B cells, which might be a crucial viral contribution to lymphomagenesis [13].
The physiological expansion of B cells during the germinal center reaction increases the size of the pool of EBV-infected cells, which go on to differentiate into long-lived memory B cells, the reservoir for virus in vivo [51,52]. Because memory B cells originate from functional survivors of the germinal center passage, we favor the notion that during this passage EBV's antiapoptotic viral genes provide a survival advantage to the infected cells favoring their developing into memory B cells in which EBV is latent and can thereby establish a lifetime relationship with its human host.
Materials and Methods
Construction of viral mutants
The BALF1
− and BHRF1− single deletion mutants were constructed on the basis of the wild-type B95.8 EBV strain cloned onto the F-factor plasmid p2089 [32]. The inactivation of the vBcl-2 genes or the EBNA2 gene in the EBV genome was carried out with two different genetic procedures as described in a recent review [57]. To promote the integration of linear DNA fragments in the DH10B Escherichia coli host, the conditional recA/redγ expression plasmid p2650 based on the pST76-amp plasmid was employed. To generate the vBcl-2 targeting plasmids for homologous recombination, subgenomic EBV DNA fragments of about 10 kilobasepairs (kbp) in size, spanning either the BALF1 or the BHRF1 locus, were subcloned to yield the plasmids p1352 and p528, respectively. The BALF1 targeting plasmid p2564 was generated by inserting the tetracycline resistance gene from pCP16 [58] into the BALF1 open reading frame in p1352. The kanamycin phosphotransferase gene from pCP15 [58] was inserted into the BHRF1 coding sequence of p528, giving rise to the targeting plasmid p2632. Homologous recombinations were carried out with the p2089 maxi-EBV genome to yield the EBV mutant genomes p2512 and p2637, which carry singly inactivated BALF1 and BHRF1 genes, respectively (see Figure 1A and 1B). The two BALF1−/BHRF1− and BHRF1−/BALF1− mutants p2636 and p2765 were generated in a second round of homologous recombination with the two targeting plasmids p2632 and p2564 (see Figure 1A and 1B). The BALF1+/BHRF1− revertant maxi-EBV p3227 with a reconstructed wild-type BALF1 gene was generated with the aid of the pST76-amp-based shuttle plasmid [59] carrying the wild-type BALF1 gene flanked by homologous EBV sequences. A cointegrate of the shuttle plasmid and the p2636 EBV plasmid was formed with the “chromosomal-building” technique [60], which led to the maxi-EBV construct p3227 after routine genetic manipulations. The EBNA2− EBV mutant 2491 was also constructed with the chromosomal-building technique and the shuttle plasmid p2419.1 to delete the entire coding sequences of the EBNA2 gene in the context of the maxi-EBV genome 2089. The BZLF1– EBV mutant 2809 was described [39]. The genetic compositions of the modified EBV genomes were verified by restriction enzyme analysis, Southern blot hybridization, and partial DNA sequencing. Details for the generation of all EBV mutants are available upon request.
Preparation and quantification of infectious viral vector stocks
On the basis of 293HEK cells, stable cell lines were established after individual transfection of the genomic maxi-EBV DNAs and subsequent selection with hygromycin as described [29]. To obtain virus stocks, the cell lines were transiently transfected with expression plasmids encoding BZLF1 [61] and BALF4 [31] to induce EBV's lytic cycle. Four days p.i. supernatants were harvested and filtered through 1.2-μm pore filters. The different EBV vector stocks were quantified by infection of Raji cells as described [29,31]. Briefly, 3 × 105 Raji cells were incubated at 37 °C in 24-well cluster plates with different dilutions of the virus stocks to be analyzed. The absolute number of GFP+ cells was determined by ultraviolet microscopy in a defined fraction of the total Raji cell population 4 d after infection. On the basis of these data, GRUs per milliliter were calculated as a measure of the concentration of infectious maxi-EBV particles in different virus stocks. We used this titering method because other biologically relevant readouts are not feasible with nontransforming EBV mutants and very time-consuming with transformation-competent wild-type EBV. Infection and evaluation of GFP+ Raji cells underestimates the concentration of infectious EBV virions by a factor of at least ten (data not shown).
Isolation, separation, and infection of primary B lymphocytes
Human primary mononuclear cells were isolated from adenoids, depleted of T cells by rosetting with sheep erythrocytes and purified by Ficoll-Hypaque density-gradient centrifugation. The cell population was further depleted of monocytes by plastic adherence for 1 h. To quantify the efficiency of growth transformation with the different virus stocks, 3 × 105 human primary B lymphocytes per well were seeded in 96-well cluster plates on lethally irradiated Wi38 human fibroblast feeder cells and infected with serially diluted virus stocks with 2089 wild-type EBV and the different viral mutants in a total volume of 100 μl as described [29]. Forty-eight wells were infected with each single virus dilution. A 50-μl portion of medium was exchanged every week for fresh medium, and the number of wells with proliferating cells was determined 6 wk p.i. The infection experiments with different viral mutants were independently performed at least in triplicate. The experiments with activated B blasts were carried out similarly. In order to generate proliferating and activated B blasts, primary B lymphocytes were cultivated on an irradiated CD40-ligand feeder cell layer [34] in the presence of 4 ng/ml IL-4 (PAN Biotech, Aidenbach, Germany) and 1 μg/ml cyclosporine A (Novartis, Basel, Switzerland) for several weeks prior to the infection experiments. The proliferation of B blasts and their long-term survival in vitro is strictly dependent on both CD40L and IL-4 generated signals. To quantify the number of GRUs needed to yield clonal LCLs from B blasts, 3 × 105 blasts per well were seeded on an irradiated CD40-feeder layer in the absence of IL-4 and infected with serially diluted virus stocks as described above. Fifty percent of the volume of the cell culture media was exchanged on a weekly basis, and the cells were cultivated for 6 wk, during which the CD40-feeder cell layer had completely disintegrated.
RT-PCR analysis
RNA was extracted from primary B cells with the RNeasy Midi Kit (Qiagen, Valencia, California, United States); 5 μg of RNA was reverse transcribed with the Superscript III First Strand Synthesis Kit (Invitrogen, Carlsbad, California, United States) according to the manufacturer's protocol in a total volume of 20 μl. To monitor cellular DNA contamination of the RNA preparation, PCR reactions were performed with primers for the abundant cellular transcripts HPRT and ß-actin. PCR reactions for ß-actin were an initial template denaturation of 4 min at 94 °C, with amplification for 25 cycles (1 min at 94 °C, 1 min at 61 °C, 1 min at 72 °C), followed by a final elongation for 10 min at 72 °C. PCR reactions for HPRT were 4 min at 94 °C, with amplification for 25 cycles (1 min at 94 °C, 1 min at 54 °C, 1 min at 72 °C), followed by a final elongation for 10 min at 72 °C. Two out of 20 μl of the cDNA reaction were used as template. Unspliced PCR amplification products indicative of cellular DNA contamination were not detected (see Figure 2). PCR reactions for cDNA detection of three EBV genes were as follows: EBNA2 (an initial template denaturation of 4 min at 94 °C; 35 cycles of 1 min at 94 °C, 1 min at 61 °C, 1 min at 72 °C; followed by a final elongation for 10 min at 72 °C); BHRF1 (an initial template denaturation of 4 min at 94 °C; 35 cycles of 1 min at 94 °C, 1 min at 54 °C, 1.5 min at 72 °C; followed by a final elongation for10 min at 72 °C); and BALF1 (an initial template denaturation of 4 min at 94 °C; 35 cycles of 1 min at 94 °C, 1 min at 61 °C, 1 min at 72 °C; followed by a final elongation for 10 min at 72 °C). Oligonucleotide primer sequences for the EBV genes EBNA2, BALF1, and BHRF1 are provided in Figure 2E together with their nucleotide coordinates of the EBV strain B95.8. The primer sequences for HPRT cDNA detection were 5′-
GAGCTATTGTAATGACCAGTC-3′ and −5′-
CCAAACTCAACTTGAACTCTC-3′; ß-actin cDNA was amplified with the primers 5′-
CACCCTGTGCTGCTCACCGAGGCC-3′ and 5′-
ACCGCTCGTTGCCAATAGTGATGA-3′.
Cell cycle analysis and analysis of apoptosis
Primary B cells (4 × 105) were infected with 4 × 104 GRUs of the different virus stocks in 20 ml of culture medium in order to obtain an MOI of 0.1. The cells were kept on plastic, and at days 1, 3, 5, and 8 one-fourth of the culture (5 ml) was harvested and further analyzed for cell cycle status and apoptosis. Infection experiments were carried out with 2089 EBV, the EBNA2− deletion mutant 2491, and the different mutant virus stocks (see Figure 1); for a negative control the cells were left uninfected. To determine the fraction of apoptotic cells, Annexin-V staining was performed with the Annexin-V-APC kit (BioVision, Mountain View, California, United States), according to the manufacturer's protocol. For each time point, 5 ml of each sample was harvested, spun, and the cells were stained with 5 μl of APC-coupled Annexin-V and 5 μl of PI at a final concentration of 10 μg/ml. As an internal FACS calibration standard, 1 × 104 BD CaliBRITE Beads (Becton-Dickinson, Palo Alto, California, United States) were added to yield a final concentration of 2 × 104 beads/ml. The beads are very small, resulting in a high intensity in the sideward scatter channel. Since the beads also display a very bright APC fluorescence, their characteristics do not interfere with the cells to be analyzed but allow their unbiased detection and quantification. FACS analysis was carried out in a FACS-Calibur machine (Becton-Dickinson). The 3 × 103 BD CaliBRITE Beads were set as a volume standard corresponding to 3 × 104 cells plated initially to determine the absolute number of cells at any given time point without interference from the dynamics of the cell culture. To exclude cellular debris, only cells within the lymphocyte gate were analyzed (see Figure 4B).
Cells infected with the different recombinant virus stocks express GFP as early as day 1 p.i. The GFP gate was set such that only cells with a GFP signal brighter than primary B cells infected with the GFP− prototypic B95.8 EBV strain scored positive as infected GFP+ cells (see Figure 4C). EBV-infected GFP+ cells in the lymphocyte gate were analyzed for their Annexin-V and PI staining. In uninfected samples, only cells that fulfilled the criteria of lymphocytic cells by forward and sideward scatter criteria were analyzed for both Annexin-V and PI staining. The number of Annexin-V−/PI− cells in the infected samples and the uninfected control were set to 100% at day 1 p.i.
To analyze the cell cycle status of the infected primary B cells in comparison to the uninfected controls, the samples were incubated with the thymidine analog BrdU for 2 h prior to FACS analysis at each time point. The cell proliferation assays were immediately performed with the BrdU Flow Kit (BD Biosciences Pharmingen, San Diego, California, United States). The cells were stained with an APC-coupled BrdU-specific antibody after fixation and permeabilization, and the cellular DNA was counterstained with the DNA intercalating dye 7-AAD according to the manufacturer's protocol. FACS analysis was performed until 3 × 104 cells were analyzed. The recorded data were gated for cells in the G1, S, G2/M phases of the cell cycle and for cells with a subG1 DNA content. The total of all events was set to 100%.
Supporting Information
Accession Numbers
UniProtKB/TrEMBL (http://www.expasy.org/uniprot/) accession numbers for the proteins are Bax (Q07812), Bcl-XL (Q07817), Bcl-2 (P10415), CD40 receptor (P25942), LMP1 (P03230), LMP2A (Q777H4), and LMP2B (Q8AZK9); for the virus, E1B 19K (P03247); for the genes, BALF1 (Q777A6) and BHRF1 (Q777H0); for EBV nuclear antigens, EBNA1 (P03211), EBNA2 (Q69023), EBNA3A (Q8AZJ8), EBNA3B (Q777E8), EBNA3C (Q777E7), and EBNA-LP (Q8AZK7).
We thank Bill Sugden for helpful discussions, suggestions, and comments on the manuscript. We are also extremely grateful to Andreas Moosmann for providing us with B blasts and technical advice. Our work was supported by SFB455 of the Deutsche Forschungsgemeinschaft, the Bayerische Forschungsstiftung, and the Sanderstiftung and by a National Institutes of Health Public Health Service Grant CA70723.
Competing interests. The authors have declared that no competing interests exist.
Author contributions. WH conceived and designed the experiments. MA performed the experiments. MA and WH analyzed the data. MA and WH contributed reagents/materials/analysis tools. WH wrote the paper.
Citation: Altmann M, Hammerschmidt W (2005) Epstein-Barr virus provides a new paradigm: A requirement for the immediate inhibition of apoptosis. PLoS Biol 3(12): e404.
Abbreviations
BrdU5-bromo-2′-deoxyuridine
EBVEpstein-Barr virus
FACSfluorescence-activated cell sorting
GFPgreen fluorescence protein
GRUgreen Raji unit
kbpkilobasepairs
LCLlymphoblastoid cell line
MOImultiplicity of infection
p.i.postinfection
PIpropidium iodine
vBcl-2viral Bcl-2
==== Refs
References
Cuconati A White E Viral homologs of BCL-2: Role of apoptosis in the regulation of virus infection Genes Dev 2002 16 2465 2478 12368257
Boya P Pauleau AL Poncet D Gonzalez-Polo RA Zamzami N Viral proteins targeting mitochondria: Controlling cell death Biochim Biophys Acta 2004 1659 178 189 15576050
Benedict CA Norris PS Ware CF To kill or be killed: Viral evasion of apoptosis Nat Immunol 2002 3 1013 1018 12407409
Henderson S Huen D Rowe M Dawson C Johnson G Epstein-Barr virus-coded BHRF1 protein: A viral homologue of Bcl-2, protects human B cells from programmed cell death Proc Natl Acad Sci U S A 1993 90 8479 8483 8397406
Bellows DS Howell M Pearson C Hazlewood SA Hardwick JM Epstein-Barr virus BALF1 is a BCL-2-like antagonist of the herpesvirus antiapoptotic BCL-2 proteins J Virol 2002 76 2469 2479 11836425
Marshall WL Yim C Gustafson E Graf T Sage DR Epstein-Barr virus encodes a novel homolog of the bcl-2 oncogene that inhibits apoptosis and associates with Bax and Bak J Virol 1999 73 5181 5185 10233985
Irusta PM Chen YB Hardwick JM Viral modulators of cell death provide new links to old pathways Curr Opin Cell Biol 2003 15 700 705 14644194
Hilleman MR Strategies and mechanisms for host and pathogen survival in acute and persistent viral infections Proc Natl Acad Sci U S A 2004 101 Suppl 2 14560 14566 15297608
Roulston A Marcellus RC Branton PE Viruses and apoptosis Annu Rev Microbiol 1999 53 577 628 10547702
de Lima BD May JS Marques S Simas JP Stevenson PG Murine gammaherpesvirus 68 bcl-2 homologue contributes to latency establishment in vivo J Gen Virol 2005 86 31 40 15604429
Gangappa S Kapadia SB Speck SH Virgin HW Antibody to a lytic cycle viral protein decreases gammaherpesvirus latency in B-cell-deficient mice J Virol 2002 76 11460 11468 12388707
Gangappa S van Dyk LF Jewett TJ Speck SH Virgin HW Identification of the in vivo role of a viral bcl-2 J Exp Med 2002 195 931 940 11927636
Küppers R B cells under influence: Transformation of B cells by Epstein-Barr virus Nat Rev Immunol 2003 3 801 812 14523386
Babcock GJ Hochberg D Thorley-Lawson AD The expression pattern of Epstein-Barr virus latent genes in vivo is dependent upon the differentiation stage of the infected B cell Immunity 2000 13 497 506 11070168
Kieff E Rickinson AB Knipe DM Howley PM Griffin DE Martin MA Lamb RA Epstein-Barr virus and its replication Fields' virology, 4th ed 2001 Philadelphia Lippincott–Williams and Wilkins 2511 2573
Rickinson AB Kieff E Knipe DM Howley PM Griffin DE Martin MA Lamb RA Epstein-Barr virus Fields' virology, 4th ed 2001 Philadelphia Lippincott–Williams and Wilkins 2575 2627
[Anonymous] Epstein-Barr virus and Kaposi's sarcoma herpesvirus/human herpesvirus 8; International Agency for Research on Cancer 1997 Lyon (France) IARC
Dirmeier U Hoffmann R Kilger E Schultheiss U Briseno C Latent membrane protein 1 of Epstein-Barr virus coordinately regulates proliferation with control of apoptosis Oncogene 2005 24 1711 1717 15674340
Cahir-McFarland ED Davidson DM Schauer SL Duong J Kieff E NF-kappa B inhibition causes spontaneous apoptosis in Epstein-Barr virus-transformed lymphoblastoid cells Proc Natl Acad Sci U S A 2000 97 6055 6060 10811897
Cahir-McFarland ED Carter K Rosenwald A Giltnane JM Henrickson SE Role of NF-kappa B in cell survival and transcription of latent membrane protein 1-expressing or Epstein-Barr virus latency III-infected cells J Virol 2004 78 4108 4119 15047827
Feuillard J Schuhmacher M Kohanna S Asso-Bonnet M Ledeur F Inducible loss of NF-kappaB activity is associated with apoptosis and Bcl-2 down-regulation in Epstein-Barr virus-transformed B lymphocytes Blood 2000 95 2068 2075 10706876
Kilger E Kieser A Baumann M Hammerschmidt W Epstein-Barr virus-mediated B-cell proliferation is dependent upon latent membrane protein 1, which simulates an activated CD40 receptor EMBO J 1998 17 1700 1709 9501091
Kulwichit W Edwards RH Davenport EM Baskar JF Godfrey V Expression of the Epstein-Barr virus latent membrane protein 1 induces B cell lymphoma in transgenic mice Proc Natl Acad Sci U S A 1998 95 11963 11968 9751773
Hammerschmidt W Sugden B Genetic analysis of immortalizing functions of Epstein-Barr virus in human B lymphocytes Nature 1989 340 393 397 2547164
Cohen JI Wang F Mannick J Kieff E Epstein-Barr virus nuclear protein 2 is a key determinant of lymphocyte transformation Proc Natl Acad Sci U S A 1989 86 9558 9562 2556717
Kaye KM Izumi KM Kieff E Epstein-Barr virus latent membrane protein 1 is essential for B-lymphocyte growth transformation Proc Natl Acad Sci U S A 1993 90 9150 9154 8415670
Tomkinson B Robertson E Kieff E Epstein-Barr virus nuclear proteins EBNA-3A and EBNA-3C are essential for B-lymphocyte growth transformation J Virol 1993 67 2014 2025 8445720
Kempkes B Pich D Zeidler R Sugden B Hammerschmidt W Immortalization of human B lymphocytes by a plasmid containing 71 kilobase pairs of Epstein-Barr virus DNA J Virol 1995 69 231 238 7983714
Dirmeier U Neuhierl B Kilger E Reisbach G Sandberg ML Latent membrane protein 1 is critical for efficient growth transformation of human B cells by Epstein-Barr virus Cancer Res 2003 63 2982 2989 12782607
Humme S Reisbach G Feederle R Delecluse HJ Bousset K The EBV nuclear antigen 1 (EBNA1) enhances B cell immortalization several thousandfold Proc Natl Acad Sci U S A 2003 100 10989 10994 12947043
Neuhierl B Feederle R Hammerschmidt W Delecluse HJ Glycoprotein gp110 of Epstein-Barr virus determines viral tropism and efficiency of infection Proc Natl Acad Sci U S A 2002 99 15036 15041 12409611
Delecluse HJ Hilsendegen T Pich D Zeidler R Hammerschmidt W Propagation and recovery of intact, infectious Epstein-Barr virus from prokaryotic to human cells Proc Natl Acad Sci U S A 1998 95 8245 8250 9653172
Frisan T Levitsky V Masucci M Limiting dilution assay Methods Mol Biol 2001 174 213 216 11357647
Garrone P Neidhardt EM Garcia E Galibert L van Kooten C Fas ligation induces apoptosis of CD40-activated human B lymphocytes J Exp Med 1995 182 1265 1273 7595197
Zarnegar B He JQ Oganesyan G Hoffmann A Baltimore D Unique CD40-mediated biological program in B cell activation requires both type 1 and type 2 NF-kappaB activation pathways Proc Natl Acad Sci U S A 2004 101 8108 8113 15148378
Pearson GR Luka J Petti L Sample J Birkenbach M Identification of an Epstein-Barr virus early gene encoding a second component of the restricted early antigen complex Virology 1987 160 151 161 2820125
Kempkes B Spitkovsky D Jansen-Durr P Ellwart JW Kremmer E B-cell proliferation and induction of early G1-regulating proteins by Epstein-Barr virus mutants conditional for EBNA2 EMBO J 1995 14 88 96 7828599
Schepers A Pich D Hammerschmidt W Activation of oriLyt, the lytic origin of DNA replication of Epstein-Barr virus, by BZLF1 Virology 1996 220 367 376 8661388
Feederle R Kost M Baumann M Janz A Drouet E The Epstein-Barr virus lytic program is controlled by the co-operative functions of two transactivators EMBO J 2000 19 3080 3089 10856251
Sinclair A Palmero I Peters G Farrell PJ EBNA-2 and EBNA-LP cooperate to cause G0 to G1 transition during immortalization of resting human B lymphocytes by Epstein-Barr virus EMBO J 1994 13 3321 3328 8045261
Hardwick JM Bellows DS Viral versus cellular BCL-2 proteins Cell Death Differ 2003 10 Suppl 1 S68 S76 12655348
Shen Y Shenk TE Viruses and apoptosis Curr Opin Genet Dev 1995 5 105 111 7749317
Marchini A Tomkinson B Cohen JI Kieff E BHRF1, the Epstein-Barr virus gene with homology to Bcl2, is dispensable for B-lymphocyte transformation and virus replication J Virol 1991 65 5991 6000 1656084
Lee MA Yates JL BHRF1 of Epstein-Barr virus, which is homologous to human proto-oncogene bcl2, is not essential for transformation of B cells or for virus replication in vitro J Virol 1992 66 1899 1906 1312610
Foghsgaard L Jaattela M The ability of BHRF1 to inhibit apoptosis is dependent on stimulus and cell type J Virol 1997 71 7509 7517 9311830
Murray PG Swinnen LJ Constandinou CM Pyle JM Carr TJ BCL-2 but not its Epstein-Barr virus-encoded homologue, BHRF1, is commonly expressed in posttransplantation lymphoproliferative disorders Blood 1996 87 706 711 8555494
Flano E Kim IJ Woodland DL Blackman MA Gamma-herpesvirus latency is preferentially maintained in splenic germinal center and memory B cells J Exp Med 2002 196 1363 1372 12438427
Willer DO Speck SH Long-term latent murine Gammaherpesvirus 68 infection is preferentially found within the surface immunoglobulin D-negative subset of splenic B cells in vivo J Virol 2003 77 8310 8321 12857900
Gibson W Roizman B Compartmentalization of spermine and spermidine in the herpes simplex virion Proc Natl Acad Sci U S A 1971 68 2818 2821 5288261
Booy FP Newcomb WW Trus BL Brown JC Baker TS Liquid-crystalline, phage-like packing of encapsidated DNA in herpes simplex virus Cell 1991 64 1007 1015 1848156
Thorley-Lawson DA Epstein-Barr virus: Exploiting the immune system Nat Rev Immunol 2001 1 75 82 11905817
Thorley-Lawson DA Babcock GJ A model for persistent infection with Epstein-Barr virus: The stealth virus of human B cells Life Sci 1999 65 1433 1453 10530796
Küppers R Molecular biology of Hodgkin's lymphoma Adv Cancer Res 2002 84 277 312 11883530
Liu YJ Joshua DE Williams GT Smith CA Gordon J Mechanism of antigen-driven selection in germinal centres Nature 1989 342 929 931 2594086
Caldwell RG Wilson JB Anderson SJ Longnecker R Epstein-Barr virus LMP2A drives B cell development and survival in the absence of normal B cell receptor signals Immunity 1998 9 405 411 9768760
Mancao C Altmann M Jungnickel B Hammerschmidt W Rescue of ‘crippled' germinal center B cells from apoptosis by Epstein-Barr virus Blood 2005 E-pub ahead of print
Neuhierl B Delecluse HJ Molecular genetics of DNA viruses: Recombinant virus technology Methods Mol Biol 2005 292 353 370 15507720
Cherepanov PP Wackernagel W Gene disruption in Escherichia coli TcR and KmR cassettes with the option of Flp-catalyzed excision of the antibiotic-resistance determinant Gene 1995 158 9 14 7789817
Posfai G Koob MD Kirkpatrick HA Blattner FR Versatile insertion plasmids for targeted genome manipulations in bacteria: Isolation, deletion, and rescue of the pathogenicity island LEE of the Escherichia coli O157:H7 genome J Bacteriol 1997 179 4426 4428 9209066
O'Connor M Peifer M Bender W Construction of large DNA segments in Escherichia coli
Science 1989 244 1307 1312 2660262
Hammerschmidt W Sugden B Identification and characterization of oriLyt, a lytic origin of DNA replication of Epstein-Barr virus Cell 1988 55 427 433 2846181
Takada K Horinouchi K Ono Y Aya T Osato T An Epstein-Barr virus-producer line Akata: Establishment of the cell line and analysis of viral DNA Virus Genes 1991 5 147 156 1647567
|
16277553
|
PMC1283332
|
CC BY
|
2021-01-05 08:21:46
|
no
|
PLoS Biol. 2005 Dec 15; 3(12):e404
|
utf-8
|
PLoS Biol
| 2,005 |
10.1371/journal.pbio.0030404
|
oa_comm
|
==== Front
PLoS BiolPLoS BiolpbioplosbiolPLoS Biology1544-91731545-7885Public Library of Science San Francisco, USA 10.1371/journal.pbio.0030417SynopsisBioinformatics/Computational BiologyEvolutionGenetics/Genomics/Gene TherapyNeuroscienceStatisticsHomo (Human)PrimatesSelection on a Neural Gene Regulator Sheds Light on Human Evolution Synopsis12 2005 15 11 2005 15 11 2005 3 12 e417Copyright: © 2005 Public Library of Science.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
Ancient and Recent Positive Selection Transformed Opioid cis-Regulation in Humans
==== Body
With humans and chimpanzees differing by just 1.2% at the DNA level, it's clear that our differences do not arise from gene variation alone. Thirty years ago, Mary-Claire King and Alan Wilson pointed to our extensive protein similarities as evidence that those investigating the genetic basis of human origins should focus on the regulators of gene expression rather than on the genes themselves.
DNA sequences that regulate gene expression, called cis-regulatory elements, occur on the same DNA molecule as the regulated gene. By recruiting proteins that initiate or block transcription, cis-regulatory elements influence the rate at which genes are transcribed, in which cells, and under what conditions. But since sequence inspection doesn't reveal whether regulatory sequence changes are functional or neutral, finding evidence for human-specific changes and positive selection in these sequences is far harder than finding similar evidence in protein sequences or in gene sets. Now, Matthew Rockman, Gregory Wray, and their colleagues provide further support for King and Wilson's predictions by showing that positive selection altered the cis-regulation of a gene expressed in the human brain.
To find evidence of regulatory changes underlying uniquely human traits, Rockman et al. examined the regulatory evolution of prodynorphin, a gene expressed in multiple brain and endocrine cell types. The protein encoded by prodynorphin is a precursor molecule for a suite of neuropeptides that bind to opiate receptors and affect perception, pain sensation, emotion, and learning. In humans, prodynorphin's promoter contains what's called a 68 base-pair tandem repeat polymorphism—individuals can have up to four copies of the 68 DNA base-pair element, which occur side by side. The polymorphism, which affects how many transcripts of the gene are produced, has been tentatively linked to schizophrenia, cocaine addiction, and epilepsy.
To explore how this functional variation evolved, Rockman et al. first sequenced and analyzed prodynorphin regulatory DNA from 74 human chromosomes and 32 nonhuman primate chromosomes (chimp, bonobo, gorilla, orangutan, baboon, and two macaque species). The duplication leading to tandem repeats appears unique to humans, since all the monkeys and other great apes carry only one copy of the 68 base-pair element. Further distinguishing humans from the last common ancestor of humans and chimps, the human copies also carry five mutations, or substitutions, far more than would be expected if the mutations were neutral (that is, had no effect on fitness). Three nearby polymorphisms also occurred at a higher-than-expected frequency in humans, a sign that selection acted on the linked neighboring sequences. The protein-coding sequence of prodynorphin, on the other hand, appears to have undergone negative selection, discarding harmful mutations that would disrupt its function.
Positive natural selection altered the regulation of the human prodynorphin gene, which produces a product that interacts with the same receptors that opium targets
To determine the functional effects of the human substitution, the authors attached a bioluminescent enzyme to human and chimp prodynorphin
cis-regulatory DNA, and introduced the modified DNA into human cell lines so they could measure transcription levels. Only the human 68 base-pair element significantly increased transcription of prodynorphin, and this increase was seen only in brain cells.
Because different numbers of repeats are associated with different effects, such as protection against cocaine addiction and neurological disease, Rockman et al. searched for signs of recent selection. If selection had occurred, divergence among populations should be increased and variation within populations reduced, relative to the neutral case. And that's what the authors found. The frequency of three-repeat versions of the 68 base-pair element was less than 10% in Chinese and Papua New Guinean populations and over 60% in Italy and Ethiopia. And variation at genome markers called microsatellites was significantly low overall across most of the populations studied (which also included India). Because microsatellites typically undergo high mutation rates, if a microsatellite is linked to an element under positive selection, it should show reduced variation.
The observed pattern of variations within and among human populations, the authors argue, suggests that recent selection has favored different versions of prodynorphin-regulatory elements in different regions of the world. These results support the longstanding notion that changes in gene regulation had major impacts on the evolution of novel traits, and may well hold the key to that eternal question, what makes us human? —Liza Gross
|
0
|
PMC1283333
|
CC BY
|
2021-01-05 08:21:45
|
no
|
PLoS Biol. 2005 Dec 15; 3(12):e417
|
utf-8
|
PLoS Biol
| 2,005 |
10.1371/journal.pbio.0030417
|
oa_comm
|
==== Front
PLoS BiolPLoS BiolpbioplosbiolPLoS Biology1544-91731545-7885Public Library of Science San Francisco, USA 10.1371/journal.pbio.0030424SynopsisBioinformatics/Computational BiologyBiotechnologyMolecular Biology/Structural BiologyBiochemistryCaenorhabditisDrosophilaHomo (Human)YeastA Systematic Way to Find Linear Motifs Mediating Protein–Protein Interactions Synopsis12 2005 15 11 2005 15 11 2005 3 12 e424Copyright: © 2005 Public Library of Science.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
Systematic Discovery of New Recognition Peptides Mediating Protein Interaction Networks
==== Body
If John Donne had been a systems biologist, instead of writing “No man is an island,” he might have written “No protein is an island.” Proteins in cells interact with numerous partners, forming complex networks that keep cells ticking over, allow them to respond to external stimuli, and make sure that essential processes like cell division go smoothly.
Interacting proteins can be identified experimentally in many ways. For example, an antibody recognizing one protein can be used to pull that protein plus any proteins that are bound to it out of cell extracts. High-throughput assays have also been developed that can screen whole genomes for genes encoding interacting proteins. But we remain a long way from having an accurate picture of all the protein–protein interactions in even one cell, let alone a whole organism.
Proteins interact with each other in two main ways. The first is through globular domains, which are formed when linear strings of 100–200 amino acids fold into specific shapes determined by their amino acid sequence. The second is through a globular domain in one protein and a short linear sequence (motif) of three to eight amino acids in the other. The structures of globular domains are relatively easy to solve, and how they interact in well-established pairs of interacting proteins can be visualized. Furthermore, other potentially interacting domains can then be inferred through sequence similarity to known globular domains. By contrast, the linear motifs involved in protein–protein interactions can't be easily identified by sequence comparisons. Until now, the only way to find them has been through time-consuming experiments, and consequently, only a few hundred linear motifs are known.
Now, a systematic way to find linear motifs in the billions of sequences stored in databases, devised and tested by Victor Neduva et al., ushers in a new era in understanding protein–protein interaction networks. The researchers started with the hypothesis that a set of proteins with a common interacting partner will share a feature that mediates binding. This feature could be a domain or a linear motif. To find the latter, they stripped away the sequences of globular domains and of long repetitive regions to leave behind the nonstructured parts of the proteins, regions where they believe linear motifs are most likely to lie. They then determined whether any short sequences in these protein remnants occurred more frequently within the set of interacting proteins than would be expected by chance. These statistically significant short sequences are potential linear motifs involved in protein binding.
The researchers tested their approach on protein sets sharing known motifs, and showed that it efficiently identified these motifs while minimizing the number of false positives. They then extracted protein sets that shared a common interaction partner from four species-specific datasets of protein–protein interactions and ran their protocol for finding linear motifs. From a fly dataset, for example, they identified 26 protein sets with one or more linear motifs that occurred more frequently than expected by chance. That nine of these motifs were already known provided important validation of their approach, but the researchers also checked several of the new motifs in direct binding experiments. For example, they showed that a motif predicted to bind to the fly protein Translin did in fact bind to it; a mutated version of the motif did not.
From their results, Neduva et al. estimate that hundreds of linear motifs may remain to be discovered. Given the central role that linear motifs play in protein–protein interactions, their systematic method for finding them should rapidly improve our understanding of the complex network of protein–protein interactions that drives the everyday lives of cells. —Jane Bradbury
|
0
|
PMC1283334
|
CC BY
|
2021-01-05 08:21:46
|
no
|
PLoS Biol. 2005 Dec 15; 3(12):e424
|
utf-8
|
PLoS Biol
| 2,005 |
10.1371/journal.pbio.0030424
|
oa_comm
|
==== Front
PLoS BiolPLoS BiolpbioplosbiolPLoS Biology1544-91731545-7885Public Library of Science San Francisco, USA 10.1371/journal.pbio.0030430SynopsisMicrobiologyIn VitroVirusesVirus Proteins Prevent Cell Suicide Long Enough to Establish Latent Infection Synopsis12 2005 15 11 2005 15 11 2005 3 12 e430Copyright: © 2005 Public Library of Science.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
Epstein-Barr Virus Provides a New Paradigm: A Requirement for the Immediate Inhibition of Apoptosis
==== Body
Little more than a small genome encased in protein, a virus can't reproduce without help from the cell it may ultimately destroy. After attaching to the cell membrane, a virus slips inside the cell, then co-opts its transcription and replication machinery to reproduce. Some viruses, such as the Epstein-Barr virus (EBV), enter a latent stage before they start to reproduce. During latency, the virus expresses its own genes, which help maintain the viral genome until replication begins.
In many viral life cycles, including EBV, reproduction continues until the cell bursts and releases a new crop of viruses. As a defense, cells undergo programmed cell death, or apoptosis, which protects other cells by containing the invader. But just as the host depends on apoptosis to survive, viral survival depends on preventing apoptosis. To do this, many viruses use relatives of a protein called Bcl-2 (the viral version is called vBcl-2). Though many vBcl-2–like proteins, or homologs, can trigger apoptosis, all the viral Bcl-2 homologs identified so far inhibit apoptosis during viral reproduction.
In a new study, Markus Altmann and Wolfgang Hammerschmidt use EBV as a model system to explore whether viral proteins play a role in latent infection. EBV, a tumor-causing herpes virus, targets B lymphocytes, the immune system's antibody-producing cells. While most herpes viruses encode one Bcl-2 gene, EBV encodes two, BALF1 and BHRF1. BHRF1 is known to prevent apoptosis in cells, but there is no consensus on BALF1 function.
Altmann and Hammerschmidt created several viral mutants to probe BHRF1 and BAFL1 function. Interestingly, they found that blocking the function of both genes had no effect during viral reproduction. But what about the latent stage of the EBV life cycle? To investigate this question, the authors compared BHRF1 and BAFL1 activity with that of EBV genes expressed during latency. While 11 latency genes are expressed throughout the latent phase of EBV's life cycle, Altmann and Hammerschmidt found that both BHRF1 and BAFL1 act transiently—their transcripts were detected just 24 hours after infection, but not three weeks later, when the other genes were still active.
This electron microscopic image of two Epstein Barr Virus virions (viral particles) shows round capsids—protein-encased genetic material—loosely surrounded by the membrane envelope
This unexpected temporal regulation of gene expression suggested that BHRF1 and BAFL1 activity might initially regulate apoptosis in the infected cells. One of the genes linked to latent EBV infection, called EBNA2, regulates other latent genes and is required for the establishment of latency. To tease apart the effects of EBNA2 from those of BHRF1 and BAFL1, the authors infected nondividing primary B lymphocytes with mutant viruses lacking EBNA2 in one set of experiments, and with double mutants lacking both BHRF1 and BAFL1 in another set. Three days after infection, some 70% of cells infected with the double-mutant virus were dying. “In stark contrast,” the authors noted, all the B lymphocytes infected with the EBNA2 mutants—which expressed BHRF1 and BAFL1 right after infection—were still alive.
Once the virus is persistent in B lymphocytes, it can transform the cells into a cancerous state called lymphoblastoid cell line. This transformation did not occur when both BHRF1 and BAFL1 were nonfunctional. But when the authors reconstructed the double-mutant strain with just one functional gene, the virus regained its ability to transform the cells. Thus, the two genes are redundant—if one is disabled, the other can take its place.
Altogether, these results show that BHRF1 and BAFL1 play a critical role in establishing latent infection by preventing apoptosis until the latency genes can be activated. After the virus penetrates into the cell, either BHRF1 or BAFL1 keeps the cell alive initially, then EBNA2 activates the latency genes, allowing the virus to persist and promote B lymphocyte transformation. Because vBcl2 homologs had not been directly linked to latent infection before, this study shines new light on the resources viruses use to bypass the host's defenses. Future studies can begin to explore the mechanisms regulating BHRF1 and BAFL1 expression, which may ultimately suggest ways to disrupt their activity and interfere with the virus's success. —Liza Gross
|
0
|
PMC1283335
|
CC BY
|
2021-01-05 08:21:45
|
no
|
PLoS Biol. 2005 Dec 15; 3(12):e430
|
utf-8
|
PLoS Biol
| 2,005 |
10.1371/journal.pbio.0030430
|
oa_comm
|
==== Front
PLoS BiolPLoS BiolpbioplosbiolPLoS Biology1544-91731545-7885Public Library of Science San Francisco, USA 10.1371/journal.pbio.0030432SynopsisAnimal BehaviorNeurosciencePhysiologyRattus (Rat)Synchronized Brain Interactions Associated with Memory and Decision-Making Synopsis12 2005 15 11 2005 15 11 2005 3 12 e432Copyright: © 2005 Public Library of Science.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
Theta Rhythms Coordinate Hippocampal-Prefrontal Interactions in a Spatial Memory Task
==== Body
Next time you lose your keys, you might consider the Clark's nutcracker. During the fall, this woodland resident collects over 30,000 seeds, buries them in discrete locations, then returns over the winter to retrieve its cache. This improbable behavior requires the coordinated activity of different brain structures to integrate spatial coordinates encoded in the hippocampus with memories of how to find the seed stash. As it turns out, food-storing birds have a significantly larger hippocampus—a brain region involved in spatial organization and memory—than nonhording species.
In the laboratory, rats learning maze tasks also rely on hippocampal spatial information, which the prefrontal cortex integrates with memory of the route, task rules, and other relevant cues to direct navigational decisions. How the brain coordinates this activity is an active area of research. When neuron populations fire in sync, they produce oscillations in brain wave patterns (measured as local field potentials) that operate at many different frequencies. Brain wave frequencies called theta rhythms, which are prevalent in the rat hippocampus, are associated with working memory and decision-making in both animals and humans. Theta rhythms—which oscillate at about eight cycles per second—appear to act like a metronome for individual neurons that “phase lock” their firing in time with the theta rhythm.
Whether the synchronized activity of neuron populations across different brain structures correlates with functions like decision-making, and whether phase-locking somehow coordinates these diverse structures has remained an open question. But now, by training rats on a spatial working memory task—navigating a maze to a food reward—Matthew Jones and Matthew Wilson demonstrate a clear correlation between coordinated hippocampal and prefrontal cortex activity and memory or decision-making processes.
As rats ran through this maze, their brain activity was recorded to study the neural basis of memory and decision-making
Jones and Wilson first trained rats on a simple maze task. The maze was shaped like a stretched-out “H” (see diagram). Rats were trained to shuttle back and forth across the long central arm for about 20 trials per day. At one end of this central arm, a moveable barrier directed the rats to turn either left or right toward a chocolate reward. At the opposite end, rats encountered a free choice at the T-junction: the correct turn (leading to more chocolate) was contingent upon the direction in which they were previously directed by the barrier at the “forced turn” end of the maze. As rats ran toward this choice point, they therefore had to “hold in mind” both task rules and information about the preceding forced turn in order to decide upon the correct route. Like the nutcracker, flying from seed stash to seed stash in search of its food, the rats' performance presumably relies upon coordination of spatial information stored in the hippocampus with connected brain regions that guide behavior.
After rats had learned to correctly navigate the maze over 80% of the time for two straight days, they were outfitted with electrodes to search for neurons showing task-related activity. The authors recorded action potentials (activation signals) of groups of individual neurons, and local field potentials, from the medial prefrontal cortex (mPFC), which is associated with working memory and decision-making, and a hippocampal region called CA1 (named after the Egyptian god Ammon's horns, cornu Ammonis in Latin). It has been known since the early seventies that neurons in CA1 show spatially selective activity—that is, each neuron fires action potentials only in restricted regions of an animal's environment.
Firing rates of individual neurons in both CA1 and mPFC were indeed task-related: they distinguished between the directions of runs across the central arm, and between the different routes between reward points during the choice stages. The firing rates of CA1–mPFC neuron pairs coactivated during central arm crossings showed the highest correlations as rats ran toward the decision point. This correlated activity between the neuron pairs was significantly reduced when rats made mistakes and chose the wrong direction. Such synchronized activity, the authors explain, may represent the transfer of spatial information from the hippocampus to a working memory system in the mPFC.
Jones and Wilson go on to show that many CA1 and mPFC neurons were phase-locked to theta rhythms, with enhanced phase-locking during trials requiring working memory and decision-making. This effectively means that the firing of neurons in both structures was aligned to the same theta rhythm “metronome.” This, in turn, means that CA1–mPFC activities became correlated during distinct portions of the task. These correlations suggest that, as expected, the coordination and function of different brain regions depends on the task at hand. Additionally, this study shows that theta rhythms can be used as a reference against which to coordinate hippocampal and mPFC activity in accordance with behavioral demands of this maze task. Beyond shedding light on the neurobiology of behavior, these findings suggest that theta rhythms may contribute to diseases that involve disruptions in prefrontal cortex connectivity, such as schizophrenia—which, interestingly, can impair the spatial working memory of patients. —Liza Gross
|
0
|
PMC1283336
|
CC BY
|
2021-01-05 08:21:45
|
no
|
PLoS Biol. 2005 Dec 15; 3(12):e432
|
utf-8
|
PLoS Biol
| 2,005 |
10.1371/journal.pbio.0030432
|
oa_comm
|
==== Front
PLoS BiolPLoS BiolpbioplosbiolPLoS Biology1544-91731545-7885Public Library of Science San Francisco, USA 10.1371/journal.pbio.0030433SynopsisCancer BiologyMus (Mouse)A Surprising New Path to Tumor Development Synopsis12 2005 15 11 2005 15 11 2005 3 12 e433Copyright: © 2005 Public Library of Science.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
Impaired DNA Replication within Progenitor Cell Pools Promotes Leukemogenesis
==== Body
In the long-established model of tumorigenesis, cells acquire specific mutations that disrupt the normal regulatory constraints on unlimited cell division. These mutations allow cells to proliferate at the expense of healthy tissues and develop into tumors. As a result, most cancer treatments use agents that curb tumor growth by inhibiting cell divisions. Surprisingly, a few conditions known to reduce a cell's ability to divide—including genetic traits, some diets, and chemotherapies—seem to increase the chances of tumor occurrence. In a new study, Ganna Bilousova and her colleagues shed new light on this troubling paradox by examining the development of mouse blood cell tumors (leukemias) in contexts that reduce cell division rates.
All circulating blood cells are descendents of progenitor cells that reside in the bone marrow. The ultimate progenitors, called hematopoietic stem cells, give rise to all blood cells by producing more specialized progenitors that initiate distinct lineages, such as the red and white blood cells. Progenitors divide to ensure the constant renewal of blood cells, but their cell divisions must be tightly controlled to avoid generating too many blood cells and increasing the risk of tumors.
Overproliferation of blood cell progenitors seems to underlie chronic myelogenous leukemia (CML) and acute lymphoblastic leukemia (ALL). The blood progenitors of both leukemias harbor a genetic anomaly known as the Philadelphia chromosome. This aberrant joining of Chromosomes 9 and 22 produces an aberrant protein called Bcr-abl, which promotes both cell proliferation and mutations. Either property could account for Bcr-abl's cancer-promoting, or oncogenic, potential. Yet, as the authors show, simply introducing a few blood progenitors carrying the Bcr-abl protein is not enough to produce leukemias in mice. Only when cell divisions are impaired in the mice blood progenitors does the proliferative advantage, and oncogenic potential, of Bcr-abl become obvious.
Their experimental setup mimics the conditions of leukemia onset—in which an oncogenic mutation presumably arises in a single (or a few) blood progenitors—yet eventually allows the cell to supersede its nonmutant counterparts. The authors transplant a mixture of two types of progenitors—some express Bcr-abl and some don't—into mice, and examine their relative contribution to the recipients' blood a few weeks later. (The rodents' own blood progenitors had been previously destroyed by irradiation.) The authors find that whether they slow cell divisions genetically (using stem cells with mutations in key regulators of cell divisions) or chemically (feeding transplanted mice hydroxyurea, a drug commonly used in cancer therapy), the progenitor cells that carry Bcr-abl can overcome the hurdle and produce mature blood cells and, eventually, tumors. In the absence of this challenge, however, Bcr-abl cells barely contribute to the recipients' blood, and far fewer cancers arise. What's more, the authors can quench the oncogenicity of Bcr-abl progenitor cells simply by cotransplanting normal progenitor cells (free of Bcr-abl and able to divide).
Leukemia cells
The implication of these observations is that nonmutant progenitor cells normally outcompete a Bcr-abl cell arising in a bone marrow niche. But under conditions that impair cell divisions, whether genetic, environmental, or therapeutic, the mutated cell has the advantage and might eventually take over. Overcoming cell division blockers is a tall order that not all oncogenes are expected to fill. Yet mutations in p53, an oncogene linked to a wide variety of cancers, confer the same advantage as Bcr-abl in this experimental setting, which suggests that the authors' model might apply to other tumors as well. While much more work needs to be done for the implications of this study to affect people, Bilousova and colleagues speculate that preventing cancer not only involves avoiding exposure to mutation-causing substances, but also a lifestyle that promotes healthy cell division behavior among tissue progenitors. —Francoise Chanut
|
0
|
PMC1283337
|
CC BY
|
2021-01-05 08:21:45
|
no
|
PLoS Biol. 2005 Dec 15; 3(12):e433
|
utf-8
|
PLoS Biol
| 2,005 |
10.1371/journal.pbio.0030433
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1627755410.1371/journal.pmed.0020353Research ArticlePharmacology/Drug DiscoveryEpidemiology/Public HealthCardiovascular MedicineEpidemiologyPharmacology and toxicologyUnderutilization of Aspirin Persists in US Ambulatory Care for the Secondary and Primary Prevention of Cardiovascular Disease Aspirin Use in Patients at Risk of CVDStafford Randall S
1
*Monti Veronica
1
Ma Jun
1
1Program on Prevention Outcomes and Practices, Stanford Prevention Research Center, Stanford University School of Medicine, Stanford, California, United States of AmericaBaigent Colin Academic EditorRadcliffe InfirmaryUnited Kingdom*To whom correspondence should be addressed. E-mail: [email protected]
Competing Interests: The authors have declared that no competing interests exist.
Author Contributions: RSS obtained study funding. RSS and JM designed the study. RSS, VM, and JM analyzed the data and contributed to writing the paper.
12 2005 15 11 2005 2 12 e35321 4 2005 25 8 2005 Copyright: © 2005 Stafford et al.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
Flying in the Face of the Evidence; Low Aspirin Use in the US Outpatient Setting
Background
Despite the proven benefits of aspirin therapy in the primary and secondary prevention of cardiovascular disease (CVD), utilization rates of aspirin remain suboptimal in relation to recommendations. We studied national trends of aspirin use among intermediate- to high-risk patients in the US ambulatory care settings and compared the priority given to aspirin versus statins for CVD risk reduction. We also examined patient and health care provider contributors to the underuse of aspirin.
Methods and Findings
We used the 1993–2003 US National Ambulatory Medical Care Survey and National Hospital Ambulatory Medical Care Survey to estimate aspirin use by cardiovascular risk. Physician-noted cardiovascular diseases defined high risk. Intermediate risk was defined as having diabetes mellitus or multiple major risk factors. The proportion of patient visits in which aspirin was reported increased from 21.7% (95% confidence interval: 18.8%–24.6%) in 1993–1994 to 32.8% (25.2%–40.4%) in 2003 for the high-risk category, 3.5% (2.0%–5.0%) to 11.7% (7.8%–15.7%) for visits by patients diagnosed with diabetes, and 3.6% (2.6%–4.6%) to 16.3% (11.4%–21.2%) for those with multiple CVD risk factors. Beginning in 1997–1998, statins were prioritized over aspirin as prophylactic therapy for reducing CVD risk, and the gaps remained wide through 2003. In addition to elevated CVD risk, greater aspirin use was independently associated with advanced age, male gender, cardiologist care, and care in hospital outpatient departments.
Conclusion
Improvements in use of aspirin in US ambulatory care for reducing risks of CVD were at best modest during the period under study, particularly for secondary prevention, where the strongest evidence and most explicit guidelines exist. Aspirin is more underused than statins despite its more favorable cost-effectiveness. Aggressive and targeted interventions are needed to enhance provider and patient adherence to consensus guidelines for CVD risk reduction.
Aspirin is underused compared with statins especially for secondary prevention of cardiovascular disease, despite being more cost-effective.
==== Body
Introduction
Cardiovascular disease (CVD), including myocardial infarction and stroke, is the leading cause of morbidity and mortality in the United States. A broad array of randomized trials have demonstrated the benefits of low doses of aspirin (75–325 mg) [1,2] for both the primary [3–7] and secondary prevention [8–11] of CVD. Most trials demonstrate a 15%–40% reduction in cardiovascular events with chronic aspirin use. Aspirin is unequivocally recommended as a secondary prevention strategy in non-contraindicated patients with known CVD [12,13]. As for primary prevention, the American Diabetic Association recommends regular aspirin for men and women with diabetes mellitus (DM) who are older than 40 y or have additional cardiovascular risk factors [14]. In addition, aspirin is indicated for apparently healthy individuals without CVD or DM but otherwise with an increased cardiovascular risk, which is defined as a 3% or greater risk in 5 y by the US Preventive Services Task Force [2] or a 10% or greater risk in 10 y by the American Heart Association [1]. However, the latest results from the Women's Health Study [7] suggest that careful ascertainment of the absolute benefit and risk on a case-by-case basis is essential to deciding on the use of aspirin therapy in men and, even more so, in women who have showed no clinical manifestations of CVD or diabetes.
Despite the proven benefits of aspirin therapy for reducing cardiovascular risk, aspirin use falls considerably short of recommendations. National surveys of the prescribing of cardiac medications found that aspirin use in visits by patients with coronary heart disease (CHD) increased significantly from 5% in 1980 to 32% in 1995, but then remained unchanged or even declined in subsequent years [15,16]. The Third National Health and Nutrition Examination Survey (also called NHANES III) data showed that among patients with DM, only 37% of those with CHD and 13% of those with risk factors for CHD were regular aspirin users [17]. While aspirin underutilization is also present in other countries [18,19], some evidence suggests that the problem is more prominent in the US. For instance, outpatient use of aspirin for secondary prevention ranged from approximately 40% to 90% in many European countries, in comparison to approximately 24% in the US [15,20–23]. Greater aspirin use is associated with middle to older age (55–75 y old), male gender, diagnosis of hyperlipidemia, smoking, having medical insurance, revascularization or coronary angioplasty, and use of other medications [24–28].
Despite ample evidence of aspirin underutilization, research on national trends of outpatient aspirin use by CVD risk category is limited. Using two companion national datasets on ambulatory care in the US, our study tracked changes from 1993–2003 in reported aspirin use by CVD risk status, distinguishing between secondary and primary prevention. Multiple reasons may account for the widespread aspirin underutilization, one being lower priority assigned to aspirin therapy compared to other medications available for CVD risk reduction. To explore this possibility, we examined the priority given to aspirin in comparison to statins. We also examined patient and physician contributors to shortfalls in aspirin use.
Methods
The Stanford University Institutional Review Board determined that this study was exempt from “human subjects” requirements.
Data Sources
We obtained annual data 1993–2003 from the National Ambulatory Medical Care Survey (NAMCS) and the Outpatient Department component of the National Hospital Ambulatory Medical Care Survey (NHAMCS). The National Center for Health Statistics provides complete descriptions of both surveys and yearly data at http://www.cdc.gov/nchs/about/major/ahcd/ahcd1.htm. These surveys, particularly NAMCS, have been validated against other data sources [29,30] and utilized in past research on aspirin use for CVD risk reduction [16,25].
In brief, NAMCS captured health-care services provided by private office-based physicians, while NHAMCS captured services offered at hospital outpatient departments. The sampling universe for NAMCS was office-based, patient-care physicians in 15 specialty strata from the master files maintained by the American Medical Association and American Osteopathic Association. The sampling frame for NHAMCS included short-stay (shorter than 30 d) hospitals, or general-specialty (medical or surgical) or children's general hospitals. Both surveys utilized multistage probability sampling procedures, which enable researchers to generate nationally representative estimates. Between 1993 and 2003, annual participation rates among physicians selected for NAMCS averaged 70%, while the participation rate of selected hospitals with outpatient departments was 90% in NHAMCS. Standard encounter forms were completed for a systematic random sample of patient visits during randomly assigned reporting periods. Yearly encounter forms varied slightly between NAMCS and NHAMCS and were revised every 2 y. We based this study on variables common to NAMCS and NHAMCS over time, including patient demographics, visit characteristics, reasons for visit (up to three), diagnoses (up to three), and new and continuing medications (up to five in 1993–1994, six in 1995–2002, and eight in 2003). Item nonresponse rates were mostly 5% or less in both surveys for all years.
Patients
CHD risk categorization
We defined four mutually exclusive categories of CVD risk based on the presence of specific diagnoses and risk factors. The presence of CHD, myocardial infarction, stroke or transient ischemic attack, peripheral vascular disease, claudication, or angina defined high CVD risk. In the absence of known CVD, visits by patients with DM who were older than 40 y or had additional risk factors (i.e., hypertension, smoking, dyslipidemia, and/or albuminuria) were defined as intermediate risk. The remaining patients were defined in a second intermediate risk category if they met either of the following criteria: (1) Two or more major CVD risk factors (i.e., hypertension, smoking, and/or dyslipidemia) among men age 45–54 and women age 55–64; or (2) One or more risk factors among men older than 55 and women older than 65. Patient visits ineligible for any of the former three categories were considered low risk. The absence of data elements such as family history of premature CHD or levels of high-density lipoprotein cholesterol precluded more accurate risk stratification according to the Framingham risk scoring [31].
Patient visit characteristics
We included the following patient visit characteristics: patient age, gender, race/ethnicity, health care insurance status, visit status, US census region, metropolitan area status, and physician specialty. Health care insurance was classified as private/commercial insurance, public insurance (i.e., Medicare and Medicaid), and other insurance (e.g., workers' compensation and self-pay). Visit status distinguished first-time visits from return visits to a physician's practice. Physician specialty was available only from NAMCS, which contributed more than 90% of the total weighted visits for each of the study years. We categorized physician specialties into cardiology, internal medicine, general and family practice, and a category encompassing all others.
Measures
Our primary analytic goals were to assess the probability of aspirin use by CVD risk and its relationship to patient visit characteristics. The probability of aspirin use was defined as the proportion of non-contraindicated patient visits in which aspirin or a therapeutically equivalent medication was reported as a new or continuing medication. Measuring the probability of use by CVD risk provided a means to estimate the gaps between current practice and evidence-based medicine regarding aspirin therapy. We defined aspirin therapy as reported use of generic or brand name forms of aspirin, clopidogrel, ticlopidine, or non-narcotic combination analgesics containing aspirin. The number of patient visits in which clopidogrel or ticlopidine was reported is too small to allow their use over time being tracked separately. Oral anticoagulants are not considered aspirin equivalents and are not recommended for the primary or secondary prevention of CVD in a vast majority of patients. Moreover, judging the appropriateness of using or avoiding aspirin for someone who is already on anticoagulant therapy required more clinical detail than our data sources can provide. Therefore, we felt it was appropriate to exclude patients on anticoagulant therapy. We were unable to assess patients' use of over-the-counter aspirin if it was not reported on the encounter form. We excluded visits by patients younger than 21 y and those with bleeding tendency, gastrointestinal bleeding, anticoagulant therapy, or clinically active hepatic disease.
Statistical Analyses
Statistical analyses accounting for sampling weights and the complex sample designs of NAMCS and NHAMCS were performed using SAS for Windows software (SAS Institute, Cary, North Carolina, United States) and SAS-callable SUDAAN software (RTI, Research Triangle Park, North Carolina, United States). The unit of analysis in both surveys was the patient visit. Comparisons of NAMCS and NHAMCS suggested limited differences on key outcome measures. We therefore combined the two surveys to obtain a wider range of outpatient settings and a broader socioeconomic spectrum of patients seeking ambulatory care. Also, to minimize random fluctuations between years, we analyzed data in 2-y groupings, except for 2003, for depicting temporal trends in aspirin use. The SAS SURVEYMEANS procedure was performed, which generated national estimates of the probability of aspirin use by CVD risk and corresponding 99% confidence intervals (CIs). We chose to report 99% CIs in following National Center for Health Statistics analytical guidelines and also as a conservative measure to avoid over-interpretation of the findings. Chi-square tests were performed using PROC CROSSTAB in SUDAAN to examine the association of aspirin use with each patient visit characteristic based on combined 1993–2003 NAMCS and NHAMCS data. The independent effect of each patient visit characteristic on aspirin use after controlling for all other characteristics was assessed with a multivariate logistic regression model using PROC RLOGISTIC in SUDAAN. The model produced adjusted odds ratios and 99% CIs.
Results
The volume of outpatient visits by patients identified as being at elevated risk for future CVD events, particularly those at intermediate risk, rose markedly over the study period. The number of high-risk patient visits increased by 33% from 44.2 (99% CI, 41.0–47.4) million in 1993–1994 to 58.8 (54.0–63.6) million in 2001–2002. The number of intermediate-risk patient visits in which a diagnosis of DM was noted more than doubled, from 40.5 (37.1–43.9) million to 83.3 (77.4–89.3) million, and for those with multiple risk factors the increase was 57%, from 70.2 (65.7–74.7) million to 110.4 (102.8–118.0) million. The number of low-risk patient visits rose by 23%, from 975.4 (962.6–988.2) million to 1.20 (1.18–1.22) billion. In 2003, the number of patient visits in each of the four risk categories was 29.5 (22.5–36.6) million for high-risk patients, 39.9 (32.0–47.9) million for intermediate-risk patients, 55.8 (45.5–66.2) million for those with multiple risk factors, and 626.9 (537.1–716.7) million for those with low risk.
Trends over time showed improved, though still substantially suboptimal, aspirin use in the high and intermediate risk categories, with sustained improvements seen beginning in 1999–2000 (Figure 1). The probability of aspirin use among patient visits in 1993–1994 was 21.7% (18.8%–24.6%) for the high-risk category, 3.5% (2.0%–5.0%) for the diabetic, intermediate-risk category, and 3.6% (2.6%–4.6%) for the other intermediate-risk category. The probabilities for these three risk categories fluctuated somewhat but remained essentially unchanged through 1999–2000. Increases were observed in 2001–2002 and persisted in 2003. The probability of aspirin use in 2003 was 32.8% (25.2%–40.4%) for the high-risk category, 11.7% (7.8%–15.7%) for the diabetic, intermediate-risk category, and 16.3% (11.4%–21.2%) for the other intermediate-risk category. Aspirin use remained 1%–3% among low-risk patient visits.
Figure 1 National Trends in Aspirin Use in Patient Visits Defined as Low Risk, Intermediate Risk, DM, or High Risk
To explore the relative priority assigned to aspirin and statins, we examined trends in the co-prescribing of the medications. For this series of analyses, the number of visits by patients with DM was relatively small and were therefore grouped with those with known CVD to compose the high-risk category. Both aspirin and statins were used more frequently when the other therapy was present; however, improvements over time were more evident for statin use among aspirin-treated patient visits than for aspirin use among statin-treated patients. Specifically, the proportion of visits by high-risk patients on aspirin while a statin was used declined modestly from 36.5% (24.9%–48.2%) in 1993–1994 to 25.6% (20.1%–31.1%) in 1999–2000 but then rebounded to 43.9% (35.1%–52.8%) in 2003 (Figure 2). In contrast, statin use among visits by high-risk patients on aspirin grew successively from 11.6% (7.4%–15.7%) to 54.3% (45.7%–63.0%) (Figure 2). Of visits by intermediate-risk patients, the probability of aspirin use when on a statin increased from 6.0% (1.4%–10.6%) in 1993–1994 to 33.8% (21.5%–46.0%) in 2003, while the probability of statin use when on aspirin rose from 8.8% (2.2%–15.3%) to 48.1% (35.2%–61.0%) (Figure 3).
Figure 2 National Trends in Aspirin and Statin Use When the Other Therapy Is Present among High-Risk Patient Visits
Vertical bars indicate 99% CIs.
Figure 3 National Trends in Aspirin and Statin Use When the Other Therapy Is Present among Intermediate-Risk Patient Visits
Vertical bars indicate 99% CIs.
The association of greater aspirin use with higher CVD risk was confirmed by multivariate logistic regression (Table 1). After adjusting for patient visit characteristics and the number of medications reported, aspirin use was over four time as likely among visits by high-risk patients and approximately two times as likely among visits by patients with multiple risk factors as it was among low-risk patient visits. The odds ratio was marginally significant for the diabetic, intermediate-risk category. The significance of increases in aspirin use over time did not sustain in the multivariate logistic regression. As for patient visit characteristics, lower probability of aspirin use was found among 20- to 44-y-olds (versus those 45 y or older), women (versus men), visits to noncardiologists (versus visits to cardiologists), return visits (versus first-time visits), and visits to private physician offices (versus visits to hospital outpatient departments). Finally, the probability of aspirin use was positively associated with the number of medications reported (odds ratio, 1.71; 99% CI, 1.65–1.76).
Table 1 Predictors of Aspirin Use With Combined 1995–2002 NAMCS and NHAMCS Data
Discussion
This study documents national trends in the probability of aspirin use by CVD risk category among patient visits to office-based physicians and hospital outpatient departments. Some improvements were observed over time in the use of aspirin for both the secondary and primary prevention of CVD. However, the magnitude of those improvements is minimal relative to the substantial gaps between clinical practice and evidence-based recommendations. The gaps observed with secondary prevention are particularly concerning, given the existence of conclusive clinical evidence and unequivocal practice guidelines. The use of aspirin among primary prevention patients, including those with diabetes, also appears to be suboptimal, but additionally may reflect uncertainty about the evidence. Our analysis also suggests that despite aspirin's more favorable cost-effectiveness, statins have been prioritized ahead of aspirin as therapy for reducing CVD risk.
While ample evidence attests to the underuse of aspirin in reducing risks of CVD, this study uniquely provides an 11-y trajectory of aspirin use in US outpatient settings and reveals that improvements have been at best modest. The magnitude of improvements seems particularly small in the context of often-repeated national guidelines and abundant clinical evidence supporting aspirin use for the prevention of CVD, particularly in patients with known CVD. Even in 2003, aspirin use was reported in only one-third of the visits by patients having CVD, which points to widespread under-appreciation of aspirin as an efficacious and cost-effective secondary prevention therapy. The usage was 12% among visits by diabetics, a group at increased cardiovascular risk. This was lower than the 16% found among visits by patients with multiple major cardiovascular risk factors for whom evidence supporting prophylactic aspirin therapy is less definitive. The continued increases in aspirin use since 1999–2000 may reflect heightened awareness of the benefits of aspirin in reducing cardiovascular morbidity and mortality, mediated through intensified dissemination of national guidelines and clinical trial findings [2,9–11,32]. We did not find evidence of aspirin overuse in low-risk patients.
Compared with clinical practice in Europe [22,23], our study results add support to the observation that underuse of aspirin is more problematic in the US. The genesis of this gap is likely multifactorial and open to postulation. For instance, US physicians may face greater pressure than their European colleagues to prescribe newer medications as a result of less-restrictive regulations on drug advertising. Also, direct-to-consumer advertising has been shown to change patient-physician relationships and physician prescribing behavior.
The widespread aspirin underutilization could be partly due to uncertainties in risk assessment. Health care providers show little consistency as to how much risk of excess bleeding is acceptable, which may partly account for the variability in aspirin prescribing [33]. Data indicate that aspirin use is linked to approximately 2.5%–4.5% of the annual upper gastrointestinal events (symptomatic ulcers) and 1%–1.5% of serious complications, such as severe bleeding, perforation, and obstruction [34]. These risk estimates should be evaluated in the context of average reductions of 15−40% in cardiovascular events when aspirin is used as a preventive therapy [3,10,11,14,35]. Accurate risk assessment can be difficult at the individual patient level, especially when discrepancies arise between verbal and written medical history information [36]. Aspirin resistance may also limit the rates of aspirin use. However, the frequency of aspirin resistance is less well known and may range from 5% to 60% [37]. In some patients it may be dose related. Lee et al. [38] indicate that even a low-dose aspirin of 100 mg or less may increase aspirin resistance in patients with coronary artery disease.
Past research also suggests that physicians may assign lower priority to aspirin therapy than to other cardiovascular risk-lowering therapies [25,26,36], and our evaluations of the co-prescription of aspirin and statins support this assessment. We found that aspirin and statin use was significantly higher when the other therapy was present; however, the incremental use became progressively greater for statins over time. Beginning in 1997–1998, statin use in the presence of aspirin transcended aspirin use in the presence of statins for both the high- and intermediate-risk categories, and the gaps remained wide through 2003. These results suggest that even though statins themselves may be underused, aspirin is given even lower priority for lowering cardiovascular risk. These findings are intriguing because both therapies reduce cardiovascular risk by similar magnitudes but differ vastly in cost; statins are prioritized despite the far greater cost-effectiveness of aspirin [39–41]. Also, secondary analyses of clinical trial data indicate that aspirin and statins used in combination may be more effective at reducing the relative risk of CVD events than when used alone [42].
Statins are newer and more intensely advertised than aspirin, which may partly explain the preferential use of these drugs. Lipid-lowering medications already ranked the fifth most promoted drug class in the US in 1998 [43]. Statins are proven effective for both the primary and secondary prevention of CVD, whereas the effectiveness of aspirin in primary prevention is less certain. Also, while they are increasingly used as a prophylactic treatment, statins are still most commonly prescribed to people with hyperlipidemia. In contrast, use of aspirin is not specific to any risk factor in the prevention of CVD and therefore may be neglected by many physicians who are trained to perform in an overly acute-care-centered health care system. In addition, statins may be perceived to have a more favorable side-effect profile than aspirin, which has been shown to increase the risk of severe gastrointestinal and cerebral hemorrhage [34]. Finally, our comparison of aspirin and statin use is confounded by the likelihood of underreporting of over-the-counter aspirin use by participating physicians and clinical staff.
In agreement with previous findings, lower aspirin use is associated with female gender, younger age, noncardiologist care, and care in the private office setting [15,16,27,28,44,45]. The appropriateness of prophylactic aspirin therapy among women, particularly those under 65 y of age, is yet to be determined in light of the new evidence from the Women's Health Study [7]. However, variations of aspirin use by physician specialty and type of health care setting raise questions about equity in the process of care. As a result of high penetration of managed care, patients are increasingly less likely to see a specialist such as cardiologist, unless referred by their primary care provider [45]. Primary care providers, including those who practice in private offices, are expected to adhere more diligently to practice guidelines in this area that was previously the domain of specialists.
Our findings should be interpreted in the context of several limitations of the data sources used. Both NAMCS and NHAMCS are designed to produce national estimates on the basis of patient visits, and they provided no way to link multiple visits by the same patient. The per-patient visit nature of our analysis may lead to overestimation of aspirin use, particularly for high-risk patients, due to more frequent visits by sicker patients and indiscriminate reporting of sporadic and long-term use of aspirin. Individuals who have visited an ambulatory care facility may differ from those who fail to do so or do so less frequently. However, observed aspirin use may underestimate actual administration due to its low-cost, over-the-counter availability, although participating physicians and clinical staff are instructed to record nonprescription medication. In an attempt to indirectly gauge the potential of underreporting of aspirin use, Stafford [16] studied the reporting of multivitamin use during pregnancy and nonprescription analgesic use for osteoarthritis, and concluded that these surveys capture a reasonably substantial proportion of nonprescription medication use. By limiting the number of medications reported to six or fewer, some medications, particularly those perceived as less critical for the treatment of primary diagnoses, may not be reported. When we compared patient visits in which the maximum number of medications were reported with visits in which fewer were reported, we found no differences in the likelihood of aspirin use. If aspirin is, in fact, under-reported, less clinical attention and priority may be given to aspirin use compared to other therapies.
While these data limitations present certain difficulties in interpreting the absolute usage of aspirin, they should have limited impact on our trend analysis. The extent of under-reporting may have attenuated over time due to increased awareness of its effectiveness in cardiovascular risk reduction, which could partly explain the increasing trends in use that we observed. We have no reason to believe that under-reporting varies so substantially by patient visit characteristics that it could have confounded our multivariate logistic analysis.
In conclusion, improvements in aspirin use for reducing risks of CVD among US outpatients are at best modest, and substantial treatment gaps persist, particularly in secondary prevention, for which definitive evidence of benefits is available. Aspirin is more underused than statins despite its more favorable cost-effectiveness. Marked changes in clinical practice are unlikely to occur unless more aggressive, innovative means are implemented to enhance health care provider and patient adherence to consensus guidelines on aspirin therapy to prevent CVD events. In particular, targeted interventions may be warranted in patient subpopulations in which aspirin use is lower than average, including women, young adults, and ethnic minorities. Targeted continuing medical education for primary care providers especially in solo or small-group practices, may introduce greater consistency into practice by specialty and practice setting.
Patient Summary
Background
Aspirin is known to be effective in lessening the chance of heart attack, stroke, or other cardiovascular diseases that may occur when blood vessels are blocked by blood clots. Therefore, guidelines recommend that certain groups of people take aspirin regularly either to prevent such clots forming in the first place, or after such a clot has formed to prevent further clots. However, aspirin may increase the chance of bleeding in some people; hence it is important that the benefits of taking aspirin are balanced against possible side effects.
Why Was This Study Done?
The researchers wanted to investigate temporal patterns of aspirin use among patients who would potentially benefit from taking it, and ask whether there were any particular reasons—for either patients or their health care providers—that influenced such use.
What Did the Researchers Do and Find?
They used data over 11 years from two nationwide surveys in the US that study prescribing patterns in outpatients. Some improvements were observed between 1993 and 2003 in the use of aspirin among patients with known CVD and those without. However, the magnitude of those improvements is minimal relative to the substantial gaps between clinical practice and evidence-based recommendations. From 1997 to 1998 onward, statins were used more frequently compared with aspirin as prophylactic therapy for reducing cardiovascular disease risk. Greater aspirin use was seen most frequently in people of advanced age, who were male, who were being cared for by cardiologists (rather than general physicians or other specialists), and who were being seen in hospital outpatient departments (rather than private practices).
What Do These Findings Mean?
Although there is very good evidence that aspirin is particularly useful when given after a cardiovascular event—so-called secondary prevention—there were only modest increases in the use of aspirin in this period. Aspirin is less frequently used than statins, despite its greater cost-effectiveness. Innovative interventions are needed to enhance patients' and health care providers' understanding of and adherence to the guidelines that have been developed on reducing the risk of cardiovascular disease.
Where Can I Get More Information Online?
MedlinePlus has information on aspirin and related drugs:
http://www.nlm.nih.gov/medlineplus/druginfo/uspdi/202515.html
Omni is a UK-based free catalog of hand-selected and evaluated Internet resources in health and medicine, including a page of links on aspirin:
http://omni.ac.uk/browse/mesh/D001241.html
This research was sponsored by an unrestricted grant from Bayer Pharmaceutical Corporation. Additional support was provided by a research award from the Agency for Healthcare Research and Quality (R01-HS11313). Both funding sources had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Citation: Stafford RS, Monti V, Ma J (2005) Underutilization of aspirin persists in US ambulatory care for the secondary and primary prevention of cardiovascular disease. PLoS Med 2(12): e353.
Abbreviations
DMdiabetes mellitus
CHDcoronary heart disease
CIconfidence interval
CVDcardiovascular disease
NAMCSNational Ambulatory Medical Care Survey
NHAMCSNational Hospital Ambulatory Medical Care Survey
==== Refs
References
Berg AO Atkins D Aspirin for the primary prevention of cardiovascular events. US Preventive Services Task Force Ann Intern Med 2002 136 157 160 11790071
Pearson TA Blair SN Daniels SR Eckel RH Fair JM AHA guidelines for primary prevention of cardiovascular disease and stroke: 2002 update: Consensus panel guide to comprehensive risk reduction for adult patients without coronary or other atherosclerotic vascular diseases. American Heart Association Science Advisory and Coordinating Committee Circulation 2002 106 388 391 12119259
Hansson L Zanchetti A Carruthers S Dahlof B Elmfeldt D Effects of intensive blood-pressure lowering and low-dose aspirin in patients with hypertension: principal results of the Hypertension Optimal Treatment (HOT) randomized trial. HOT Study Group Lancet 1998 351 1755 1762 9635947
Steering Committee for the Physician's Health Study Research Group Final report on the aspirin component of the ongoing Physician's Health Study. The Steering Committee for the Physician's Health Study Research Group N Engl J Med 1989 321 129 135 2664509
Medical Research Council's General Practice Research Framework Thrombosis prevention trial: Randomized trial of low-intensity oral anticoagulation with warfarin and low-dose aspirin in the primary prevention of ischaemic heart disease in men at increased risk. The Medical Research Council's General Practice Research Framework Lancet 1998 351 233 241 9457092
Collaborative Group of the Primary Prevention Project Low-dose aspirin and vitamin E in people at cardiovascular risk: A randomized trial in general practice. Collaborative Group of the Primary Prevention Project Lancet 2001 357 89 95 11197445
Ridker PM Cook NR Lee IM Gordon D Gaziano JM Manson JE A randomized trial of low-dose aspirin in the primary prevention of cardiovascular disease in women N Engl J Med 2005 352 1293 1304 15753114
ISIS-2 Collaborative Group Randomized trial of intravenous streptokinase, oral aspirin, both, or neither among 17,187 cases of suspected acute myocardial infarction: ISIS-2. ISIS-2: Second International Study of Infarct Survival, Collaborative Group Lancet 1988 2 349 360 2899772
Anti-Platelet Trialists' Collaboration Collaborative meta-analysis of randomized trials of anti platelet therapy for prevention of death, myocardial infarction, and stroke in high risk patients: Anti-Platelet Trialists' Collaboration Br Med J 2002 324 71 86 11786451
Weisman SM Graham DY Evaluation of the benefits and risks of low-dose aspirin in the secondary prevention of cardiovascular and cerebrovascular events Arch Intern Med 2002 162 2197 2202 12390062
Chen Z Sandercock P Pan H Counsell C Collins R Indications for early aspirin use in acute ischemic stroke. A combined analysis of 40,000 randomized patients from the Chinese Acute Stroke Trial and the International Stroke Trial Stroke 2000 31 1240 1249 10835439
Harrington RA Becker RC Ezekowitz M Meade TW O'Connor CM Antithrombotic therapy for coronary artery disease: The Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy Chest 2004 126 513S 548S 15383483
Albers GW Amarenco P Easton JD Sacco RL Teal P Antithrombotic and thrombolytic therapy for ischemic stroke: The Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy Chest 2004 126 483S 512S 15383482
American Diabetes Association Aspirin therapy in diabetes Diabetes Care 2004 27 S72 S73 14693931
Stafford RS Radley DC The underutilization of cardiac medications of proven benefit, 1990 to 2002 J Am Coll Cardiol 2003 41 69 72 12570947
Stafford RS Aspirin use is low among United States outpatients with coronary artery disease Circulation 2000 101 1097 1101 10715254
Rolka DB Fagot-Campagna A Narayan KM Aspirin use among adults with diabetes. Estimates from the Third National Health and Nutrition Examination Survey Diabetes Care 2001 24 197 201 11213865
Kramer JM Newby LK Chang WC Simes RJ Van de Werf F International variation in the use of evidence based medicines for acute coronary syndromes Eur Heart J 2003 24 2133 2141 14643274
Venturini F Romero M Tognoni G Patterns of practice for acute myocardial infarction in a population from ten countries Eur J Clin Pharmacol 1999 54 877 886 10027664
O'Connor GT Quinton HB Traven ND Ramunno LD Dodds TA Geographic variation in the treatment of acute myocardial infarction: The Cooperative Cardiovascular Project JAMA 1999 281 627 633 10029124
Bennett KE Williams D Feely J Under-prescribing of cardiovascular therapies for diabetes in primary care Eur J Clin Pharmacol 2003 58 835 841 12698311
EUROASPIRE II Lifestyle and risk factor management and use of drug therapies in coronary patients from 15 countries: Principal results from EUROASPIRE II Euro Heart Survey Programme. EUROASPIRE II Study Group Eur Heart J 2001 22 526 528 11259137
Steffenino G Galliasso M Gastaldi C Ricca N Mangiacotti B Nurses' observational study on the practice of secondary prevention in a cardiovascular department Ital Heart J 2003 4 473 478 14558299
Avezum A Makdisse M Spencer F Gore JM Fox KAA Impact of age on management and outcome of acute coronary syndrome: observations from the Global Registry of Acute Coronary Events (GRACE) Am Heart J 2005 149 67 73 15660036
Meigs JB Stafford RS Cardiovascular disease prevention practices by U.S. physicians for patients with diabetes J Gen Intern Med 2000 15 220 228 10759996
Califf RM DeLong ER Ostbye T Muhlbaier LH Chen A Underuse of aspirin in a referral population with documented coronary artery disease Am J Cardiol 2002 89 653 661 11897205
Ganz DA Lamas GA Orav EJ Goldman L Gutierrez PR Age related differences in management of heart disease: A study of cardiac medication use in an older cohort. Pacemaker Selection in the Elderly (PASE) Investigators J Am Geriat Soc 1999 47 145 150 9988284
Krumholz HM Radford MJ Ellerbeck EF Hennen J Meehan TP Aspirin in the treatment of acute myocardial infarction in elderly Medicare beneficiaries. Patterns of use and outcomes Circulation 1995 92 2841 2847 7586250
Zell ER McCaig LF Kupronis BA Besser RE Schuchat A A comparison of the National Disease and Therapeutic Index and the National Ambulatory Medical Care Survey to evaluate antibiotic usage Proceedings of the survey research methods section, American Statistical Association 2000 Alexandria (Virginia) American Statistical Association 840 845
Gilchrist VJ Stange KC Flocke SA McCord G Bourguet CC A comparison of the National Ambulatory Medical Care Survey (NAMCS) measurement approach with direct observation of outpatient visits Medical Care 2004 42 276 280 15076827
Wilson PWF D'Agostino RB Levy D Belanger AM Silbershatz H Prediction of coronary heart disease using risk factor categories Circulation 1998 97 1837 1847 9603539
Colwell JA Aspirin therapy in diabetes is underutilized Diabetes Care 2001 24 194 196
Devereaux PJ Anderson DR Gardner MJ Putnam W Flowerdew GJ Differences between perspectives of physicians and patients on anticoagulation in patients with atrial fibrillation: Observational study Br Med J 2001 323 1218 1222 11719412
Laine L The gastrointestinal effects of nonselective NSAIDs and COX-2-selective inhibitors Semin Arthritis Rheum 2002 32 25 32 12528071
ETDRS Aspirin effects on mortality and morbidity in patients with diabetes mellitus. Early Treatment Diabetic Retinopathy Study report 14: ETDRS Investigators JAMA 1992 268 1292 1300 1507375
Short D Frischer M Bashford J Ashcroft D Why are eligible patients not prescribed aspirin in primary care? A qualitative study indicating measures for improvement BMC Fam Pract 2003 4 9 12871601
Martin CP Talbert RL Aspirin resistance: an evaluation of current evidence and measurement methods Pharmacotherapy 2005 25 942 953 16006273
Lee PY Chen WH Ng W Cheng X Kwok JY Tse HF Lau CP Low-dose aspirin increases resistance in patients with coronary artery disease Am J Med 2005 118 723 727 15989905
Marshall T Coronary heart disease prevention: Insights from modelling incremental cost effectiveness Br Med J 2003 327 1 5 12842922
Drummond A Kwok S Morgan J Durrington PN Costs of aspirin and statins in general practice QJM 2001 95 23 26
Probstfield JF How cost-effective are new preventive strategies for cardiovascular disease? Am J Cardiol 2003 91 22G 27G 12505566
Hennekens CH Sacks FM Tonkin A Jukema JW Byington RP Additive benefits of pravastatin and aspirin to decrease risks of cardiovascular disease Arch Intern Med 2004 164 40 44 14718320
Ma J Stafford RS Cockburn IM Finkelstein SN A statistical analysis of the magnitude and composition of drug promotion in the United States in 1998 Clin Ther 2003 25 1503 1517 12867225
Frances CD Go AS Dauterman KW Deosaransingh K Jung DL Outcome following acute myocardial infarction: Are differences among physician specialties the result of quality of care or case mix? Arch Intern Med 1999 159 1429 1436 10399894
Akosah KO Larson DE Brown WM Paul K Schaper A Using a systemwide care path to enhance compliance with guidelines for acute myocardial infarction Jt Comm J Qual Saf 2003 29 245 259
|
16277554
|
PMC1283363
|
CC BY
|
2021-01-05 10:39:28
|
no
|
PLoS Med. 2005 Dec 15; 2(12):e353
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020353
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1627984010.1371/journal.pmed.0020355Research ArticleNeuroscienceOtherNeurology/NeurosurgeryPathologyNeurologyDementiaPathologyPersistent Amyloidosis following Suppression of Aβ Production in a Transgenic Model of Alzheimer Disease Persistent Amyloid in Tet-Off APP MiceJankowsky Joanna L
1
2
*Slunt Hilda H
1
¤Gonzales Victoria
1
Savonenko Alena V
1
Wen Jason C
3
Jenkins Nancy A
4
Copeland Neal G
4
Younkin Linda H
5
Lester Henry A
2
Younkin Steven G
5
Borchelt David R
1
6
*¤1Department of Pathology, The Johns Hopkins University School of Medicine, Baltimore, Maryland, United States of America,2Division of Biology, California Institute of Technology, Pasadena, California, United States of America,3Department of Cell Biology, The Johns Hopkins University School of Medicine, Baltimore, Maryland, United States of America,4Mouse Cancer Genetics Program, National Cancer Institute Frederick Cancer Research and Development Center, Frederick, Maryland, United States of America,5Mayo Clinic Jacksonville, Jacksonville, Florida, United States of America,6Department of Neuroscience, The Johns Hopkins University School of Medicine, Baltimore, Maryland, United States of AmericaAguzzi Adriano Academic EditorZürich UniversitySwitzerland*To whom correspondence should be addressed. E-mail: [email protected] (JLJ); E-mail: [email protected] (DRB)¤ Current address: Department of Neuroscience, McKnight Brain Institute, University of Florida, Gainesville, Florida, United States of America
Competing Interests: The authors have declared that no competing interests exist.
Author Contributions: JLJ and DRB designed the study and wrote the manuscript, JLJ, HHS, VG, JCW, LHY, SGY and DRB performed experiments, NAJ and NGC generated transgenic founders, HAL assisted with data interpretation, and AVS performed statistical analyses.
12 2005 15 11 2005 2 12 e35514 6 2005 22 8 2005 Copyright: © 2005 Jankowsky et al.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
Will Stopping Aβ Production Reverse the Damage in Alzheimer Disease?
Background
The proteases (secretases) that cleave amyloid-β (Aβ) peptide from the amyloid precursor protein (APP) have been the focus of considerable investigation in the development of treatments for Alzheimer disease. The prediction has been that reducing Aβ production in the brain, even after the onset of clinical symptoms and the development of associated pathology, will facilitate the repair of damaged tissue and removal of amyloid lesions. However, no long-term studies using animal models of amyloid pathology have yet been performed to test this hypothesis.
Methods and Findings
We have generated a transgenic mouse model that genetically mimics the arrest of Aβ production expected from treatment with secretase inhibitors. These mice overexpress mutant APP from a vector that can be regulated by doxycycline. Under normal conditions, high-level expression of APP quickly induces fulminant amyloid pathology. We show that doxycycline administration inhibits transgenic APP expression by greater than 95% and reduces Aβ production to levels found in nontransgenic mice. Suppression of transgenic Aβ synthesis in this model abruptly halts the progression of amyloid pathology. However, formation and disaggregation of amyloid deposits appear to be in disequilibrium as the plaques require far longer to disperse than to assemble. Mice in which APP synthesis was suppressed for as long as 6 mo after the formation of Aβ deposits retain a considerable amyloid load, with little sign of active clearance.
Conclusion
This study demonstrates that amyloid lesions in transgenic mice are highly stable structures in vivo that are slow to disaggregate. Our findings suggest that arresting Aβ production in patients with Alzheimer disease should halt the progression of pathology, but that early treatment may be imperative, as it appears that amyloid deposits, once formed, will require additional intervention to clear.
In a transgenic mouse that overexpressed amyloid beta, turning off production via a tetracycline sensitive switch did not decrease the number of amyloid plaques present
==== Body
Introduction
Over a decade ago the amyloid cascade hypothesis predicted that increased levels of amyloid-β (Aβ) peptide lead to secondary pathologies that ultimately culminate in the onset of Alzheimer disease (AD) [1]. Early support for this hypothesis came from genetic studies linking early-onset AD to mutations in the amyloid precursor protein (APP), from which Aβ is derived, and presenilins 1 and 2, which are interchangeable components of a endoprotease complex that releases Aβ from APP (for review see [2,3]). If, as predicted, overproduction of Aβ initiates the cascade of events leading to disease, then therapeutic strategies that lower Aβ levels should either arrest or reverse the progression from peptide to dementia.
Early evidence from clinical trials of antibody-mediated clearance, one of the first Aβ-lowering approaches tested in humans, suggested that treatments designed to reduce amyloid burden may indeed be beneficial. Although the trials were halted because of adverse effects in a subset of volunteers [4,5], further analysis of several patients found evidence that amyloid pathology, and to a lesser degree cognitive function, was improved in proportion to the patient's titer of Aβ-specific antibody [6,7]. While this approach is promising, constant exposure to antibodies that recognize an epitope highly enriched in the brain may have unexpected side effects that will limit its long-term use.
An alternative approach that is being actively pursued for future treatment of AD seeks to lower Aβ levels by limiting its production from the precursor protein APP. Peptide Aβ is released from APP by the action of two enzymes, the β-APP cleaving enzyme 1 (BACE1) and γ-secretase, which cleave the holoprotein at the N- and C-termini of Aβ, respectively. Several inhibitors of γ-secretase have already been produced [8,9], and small molecule inhibitors of β-APP cleaving enzyme 1 are currently being developed [10,11]. The long-term effectiveness of this approach in either humans or model systems, however, has not been reported. Although loss of β-APP cleaving enzyme 1 function can prevent the development of plaques in transgenic mouse models for AD (F. Laird, H. Cai, P. C. Wong, personal communication), it is not known whether the brain can clear pre-existing amyloid deposits once production of Aβ has been suppressed.
Clearly, the amyloid-lowering approach should be rigorously examined in animal models before these reagents are tested in patients. However, the chemical secretase inhibitors most likely to reach human trials are still in development. Therefore, we developed a mouse model of Alzheimer-type amyloid that expresses a controllable APP transgene. This system, commonly known as the tet-off system, can be regulated by analogs of tetracycline administered in food or water [12,13]. The strong expression levels produced with the tet-off vectors, combined with the ability to reduce this expression by several orders of magnitude with tetracycline [14], allowed for a stringent test of how a highly effective pharmaceutical inhibitor of Aβ production would impact the progression of amyloid pathology and whether reversal of these lesions might be possible following such treatment.
Methods
Transgene Construction
We created a tetracycline-responsive chimeric mouse/human APP695Swedish/Indiana (swe/ind) vector by replacing the promoter region of the moPrP.XhoI vector (also known as pPrPpE1/E2,3sal [15]) with the tetracycline-responsive promoter of pTetSplice (Life Technologies, Rockville, Maryland, United States), and then ligating mouse APP with a humanized Aβ domain (mo/huAPP695) cDNA into the new vector. We began by cloning the tetracycline-responsive promoter (bp 6–481) from pTetSplice by PCR using primers that added external BamHI and NotI sites to the 5′ end and a BamHI site to the 3′ end, while destroying XhoI and BamHI sites within the promoter (forward: GCC GGA TCC GCG GCC GCC GTC GAG TTT ACC ACT CCC TAT C; reverse: GCC GGA TCC ACT CTA GAA GAT CCC CGG GTA CCG). We then isolated the moPrP.XhoI intron by amplification with primers that added an external BamHI site to the 5′ end of exon 1 and ran through the Asp718 site of exon 2 (forward: GCC GGA TCC GAT CAG CAG ACC GAT TCT GG; reverse: GCC GGT ACC ACT AGG AAG GCA GAA TGC). This 2-kb intron fragment was cloned into a TA cloning vector (Invitrogen, Carlsbad, California, United States), then excised by Asp718 digestion and ligated to the 6.8-kb Asp718 fragment of moPrP.XhoI containing exon 2, exon 3, the 3′ UTR, and pBluescript to generate an intermediate vector with all three exons and a central intron but no promoter. This vector was then opened at the BamHI site introduced by the intron cloning primer, and ligated to the 0.5-kb BamHI-cut tetracycline promoter fragment. This ligation generated a 9.3-kb vector encoding the tetracycline promoter from pTetSplice with two exons, one intron, and the original 3′ UTR of the moPrP.XhoI vector, all carried in the pBluescript cloning vector.
We incorporated the Swedish (KM570/571NL) and Indiana (V617F) mutations into the mo/huAPP695 cDNA (in BS-KS) by PCR using a four-primer strategy: first, two partially overlapping products were generated in separate reactions using primers that encode the desired mutations (Swedish forward: GGA GAT CTC TGA AGT GAA TCT GGA TGC AGA ATT CCG/Indiana reverse: GGG TGA TGA AAA TCA CGG TTG C; Indiana forward: CAA CCG TGA TTT TCA TCA CCC TGG/M13 reverse). The two PCR products were ligated, digested with BglII and ApaI and cloned back into the original mo/huAPP695-BS-KS vector. Finally, the new APP695swe/ind was subcloned into the XhoI site of the moPrP-tetP vector from above to complete the construct.
Pronuclear Injection, Screening of Founders, and Maintenance of the Lines
The moPrP-tetP-mo/huAPP695swe/ind vector was linearized and the pBluescript domain excised by digestion with NotI. The purified vector was injected into the pronucleus of fertilized eggs from C57BL/6J × C3HeJ F1 matings. Founder animals were screened for the presence of the transgene by three-way PCR using the S36 and PrP-S/PrP-AS primers described below. Transgene-positive founders were bred to animals expressing the tetracycline transactivator (tTA) under control of the calcium-calmodulin kinase IIα (CaMKIIα) promoter obtained from Jackson Laboratory [16] (Bar Harbor, Maine, United States; stock # 3010; B6;CBA-TgN[Camk2a-tTA]1Mmay). The colony was thereafter maintained by crossing single transgenic tTA and APP offspring for each of the four APP lines. All mice were provided fresh food and water ad libitum. Animal protocols were approved by both the Johns Hopkins University and the California Institute of Technology Institutional Animal Care and Use Committees.
Doxycycline Administration
Doxycycline (dox) was administered through commercially available dox-containing chow (BioServ, Frenchtown, New Jersey, United States). The chow contained 200 mg/kg of antibiotic; based on estimated consumption of 5 g per mouse per day, the expected dose to each animal was 1 mg dox per day. The average 25-g animal therefore received 40 μg of dox per gram body weight per day. Chow was changed 1–2 times per week to prevent breakdown of the antibiotic.
Genotyping
Offspring were genotyped for the presence of each transgene by PCR amplification of genomic DNA extracted from a 5-mm tail biopsy. Tails were heated to 95 °C for 45 min in 250 μl of 50 mM NaOH, vortexed, then neutralized with an equal volume of 0.5 M Tris-HCl (pH 5.5). Debris was sedimented by centrifugation, and 3 μl of supernatant was used for amplification.
Genotyping for APP and tTA transgenes was performed in the same PCR reaction, using five separate primers. APP was amplified using forward primer S36 located in the 3′ end of the APP cDNA (CCG AGA TCT CTG AAG TGA AGA TGG ATG) and reverse primer PrP-AS-J located in the 3′ UTR of the vector (CCA AGC CTA GAC CAC GAG AAT GC). The tTA transgene was detected using a primer set that amplified across its two subdomains with tet forward located within the Tn10 tetracycline repressor (CGC TGT GGG GCA TTT TAC TTT AG) and tet reverse within the HSV1 VP16 (CAT GTC CAG ATC GAA ATC GTC). All reactions, whether transgene-positive or not, amplified a segment of the endogenous prion protein gene as a control for DNA quality using a forward primer, PrP-S-J, specific to the mouse PrP open reading frame (GGG ACT ATG TGG ACT GAT GTC GG) and a reverse primer, PrP-AS-J, shared by the 3′ UTR of the endogenous PrP gene and the transgene vector. Amplification reactions were run for 37 cycles at 94 °C for 30 s, 64 °C for 1 min, and 72 °C for 1 min. All samples, transgenic and wild-type, gave a 750-bp product from the endogenous PrP gene. The APP transgene yielded an additional band at 400 bp; the tTA product fell in between at 480 bp.
Immunoblotting/Quantitation
Frozen cortical or whole forebrain tissue was homogenized by sonication in five volumes of phosphate-buffered saline (PBS) with 5 mM EDTA and protease inhibitors (Mammalian cell cocktail, Sigma, St. Louis, Missouri, United States), using a probe sonicator set to 50% output (TEKMAR, Cincinnati, Ohio, United States). After dilution with an equal volume of PBS/EDTA/protease inhibitor, the samples were centrifuged briefly and the supernatant used for analysis. Fifty micrograms (6E10 and CT15) or 5 μg (22C11) of brain homogenate was loaded per lane onto 7.5%, 10%–20%, or 4%–20% Tris-HCl PAGE gels (Bio-Rad Laboratories, Hercules, California, United States) and electrophoresed for several hours in 1× Tris-glycine–sodium dodecyl sulfate (1×TG-SDS) buffer (6E10 and 22C11; Amresco, Solon, Ohio, United States) or 1× Tris-tricine-SDS buffer (CT15; Invitrogen, Carlsbad, California, United States). Proteins were transferred overnight to 0.45-μm Optitran nitrocellulose (Schleicher and Schuell, Keene, New Hampshire, United States) in 1× TG buffer (Amresco). Blots were blocked in PBS containing 5% nonfat dry milk powder, and incubated for 3 h at room temperature in blocking solution with one of the following antibodies: mouse monoclonal 22C11 (kind gift of Konrad Beyreuther and Andreas Weidemann; [17]) diluted 1:1,000, mouse monoclonal 6E10 (Signet Laboratories, Dedham, Massachusetts, United States) diluted 1:2,500, rabbit polyclonal anti-superoxide dismutase 1 (m/hSOD1) [18] diluted 1:2,500 to 1:4,000, or rabbit polyclonal CT15 (kind gift of Ed Koo; [19]) diluted 1:1,000. Subsequently, the blots were washed with PBS containing 0.1% Tween-20, and then incubated with either goat anti-mouse– or goat anti-rabbit–HRP conjugated secondary antibody diluted 1:1,000 in blocking solution. After several additional rinses in PBS with 0.1% Tween-20, blots were developed with enhanced chemiluminescence reagent and imaged with the Bio-Rad Molecular Imager FX system.
Staining intensity within each lane was quantified using the Quantity One image analysis software (Molecular Imager FX, Bio-Rad Laboratories). Background was calculated from across the image and subtracted from the entire file. The signal intensity for each band (corrected signal intensity × pixel number) was then calculated using the Volume report tool.
Slot Blot mRNA Analysis
Five micrograms per sample of total RNA extracted from fresh-frozen brain, liver, kidney, heart, lung, spleen, and skeletal muscle was vacuum-filtered through 0.45-μm Optitran nitrocellulose. After several washes through the manifold with 10× SSC, blots were UV-cross-linked and probed with a radiolabeled ∼350-bp BglII–XhoI cDNA fragment of mo/huAPP695 cDNA. After hybridizing overnight at 65 °C in 1% BSA/1 mM EDTA/0.5 M sodium phosphate buffer (pH 7.2)/7% SDS [20], the blots were washed twice at 65 °C for 30 min each in 0.1% BSA/1 mM EDTA/40 mM sodium phosphate buffer (pH 7.2)/5% SDS before two final 30-min washes at 65 °C with 1 mM EDTA/40 mM sodium phosphate buffer (pH 7.2)/1% SDS. Blots were wrapped wet and exposed to phosphorscreens overnight at room temperature.
Amyloid Histology
Mice were euthanized by ether inhalation and brains removed for immersion fixation in 4% paraformaldehyde/1× PBS. After 48 h in fixative at 4 °C, brains were transferred to PBS, dehydrated in alcohols, treated with cedarwood oil and methylsalicylate, and embedded in paraffin for sectioning.
Hirano silver stain
Silver impregnation histology was performed on 10-μm paraffin-embedded sections by Hirano's modification of the Bielschowsky method [21]. Briefly, sections were deparaffinized through xylene and alcohols into tap water before being placed into fresh 20% silver nitrate solution for 20 min. After being washed thoroughly with distilled water, slides were immersed in 20% silver nitrate solution titrated with fresh ammonium hydroxide. After 20 min, slides were washed with ammonia water before being individually developed with 100 μl of developer (20 ml of 37% formaldehyde, 100 ml of distilled water, 50 μl of concentrated nitric acid, and 0.5 g of citric acid) added to 50 ml of titrated silver nitrate solution. Slides were then rinsed in tap water, fixed in 5% sodium thiosulfate, and dehydrated through alcohols and xylene.
Thioflavin-S staining
Following deparaffinization of sections through xylene and alcohols, amyloid impregnation with thioflavin-S was performed according to the Guntern modification of the standard protocol. Slides holding 10-μm paraffin sections were washed twice in distilled water, then immersed for 5 min in a 0.25% potassium permanganate solution, followed by 5 min in a 1% potassium metabisulfate/1% oxalic acid solution. After this preparation, slides were placed into a filtered aqueous 0.02% thioflavin-S solution (Chroma-Gesellschaft Schmid, Kongen, Germany) for 8 min. Excess stain was removed by two brief rinses in 80% ethanol, then two in distilled water, after which slides were finished in aqueous mounting medium for florescence photomicrography.
Ubiquitin, glial fibrillary acidic protein, and Aβ immunohistochemistry
Prior to immunostaining, slides were deparaffinized by oven heating followed by immersion in xylene. After rehydration through graded alcohols into tap water, endogenous peroxidase activity was quenched by incubation with 3% hydrogen peroxide in methanol. Slides were microwaved for 5–7 min in water, cooled for 5 min, then washed in TBS. Nonspecific staining was blocked for 1 h with 3% normal goat serum and 0.1% Triton-X 100 in TBS. Slides were then placed into primary antibody (rabbit anti-Aβ peptide polyclonal antibody, Zymed Laboratories, South San Francisco, California, United States; rabbit anti-ubiquitin and rabbit anti–glial fibrillary acidic protein (GFAP) polyclonal antibodies, Dako, Carpinteria, California, United States) diluted 1:500 in TBS with 2% normal goat serum and incubated overnight at room temperature. After being washed of excess primary antibody with several changes of TBS, slides were incubated with either the Vectastain Elite anti-rabbit secondary system (anti-Aβ; Vector Laboratories, Burlingame, California, United States) or peroxidase/anti-peroxidase reagents (anti-ubitquitin and anti-GFAP; Sternberger Monoclonals, Lutherville, Maryland, United States) according to the manufacturers' directions. Antibody binding was visualized with diaminobenzidene, and sections were counterstained with hematoxylin.
Filter Trap Assay
An aliquot of each cortical homogenate used for Western blotting above was partially solubilized by the addition of SDS to a final concentration of 1%. Serial 1:2 dilutions were made with 1× PBS/1% SDS, and 100 μl of each dilution was then vacuum-filtered through a pre-wet 0.22-μm cellulose acetate membrane (Schleicher and Schuell) [22]. Each well was washed several times with PBS, after which blots were incubated overnight with polyclonal anti-Aβ antibody (Zymed Laboratories) diluted 1:600 in a blocking solution of 1× TBS/5% nonfat dry milk powder. After washing the blots three times for 10 min each in 1× TBS/0.1% Tween-20, the membrane was incubated for 1 h with HRP-conjugated protein A (Sigma) diluted 1:5,000 in blocking solution. The membranes were again washed three times with 1× TBS/0.1% Tween-20, before antibody binding was detected with enhanced chemiluminescence (PerkinElmer, Boston, Massachusetts, United States). Digital images of each blot were captured with a Molecular Imager FX gel documentation system, and the intensity of Aβ staining was quantified using Quantity One image analysis software.
Aβ ELISA
An aliquot of cortical homogenate generated for Western analysis described above was subjected to a three-step sequential extraction using PBS, 2% SDS, and 70% formic acid (FA). At each step, the sample was sonicated in appropriate buffer and centrifuged at 100,000g for 30 min (1- to 1.5-mo samples) or 60 min (6- to 12-mo samples) at 4 °C as previously described [23–25]. The supernatant was removed for analysis, and the pellet was sonicated in the next solution in sequence. The 2% SDS extracts were diluted in EC buffer, and the FA extracts neutralized with 1M Tris-phosphate buffer (pH 11) then diluted with EC buffer prior to testing. Human Aβ was measured in each fraction using BAN50 for capture (epitope Aβ1–16) and BA27 and BC05 for detection (Aβ40 and Aβ42, respectively) (Takeda Chemical Industries, Osaka, Japan). Total Aβ (mouse + human; 1- to 1.5-mo samples only) was measured in each fraction using BNT77 for capture (epitope Aβ11–28) and BA27 and BC05 for detection. All values were calculated as picomoles per gram based on the initial weight of cortical tissue.
Activity Monitoring
Daily basal activity was studied in 28 CaMKIIα-tTA × tet-APPswe/ind line 107 mice between 4 and 5 mo of age. Animals were separated into individual cages immediately before the start of each experiment (n = 3–6 per genotype untreated, 2–5 per genotype dox-reared). The cages were placed inside activity-monitoring frames designed to count every time the animal passed through one of three photobeams spanning the width of the cage (San Diego Instruments, San Diego, California, United States). Experiments were started midway through the light phase of the day, and data were collected in 1-h bins for the following 48 h. Testing rooms were maintained on the same 13:11 h day:night cycle as the main animal housing areas and were closed to entry during the experiment.
Statistical Analyses
Statistical analyses of protein expression, ELISA data, and filter trap assays were performed by ANOVA with Tukey's honest significant difference post-hoc test applied to significant main effects or interactions (Statistica 6.0, StatSoft, Tulsa, Oklahoma, United States). In cases of positively skewed data distribution, log10(x + 0.5) transformation was applied to the raw data before submitting them to ANOVA.
Results
We used the tet-off transgene system to express a double mutant version of chimeric mo/huAPP695 (swe/ind KM570, 571NL, and V617F) from a tetracycline-responsive promoter [12,13]. Transgenic APP expression was activated by crossing the APPswe/ind mice to animals producing tTA under control of the CaMKIIα promoter [16]. After initial screening of founders, we identified four lines of tet-APPswe/ind mice that produced very high levels of transgene product in offspring coexpressing tTA (Figures 1 and Figure S1). Compared to a standard APP transgenic line used for previous amyloid studies by our laboratory (line C3–3; [15,26,27]), we estimated that the four controllable lines produce transgenic APP protein at 10- to 30-fold over endogenous levels (Figure S1). This estimate was confirmed by direct comparison of APP levels in nontransgenic and tet-off APP mice using an antibody that recognizes both endogenous APP (and amyloid precursor-like protein 2) and the transgenic protein (monoclonal antibody 22C11; Figure 1D).
Figure 1 Control of Transgenic APP Expression by Dox
(A) Western blotting for transgenic APP using the human-specific 6E10 antibody shows expression of full-length protein in forebrain tissue from young predeposit double transgenic animals (line 107) and its suppression following dox treatment. Untreated animals show high levels of transgene expression; protein levels drop dramatically in animals acutely treated with dox for 2 wk. A faint, but detectable band of full-length protein remains in acutely treated animals that can be eliminated in mice born and raised on dox.
(B) Immunoblotting with the N-terminal antibody 22C11 to detect both transgenic and endogenous protein shows that dox treatment reduces APP/APLP to levels found in nontransgenic mice.
(C) Measurement of signal intensity from the Western blot in (A) shows transgenic APP levels are decreased more than 95% by dox in both acutely and chronically treated animals (97.2% for 4 wk + 2 wk dox, 98.0% for reared on dox versus 4 wk untreated; ANOVA effect of treatment group F
4,15 = 85.55, p < 0.001). APP levels in 4 wk + 2 wk dox, reared on dox, and nontransgenic (NTg) were not significantly different (p > 0.9, Tukey post-hoc test).
(D) Measurement of signal intensity from the Western blot in (B) shows total APP/APLP levels in dox-treated tTA/APP mice are significantly lower than in 4-wk-old untreated mice (ANOVA effect of treatment group F
4,15 = 84.41, p < 0.001) and indistinguishable from those of nontransgenic animals (p > 0.9, Tukey post-hoc test). *, p < 0.001 versus untreated 4-wk-old mice, Tukey post-hoc test applied to significant effect of group ANOVA.
Importantly, all four new lines of tet-off APP mice showed nearly complete suppression of the transgene following dox treatment (Figures 1 and Figure S1). We focused on one of the four lines, line 107, to examine in more detail the time dependence and extent of transgene suppression following either acute or chronic treatment with dox. Two dox-treated groups were compared to two untreated groups: one group of mice was born and raised on dox, a second group was treated with dox for 2 wk starting at 1 mo of age (4 wk + 2 wk dox); two untreated groups kept on normal chow were harvested at either 4 or 6 wk of age. Animals born and raised on dox harbored no transgenic APP (Figure 1A). Following as little as 2 wk of dox treatment, transgenic APP expression was reduced by more than 95% compared to pre-dox levels. The residual expression remaining in acutely treated mice represents less than 4% of the transgenic protein produced in the absence of dox (Figure 1C), and likely results from slight leakage at the level of transcription (data not shown). Importantly, the total amount of APP (endogenous plus transgenic) and related APLPs in both acute and chronically treated animals was statistically indistinguishable from that in nontransgenic mice (Figure 1D; statistical analyses for experiments throughout the study are presented in the accompanying figure legends).
To ensure that Aβ production was suppressed in concert with the dox-mediated inhibition of its precursor APPswe/ind, we measured Aβ40 and Aβ42 levels by ELISA in forebrain homogenates from young tet-off animals. At 1 mo of age, the mice lacked visible amyloid aggregates that might act as an intractable reservoir of peptide remaining in the brain after the transgene had been suppressed. To further ensure we could detect any such insoluble aggregates that might bias our measure of changes in peptide synthesis, we performed a sequential three-step extraction with PBS, 2% SDS, and 70% FA that would separate peptides by solubility. We compared the levels of human transgene-derived Aβ40 and Aβ42 in untreated mice at 4 and 6 wk of age to animals that had either been born and raised on dox or that had been left untreated for 4 wk and then placed on dox chow for 2 wk prior to harvest (the same groups described above for immunoblot analysis of APPswe/ind levels, line 107). Consistent with the reduction in full-length APPswe/ind synthesis shown by immunoblot (see Figure 1), we found that transgene-derived Aβ levels were completely suppressed in animals born and raised on dox, and were sharply reduced following acute (2 wk) antibiotic treatment. Compared to the levels in untreated 4-wk-old mice, PBS-soluble Aβ42 dropped by 95.2% following 2 wk of dox treatment and by 99.2% with chronic treatment (Figure 2A). Similarly, SDS-soluble Aβ42 decreased by 75.2% and 94.8% following 2-wk or lifelong dox treatment (Figure 2B). Only the FA fraction revealed a small dox-resistant pool of peptide in acutely treated animals that we believe represents stable predeposit aggregates that have already accumulated by 4 wk of age when treatment was begun (Figure 2C). Indeed, animals that were born and raised on dox did not harbor this reservoir of treatment-resistant peptide, with 96.3% less Aβ42 than untreated 4-wk-old mice. Measurement of total Aβ in chronically treated mice, including endogenous and transgene-derived peptide, demonstrated that Aβ levels in tet-off APP mice were reduced to the level of endogenous peptide found in nontransgenic animals (Figure 2D). Taken together with the immunoblotting data for full-length APPswe/ind, the ELISA measurements indicate that dox-mediated suppression of transgenic APPswe/ind synthesis leads to parallel reduction of Aβ levels.
Figure 2 Aβ Levels Are Dramatically Reduced by Transgene Suppression
Cortical homogenates from young, predeposit tTA/APP mice used for Western blot in Figure 1 (line 107) were fractionated by sequential multi-step extraction with PBS, 2% SDS, and 70% FA followed by human-specific Aβ ELISA to measure transgene-derived peptide in each fraction. Aβ40 is shown in white, Aβ42 in black.
(A) PBS-soluble Aβ levels are substantially reduced by both acute and chronic dox treatment (ANOVA, effect of treatment group F
4,24 = 137.10 and 386.01, p < 0.001, for Aβ40 and Aβ42, respectively). Aβ levels in treated animals are indistinguishable from nontransgenic (NTg) animals (p > 0.3, Tukey post-hoc test).
(B) In the young animals tested here prior to the formation of visible amyloid deposits, most Aβ is extracted into the SDS fraction (84% and 76% of all transgene-derived Aβ40 and Aβ42, respectively). As in the PBS-soluble fraction, Aβ levels in the SDS fraction are significantly lowered by dox treatment compared to untreated animals (ANOVA effect of group F
4,24 = 197.57 and 163.48, p < 0.001, for Aβ40 and Aβ42, respectively). Acutely treated animals retained a small (although significant) amount of residual peptide (p < 0.001 compared to nontransgeinc, Tukey post-hoc test), whereas Aβ levels in mice born and raised on dox were reduced to levels indistinguishable from nontransgenic (p > 0.8, Tukey post-hoc test).
(C) The FA-soluble fraction already contains a small but significant pool of aggregated Aβ42 in untreated animals by 4 wk of age (p < 0.05 versus nontransgenic; Tukey post-hoc test applied to significant effect of group ANOVA F
4,24 = 17.11, p < 0.001). By 6 wk of age, the amount of Aβ in the FA fraction is increased significantly preceding the appearance of visible deposits 2 wk later. The FA pool is the only peptide fraction not lowered by acute dox treatment (4 wk untreated = 4 wk + 2 wk dox, p > 0.9, Tukey post-hoc test), consistent with poor turnover of aggregated Aβ species.
(D) Measurements of total Aβ, including both endogenous and transgene-derived peptides, show that animals born and raised on dox harbor Aβ levels identical to nontransgenic animals (p > 0.9, Tukey post-hoc test, effect of group ANOVA F
4,24 = 39.13 and 35.29, p < 0.001, for Aβ40 and Aβ42, respectively). Whereas chronic transgene suppression fully prevents synthesis of both peptides, acute dox treatment fully suppresses Aβ40 levels (p > 0.8, Tukey post-hoc test), but leaves a small amount of nonsuppressed Aβ42. The residual Aβ42 observed in acutely treated young animals derives from uncleared aggregates extracted in the SDS and FA fractions. *, p < 0.05; **, p < 0.005; ***, p < 0.001 versus 4-wk-old untreated mice, Tukey post-hoc applied to significant effect of group ANOVA.
The ELISA data also confirmed that incorporation of the Swedish and Indiana mutations led to high levels of Aβ42, which we predicted would induce rapid plaque formation in untreated animals. Histological characterization of double transgenic (CaMKIIα-tTA × tet-APPswe/ind) mice revealed early-onset amyloid formation in all four new lines. Amyloid plaques were seen in mice as young as 8 wk of age (data not shown). Plaques were limited to the forebrain, including the cortex, hippocampus, olfactory bulb, and striatum, where the CaMKIIα promoter is known to be most active [16,28] (Figure S2). By 6 mo of age, amyloid burden became severe, covering large areas of the cortex and hippocampus (Figure S3). No lesions were seen in the cerebellum or brain stem even at late ages, consistent with CaMKIIα-controlled transgene expression. Unlike what is thought to occur in the human disease, the first visible plaques in the tet-off APP mice are fibrillar-cored deposits. We have noted the same early appearance of cored deposits in other lines of APP transgenic mice that harbor the Swedish mutation [27]. Diffuse plaques were apparent in 6-mo-old tTA/APP mice, and became relatively abundant by 9 mo of age. At older ages (9–12 mo) amyloid deposits were visible in the thalamus, which has also been observed in mice expressing mutant APP via the Thy-1 promoter. The presence of amyloid pathology in this region has been attributed to axonal transport of APP/Aβ to the terminals of cortical neurons in the thalamus [29]. Most importantly, only double transgenic mice, expressing both the tTA and APP transgenes, developed amyloid lesions. Single transgenic mice up to 15 mo of age showed no sign of pathology (Figure S3). Similarly, amyloid pathology can be completely prevented in double transgenic animals born and raised on dox. Animals from our highest expressing line (line 885) maintained on dox for up to 1 y harbored no amyloid pathology (data not shown), indicating that residual leakage of transgene expression in the presence of dox does not provide sufficient Aβ peptide to induce amyloid formation even over long periods.
To mimic therapeutic intervention with inhibitors of Aβ production, we raised a group of 25 double transgenic mice (CaMKIIα-tTA × APP line 107) on normal food until 6 mo of age, when we knew amyloid formation was already well underway in the brain. At 6 mo, half of the animals were switched from normal chow to food containing dox at 200 mg/kg until they were sacrificed at 9 or 12 mo of age. The remaining control animals were kept on standard chow (untreated). In all, four cohorts were created: 6 mo untreated (n = 7), 9 mo untreated (n = 5), 6 mo + 3 mo treated (n = 8), and 6 mo + 6 mo treated (n = 5). Full suppression (>95%) of transgenic APPswe/ind levels in the dox-treated animals was confirmed by immunoblot (Figure 3). To ensure that the transgene could be suppressed as rapidly in 6-mo-old mice with fulminant pathology as it can in young, predeposit animals, we treated an additional set of 6-mo-old animals with dox for 1 wk prior to harvest. Importantly, both APPswe/ind and APP–C-terminal fragment levels were fully suppressed after only 1 wk of treatment, indicating that the in vivo half-life of APPswe/ind and its processed C-terminal fragments are relatively short (Figure 3D).
Figure 3 Robust Transgene Suppression in Older Mice with Preexisting Amyloid Pathology
(A) Cortical homogenates from 6- to 12-mo-old animals used for pathology studies described below (line 107) were immunoblotted with human-specific antibody 6E10 to examine transgene suppression following 3 or 6 mo of dox treatment. The blot was co-immunostained for endogenous superoxide dismutase 1 (SOD1) as a control for loading.
(B) Quantitation of signal intensity from the Western blot shown in (A). Transgenic APP levels are significantly suppressed following 3 or 6 mo of dox treatment (96.9% and 97.6%, respectively). *, p < 0.001 compared to 6-mo-old untreated animals, Tukey post-hoc test applied to significant effect of group ANOVA F
3,12 = 107.22, p < 0.001. These data demonstrate that strong transgene suppression is attained both before and after the onset of amyloid pathology (see Figure 1 for predeposit experiments).
(C) Experimental design. To examine the effects of chronic Aβ suppression on amyloid pathology after the onset of deposition, we compared untreated controls harvested at 6 and 9 mo of age to animals placed on dox at 6 mo of age and harvested after 3 or 6 mo of treatment.
(D) Dox treatment leads to rapid transgene suppression even in 6-mo-old tTA/APP mice. Immunostaining with 6E10 shows APPswe/ind levels are dramatically reduced in 6-mo-old mice treated for 1 wk with dox (upper panel). A separate blot was immunostained for APP C-terminal fragments with CT15 antibody to show that the precursors to Aβ cleavage are decreased in parallel with the full-length protein (middle panel). Costaining for superoxide dismutase 1 was used as an internal control for loading (lower panel, taken from bottom half of 6E10 blot).
Tissue sections from each animal in the four treatment groups were stained for amyloid pathology by Hirano silver, Campbell-Switzer silver, thioflavin-S, and Aβ immunohistochemistry. As expected, the 6 mo untreated cohort displayed moderate amyloid pathology, and the 9 mo untreated cohort progressed to a severe amyloid burden. In contrast, the extent of amyloid pathology in mice from the 6 mo + 3 mo treated or 6 mo + 6 mo treated cohorts closely resembled that of the 6 mo untreated cohort, despite the significant age difference between the treated and untreated groups (Figures 4 and Figure S3). Well-formed plaques remained in the treated animals after 6 mo of transgene suppression, even though as much time was given to clear the lesions as they had taken to form. Moreover, both types of amyloid, diffuse and fibrillar, remained intact throughout treatment. Using the Campbell-Switzer silver stain to distinguish different forms of amyloid, we found diffuse plaques were as persistent as cored deposits (Figure S4). It was nevertheless clear that dox-induced suppression of transgenic APP had completely halted the progression of pathology.
Figure 4 Suppression of Transgenic APP Arrests Progression of Amyloid Pathology
(A) Aggregated Aβ was quantified in cortical tissue from dox-treated and control tTA/APP mice (line 107) using a filter trap assay. Serial dilutions of protein homogenate were passed through a cellulose acetate filter; protein aggregates larger than the pore size were trapped and immunostained for Aβ.
(B) Quantitation of signal intensity in the linear range of each filter trap dilution series (arrow in [A]) was used to compare aggregate load across treatment groups. Aggregated Aβ increased significantly between 6 and 9 mo of age in untreated mice (significant effect of group ANOVA F
3,18 = 7.85, p < 0.002). This progression of pathology was completely prevented by transgene suppression. The amount of aggregated Aβ was identical in untreated mice at 6 mo of age to that in 9- or 12-mo-old animals treated with dox (p > 0.9, Tukey post-hoc test). Single transgenic tTA samples were included as negative controls and showed no signal above background. *, p < 0.01; **, p < 0.005 versus 9-mo-old untreated mice, Tukey post-hoc test; ***, p < 0.001 versus 9-mo-old untreated mice, Student's t-test.
(C) Amyloid pathology in the hippocampus of representative mice from each treatment group: Hirano silver stain (top row), thioflavin-S (middle row), and Aβ immunohistochemistry (bottom row). Amyloid burden increases dramatically between 6 and 9 mo of age in untreated animals, but remains stable in transgene-suppressed mice over the same period (6 mo + 3 mo dox and 6 mo + 6 mo dox). Single transgenic animals (tTA only shown here) show no sign of amyloid pathology at any age tested.
To confirm that the arrest of plaques without any sign of clearance was not unique to the line 107 mice, we repeated the dox-suppression experiment in a second line of tet-off APP mice (CaMKIIα-tTA × tet-APPswe/ind line 18; n = 22). Again, long-term dox treatment was begun at 6 mo of age, and mice were harvested after 3 mo of transgene suppression (6 mo untreated, n = 8; 9 mo untreated, n = 6; 6 mo + 3 mo treated, n = 8). Immunoblotting for APP confirmed full transgene suppression in the treated animals (Figure S5). As in the line 107 mice described above, amyloid burden worsened substantially in the untreated mice between 6 and 9 mo of age. Suppression of transgene expression abruptly arrested progression of pathology (Figure S6), but again without any sign of reduction. Both silver- and thioflavin-S-positive plaques could still be found in each of the dox-treated animals.
We biochemically measured the amount of aggregated Aβ in the brains of our mice before and after transgene suppression using filter trap analysis of cortical tissue from each animal. In this assay, serial dilutions of protein homogenate are passed through a cellulose acetate filter; particles larger than the pore size of the filter become trapped in the membrane and are revealed by immunoblotting [22]. Consistent with our visual analysis of the histological sections, line 107 tTA/APP mice treated with dox for 3 or 6 mo had the same amount of aggregated Aβ as when they started treatment at 6 mo of age (Figure 4A and 4B). In contrast, untreated 9-mo-old mice had almost twice as much aggregated Aβ as either of the treated groups. Filter trap analysis of line 18 tTA/APP mice yielded similar results: the increase in aggregated Aβ observed in untreated animals between 6 and 9 mo of age was completely arrested by transgene suppression (Figure S5C and S5D).
We next used ELISA to measure total Aβ in the brains of each group to determine whether any change in the amount or solubility of peptide occurred while APPswe/ind expression was suppressed. Cortical homogenates were sequentially extracted to separate peptide into PBS-, SDS-, and FA-soluble fractions, then transgene-derived Aβ40 and Aβ42 were measured by human-specific ELISA [23]. In all animals harboring amyloid deposits, we found that the vast majority of Aβ (>99%) was extracted into the SDS and FA fractions (Figure 5A and 5B). Consistent with the filter trap results presented above, there were no significant differences in SDS- or FA-soluble Aβ between the 6 mo untreated cohort and either the 6 mo + 3 mo treated or 6 mo + 6 mo treated cohorts. However, brains of both 6 mo + 3 mo and 6 mo + 6 mo treated cohorts contained roughly twice as much PBS-soluble Aβ40 as untreated 6-mo-old mice (Figure 5C). Levels of Aβ42 showed a similar trend, but did not reach statistical significance. In fact, levels of PBS-soluble Aβ40 and Aβ42 in the 6 mo + 3 mo and 6 mo + 6 mo treated cohorts were most similar to that of the 9 mo untreated cohort, suggesting that age, as opposed to synthetic rate (which would be negligible in the treated animals), may determine the fraction of PBS-soluble Aβ in these animals.
Figure 5 Aβ ELISA Confirms Arrest of Progression without Clearance of Peptide in Mice with Preexisting Aggregates
Aβ levels in untreated 6- and 9-mo-old tTA/APP line 107 mice (shown in Figure 4) were compared to those in 9- and 12-mo-old animals treated with dox from the age of 6 mo. Single transgenic APP samples were included as negative controls. Cortical homogenates were fractionated by sequential multi-step extraction with PBS, 2% SDS, and 70% FA followed by human-specific Aβ ELISA to measure transgene-derived peptide in each fraction. Aβ40 is shown in white, Aβ42 in black.
(A and B) Most Aβ in the brains of plaque-bearing mice is extracted into the FA and SDS fractions. Consistent with amyloid burden (Figures 4 and Figure S3), SDS- and FA-extracted Aβ levels in untreated 9-mo-old mice were significantly higher than in untreated 6-mo-old mice (Tukey post-hoc test applied to significant effect of group ANOVA for SDS and FA fractions F
3,18 = 4.72–12.92, p < 0.02). In contrast, 3 or 6 mo of transgene suppression held Aβ at levels equivalent to those harbored when treatment was started (p > 0.2 compared to 6 mo untreated mice, Tukey post-hoc test). *, p < 0.05; **, p < 0.005; ***, p < 0.001 versus 9-mo-old untreated mice, Tukey post-hoc test. Significance for APP versus 9-mo-old untreated mice is based on Student's t-test.
(C) The PBS fraction represents less than 0.1% of total Aβ (note the change in y-axis from [A] and [B]), but only here do Aβ levels in the dox-treated mice differ from those in younger untreated mice. Although both peptides appear elevated in the treated groups compared to the untreated 6-mo-old mice, only Aβ40 reaches statistical significance (p < 0.05, Tukey post-hoc test applied to significant effect of group ANOVA for Aβ40 F
3,18 = 4.60, p < 0.02). A similar trend was seen for Aβ42, where ANOVA yielded a significant effect of group for PBS-soluble Aβ42 (F
3,18 = 3.75, p < 0.03), however this was due only to differences between the untreated 6- and 9-mo-old groups. •, p < 0.05 versus 6-mo-old untreated mice, Tukey post-hoc test; •••, p < 0.001 versus 6-mo-old untreated mice, Student's t-test.
We also assessed neuritic and glial pathology surrounding the plaques to determine whether there were any changes in nearby tissue following long-term transgene suppression. Both Hirano silver stain and ubiquitin immunostaining showed neuritic pathology in all treatment groups (Figure 6). Similarly, activated astrocytes immunostained for GFAP were found near plaques in all animals (Figure 6). Neuritic and glial pathology were more severe in the older untreated mice. In contrast, transgene suppression prevented the growth of individual deposits apparent in untreated mice, and limited the surrounding pathology to what was already present when treatment began.
Figure 6 Neuritic and Glial Pathology Are Unchanged following Transgene Suppression
Dystrophic neurites and activated astrocytes surround most compact plaques in tet-off APP mice (line 107). Dark-stained, ubiquitin-filled neurites and reactive astrocytes form a halo around cored, fibrillar deposits by 6 mo of age that worsens with time in untreated mice. Both plaque-associated pathologies are arrested, although not reversed, by transgene suppression. Hirano silver stain (top row); GFAP immunohistochemistry (middle row); ubiquitin immunohistochemistry (bottom row).
An obvious question we sought to address was whether the deposition of Aβ diminished cognitive ability in untreated mice, and what might happen to cognition when the process was interrupted. Unfortunately, efforts to characterize cognitive behavior were compromised by severe hyperactivity in untreated double transgenic mice. The tTA/APP animals were often seen running in circles around the perimeter of their cages, and a similar swimming pattern was noted when the mice were tested in the Morris water maze. In the radial water maze, repetitive swim patterns were noted with no evidence of choice-motivated actions. Other studies have dealt with similar problems by excluding animals that do not show adequate attention to the task, retaining only those mice that meet certain performance criteria [30]. In our case, the penetrance of hyperactivity was close to 100%, leaving us with no testable animals. This phenotype has not affected previous lines of APPswe mice we have produced, such as lines E1–2 or C3–3, that express lower levels of transgenic protein. Indeed, in past studies where hyperactivity was not a factor, we established a clear relationship between amyloid load and cognitive ability [31]. However, in the current study, we feel that although the poor performance of the tTA/APP mice in the maze tests could technically be scored as cognitive impairment, the animals' severe hyperactivity made interpretation of the cognitive tasks impossible.
In order to understand the nature and extent of hyperactivity in the tTA/APP mice, we quantified daily activity levels in double transgenic animals along with their single transgenic and nontransgenic siblings using four-beam frames designed to monitor ambulation within an enclosed cage. As shown in Figure 7, the double transgenic mice were up to 10-fold more active during the dark phase of the day–night cycle than any of the control groups. Activity levels appeared to follow a relatively normal diurnal cycle, decreasing substantially during the daylight hours. However, even during the light phase, the tTA/APP mice remained many-fold more active than normal controls. This behavior was partially, but not consistently, reversed by 1 mo of transgene suppression beginning at 4–5 mo of age (data not shown). In contrast, hyperactivity was completely abolished by rearing tTA/APP mice on dox. Animals born and raised on dox showed activity levels similar to the untreated controls (Figure 7C). Intriguingly, all of the dox-reared animals, both transgenic and wild-type, showed altered circadian rhythms with far less distinction between their day- and nighttime activity levels.
Figure 7 Transgene Suppression Attenuates Hyperactivity in tTA/APP Mice
(A) A 48-h measure of ambulation records extreme hyperactivity in untreated double transgenic mice compared to single transgenic and nontransgenic controls (line 107). This phenotype is completely eliminated by rearing the double transgenic mice on dox.
(B) The same data shown in (A) are replotted to magnify data from untreated control and dox-treated groups.
(C and D) Activity levels in the combined control groups of (A) and (B) are here separated by genotype. None of the single transgenic or nontransgenic control groups display the hyperactivity present in untreated tTA/APP animals. Again, note the y-axes have been enlarged for detail compared to (A).
Discussion
We present a new mouse model for AD that was designed to test the consequences of inhibiting Aβ production after the onset of amyloid pathology. New lines of transgenic mice were developed for this study that express high levels of APPswe/ind under the control of a tetracycline-responsive promoter. We demonstrate that treatment with dox suppresses steady-state levels of both APPswe/ind and its C-terminal fragments, indicating that the mutant proteins have a relatively short half-life in vivo. Transgenic expression of APPswe/ind and consequent overproduction of Aβ42 cause early-onset amyloid deposition in untreated mice, in which deposits appear as early as 2 mo of age. Amyloid burden worsens significantly with age, and by 9 mo, the hippocampus and cortex of untreated mice are largely filled with aggregated peptide. We find that suppression of transgenic APP by more than 95% abruptly halts the progression of amyloid pathology. Importantly, this outcome occurs in animals already harboring considerable amyloid pathology, a situation similar to what might be expected in patients to be treated with secretase inhibitors. Somewhat unexpectedly, we observe no appreciable clearance of deposited amyloid even following periods of transgene suppression equal to the time taken for plaques to form. This latter finding indicates that compared to other disease-associated protein aggregates such as mutant huntingtin, which clear in less than 3 mo [32], the disaggregation of extracellular amyloid is relatively slow.
Notably, pharmaceutical γ-secretase inhibitors published to date show less strict regulation of Aβ production following chronic administration than we have attained here. Two independent compounds tested in Tg2576 [33] and TgCRND8 [34] transgenic mice show no more than 85% suppression of Aβ40 levels, and where measured, even less suppression of Aβ42 (approximately 60%) [35,36]. Thus, even after accounting for the much higher APP expression levels in our mice than in the Tg2576 and TgCRND8 lines, we have achieved better absolute suppression of Aβ production with the tet-off system than is currently possible with published γ-secretase inhibitors. Since even the most advanced future pharmaceutical agents are unlikely to attain more complete control of Aβ production than achieved here, this system provides a salient test of therapeutic intervention with Aβ-lowering compounds.
Although the progression of amyloid deposition was sharply arrested by this approach, we found that a substantial amyloid burden remained even after long periods of transgene suppression. We examined a small number of animals after 12 mo of dox treatment (beginning at 6 mo of age), and found that amyloid deposits were still relatively abundant. Longer-term treatments are now in progress. At the latest treatment interval analyzed by ELISA, animals administered dox for 6 mo showed elevated levels of PBS-soluble Aβ (see Figure 5) that could be interpreted as an indication that the plaques are slowly releasing peptide (or oligomeric Aβ) into the soluble pool and might eventually dissolve. Whether inhibiting Aβ production longer than 6 (or 12) mo may ultimately result in clearance of amyloid is under investigation; unfortunately, the life span of the model eventually limits this experiment. Therapeutics used in humans will have considerably more time to act than is possible within the life span of rodent models. Long-term treatments would certainly be possible and could be a key to effective therapy. Overall, however, we interpret our findings as evidence that AD therapies that significantly lower the production of Aβ (by either inhibiting secretase activity or inhibiting APP expression) may not quickly reverse preexisting pathology, but should effectively halt further deposition of amyloid.
In interpreting our study, it should be remembered that the earliest plaques to appear in these mice, like other APP transgenics harboring the Swedish mutation [27], are predominantly fibrillar deposits, which may be less tractable than the diffuse aggregates thought to come first in the course of the human disease. However, our data suggest that once diffuse deposits are formed in these mice, they are no more easily cleared in our system than cored plaques (see Figure S4). An additional consideration we recognize is that a small amount of transgene expression continues in the presence of dox and that endogenous mouse Aβ continues to be produced. It is possible that the combined low levels of endogenous mouse Aβ and nonsuppressed human peptide are sufficient to maintain existing deposits. However, these low levels of peptide are not sufficient to induce new amyloid formation, as CaMKIIα-tTA × tetAPPswe/ind mice raised on dox for up to a year do not develop amyloid lesions (data not shown). It is also clear that in this genetic system, we have raised the production of Aβ to levels not found in humans to accelerate pathology into an experimentally feasible time frame. This system allowed us to create an approximately 20-fold differential between APP/Aβ synthetic rates before and after treatment, yet the in vivo equilibrium between aggregated and disaggregated states of Aβ still favored the maintenance of amyloid deposits. In our opinion, it seems unlikely that amyloid deposits in human brain would be inherently any less stable than those formed in mouse brain. However, the human brain may harbor clearance mechanisms not shared by mice that would allow more efficient removal of preexisting amyloid.
One potential mechanism by which amyloid may be more efficiently removed in the human disease than in the mouse models is through microglial phagocytosis. Resident microglia in transgenic mouse models localize to tissue surrounding plaques but show little evidence of amyloid engulfment [37–40]. In contrast, microglia surrounding amyloid plaques in human brain show a much higher state of activation with greater expression of complement receptor [40]. Thus, the role of microglia in amyloid metabolism is minor in transgenic models compared to the human condition. Somewhat paradoxically, several studies further demonstrate that treatment with anti-inflammatory drugs to reduce microglial activation actually lowers amyloid load in APP transgenic mice, suggesting a role for mouse microglia in the formation and maintenance of amyloid aggregates [41–43]. However, this outcome may be alternatively explained by direct effects of many anti-inflammatory drugs on γ-secretase cleavage [44–47]. Nonetheless, the role of microglia in both the human condition and the mouse models is poorly understood, and differences in microglial reactivity between the two could lead to significantly faster amyloid clearance in the brains of patients with AD than we observe in the tet-off APP mice.
Given the relatively minor role played by microglia in other mouse models of amyloidosis, we think it unlikely that these cells have influenced the rate of amyloid clearance in the tet-off APP mice. Even so, we considered the possibility that chronic dox treatment may have altered the activation state of microglia in our treated mice. Dox is structurally similar to minocycline, a reported anti-inflammatory drug and inhibitor of microglial activation [48]. However, if dox does have anti-inflammatory activity, then, based on previous studies with other anti-inflammatories, we would have expected to find less amyloid in the dox-treated animals. Clearly, that was not the case. While it is possible that dox acts in some other way to slow amyloid clearance, data from multiple studies demonstrate that microglial responses are normally weak in the mouse AD models [37–40], and thus it is doubtful that dox-mediated microglial inhibition affected the outcome of our study.
The persistence and stability of amyloid deposits in our system is unexpected given the speed with which Aβ aggregates are cleared in other mouse models of therapeutic intervention. Anti-Aβ antibodies injected directly into the brain have been shown to eliminate amyloid deposits in as little as 1 wk after treatment [49–51]. Peripheral antibody injection decreases amyloid load more broadly, and although it does not appear to act as quickly as local injection, can significantly reduce amyloid load within 2 mo of initial treatment [52,53]. More recently, an alternative approach has shown that lentiviral transfer of neprilysin can also reduce the number of aggregates in the area of the injection site [54]. Careful study of the mechanism behind several of the antibody-mediated therapies has suggested that activated microglia play an important role in the removal of fibrillar plaques after immunization [50,52,55]. However, it has been noted that deletion of the Fc receptor (the primary receptor for microglial opsinization of antibody–antigen complexes) in APP transgenic mouse models has no impact on the effectiveness of antibody-mediated therapy [56,57]. It is, nevertheless, possible that lack of microglia activation is the major difference between the slow clearance described here, where no perturbation of the immune system is expected, and the rapid clearance described in studies involving antibody or viral injection. In isolation, mild activation of microglia by injection damage or opsinization may not be adequate to induce substantial phagocytosis, but when combined with an Aβ-lowering agent, such as neprilysin or Aβ-targeted antibodies, the two may work in concert to clear peptide deposits. Consistent with this hypothesis, strong activation of microglia through transgenic expression of TGFβ [58] or central injection of lipopolysaccharide [59,60] can by itself substantially reduce plaque burden in APP transgenic mice. But in the case of acute antibody- and/or injury-mediated activation, once the inflammation has passed, and the antibody and bound peptide have been cleared and degraded, the remaining Aβ quickly reaggregates and amyloid pathology is reestablished [49]. This finding reinforces the notion that without continued stimulation, microglia in mouse models do not maintain the same level of sustained activation that may occur in humans.
SantaCruz et al. recently published a study of mice that express P301L human tau via a similar vector system [61]. As in our tet-off APP mice, SantaCruz et al. found that tau neurofibrillary tangles, like amyloid plaques, are not cleared efficiently following transgene suppression. The lack of clearance in both models of AD pathology comes as a stark contrast to the rapid removal of protein aggregates found in similar tet-off mouse models of Huntington [32] and prion disease [62]. In these cases, disrupting the input of new monomer to the system via dox-mediated transgene suppression led to relatively rapid clearance of protein aggregates. By contrast, our study and that of SantaCruz et al. suggest that protein aggregates in AD may be more tenacious than in other neurodegenerative disorders. Perhaps once aggregated, Aβ and tau are either inherently more stable than other protein aggregates or more resistant to intra- and extracellular clearance mechanisms.
One question we were not able to address in this study is whether abrogating synthesis of new Aβ halts the progression of cognitive decline. Studies from the tet-off tau mice suggest that protein clearance may in fact not be required for cognitive improvement following transgene suppression [61]. At present, because of unexpected noncognitive behavioral abnormalities, it is not clear whether the tet-off APP mice can be used to address the same question in the context of amyloid pathology. Both lines of tTA/APP mice we studied here display extreme hyperactivity visible as cage circling and quantified by activity monitoring (see Figure 7). Many of the double transgenic mice showed similar circular patterns of swimming near the edge of the tank when tested in the Morris water maze. Expression of this phenotype makes standard tests of learning and memory uninterpretable. Hyperactivity nonspecifically inhibits choice-driven changes in movement, the key element behind all cognitive behavioral paradigms. We are currently working to determine whether hyperactivity correlates with expression of the APPswe/ind holoprotein or its proteolytic derivatives. Preliminary studies suggest that hyperactivity does not appear quickly when dox-reared mice are shifted to nonmedicated diets (J. L. J., unpublished data). These data may indicate that the neuroactive culprit is not immediately present after transgenic APP synthesis is initiated, but requires additional time to develop. Alternatively, hyperactivity may be caused by neuronal alterations due to transgene expression during early postnatal development. Further experiments are needed to distinguish between these possibilities.
In summary, we demonstrate that abrogating Aβ production halts the progression of pathologic changes in a transgenic mouse model of Alzheimer-type amyloidosis. However, despite dramatic reductions in Aβ synthesis, neuritic plaques are stable structures in vivo that do not quickly disaggregate. It is possible that a combination of therapies to limit Aβ production, increase Aβ degradation, and enhance phagocytosis of deposited amyloid may be required to reverse damage associated with AD. However, if started early enough in the course of disease, secretase inhibitors alone could provide substantial benefit in slowing pathogenic processes linked to amyloid deposition. Even at later stages in the disease, the presence of substantial microglial activation in human AD [40] suggests that simply slowing the formation of new amyloid deposits may allow ongoing phagocytosis to diminish preexisting lesions. However, the development of safe and effective secretase inhibitors will ultimately be required to determine whether the human brain has the capacity to repair amyloid-associated damage of AD once the progression of pathology is arrested.
Supporting Information
Figure S1 Transgenic APP Expression and Suppression by Dox in the Four New Tet-Off APP Lines
Western blotting with human-specific antibody 6E10 reveals transgene-derived full-length APP in cortical homogenates from untreated animals (left lanes of each panel). The new lines produce exceptionally high levels of transgene expression; an equal amount of brain homogenate from a standard transgenic APP line is shown for comparison (extreme left lane, Standard Tg line C3–3; [15,63]). After 1 mo of dox treatment, transgenic protein in all four new tet-off APP lines is reduced to residual levels (+ dox; right lanes of each panel).
(474 KB TIF).
Click here for additional data file.
Figure S2 Transgenic APP mRNA Is Brain-Specific
A slot blot of mRNA harvested from various tissues in three of the four new tet-off APP lines and a nontransgenic control was used to examine transgenic mRNA expression. Hybridization is seen only in the brain; no signal above background is seen in any other tissue.
(781 KB PSD).
Click here for additional data file.
Figure S3 Amyloid Pathology in the Cortex Reiterates That in the Hippocampus
Amyloid histology was performed on sections from line 107 double transgenic mice by Hirano silver stain (top row), thioflavin-S (middle row), and Aβ immunohistochemistry (bottom row) to examine the persistence of pathology following transgene suppression. As in the hippocampus (see Figure 4 and text), the progression of amyloid pathology in the cortex worsens substantially between 6 and 9 mo of age in untreated mice. This progression is completely prevented by suppression of the transgene with dox. For comparison, normal neurohistology is shown in an age-matched single transgenic (tTA only) animal. No amyloid pathology has been detected in either APP or tTA single transgenic animals up to 15 mo of age.
(4.8 MB PSD).
Click here for additional data file.
Figure S4 Diffuse Deposits Do Not Disperse During Aβ Suppression
Campbell–Switzer silver stain was used to differentiate cored (brown) from diffuse (black) deposits in line 107 tTA/APP mice. This stain demonstrates that both types of deposit persist throughout long periods of transgene suppression. The lower panels, showing low-power images (10×) of frontal cortex from each condition, reveal little change in the extent of diffuse amyloid following up to 6 mo of Aβ suppression. High-power images (40×) in the upper panels show that the diffuse halo surrounding individual cored deposits remains relatively unchanged in treated mice. Untreated tTA single transgenic animals are shown as a negative control. Protocol for the Campbell-Switzer Alzheimer's Method was kindly shared by Robert Switzer, III (NeuroScience Associates, Knoxville, Tennessee, United States), and can be downloaded at http://www.nsalabs.com/Documents/publications/campbell-switzer_protocol.htm [64,65].
(923 KB JPG).
Click here for additional data file.
Figure S5 Chronic Transgene Suppression and Arrest of Aβ Aggregate Formation in an Independent Line of Tet-Off APP Mice (CaMKIIα-tTA × tet-APPswe/ind Line 18)
(A) The experiment presented in the text for line 107 tet-off APP was repeated with a second tet-off APP line (line 18) to control for integration site artifacts. Cortical homogenates from untreated control and dox-treated double transgenic mice were immunoblotted for full-length APP with the human-specific antibody 6E10 to confirm transgene suppression at the time of harvest. Immunostaining for endogenous superoxide dismutase (SOD1) was included as a loading control.
(B) Quantitation of signal intensity from the Western blot in (A) shows transgenic APP levels in line 18 are suppressed by more than 98% following 3 mo of dox treatment (significant effect of group ANOVA F
2,8 = 1559.7, p < 0.001). This level of suppression was equal to or better than that attained in line 107 (see Figure 3B).
(C) Serial dilution filter trap assay was used to quantify aggregated Aβ in cortical homogenates.
(D) Quantitation of signal intensity in the linear range of the dilution series shown in (C). Consistent with the amyloid histology shown in Figure S5, aggregate formation was significantly increased between 6 and 9 mo of age in untreated mice (significant effect of group ANOVA F
2,18 = 12.14, p < 0.001). Aggregate formation was completely arrested by transgene suppression, and is identical in 9-mo-old mice treated with dox for 3 mo as in untreated animals harvested when treatment began (p > 0.5, Tukey post-hoc test). *, p < 0.05; **, p < 0.005; ***, p < 0.001 versus 9-mo-old untreated mice, Tukey post-hoc test.
(962 KB TIF).
Click here for additional data file.
Figure S6 Arrest of Amyloid Progression by Chronic Transgene Suppression in Line 18 Tet-Off APP Mice
Amyloid histology in cortical (first and third rows) and hippocampal (second and fourth rows) sections from untreated tTA/APP mice shows a dramatic progression of pathology between 6 and 9 mo of age. Suppression of transgenic APP expression arrests this progression, although without any sign of plaque clearance (6 mo + 3 mo dox). Hirano silver stain (top panels); thioflavin-S (bottom panels).
(5.8 MB PSD).
Click here for additional data file.
Patient Summary
Background
Patients with Alzheimer disease (AD) have elevated levels of a small protein called amyloid-β peptide that sticks together to form what are known as amyloid plaques in their brains. This peptide is normally made at low levels in healthy individuals, and is made when a larger protein called amyloid precursor protein (APP) is cut down in size. New treatments are now being developed that will decrease the amount of Aβ produced from APP. However, it is not clear whether lowering the production of Aβ will allow the brain to heal itself by clearing the amyloid plaques. The answer to this question may be important for deciding when Aβ-lowering drugs should be started, and may also determine how effective they are in reversing the mental symptoms of AD.
What Did the Researchers Do and Find?
Because new drugs designed to lower Aβ levels are still in development, they are not available for testing in animal models of the disease. Instead, basic questions about the effectiveness of this type of treatment must be answered using systems that mimic how the drugs work. To do this, the authors created mice that produce too much APP and that develop the same amyloid lesions as do human patients with AD. Unlike normal mice, these mice also carried a “switch” gene that allowed the researchers to turn off APP by feeding the mice special food. Turning off APP in these mice had the same effect as treating them with Aβ-lowering drugs, and so the researchers were able to ask what happened to the amyloid plaques after Aβ production was shut down. They showed that lowering Aβ production prevents the amyloid lesions from getting worse as the disease progresses. This means that treatment with Aβ-lowering drugs may be able to stop the disease from filling the brain with plaques. However, the researchers also found that the amyloid lesions that had formed before treatment was started remained intact throughout the experiment.
What Do These Findings Mean?
These results indicate that treatments designed to lower the production of Aβ may be an important part of future AD treatment, as this approach seems to prevents additional amyloid plaques from forming in the mouse brain. However, by itself, this strategy may not be able to rid the brain of plaques that have already formed in the brain before treatment is started. The findings suggest that early treatment may be important for this approach to succeed.
Where Can I Get More Information Online?
MedlinePlus has several Web pages of information on Alzheimer disease:
http://www.nlm.nih.gov/medlineplus/alzheimersdisease.html
The ADEAR Center of the US Government's National Institute on Aging also has information on Alzheimer disease:
http://www.alzheimers.org/
The Alzheimer's Association Web site contains information on both caregiving and research:
http://www.alz.org
We thank Patrick Tremblay for helpful advice on the tet system at a critical time in the project, and Mark Mayford for sharing the CaMKIIα-tTA mice through Jackson Laboratory. We also thank Fraser Moss for saving several immunoblots with last-minute shipments, Andy Groves for many thoughtful discussions, Neil Segil for generously sharing his laboratory and equipment, Beth Olson, Natasha Bouey, and Yolanda Jackson for outstanding animal care, Debbie Swing for expert microinjection, and Dave Fromholt for genotyping and dissection. We gratefully acknowledge Takeda Chemical Industries for providing antibodies BAN50, BA27, and BC05, Konrad Beyreuther and Andreas Weidemann for providing 22C11 antibody, and Ed Koo for sharing CT15 antibody. This work was supported by grants from the Johns Hopkins Alzheimer's Disease Research Center (JLJ), the National Alliance for Research on Schizophrenia and Depression (Young Investigator Award [JLJ]), the Rose Hills Foundation (JLJ), the Alzheimer's Association (Zenith Award [DRB]), the National Institute of Aging (K01 AG26144–01 [JLJ], P50 AGO5146–20 [DRB], R01 AG006656–16 [SGY], and P01 AG015453 [SGY]), the National Institute of Neurologic Disease and Stoke (R01 NS 047225 [DRB]), and the National Cancer Institute (NAJ and NGC). The funding agencies generously provided for research supplies, animal care, and salary support; the funders of this work had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Citation: Jankowsky JL, Slunt HH, Gonzales V, Savonenko AV, Wen JC, et al. (2005) Persistent amyloidosis following suppression of Aβ production in a transgenic model of Alzheimer disease. PLoS Med 2(12): e355.
Abbreviations
Aβamyloid-β
ADAlzheimer disease
APLPamyloid precursor–like protein
APPamyloid precursor protein
CaMKIIαcalcium-calmodulin kinase IIα
doxdoxycycline
FAformic acid
GFAPglial fibrillary acidic protein
mo/huAPP695mouse APP with a humanized Aβ domain
PBSphosphate-buffered saline
SDSsodium dodecyl sulfate
swe/indSwedish/Indiana
tTAtetracycline transactivator
==== Refs
References
Hardy JA Higgins GA Alzheimer's disease: The amyloid cascade hypothesis Science 1992 256 184 185 1566067
Selkoe DJ Podlisny MB Deciphering the genetic basis of Alzheimer's disease Annu Rev Genomics Hum Genet 2002 3 67 99 12142353
Rademakers R Cruts M Van Broeckhoven C Genetics of early-onset Alzheimer dementia ScientificWorldJournal 2003 3 497 519 12847300
Orgogozo JM Gilman S Dartigues JF Laurent B Puel M Subacute meningoencephalitis in a subset of patients with AD after Aβ42 immunization Neurology 2003 61 46 54 12847155
Check E Nerve inflammation halts trial for Alzheimer's drug Nature 2002 415 462
Hock C Konietzko U Streffer JR Tracy J Signorell A Antibodies against β-amyloid slow cognitive decline in Alzheimer's disease Neuron 2003 38 547 554 12765607
Nicoll JA Wilkinson D Holmes C Steart P Markham H Neuropathology of human Alzheimer disease after immunization with amyloid-β peptide: A case report Nat Med 2003 9 448 452 12640446
Tsai JY Wolfe MS Xia W The search for γ-secretase and development of inhibitors Curr Med Chem 2002 9 1087 1106 12052174
Wolfe MS Therapeutic strategies for Alzheimer's disease Nat Rev Drug Discov 2002 1 859 866 12415246
Cumming JN Iserloh U Kennedy ME Design and development of BACE-1 inhibitors Curr Opin Drug Discov Devel 2004 7 536 556
Citron M β-secretase inhibition for the treatment of Alzheimer's disease—Promise and challenge Trends Pharmacol Sci 2004 25 92 97 15102495
Gossen M Bujard H Tight control of gene expression in mammalian cells by tetracycline-responsive promoters Proc Natl Acad Sci U S A 1992 89 5547 5551 1319065
Furth PA St Onge L Boger H Gruss P Gossen M Temporal control of gene expression in transgenic mice by a tetracycline-responsive promoter Proc Natl Acad Sci U S A 1994 91 9302 9306 7937760
Kistner A Gossen M Zimmermann F Jerecic J Ullmer C Doxycycline-mediated quantitative and tissue-specific control of gene expression in transgenic mice Proc Natl Acad Sci U S A 1996 93 10933 10938 8855286
Borchelt DR Davis J Fischer M Lee MK Slunt HH A vector for expressing foreign genes in the brains and hearts of transgenic mice Genet Anal 1996 13 159 163 9117892
Mayford M Bach ME Huang YY Wang L Hawkins RD Control of memory formation through regulated expression of a CaMKII transgene Science 1996 274 1678 1683 8939850
Weidemann A Konig G Bunke D Fischer P Salbaum JM Identification, biogenesis, and localization of precursors of Alzheimer's disease A4 amyloid protein Cell 1989 57 115 126 2649245
Borchelt DR Lee MK Slunt HS Guarnieri M Xu ZS Superoxide dismutase 1 with mutations linked to familial amyotrophic lateral sclerosis possesses significant activity Proc Natl Acad Sci U S A 1994 91 8292 8296 8058797
Sisodia SS Koo EH Hoffman PN Perry G Price DL Identification and transport of full-length amyloid precursor proteins in rat peripheral nervous system J Neurosci 1993 13 3136 3142 8331390
Church GM Gilbert W Genomic sequencing Proc Natl Acad Sci U S A 1984 81 1991 1995 6326095
Yamamoto T Hirano A A comparative study of modified Bielschowsky, Bodian and thioflavin S stains on Alzheimer's neurofibrillary tangles Neuropathol Appl Neurobiol 1986 12 3 9 2422580
Xu G Gonzales V Borchelt DR Rapid detection of protein aggregates in the brains of Alzheimer patients and transgenic mouse models of amyloidosis Alzheimer Dis Assoc Disord 2002 16 191 195 12218651
Kawarabayashi T Younkin LH Saido TC Shoji M Ashe KH Age-dependent changes in brain, CSF, and plasma amyloid β protein in the Tg2576 transgenic mouse model of Alzheimer's disease J Neurosci 2001 21 372 381 11160418
Suzuki N Cheung TT Cai XD Odaka A Otvos L An increased percentage of long amyloid β protein secreted by familial amyloid β protein precursor (βAPP717) mutants Science 1994 264 1336 1340 8191290
Gravina SA Ho L Eckman CB Long KE Otvos L Amyloid β protein (Aβ) in Alzheimer's disease brain. Biochemical and immunocytochemical analysis with antibodies specific for forms ending at Aβ40 or Aβ42(43) J Biol Chem 1995 270 7013 7016 7706234
Jankowsky JL Fadale DJ Anderson J Xu GM Gonzales V Mutant presenilins specifically elevate the levels of the 42 residue β-amyloid peptide in vivo: Evidence for augmentation of a 42-specific γ-secretase Hum Mol Genet 2004 13 159 170 14645205
Borchelt DR Ratovitski T van Lare J Lee MK Gonzales V Accelerated amyloid deposition in the brains of transgenic mice coexpressing mutant presenilin 1 and amyloid precursor proteins Neuron 1997 19 939 945 9354339
Herms J Zurmohle U Brysch W Schlingensiepen KH Ca2+/calmodulin protein kinase and protein kinase C expression during development of rat hippocampus Dev Neurosci 1993 15 410 416 7835246
Phinney AL Deller T Stalder M Calhoun ME Frotscher M Cerebral amyloid induces aberrant axonal sprouting and ectopic terminal formation in amyloid precursor protein transgenic mice J Neurosci 1999 19 8552 8559 10493755
Kotilinek LA Bacskai B Westerman M Kawarabayashi T Younkin L Reversible memory loss in a mouse transgenic model of Alzheimer's disease J Neurosci 2002 22 6331 6335 12151510
Savonenko A Xu GM Melnikova T Morton JL Gonzales V Episodic-like memory deficits in the APPswe/PS1dE9 mouse model of Alzheimer's disease: Relationships to β-amyloid deposition and neurotransmitter abnormalities Neurobiol Dis 2005 18 602 617 15755686
Yamamoto A Lucas JJ Hen R Reversal of neuropathology and motor dysfunction in a conditional model of Huntington's disease Cell 2000 101 57 66 10778856
Hsiao K Chapman P Nilsen S Eckman C Harigaya Y Correlative memory deficits, Aβ elevation, and amyloid plaques in transgenic mice Science 1996 274 99 102 8810256
Chishti MA Yang DS Janus C Phinney AL Horne P Early-onset amyloid deposition and cognitive deficits in transgenic mice expressing a double mutant form of amyloid precursor protein 695 J Biol Chem 2001 276 21562 21570 11279122
Wong GT Manfra D Poulet FM Zhang Q Josien H Chronic treatment with the gamma-secretase inhibitor LY-411,575 inhibits β-amyloid peptide production and alters lymphopoiesis and intestinal cell differentiation J Biol Chem 2004 279 12876 12882 14709552
Barten DM Guss VL Corsa JA Loo A Hansel SB Dynamics of β-amyloid reductions in brain, cerebrospinal fluid, and plasma of β-amyloid precursor protein transgenic mice treated with a γ-secretase inhibitor J Pharmacol Exp Ther 2005 312 635 643 15452193
Sasaki A Shoji M Harigaya Y Kawarabayashi T Ikeda M Amyloid cored plaques in Tg2576 transgenic mice are characterized by giant plaques, slightly activated microglia, and the lack of paired helical filament-typed, dystrophic neurites Virchows Arch 2002 441 358 367 12404061
Wegiel J Imaki H Wang KC Wegiel J Rubenstein R Cells of monocyte/microglial lineage are involved in both microvessel amyloidosis and fibrillar plaque formation in APPsw tg mice Brain Res 2004 1022 19 29 15353209
Wegiel J Wang KC Imaki H Rubenstein R Wronska A The role of microglial cells and astrocytes in fibrillar plaque evolution in transgenic APPsw mice Neurobiol Aging 2001 22 49 61 11164276
Schwab C Hosokawa M McGeer PL Transgenic mice overexpressing amyloid β protein are an incomplete model of Alzheimer disease Exp Neurol 2004 188 52 64 15191802
Jantzen PT Connor KE DiCarlo G Wenk GL Wallace JL Microglial activation and β-amyloid deposit reduction caused by a nitric oxide-releasing nonsteroidal anti-inflammatory drug in amyloid precursor protein plus presenilin-1 transgenic mice J Neurosci 2002 22 2246 2254 11896164
Lim GP Yang F Chu T Chen P Beech W Ibuprofen suppresses plaque pathology and inflammation in a mouse model for Alzheimer's disease J Neurosci 2000 20 5709 5714 10908610
Yan Q Zhang J Liu H Babu-Khan S Vassar R Anti-inflammatory drug therapy alters β-amyloid processing and deposition in an animal model of Alzheimer's disease J Neurosci 2003 23 7504 7509 12930788
Eriksen JL Sagi SA Smith TE Weggen S Das P NSAIDs and enantiomers of flurbiprofen target gamma-secretase and lower Aβ42 in vivo J Clin Invest 2003 112 440 449 12897211
Weggen S Eriksen JL Das P Sagi SA Wang R A subset of NSAIDs lower amyloidogenic Aβ42 independently of cyclooxygenase activity Nature 2001 414 212 216 11700559
Weggen S Eriksen JL Sagi SA Pietrzik CU Ozols V Evidence that nonsteroidal anti-inflammatory drugs decrease amyloid β 42 production by direct modulation of γ-secretase activity J Biol Chem 2003 278 31831 31837 12805356
Lleo A Berezovska O Herl L Raju S Deng A Nonsteroidal anti-inflammatory drugs lower Aβ42 and change presenilin 1 conformation Nat Med 2004 10 1065 1066 15448688
Yrjanheikki J Keinanen R Pellikka M Hokfelt T Koistinaho J Tetracyclines inhibit microglial activation and are neuroprotective in global brain ischemia Proc Natl Acad Sci U S A 1998 95 15769 15774 9861045
Oddo S Billings L Kesslak JP Cribbs DH LaFerla FM Aβ immunotherapy leads to clearance of early, but not late, hyperphosphorylated tau aggregates via the proteasome Neuron 2004 43 321 332 15294141
Wilcock DM DiCarlo G Henderson D Jackson J Clarke K Intracranially administered anti-Aβ antibodies reduce beta-amyloid deposition by mechanisms both independent of and associated with microglial activation J Neurosci 2003 23 3745 3751 12736345
Bacskai BJ Kajdasz ST Christie RH Carter C Games D Imaging of amyloid-β deposits in brains of living mice permits direct observation of clearance of plaques with immunotherapy Nat Med 2001 7 369 372 11231639
Bard F Cannon C Barbour R Burke RL Games D Peripherally administered antibodies against amyloid β-peptide enter the central nervous system and reduce pathology in a mouse model of Alzheimer disease Nat Med 2000 6 916 919 10932230
Wilcock DM Rojiani A Rosenthal A Levkowitz G Subbarao S Passive amyloid immunotherapy clears amyloid and transiently activates microglia in a transgenic mouse model of amyloid deposition J Neurosci 2004 24 6144 6151 15240806
Marr RA Rockenstein E Mukherjee A Kindy MS Hersh LB Neprilysin gene transfer reduces human amyloid pathology in transgenic mice J Neurosci 2003 23 1992 1996 12657655
Wilcock DM Munireddy SK Rosenthal A Ugen KE Gordon MN Microglial activation facilitates Aβ plaque removal following intracranial anti-Aβ antibody administration Neurobiol Dis 2004 15 11 20 14751766
Das P Howard V Loosbrock N Dickson D Murphy MP Amyloid-β immunization effectively reduces amyloid deposition in FcRγ-/- knock-out mice J Neurosci 2003 23 8532 8538 13679422
Bacskai BJ Kajdasz ST McLellan ME Games D Seubert P Non-Fc-mediated mechanisms are involved in clearance of amyloid-β in vivo by immunotherapy J Neurosci 2002 22 7873 7878 12223540
Wyss-Coray T Lin C Yan F Yu GQ Rohde M TGFβ1 promotes microglial amyloid-β clearance and reduces plaque burden in transgenic mice Nat Med 2001 7 612 618 11329064
DiCarlo G Wilcock D Henderson D Gordon M Morgan D Intrahippocampal LPS injections reduce Aβ load in APP+PS1 transgenic mice Neurobiol Aging 2001 22 1007 1012 11755009
Herber DL Roth LM Wilson D Wilson N Mason JE Time-dependent reduction in Aβ levels after intracranial LPS administration in APP transgenic mice Exp Neurol 2004 190 245 253 15473997
SantaCruz K Lewis J Spires T Paulson J Kotilinek L Tau suppression in a neurodegenerative mouse model improves memory function Science 2005 309 476 481 16020737
Safar JG DeArmond SJ Kociuba K Deering C Didorenko S Prion clearance in bigenic mice J Gen Virol 2005 In press
Borchelt DR Thinakaran G Eckman CB Lee MK Davenport F Familial Alzheimer's disease-linked presenilin 1 variants elevate Aβ1–42/1–40 ratio in vitro and in vivo Neuron 1996 17 1005 1013 8938131
Campbell SK Switzer RC Martin TL Alzheimer's plaques and tangles: A controlled and enhanced silver staining method Soc Neurosci Abstr 1987 13 189.9
Switzer RC Campbell SK Murdock TM A histologic method for staining Alzheimer pathology 1993 March 9 United States patent 5,192,688
|
16279840
|
PMC1283364
|
CC BY
|
2021-01-05 10:39:27
|
no
|
PLoS Med. 2005 Dec 15; 2(12):e355
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020355
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1627984110.1371/journal.pmed.0020381Research ArticleCell BiologyInfectious DiseasesMicrobiologyRespiratory MedicineInfectious DiseasesTuberculosisImmunology and allergyRespiratory MedicineDC-SIGN Induction in Alveolar Macrophages Defines Privileged Target Host Cells for Mycobacteria in Patients with Tuberculosis DC-SIGN in Human TuberculosisTailleux Ludovic
1
*Pham-Thi Nhan
2
Bergeron-Lafaurie Anne
3
Herrmann Jean-Louis
4
Charles Patricia
1
Schwartz Olivier
5
Scheinmann Pierre
2
Lagrange Philippe H
4
de Blic Jacques
2
Tazi Abdellatif
3
Gicquel Brigitte
1
Neyrolles Olivier
1
6
*1Institut Pasteur, Unité de Génétique Mycobactérienne, Paris, France,2Hôpital Necker-Enfants-Malades, AP-HP, Service de Pneumologie et d'Allergologie Pédiatrique, Paris, France,3Hôpital Saint-Louis, AP-HP, Service de Pneumologie, Paris, France,4Hôpital Saint-Louis, AP-HP, Service de Microbiologie, Paris, France,5Institut Pasteur, Groupe Virus et Immunité, Paris, France,6Centre National de la Recherche Scientifique, URA 2172, Paris, FranceKlenerman Paul Academic EditorOxford UniversityUnited Kingdom*To whom correspondence should be addressed. E-mail: [email protected] (LT), E-mail: [email protected] (ON)
Competing Interests: The authors have declared that no competing interests exist.
Author Contributions: LT and ON designed the study with the help of NPT, ABL, JLH, PS, PHL, JdB, AT, and BG. NPT, ABL, and JLH performed clinical sampling. LT, PC, OS, and ON performed the experiments. LT and ON wrote the paper with the contribution of NPT, ABL, JLH, OS, PS, PHL, JdB, AT, and BG.
12 2005 15 11 2005 2 12 e3814 4 2005 15 9 2005 Copyright: © 2005 Tailleux et al.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
DC-SIGN and Lung Pathogenesis in Patients with Tuberculosis
Background
Interplays between Mycobacterium tuberculosis, the etiological agent of tuberculosis (TB) in human and host professional phagocytes, namely macrophages (Mφs) and dendritic cells (DCs), are central to immune protection against TB and to TB pathogenesis. We and others have recently shown that the C-type lectin dendritic cell–specific intercellular adhesion molecule-3 grabbing nonintegrin (DC-SIGN; CD209) mediates important interactions between mycobacteria and human monocyte-derived DCs (MoDCs) in vitro.
Methods and Findings
In order to explore the possible role of DC-SIGN in M. tuberculosis infection in vivo, we have analysed DC-SIGN expression in broncho-alveolar lavage (BAL) cells from patients with TB (n = 40) or with other non-mycobacterial lung pathologies, namely asthma (n = 14) and sarcoidosis (n = 11), as well as from control individuals (n = 9). We show that in patients with TB, up to 70% of alveolar Mφs express DC-SIGN. By contrast, the lectin is barely detected in alveolar Mφs from all other individuals. Flow cytometry, RT-PCR, and enzyme-linked immunosorbent assay analyses of BAL-derived fluids and cells indicated that M. tuberculosis infection induces DC-SIGN expression in alveolar Mφs by a mechanism that is independent of Toll-like receptor-4, interleukin (IL)-4, and IL-13. This mechanism most likely relies on the secretion of soluble host and/or mycobacterial factors that have yet to be identified, as both infected and uninfected bystander Mφs were found to express DC-SIGN in the presence of M. tuberculosis. Immunohistochemical examination of lung biopsy samples from patients with TB showed that the bacilli concentrate in pulmonary regions enriched in DC-SIGN-expressing alveolar Mφs in vivo. Ex vivo binding and inhibition of binding experiments further revealed that DC-SIGN–expressing alveolar Mφs constitute preferential target cells for M. tuberculosis, as compared to their DC-SIGN− counterparts. In contrast with what has been reported previously in MoDCs in vitro, ex vivo DC-SIGN ligation by mycobacterial products failed to induce IL-10 secretion by alveolar Mφs, and IL-10 was not detected in BALs from patients with TB.
Conclusion
Altogether, our results provide further evidence for an important role of DC-SIGN during TB in humans. DC-SIGN induction in alveolar Mφs may have important consequences on lung colonization by the tubercle bacillus, and on pulmonary inflammatory and immune responses in the infected host.
Alveolar macrophages from patients with TB express higher levels of DC-SIGN compared with controls. Ex-vivo infection induced DC-SIGN in infected and bystander cells, which made the latter more susceptible.
==== Body
Introduction
Interactions between Mycobacterium tuberculosis, the airborne agent of tuberculosis (TB) in human and host phagocytes, namely macrophages (Mφs) and dendritic cells (DCs), are central to anti-mycobacterial immunity and to TB pathogenesis [1,2]. In particular alveolar Mφs constitute a primary niche for the tubercle bacillus [3]. At the molecular level, interactions between the bacillus and host phagocytes rely on a variety of cellular receptors, among which the C-type lectin DC-specific intercellular adhesion molecule-3 grabbing nonintegrin (DC-SIGN; CD209) has recently received particular attention. Initially described as an HIV gp120 receptor [4,5], DC-SIGN has afterwards been shown to allow monocyte-derived DCs (MoDCs) to recognize a variety of microbes, including viruses, parasites, and bacteria [6]. A set of recent reports has established that DC-SIGN plays a key role in mycobacteria interactions with DCs, at least in vitro [7,8]. Indeed, DC-SIGN allows human MoDCs to recognize members of the M. tuberculosis complex through lipoarabinomannan (LAM), an abundant lipoglycan of the mycobacterial envelope [7,9], and through other molecules, such lipomannan, arabinomannan, and the 19 kDa antigen [8]; and DC-SIGN ligation by LAM partially inhibits the maturation of bacterial lipopolysaccharide (LPS)-stimulated MoDCs and potentiates the secretion of the anti-inflammatory cytokine interleukin (IL)-10 by these cells [8]. DC-SIGN interactions with mycobacteria or mycobacterial products may thus be of benefit for either the pathogen, by down-modulating DC functions, or for the host, by limiting tissue inflammation and immunopathology [11–13]. It is noteworthy that DC-SIGN is expressed by human pulmonary DCs [7,14], and mycobacteria-derived antigens have been detected in DC-SIGN+ DCs in lymph nodes from patients with TB, suggesting that interactions between the lectin and the bacillus may occur during the natural course of infection [7].
In order to further our understanding of the role of DC-SIGN in vivo, we have analysed the expression of the lectin in broncho-alveolar lavage (BAL) cells from patients with TB, as compared to cells from other patients.
Methods
Patients and Samples
Individuals studied (n = 74; mean age 17.6 ± 19.3 y; median age 10.5 y; 54% males) included patients with TB (n = 40), sarcoidosis (n = 11), or asthma (n = 14). Other individuals had alveolar proteinosis (n = 1), tracheal obstruction (n = 1), tracheal surgery (n = 1), inactive TB (n = 5), and lymphoma (n = 1). These individuals (n = 9) are considered as controls in the rest of the study. All individuals were HIV negative and non-smokers. None were receiving anti-mycobacterial therapy and/or corticosteroids at the time of biopsy, except for the treatment of asthma. TB was diagnosed by smear observation and/or bacterial culture and/or clinical symptoms. Asthma was defined as a history of recurrent wheezing episodes and daily use of asthma medication. The diagnosis of sarcoidosis was based on previously described criteria [16]. Inactive TB refers to patients who had a previous history of TB (which was successfully treated) without reactivation. BAL fluids were performed as previously described [17] for diagnostic purpose. Blood mononuclear cells from healthy volunteers (Etablissement Français du Sang, Paris, France) and from TB patients with smear observation were isolated by Ficoll-Paque (Pharmacia, Uppsala, Sweden) centrifugation. Only fluids in surplus were used in this study, according to institutional guidelines.
Alveolar Mφs Isolation, Treatment, and Infection
BAL cells were washed and cultured in RPMI-1640 (Invitrogen, Carlsbad, California, United States) supplemented with 10% heat-inactivated foetal calf serum (FCS; Dutscher, Brumath, France) and 2 mM L-glutamine in Petri dishes for 1 h at 37 °C before removing nonadherent cells. When required, cells were treated with IL-4 (10 ng/ml, R&D Systems), tumor necrosis factor (TNF)-α (50 ng/ml; R&D Systems, Minneapolis, Minnesota, United States), Escherichia coli–derived LPS (100 ng/ml; Sigma, St. Louis, Missouri, United States), or a green fluorescent protein (GFP)-expressing strain of M. tuberculosis H37Rv (pEGFP plasmid was a kind gift from G. R. Stewart, University of Surrey, United Kingdom) at a multiplicity of infection (MOI) of one bacterium per cell.
THP1 Cell
THP1 (
ATCC TIB-202) cells were transduced with a lentiviral DC-SIGN–encoding vector [18]. DC-SIGN–expressing cells were positively selected by flow cytometry–based cell sorting and cultivated, as THP1 cells, in RPMI-1640 supplemented with 10% heat-inactivated FCS and 2 mM L-glutamine.
Flow Cytometry
Cells were treated and analysed as previously described [19]. The following monoclonal antibodies (mAbs) were used: anti-CD11b-PE, anti-CD11c-PE, anti-CD16-PE, anti-CD32-PE, anti-CD40-PE, anti-CD64-FITC, and anti-CD206-PE (all from Beckman Coulter, Allendale, New Jersey, United States); anti-CD1a-PE, anti-CD11b-APC, anti-CD14-PE, anti-CD83-PE, anti-CD86-PE, anti-CD123-PE-Cy5, and anti–human leukocyte antigen (HLA)-DR-PE (all from BD Biosciences, San Diego, California, United States); anti-CD209 (DC-SIGN)-FITC and anti-PE (clone 120507; R&D Systems); anti-TLR2-PE, anti-TLR4-PE, and anti-TLR9-PE (all from eBioscience, San Diego, California, United States); and anti-BDCA-1-APC, anti BDCA-2-PE, and anti-BDCA-3-PE (all from Miltenyi Biotech, Bergisch Gladbach, Germany). Isotype controls were all purchased from BD Biosciences. Fluorescence was analyzed using FACScalibur and CellQuest Pro software (BD Biosciences).
Binding Experiments
Total BAL cells were pre-incubated for 30 min at 37 °C, according to a previously published procedure [20], in RPMI-1640 containing 10% FCS and eventually containing isotype controls, either a mix of PE-conjugated and unconjugated anti-DC-SIGN antibodies (clones 120507 and 1B10, a kind gift from A. Amara, Institut Pasteur, Paris, France) or a mix of APC-conjugated and unconjugated anti-CD11b antibodies (clones M1/70 [BD Biosciences] and 2LPM19c1 [Dako, Glostrup, Denmark]). All antibodies were used at 10 μg/ml. Pre-incubation at 37 °C in the presence of antibodies resulted in partial internalization of the corresponding receptors but did not decrease cell-associated fluorescence, because a mix of unlabelled and fluorescently labelled antibodies was used during this step (data not shown). Cells were then infected with GFP-expressing M. tuberculosis H37Rv, at a multiplicity of infection of five bacteria per cell, for 4 h at 4 °C to allow binding without phagocytosis. Cells were then washed in RPMI-1640 and further stained with PE-conjugated anti-DC-SIGN and APC-conjugated anti-CD11b antibodies for 30 min at 4 °C. Cells were then analyzed by flow cytometry. Binding to THP1 and THP1::DC-SIGN cells was realized as previously described for DC-SIGN–expressing HeLa cells [7]. Inhibition of binding was realized using anti–DC-SIGN antibodies at 10 μg/ml.
Enzyme-Linked Immunosorbent Assay
BAL fluids were centrifuged for 15 min at 1000 × g and supernatants were concentrated three to eight times using Ultra-15 and YM-5 concentrators (Amicon; Millipore, Billerica, Massachusetts, United States). Concentrated BAL fluids were analyzed for IL-4, IL-10, and IL-13 using high-sensitivity enzyme-linked immunosorbent assay (ELISA) kits from R&D Systems. The minimal detectable dose of IL-4, IL-10, and IL-13 was 0.11 pg/ml, <0.5 pg/ml, and <32 pg/ml, respectively. For analysis of alveolar Mφ supernatants, Mφs were cultured in Petri dishes for 1 h before removing non-adherent cells. Adherent cells were further incubated for 18 h in medium with or without E. coli LPS (10 ng/ml, Sigma), LAM (10 μg/ml; gift from G. Puzo, IPBS, Toulouse, France), anti-DC-SIGN antibodies (50 μg/ml, clones 1B10 and 120507), and isotype control (50 μg/ml). IL-10 in culture supernatant was then measured by ELISA (high-sensitivity kit, R&D Systems).
Immunohistochemistry
Lung biopsies were stained for DC-SIGN and M. tuberculosis using a DC-SIGN antibody (clone 1B10) and a homemade anti-M. bovis BCG polyclonal rabbit serum, respectively, as previously described [19].
Confocal Microscopy
Mφs were allowed to adhere for 15 min on polylysine-coated cover slips before fixation. Cells were then permeabilized with 0.05% saponin, immunostained, and analysed as described [19]. DC-SIGN was detected using clone 120507 mAb and a Cy3-conjugated secondary mAb (Amersham, Little Chalfont, United Kingdom). Examination was realized using a confocal microscope (Zeiss, Oberkochen, Germany) and the software LSM 510 v3.2 (Zeiss).
RT-PCR
Cells were recovered and homogeneized in Trizol (GIBCO BRL/Invitrogen, Carlsbad, California, United States). Total RNAs were extracted by classical chloroform procedure. Purified RNAs (100 ng) were used to amplify the DC-SIGN and GAPDH (glyceraldehyde-3-phosphate dehydrogenase) transcripts using the Onestep RT-PCR kit (Qiagen, Valencia, California, United States). Oligonucelotides hCD209F (5′-
ATTTTC
CAACTCATTTTCAGCC-3′) and hCD209R (5′-
TCTCACAGAAAGAGGAGGACAC-3′) were used to amplify a 193-base pair product in the exon 7 of DC-SIGN. Oligonucelotides GAPDHRT1 (5′-
GAGTCAACGGATTTGGTCGTAT-3′) and GAPDHRT2 (5′-
AGTCTTCTGGGTGGCAGTGAT-3′) were used to amplify a 542-base pair product from exons 2 to 8 of GAPDH. Reverse transcription was conducted for 30 min at 50 °C. PCR consisted of 15-min activation (95 °C), and 35 cycles of 1-min denaturation (94 °C), 1-min annealing (55 °C), and 1-min elongation (72 °C). RT-PCR products were analyzed on a 1% agarose gel.
Results
Alveolar CD11b-Expressing Cells Over-Express DC-SIGN in Patients with TB
Given the important role of DC-SIGN in mycobacterial interactions with human MoDCs in vitro [7–9], we asked whether the lectin might interact with the tubercle bacillus during the natural course of TB in vivo in the lungs. Flow cytometry analysis of BAL cells from TB patients typically showed two major cell populations (Figure 1A, upper left panel). One of them, noted as R1, essentially consisted of CD3+ and either CD4+ or CD4− T lymphocytes (Figure 1A, upper right panel). Cells from the other major population, noted as R2, were all CD3− (data not shown). Total cells were co-stained with antibodies directed against CD11b (type 3 complement receptor [CR3] α chain) and DC-SIGN. Cells from R2 were all CD11b+ (Figure 1A, middle left panel). In patients with TB, a variable proportion of this cell population was DC-SIGN+ (27% in the patient shown in Figure 1A, middle right panel). By contrast, R2 cells from patients without TB were nearly all CD11b+DC-SIGN− (results obtained with BAL cells from a patient with sarcoidosis are shown in Figure 1A, lower left and right panels). Results obtained for all individuals included in the study are shown in Figure 1B. The proportion of CD11b+DC-SIGN+ cells in BALs was not correlated with gender (19.8 ± 20.0% in males vs. 16.3 ± 13.7% in females; Mann-Whitney test p = 0.786), and very weakly with age, as assessed by Spearman correlation coefficient (ρ
s
2 = 0.094, p = 0.0078), which most likely reflected the bias in population sizes depending on age and pathology. Indeed, 72% of the whole population was aged ≤15 y and included 85% of the total TB cases. When the whole population was divided in individuals of ≤15 or ≥20 y of age, correlation with age was no longer observed (ρ
s
2 = 0.003, p = 0.813, and ρ
s
2 = 0.0002, p = 0.777, respectively). By contrast, the proportion of CD11b+DC-SIGN+ cells in BALs was highly related to the pathology. Indeed in patients with TB, 11%–71% of CD11b+ cells (31.2 ± 13.4%) were found to express the lectin. By contrast, CD11b+ cells from patients with other lung diseases or from control individuals barely expressed DC-SIGN (2.9 ± 1.8% in patients with asthma, Mann-Whitney test p < 10−7; 3.7 ± 4.1% in patients with sarcoidosis, Mann-Whitney test p < 10−6; 1.9 ± 2.0% in other individuals, Mann-Whitney test p < 10−5; all as compared to the group of patients with TB). Immunohistochemistry analysis of lung biopsy samples from patients with TB (n = 4) revealed that DC-SIGN+ cells localized mostly in cellular infiltrates outside granulomas and were tightly associated with bacilli (representative result from one patient is shown in Figure 1C). By contrast, and in accordance with results from flow cytometry, very few DC-SIGN–expressing cells were detected inside and outside granulomas in lung biopsy samples from patients with sarcoidosis (n = 3; representative result from one patient is shown in Figure 1D).
Figure 1 Alveolar CD11b+ Cells Over-Express DC-SIGN in Patients with TB
(A) BAL cells from a patient with TB (upper four panels) and from a patient with sarcoidosis (lower two panels) were analyzed by flow cytometry. Expression of CD3 and CD4 was analyzed on cells from R1. CD11b and DC-SIGN expression was analyzed on cells from R2.
(B) Distribution of the proportion of CD11b+DC-SIGN+ cells in BALs according to pathology and age. Black circles indicate ≤15 y of age; black triangles indicate ≥20 y; NC, no case.
(C) DC-SIGN (upper panels) and M. tuberculosis (lower panels) immunodetection in serial sections of a lung biopsy from a patient with TB. The pictures are representative of results obtained with samples from a total of four patients. G, granuloma.
(D) DC-SIGN immunodetection in a lung biopsy from a patient with sarcoidosis. The pictures are representative of results obtained with samples from a total of three patients.
In (C) and (D), magnification in left panels is 100×, and regions in squares are shown at higher magnification in right panels.
Alveolar DC-SIGN–Expressing Cells in Patients with TB Are Mφs
The ability of CD11b+DC-SIGN+ BAL cells to bind to the plastic (Figure 2A), together with their morphology (Figure 2B), strongly suggested a Mφ phenotype. In order to precisely define this phenotype, BAL cells from patients with TB were stained for DC-SIGN and for a number of markers specific to various DC subtypes, monocytes, and Mφs, and were analyzed by flow cytometry (Figure 2C and 2D). Four DC subtypes have been described in the human lungs [21], namely myeloid DCs type 1 (MDC1) expressing BDCA-1 and HLA-DR, myeloid DCs type 2 (MDC2) which are BDCA-3hiHLA-DR+CD32−CD64−, plasmacytoid DCs (pDCs) expressing BDCA-2 and CD123, and CD1a+ Langerhans cell type DCs. As shown in Figure 2C and 2D, DC-SIGN–expressing cells were clearly negative for BDCA-1 and BDCA-2. Cells were also BDCA-3loCD32+CD64+, and they were negative for the Langerhans cell markers CD1a and CD207 (langerin; data not shown). Finally the cells did not express the DC maturation marker CD83. It is thus very unlikely that these cells constitute either myeloid or plasmacytoid DCs [21]. Cells were also negative for the monocyte marker CD14 (Figure 2C). However, the cells expressed the myeloid marker CD68 (Figure 2C). Additional staining helped to further define the phenotype of these cells. Indeed DC-SIGN+ cells in R2 (see Figure 1A) clearly expressed the co-stimulation and presentation molecules CD40, CD86 (B7–2), and HLA-DR, as well as the antigen uptake molecules CD11b, CD11c (CR4 α chain), CD206 (mannose receptor), CD16 (FcγRIII), CD32 (FcγRII), CD64 (FcγRI), and the Toll-like receptors (TLRs) TLR2, TLR4, and TLR9. Altogether, these results indicate that the CD11b+DC-SIGN+ cells observed in BALs from patients with TB most likely constitute alveolar Mφs. Furthermore, DC-SIGN+ and DC-SIGN− CD11b+ alveolar Mφs were found to express HLA-DR and CD86 at similar levels (data not shown), indicating that DC-SIGN expression by these cells did likely not result from cell activation.
Figure 2 Alveolar DC-SIGN+ Cells in Patients with TB Are Mφs
(A) Total BAL cells from a patient with TB were allowed to adhere to the plastic for 1 h at 37 °C in complete medium. CD11b and DC-SIGN expression was analyzed by flow cytometry before (left) and after (right) adherence.
(B) Surface and intracellular DC-SIGN (red) expression by an adherent alveolar cell examined under the confocal microscope.
(C) Flow cytometry analysis of surface expression of BDCA-1 (CD1c), BDCA-2, BDAC-3, CD1a, CD11b, CD11c, CD14, CD68, CD83, and CD123 in DC-SIGN+ BAL cells from a patient with TB.
(D) Flow cytometry analysis of surface expression of CD40, CD86, HLA-DR, CD11b, CD11c, CD206, CD16, CD32, CD40, CD64, TLR2, TLR4, and TLR9 in DC-SIGN+ BAL cells from a patient with TB.
In (C) and (D), analysis was performed on DC-SIGN–expressing cells in R2, as shown in Figure 1.
DC-SIGN Is Induced on Resident Alveolar Mφs upon M. tuberculosis Infection
We next sought to address whether the DC-SIGN+ Mφs detected in BALs from patients with TB were resident alveolar Mφs or were derived from DC-SIGN+ monocytes recruited from the circulation during infection. Peripheral blood mononuclear cells (PBMCs) from patients with TB (n = 3) were analysed for DC-SIGN expression by flow cytometry. DC-SIGN could be detected neither in these cells nor in PBMCs from healthy blood donors (results from one representative experiment is shown in Figure 3A). To assess the induction hypothesis, DC-SIGN− alveolar Mφs were purified by adherence from patients with asthma (n = 3). Cells were then infected with M. tuberculosis at a MOI of 1 bacterium/cell, or treated with IL-4, TNF-α, or E. coli–derived LPS. IL-4 was used as a control because this cytokine has been shown to induce DC-SIGN expression in human monocytes [22]. TNF-α was used to assess whether an inflammatory context, like the one encountered in lungs from patients with TB, could induce DC-SIGN. Finally, LPS was included in order to assess whether TLR ligation may result in DC-SIGN induction, as recently reported in human circulating blood monocytes [23]. Flow cytometry analysis after 48 h showed a marked expression of DC-SIGN on the surface of the cells, both when infected with M. tuberculosis and when treated with IL-4, as compared to untreated cells (Figure 3B). By contrast, neither TNF-α- nor LPS-treated cells were found to express the lectin. Interestingly, in cells infected with a GFP-expressing strain of M. tuberculosis, both GFP+ (infected) and GFP− (uninfected) cells were found DC-SIGN+, indicating that DC-SIGN induction likely relies on a soluble factor secreted by the host and/or the pathogen and that has yet to be identified. RT-PCR analysis of the DC-SIGN transcript at various times post-infection (Figure 3C) further confirmed that M. tuberculosis infection induces DC-SIGN expression in resident alveolar Mφs as soon as 1 h after infection. Because IL-4 and IL-13 were previously shown to induce DC-SIGN expression in monocytes [22], we wished to assess whether these cytokines might account for DC-SIGN expression by alveolar Mφs in vivo in the lungs of patients with TB. We collected BAL fluids from a number of patients with TB or with other pathologies, and we measured IL-4 and IL-13 amounts by ELISA. IL-4 was detected in some samples, but without any apparent correlation with either the pathology or the amount of DC-SIGN-expressing alveolar Mφs in the corresponding samples (Table 1). IL-13 was not detectable in all samples tested (data not shown). This indicates that IL-4 and IL-13 are likely not responsible for DC-SIGN induction in vivo in the lungs of M. tuberculosis–infected patients.
Figure 3 DC-SIGN Is Induced on Resident Alveolar Mφs upon M. tuberculosis Infection
(A) Flow cytometry analysis of CD11b and DC-SIGN expression by PBMCs from a healthy donor (upper panels) and a patient with TB (lower panels).
(B) Adherent DC-SIGN− alveolar Mφs from a non-tuberculous patient were infected with a GFP-expressing strain of M. tuberculosis at a MOI of one bacterium per cell, or treated with IL-4, TNF-α, or LPS, or left untreated (ø control). After 48 h at 37 °C, cells were recovered and DC-SIGN expression was analyzed by flow cytometry. In M. tuberculosis panel, the grey area corresponds to GFP+ (infected) cells, and the plain line corresponds to GFP− (uninfected) cells.
(C) Cells infected with M. tuberculosis at a MOI of 1 for 1, 2, 4, or 24 h were analysed by RT-PCR for DC-SIGN and GAPDH mRNAs.
Table 1 IL-4 in BAL Fluids Does Not Correlate with Pathology nor with Amount of DC-SIGN-Expressing Alveolar Mφs
DC-SIGN Mediates M. tuberculosis Binding to Alveolar Mφs from Patients with TB
We next wished to evaluate the possible functional consequences of DC-SIGN induction in alveolar Mφs in patients with TB. DC-SIGN has previously been shown to be a major M. tuberculosis receptor on human MoDCs [7]. We thus evaluated lectin's contribution to mycobacterial binding to and entry into DC-SIGN+ alveolar Mφs. Total BAL cells from two patients with TB were incubated with a GFP-expressing strain of M. tuberculosis for 4 h at 4 °C in the presence or absence of blocking anti-CD11b (CR3) or anti–DC-SIGN or control isotype antibodies. The cells were stained for CD11b and DC-SIGN using phycoerythrin- or allophycocyanin-conjugated antibodies. Flow cytometry analysis revealed that among CD11b+ alveolar Mφs, DC-SIGN–expressing cells were much more prone to infection than their DC-SIGN− counterparts. After infection, almost 80% of CD11b+DC-SIGN+ versus only 20% of CD11b+DC-SIGN− cells were found GFP+ (Figure 4A and 4B). Accordingly, anti-DC-SIGN antibodies could inhibit >60% of M. tuberculosis binding to DC-SIGN–expressing Mφs (Figure 4A and 4B). The antibodies had a slight positive effect on mycobacterial binding to CD11b+DC-SIGN− cells, which was likely due to more mycobacteria available for binding to these cells in the presence of the antibody (Figure 4A and 4B). Conversely, anti-CD11b antibodies could inhibit M. tuberculosis binding to DC-SIGN− alveolar Mφs of almost 50% in average, whereas they had only a very minor effect on binding to CD11b+DC-SIGN+ cells (Figure 4A and 4B). In order to confirm that DC-SIGN constitutes a major M. tuberculosis receptor in the context of a Mφ cell and in the presence of other mycobacterial receptors, THP1 human Mφs were transduced with a lentivirus-based DC-SIGN-encoding vector and were used in M. tuberculosis–binding experiments. M. tuberculosis was found to bind to DC-SIGN–expressing THP1 Mφs by greater than 10-fold more than to THP1 cells (Figure 4C). Anti-DC-SIGN antibodies could fully inhibit mycobacterial binding to DC-SIGN-expressing cells, whereas they had virtually no effect on binding to THP1 cells. Altogether, these results indicate that DC-SIGN expression renders Mφs, and in particular alveolar Mφs in the lungs of patients with TB, more susceptible to infection than their DC-SIGN− counterparts. Confocal microscopy examination of alveolar Mφs infected with GFP-expressing M. tuberculosis and subsequently immunostained for DC-SIGN, showed, as in MoDCs [7], a marked recruitment of the lectin at the site of bacterial attachment and in the nascent phagosome (Figure 4D, upper and middle panels), followed by exclusion of the receptor from the mycobacterial vacuole once the bacillus was engulfed, most likely as a result of receptor recycling (Figure 4D, lower panels).
Figure 4 DC-SIGN Mediates M. tuberculosis Binding to Alveolar Mφs from Patients with TB
(A) Alveolar Mφs from a patient with TB were infected with GFP-expressing M. tuberculosis, in the absence (ø; upper left panel) or the presence of control isotype (upper right panel), anti-CD11b (lower left panel), or -DC-SIGN (lower right panel) blocking antibodies. In the upper panels, cells were then stained with fluorescent PE-conjugated anti-DC-SIGN and APC-conjugated anti-CD11b antibodies. In lower panels, fluorescent antibodies were added together with blocking antibodies (same clones).
(B) Proportion of GFP+ cells in DC-SIGN− (open bars) and DC-SIGN+ (grey bars) alveolar Mφs as calculated from (A) using BALs from two patients with TB. THP1 Mφs expressing or not expressing DC-SIGN (THP1::DC-SIGN) were used in a binding experiment with M. tuberculosis H37Rv, in the presence or absence of anti-DC-SIGN antibodies.
(D) Confocal microscopy examination of adherent DC-SIGN+ cells infected with GFP-expressing M. tuberculosis for various times.
DC-SIGN Does Not Mediate IL-10 Secretion in Alveolar Mφs
Besides its function in pathogen recognition and phagocytosis, DC-SIGN has been proposed to play a role in triggering intracellular signals and cytokine secretion. In particular, it has previously been reported that DC-SIGN ligation by LAM synergizes with TLR4 ligation by LPS to potentiate IL-10 secretion by MoDCs [8]. We sought to determine whether this might also be the case in alveolar Mφs, which also express TLR4 (see Figure 2C). Alveolar Mφs from three patients with TB were purified by adherence and stimulated for 18 h with LPS alone or LPS and M. tuberculosis–derived LAM, in the presence or absence of anti-DC-SIGN or control isotype antibodies. Although basal IL-10 production differed among Mφ preparations, treatment of the cells with LPS resulted in a slight increase in IL-10 secretion in all cases, as revealed by ELISA (Table 2). Treatment of the cells with LAM alone had virtually no effect on IL-10 production. By contrast with what has been reported in MoDCs [8], LAM did clearly not synergize with LPS to induce production of the cytokine by alveolar Mφs (Table 2). This was confirmed by measuring IL-10 in BAL fluids from patients with TB or with other pathologies (Table 3). IL-10 was detected in some samples but not systematically in samples from patients with TB that contained DC-SIGN-expressing Mφs.
Table 2 DC-SIGN Ligation by LAM Does Not Potentiate IL-10 (in pg/ml) Production by LPS-Stimulated Alveolar Mφs
Table 3 IL-10 Is Not Detected in BAL Fluids from Patients with TB
Discussion
Our study reveals the presence of a novel DC-SIGN–expressing subset of alveolar Mφs in the lungs of patients with TB. In patients with TB, this Mφ subset represents up to 70% of total CD11b+ Mφs. By contrast, in other patients with non-tuberculous lung diseases, as well as in control individuals, DC-SIGN Mφs represent only 3% on average of CD11b+ Mφs. Our results strongly suggest that DC-SIGN+ Mφs arise from induction of the DC-SIGN gene during infection. IL-4 and IL-13, two cytokines known to induce DC-SIGN expression in monocytes [22], are unlikely to account for DC-SIGN expression by alveolar Mφs in patients with TB, because these cytokines were either not detected (IL-13) or detected in variable amounts and independently of pathology (IL-4) in BAL fluids from a number of different patients. Recently, DC-SIGN induction has been demonstrated in human circulating blood monocytes treated with TLR agonists [23]. This is unlikely to be the case here because we have shown that LPS treatment of DC-SIGN− alveolar Mφs obtained from individuals without TB could not induce DC-SIGN expression in these cells (see Figure 3B). Inflammation alone is unlikely to induce DC-SIGN, because alveolar Mφs from patients with asthma or sarcoidosis were mostly DC-SIGN−, whereas these diseases are characterized by a marked inflammatory response in the lungs. The fact that both M. tuberculosis–infected and uninfected alveolar Mφs up-regulate DC-SIGN during ex vivo infection (see Figure 3B) is of interest, as it strongly suggests that soluble factor(s) from the host and/or the microbe can induce DC-SIGN in bystander DC-SIGN− Mφs. These factors will have to be defined in future studies. In particular, the role of M. tuberculosis–derived lipids and proteins that are secreted by the bacterium prior to uptake by host cells and from inside infected cells [24,25] will have to be investigated. In addition, the possible role of host factors, such as IL-15, will have to be examined in this respect. Indeed, IL-15 has recently been shown to induce DC-SIGN expression in human monocytes in vitro [23], and alveolar Mφs from patients with TB have been reported to produce this cytokine ex vivo [26].
So far, mycobacterial receptors on Mφs have been mostly characterized using model Mφs, namely mouse bone marrow–derived or human monocyte–derived Mφs [27]. Very few reports are available on mycobacterial receptors on alveolar Mφs, and all of them used cells from healthy individuals or laboratory animals. The main mycobacterial receptors identified on human alveolar Mφs are CRs, especially CR4 and CR3 on human cells [28–30], the surfactant protein A receptor [31], and the mannose receptor [32]. Other studies in murine alveolar Mφs confirmed these findings and added other receptors, such as scavenger receptors, to the list [33]. However, to our knowledge, ours is the first study of M. tuberculosis receptors on alveolar Mφs in patients with TB. Our results suggest a novel scenario of alveolar Mφ infection during TB. In this scenario, CRs likely mediate most of cell infection in a naive host, and DC-SIGN–expressing Mφs become privileged target cells for the bacillus once the infection is established. Furthermore, DC-SIGN induction in bystander cells may be of advantage for the tubercle bacillus to increase its intracellular territory inside the infected host.
Apart from pathogen binding, DC-SIGN may play a role in signal transduction. In particular, DC-SIGN ligation by the mycobacterial lipoglycan LAM has been reported to potentiate TLR-4–mediated IL-10 secretion by LPS-stimulated MoDCs [8]. However, treatment of LPS-stimulated alveolar Mφs with LAM did not result in an increase of IL-10 production, which remained relatively low in both stimulated and untreated cells. Moreover, IL-10 was not detected in BAL fluids from patients with TB (Table 3). This is in accordance with previous studies reporting that IL-10 is detected in comparable amounts in BALs and lung biopsies from TB patients, and from healthy individuals [34]. IL-10 production is known to be low in Mφs as compared to in DCs. The discrepancy between our results, that show no role of DC-SIGN in IL-10 secretion, and those from Geijtenbeek et al. [8], who reported such a role, likely relies on the different cell types used in the two studies. Our results do not exclude the possibility that DC-SIGN may participate in IL-10 production by DCs in patients with TB, especially in the lymph nodes. This possibility will have to be further explored and may have important local consequences, including down-modulation of the local inflammation due to infection, either directly or by driving T lymphocytes toward a regulatory phenotype [35,36].
In conclusion, our study reveals that during the natural course of TB in human lungs, soluble host and/or mycobacterial factor(s) induce DC-SIGN expression by alveolar Mφs, which renders the cells highly prone to infection by the tubercle bacillus. DC-SIGN induction in alveolar Mφs may have important consequences on lung colonization by M. tuberculosis, as well as on host immune and inflammatory responses, which will require further investigation in cell and animal models, as well as in patients with TB.
Patient Summary
Background
Tuberculosis (TB) is one of the most common major infectious disease today. It is estimated that two billion people—or one-third of the world's population—are chronically infected without active symptoms. Nine million new cases of active disease are diagnosed annually, resulting in two million deaths, mostly in developing countries. TB is predominantly a lung disease. It is caused by a microbe called Mycobacterium tuberculosis, which infects lung cells. Patients with active disease easily infect others through coughing, sneezing, and spitting.
Why Was This Study Done?
Most of what we know about how Mycobacterium infects lung cells during the early stages of TB comes from animal studies or studies in healthy volunteers. In this study, the researchers wanted to examine whether lung cells from patients with TB were different from those of healthy people or those with different lung diseases, and what this might tell us about the way the infection spreads in the lung. In particular, they looked at the surface of the lung cells, because this is the part directly involved in the first contact with Mycobacterium.
What Did the Researchers Do and Find?
They studied 74 individuals: 40 had TB, 25 had other inflammatory lung diseases, and nine had neither active TB nor lung inflammation and served as healthy “controls.” The patients underwent a procedure (called bronchoalveolar lavage) that washes out some of the secretions and cells from the lower respiratory tract. The researchers then analysed the cells in different ways. They concentrated on a type of cell called a macrophage (the natural target of Mycobacterium) and found that macrophages from patients with TB had much more of a particular protein called DC-SIGN on their surface than macrophages from patients with other diseases or from the control individuals. They then took macrophages from a control individual (which thus had very low levels of DC-SIGN) and infected them with Mycobacterium under laboratory conditions. The researchers found that shortly after infection, not only the infected cells but also some of their neighbours started to display DC-SIGN on their surface. The researchers also found that having DC-SIGN on the surface made uninfected cells much more susceptible to infection.
What Does This Mean?
The results suggest that DC-SIGN has an important function in amplifying TB infection in the lung. In the long run, understanding how Mycobacterium infects patients and either makes them ill or establishes a chronic infection without acute symptoms should help with the development of new or better ways to prevent infection or treat disease.
Where Can I Find More Information Online?
The following Web sites provide information on tuberculosis.
World Health Organization pages on TB:
http://www.who.int/tb/en/
TB Vaccine Cluster page:
http://www.tb-vac.org
Tuberculosis.net, a source for TB teaching materials:
http://www.tuberculosis.net/
Wikipedia pages on TB:
http://en.wikipedia.org/wiki/Tuberculosis
MedlinePlus pages on TB:
http://www.nlm.nih.gov/medlineplus/tuberculosis.html
We thank D. Ensergueix (Paris) for technical assistance in histology. Confocal microscopy was realised at Dynamic Imaging Platform at Institut Pasteur. We thank G. R. Stewart, A. Amara, and J. Nigou for providing pEGFP, 1B10 antibody, and ManLAM, respectively. This research project has been co-financed by Institut Pasteur and the European Commission, within the 6th Framework Programme, contract no. LSHP-CT-2003–503367. The text represents the authors' views and does not necessarily represent a position of the Commission who will not be liable for the use made of such information. LT is a fellow of the European Commission. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Citation: Tailleux L, Pham-Thi N, Bergeron-Lafaurie A, Herrmann JL, Charles P, et al. (2005) DC-SIGN induction in alveolar macrophages defines privileged target host cells for mycobacteria in patients with tuberculosis. PLoS Med 2(12): e381.
Abbreviations
BALbroncho-alveolar lavage
CRcomplement receptor
DCdendritic cell
DC-SIGNdendritic cell–specific intercellular adhesion molecule-3 grabbing nonintegrin
ELISAenzyme-linked immunosorbent assay
FCSfoetal calf serum
GFPgreen fluorescent protein
HLAhuman leukocyte antigen
ILinterleukin
LAMlipoarabinomannan
LPSlipopolysaccharide
mAbsmonoclonal antibodies
MoDCmonocyte-derived dendritic cell
Mφmacrophage
PBMCperipheral blood mononuclear cell
TBtuberculosis
TNFtumor necrosis factor
TLRToll-like receptor
==== Refs
References
Flynn JL Chan J Immunology of tuberculosis Annu Rev Immunol 2001 19 93 129 11244032
Kaufmann SH How can immunology contribute to the control of tuberculosis? Nat Rev Immunol 2001 1 20 30 11905811
Russell DG
Mycobacterium tuberculosis Here today, and here tomorrow Nat Rev Mol Cell Biol 2001 2 569 577 11483990
Curtis BM Scharnowske S Watson AJ Sequence and expression of a membrane-associated C-type lectin that exhibits CD4-independent binding of human immunodeficiency virus envelope glycoprotein gp120 Proc Natl Acad Sci U S A 1992 89 8356 8360 1518869
Geijtenbeek TB Kwon DS Torensma R van Vliet SJ van Duijnhoven GC DC-SIGN, a dendritic cell-specific HIV-1-binding protein that enhances trans-infection of T cells Cell 2000 100 587 597 10721995
Cambi A Koopman M Figdor CG How C-type lectins detect pathogens Cell Microbiol 2005 7 481 488 15760448
Tailleux L Schwartz O Herrmann J Pivert E Jackson M DC-SIGN is the major Mycobacterium tuberculosis receptor on human dendritic cells J Exp Med 2003 197 121 127 12515819
Geijtenbeek T van Vliet S Koppel E Sanchez-Hernandez M Vandenbroucke-Grauls C Mycobacteria target DC-SIGN to suppress dendritic cell function J Exp Med 2003 197 7 17 12515809
Maeda N Nigou J Herrmann JL Jackson M Amara A The cell surface receptor DC-SIGN discriminates between Mycobacterium species through selective recognition of the mannose caps on lipoarabinomannan J Biol Chem 2002 278 5513 5516 12496255
Pitarque S Herrmann JL Duteyrat JL Jackson M Stewart GR Deciphering the molecular bases of Mycobacterium tuberculosis binding to DC-SIGN reveals an underestimated complexity Biochem J 2005 Epub ahead of print
Geijtenbeek TB van Kooyk Y Pathogens target DC-SIGN to influence their fate DC-SIGN functions as a pathogen receptor with broad specificity APMIS 2003 111 698 714 12974773
Tailleux L Maeda N Nigou J Gicquel B Neyrolles O How is the phagocyte lectin keyboard played? Master class lesson by Mycobacterium tuberculosis
Trends Microbiol 2003 11 259 263 12823942
Tailleux L Gicquel B Neyrolles O
Mycobacterium tuberculosis and dendritic cells: Who's manipulating whom? Curr Immunol Rev 2005 1 101 105
Buettner M Meinken C Bastian M Bhat R Stossel E Inverse correlation of maturity and antibacterial activity in human dendritic cells J Immunol 2005 174 4203 4209 15778382
Soilleux EJ Morris LS Leslie G Chehimi J Luo Q Constitutive and induced expression of DC-SIGN on dendritic cell and macrophage subpopulations in situ and in vitro J Leuk Biol 2002 71 445 457
Costabel U Hunninghake GW ATS/ERS/WASOG statement on sarcoidosis. Sarcoidosis Statement Committee. American Thoracic Society. European Respiratory Society. World Association for Sarcoidosis and Other Granulomatous Disorders Eur Respir J 1999 14 735 737 10573213
European Society of Pneumology Task Group Technical recommendations and guidelines for bronchoalveolar lavage (BAL). Report of the European Society of Pneumology Task Group Eur Respir J 1989 2 561 585 2663535
Moris A Nobile C Buseyne F Porrot F Abastado JP DC-SIGN promotes exogenous MHC-I-restricted HIV-1 antigen presentation Blood 2004 103 2648 2654 14576049
Tailleux L Neyrolles O Honoré-Bouakline S Perret E Sanchez F Constrained intracellular survival of Mycobacterium tuberculosis in human dendritic cells J Immunol 2003 170 1939 1948 12574362
Schlesinger LS Bellinger-Kawahara CG Payne NR Horwitz MA Phagocytosis of Mycobacterium tuberculosis is mediated by human monocyte complement receptors and complement component C3 J Immunol 1990 144 2771 2780 2108212
Demedts IK Brusselle GG Vermaelen KY Pauwels RA Identification and characterization of human pulmonary dendritic cells Am J Respir Cell Mol Biol 2005 32 177 184 15576669
Relloso M Puig-Kroger A Pello OM Rodriguez-Fernandez JL de la Rosa G DC-SIGN (CD209) expression is IL-4 dependent and is negatively regulated by IFN, TGF-beta, and anti-inflammatory agents J Immunol 2002 168 2634 2643 11884427
Krutzik SR Tan B Li H Ochoa MT Liu PT TLR activation triggers the rapid differentiation of monocytes into macrophages and dendritic cells Nat Med 2005 11 653 660 15880118
Beatty WL Rhoades ER Ullrich HJ Chatterjee D Heuser JE Trafficking and release of mycobacterial lipids from infected macrophages Traffic 2000 1 235 247 11208107
Beatty WL Ullrich HJ Russell DG Mycobacterial surface moieties are released from infected macrophages by a constitutive exocytic event Eur J Cell Biol 2001 80 31 40 11211933
Zissel G Baumer I Schlaak M Muller-Quernheim J In vitro release of interleukin-15 by broncho-alveolar lavage cells and peripheral blood mononuclear cells from patients with different lung diseases Eur Cytokine Netw 2000 11 105 112 10705307
Ernst JD Macrophage receptors for Mycobacterium tuberculosis
Infect Immun 1998 66 1277 1281 9529042
Hirsch CS Ellner JJ Russell DG Rich EA Complement receptor-mediated uptake and tumor necrosis factor-alpha-mediated growth inhibition of Mycobacterium tuberculosis by human alveolar macrophages J Immunol 1994 152 743 753 8283049
Cywes C Godenir NL Hoppe HC Scholle RR Steyn LM Nonopsonic binding of Mycobacterium tuberculosis to human complement receptor type 3 expressed in Chinese hamster ovary cells Infect Immun 1996 64 5373 5383 8945590
Cywes C Hoppe HC Daffe M Ehlers MR Nonopsonic binding of Mycobacterium tuberculosis to complement receptor type 3 is mediated by capsular polysaccharides and is strain dependent Infect Immun 1997 65 4258 4266 9317035
Gaynor CD McCormack FX Voelker DR McGowan SE Schlesinger LS Pulmonary surfactant protein A mediates enhanced phagocytosis of Mycobacterium tuberculosis by a direct interaction with human macrophages J Immunol 1995 155 5343 5351 7594549
Roecklein JA Swartz RP Yeager H Nonopsonic uptake of Mycobacterium avium complex by human monocytes and alveolar macrophages J Lab Clin Med 1992 119 772 781 1593222
Stokes RW Thorson LM Speert DP Nonopsonic and opsonic association of Mycobacterium tuberculosis with resident alveolar macrophages is inefficient J Immunol 1998 160 5514 5521 9605155
Morosini M Meloni F Marone Bianco A Paschetto E Uccelli M The assessment of IFN-gamma and its regulatory cytokines in the plasma and bronchoalveolar lavage fluid of patients with active pulmonary tuberculosis Int J Tuberc Lung Dis 2003 10 994 1000
Boussiotis VA Tsai EY Yunis EJ Thim S Delgado JC IL-10-producing T cells suppress immune responses in anergic tuberculosis patients J Clin Invest 2000 105 1317 1325 10792007
Spellberg B Edwards JE Type 1/Type 2 immunity in infectious diseases Clin Infect Dis 2001 32 76 102 11118387
|
16279841
|
PMC1283365
|
CC BY
|
2021-01-05 10:39:22
|
no
|
PLoS Med. 2005 Dec 15; 2(12):e381
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020381
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 10.1371/journal.pmed.0020410SynopsisCell BiologyInfectious DiseasesMicrobiologyPathologyInfectious DiseasesTuberculosisPathologyRespiratory MedicineDC-SIGN and Lung Pathogenesis in Patients with Tuberculosis Synopsis12 2005 15 11 2005 2 12 e410Copyright: © 2005 Public Library of Science.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
DC-SIGN Induction in Alveolar Macrophages Defines Privileged Target Host Cells for Mycobacteria in Patients with Tuberculosis
==== Body
C-type lectins are carbohydrate-binding cell surface molecules with a wide range of biological functions, many of which are related to immunity. Despite its name, dendritic cell–specific intercellular adhesion molecule-3 grabbing nonintegrin (DC-SIGN) is not only expressed on dendritic cells but also on specialized macrophages in the placenta and lung. A number of pathogens are known to interact with DC-SIGN, and some (including HIV) seem to have evolved to derive advantages from these interactions.
Recent in vitro studies have shown that DC-SIGN can interact with Mycobacterium tuberculosis through a lipoglycan (a molecule composed of sugars and fatty acids) on the mycobacterial envelope called lipoarabinomannan (LAM). Trying to understand the role of DC-SIGN in tuberculosis (TB), Ludovic Tailleux and colleagues have focused on the interaction between M. tuberculosis and DC-SIGN–expressing cells in the lungs of human patients.
The researchers studied a total of 74 individuals, including 40 with TB, 11 with sarcoidosis, 14 with asthma, and nine control participants without active lung infection or inflammation. All patients underwent bronchoalveolar lavage (BAL), a procedure that yields cells and proteins from the lower respiratory tract. The researchers then examined BAL cell populations after staining for various cell-surface markers by flow cytometry, and found that, in individuals without TB, very few alveolar macrophages (an average of 3%) expressed DC-SIGN. In contrast, an average of 30% (and up to 70%) of macrophages from patients with TB expressed the lectin.
Tailleux and colleagues then incubated alveolar macrophages from a patient without TB ex vivo with M. tuberculosis, which resulted in infection of a subset of the cells. When the researchers examined DC-SIGN expression, they found that both infected and noninfected (bystander) cells in the population started to express DC-SIGN. The effect on bystander cells suggests that soluble factors from the microbe and/or the infected cells can induce DC-SIGN expression. Further functional ex vivo studies with cells from human patients indicated that DC-SIGN expression renders alveolar macrophages more susceptible to infection.
The authors propose a scenario where complement receptors mediate most of the initial infection of alveolar macrophages in a naïve host, and where—once the infection is established—DC-SIGN–expressing alveolar macrophages become preferential target cells for M. tuberculosis. Future work will be focused on identifying the soluble factors involved, and on determining whether DC-SIGN induction is an essential part of TB pathogenesis.
|
0
|
PMC1283366
|
CC BY
|
2021-01-05 10:39:26
|
no
|
PLoS Med. 2005 Dec 15; 2(12):e410
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020410
|
oa_comm
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.