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 ehp0113-a00803PerspectivesCorrespondenceSynthetic Musk Compounds: Luckenbach Responds Luckenbach Till Epel David Hopkins Marine Station of Stanford University Pacific Grove, California E-mail:
[email protected] authors declare they have no competing financial interests.
12 2005 113 12 A803 A804 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 his letter on our recent article in EHP (Luckenbach and Epel 2005), Salvito raises important questions about effects of the synthetic musk fragrances regarding a) human and environmental health effects, b) environmental concentrations of the musks, and c) uniqueness of inhibition of efflux transporters to the musks, the effect we described in our article.
a) Regarding health issues, we agree with Salvito that the available evidence indicates minimal direct affects of most synthetic musks on the health of humans and aquatic organisms. However, our data expand the definition of toxicity and detrimental effects to indirect and unanticipated consequences of these chemicals, even if the chemical itself might be nontoxic. The major point of our article (Luckenbach and Epel 2005) was that the musks inhibit efflux (drug) transporters, which act as first lines of defense to pump potentially toxic substances out of cells. These efflux transporters are ubiquitous and are found in bacteria, fungi, plants, and animals, including humans. The transporters have wide substrate specificity, and this binding to many compounds can result in inhibition of activity by competing substrates. As a consequence of transporter inhibition, cells and organisms can therefore become exposed to toxicants normally kept out of their cells.
An unexpected finding was not only that the musks inhibit these transporters in marine mussels but that the effect is long-term and persists up to 24–48 hr after removal of the musk compounds. These indirect and long-term toxicity effects are of particular concern because these chemicals are stable and bioaccumulate; for example, musk xylene has a half-life of 70 days in human tissue (Riedel and Dekant 1999).
Effects on human transporters by the musks cannot be inferred from our results, but they do point to the possibility of an interaction, considering the general property of the transporters to recognize a wide array of substrates. Therefore—and in light of accumulation of the musks in human tissue—research is needed to determine if the musks similarly inhibit the human efflux transporters, thereby compromising this defense against toxicants.
b) The musks are of environmental concern because they enter the water column from incomplete degradation in sewage plants. We agree with Salvito that the reported levels in surface waters are extremely low (picomolar range) but disagree with his conclusion that such levels indicate that musks are not a problem. In spite of these low environmental levels, there is significant bioaccumulation of these chemicals in tissues of mussels and fish, and just several months ago Nakata (2005) reported significant bioaccumulation in cetaceans. The concentrations in aquatic organisms can become quite high, being on the order of nanograms per gram fresh weight, which translates to about 0.1 μM final concentration in tissue (Nakata 2005; Rimkus 1999; Yamagishi et al. 1983).
According to Salvito, worldwide production of synthetic musks are only about one-half of the amount we cite. These lower numbers are even more worrisome because because this means that the potency of the musks to bioaccumulate is even higher.
c) Salvito points out that the inhibition of transporters is not unique to the musks. We agree and note that the observed inhibition of efflux transporter activity by the musks may be the tip of the iceberg. As with the musks, there may be many chemicals that by themselves are not toxic but similarly inhibit the efflux transporters and thereby expose the organism to normally excluded toxicants.
In summary, the available data suggest that efflux transporter inhibition could be a significant indirect, negative, and unappreciated effect of environmental chemicals. Several questions need to be answered: Do these chemicals inhibit human transporters? Are there other anthropogenic and natural products that inhibit these transporters in aquatic organisms and also in humans? Should anthropogenic chemicals be screened for inhibitory activity? If so, should there be voluntary or governmental regulations to ensure that such chemicals do not affect the health of exposed populations through these indirect actions?
==== Refs
References
Luckenbach T Epel D 2005 Nitromusk and polycyclic musk compounds as long-term inhibitors of cellular xenobiotic defense systems mediated by multidrug transporters Environ Health Perspect 113 17 24 15626642
Nakata H 2005 Occurrence of synthetic musk fragrances in marine mammals and sharks from Japanese coastal waters Environ Sci Technol 39 10 3430 3434 15952346
Riedel J Dekant W 1999 Biotransformation and toxicokinetics of musk xylene in humans Toxicol Appl Pharmacol 157 2 145 155 10366547
Rimkus GG 1999 Polycyclic musk fragrances in the aquatic environment Toxicol Lett 111 1–2 37 56 10630702
Yamagishi T Miyazaki T Horii S Akiyama K 1983 Synthetic musk residues in biota and water from Tama River and Tokyo Bay (Japan) Arch Environ Contam Toxicol 12 1 83 89 6830312
|
0
|
PMC1314935
|
CC0
|
2021-01-04 23:41:31
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A803-A804
|
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-a0080416330333PerspectivesCorrespondenceBenefits and Risks of Pesticide Testing on Humans Needleman Herbert L. School of Medicine, University of Pittsburgh, Pittsburgh, Pennsylvania, E-mail:
[email protected] J. Routt Medical University of South Carolina, Charleston, South CarolinaLandrigan Philip Department of Community Medicine, Mt. Sinai School of Medicine, New York, New YorkSass Jennifer Natural Resources Defense Council, Washington, DCBearer Cynthia School of Medicine, Case Western Reserve University, Cleveland, OhioJ.S is employed by environmental nonprofit organizations with an interest in ensuring that regulations of toxic chemicals are as health-protective as feasible. The remaining authors declare they have no competing financial interests.
12 2005 113 12 A804 A805 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 review of the history of the U.S. Environmental Protection Agency’s (EPA) response to the question of human testing of pesticides, Resnick and Portier (2005) argued that the benefits of such testing outweigh the hazards, and they attempted to refute claims that human testing is both unproductive and unethical. We consider their arguments vague, tendentious, and essentially incorrect.
Unanimously passed by both houses of Congress in 1996, with the support of pesticide manufacturers, pediatricians, and the environmental community, the Food Quality Protection Act (FQPA 1996) added a 10-fold child protective safety factor in choosing a reference dose to two earlier factors, one employed to accommodate the difference between animals and humans and one to accommodate the variance among adults. The single stimulus behind the FQPA was the growing evidence of increased childhood vulnerability, and the single reason for its unanimous, bipartisan passage was to protect children.
The pesticide industry quickly mounted a two-pronged attack on the U.S. EPA’s new guidelines and safety factors (U.S. EPA 2000), arguing that children were not more sensitive than adults. At the same time, they launched studies in which organophosphate pesticides were administered to adult “ volunteers.” This was a palpable effort to circumvent and weaken the 10-fold human/animal safety factor, and it flouted the intent of the law to stimulate the generation of data on the developmental and pediatric toxicity of pesticides. Resnik and Portier (2005), in a curious shift of responsibility, indict the FQPA as a factor in stimulating human studies with their claim that
A law that was intended to provide additional safety protection for children had the unintended effect of encouraging some companies to test toxic compounds on human beings to avoid the regulatory impact of the law.
With few exceptions the U.S. EPA has failed to use the mandated 10X factor. In June 2002 the U.S. EPA issued its cumulative assessment of the organophosphate pesticides (OPs), determining that for the 30 OPs reviewed, a 1X safety factor (that is, no factor) was used for three OPs and one metabolite, and a 3X reduction was used for the others (U.S. EPA 2002). At no time was a 10X factor used, despite the fact that the U.S. EPA possessed developmental neurotoxicity data for only 6 of the 30 OPs at the time of its assessment
In 1998 the U.S. EPA convened a special committee consisting of members of the Science Advisory Board, the Science Advisory Panel for pesticides, and outside ethicists to examine the ethics of human testing. The committee had two meetings separated by 12 months, and after five drafts, adopted a report that accepted human testing subject to rigorous or severe limitations (U.S. EPA 2000). The two pediatricians on the committee (H.L.N. and R.R.) filed a minority report that became part of the record because we objected to procedural and scientific solecisms. Resnik and Portier (2005) did not mention this report, even though Portier was a member of that committee.
Resnik and Portier (2005) recognize that past industry studies are scientifically unacceptable. They directed their comments to future, yet-to-be-specified studies. Such studies, they stated, “can be conducted only if they meet strict scientific and ethical standards and provide public health or environmental benefits.” Resnik and Portier (2005) also stated that studies of adult volunteers “could yield knowledge about the toxic effects on humans, which could promote human health” (National Research Council 2004). The reader is left to wonder what toxic effects would be better understood through human studies, or what health benefits could accrue from short-term volunteer studies. Resnik and Portier (2005) mentioned neither children’s health nor developmental toxicity. Instead they proffered the dubious hope that studies could result in stricter safety standards or new legislation that could result in reduced pesticide exposure. In today’s regulatory climate, this must be considered a slim possibility. Once more the future standards and laws remain unspecified.
Two major issues in human testing are the relevance of data obtained from adult exposure to risk estimates for children, and the scientific validity of short-term human studies as predictors of health outcomes such as neurodevelopmental deficits and carcinogenesis. It is axiomatic that a study that is poorly designed and cannot produce valid conclusions is unethical on this ground alone.
The provenance of the FQPA (1996) emerged from the growing realization that children are vastly different from adults and that the developing organism, while it is laying down and pruning back neural connections, is much more sensitive to neurotoxicants than fully formed organisms. What could possibly be learned about the risk to this group from studying the effects of toxicants on adults? Resnik and Portier (2005) did not attempt to address this question, but the answer is, very little.
One of the critical issues in evaluating the scientific validity of a study design is statistical power. On this basis alone human studies have failed. A study with inadequate power to find an effect is by definition unethical. This type of study submits subjects to some risk while providing no scientific information. There are roughly 19 million children in the United States ≤5 years of age. If a toxicant harmed 1 child in 1,000, that would place 19,000 children at risk nationwide. A study with adequate power to detect an increase in deficit from 1% to 2% would require 3,017 subjects in each group to yield a power of 0.8, at α = 0.05. Past industry studies with sample sizes < 50 had about a 3% chance of finding an effect if it were present. No mention of this power finding was made by Resnik and Portier (2005), although this was published by the U.S. EPA Science Advisory Board/Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA) Science Advisory Panel (U.S. EPA 2000) on which Portier served as a contributing member.
Resnik and Portier’s article (Resnik and Portier 2005), when examined in light of specificity, completeness, and relevance to the health of children, fails on all points. It asks the reader to accept unspecified studies on adults as productive of unspecified benefits to human health. The principal toxic target, the health of children, remains unspoken and out of awareness.
==== Refs
References
FQPA 1996. Food Quality Protection Act of 1996. Public Law 104–170.
National Research Council 2004. Intentional Human Dosing Studies for EPA Regulatory Purposes: Scientific and Ethical Issues. Washington, DC:National Academy Press.
Resnick DB Portier C 2005 Pesticide testing on human subjects: weighing benefits and risks Environ Health Perspect 113 813 817 16002367
U.S. EPA 2000. Comments on the use of data from the testing of human subjects. A Report by the Science Advisory Board and the FIFRA Scientific Advisory Panel. EPA-SAB-EC-00-017 Washington, DC:U.S. Environmental Protection Agency. Available: http://www.epa.gov/science1/pdf/ec0017.pdf [accessed 2 November].
U.S. EPA 2002. Revised OP Cumulative Risk Assessment. FQPA Safety Factor. Washington, DC:U.S. Environmental Protection Agency. Available: http://www.epa.gov/pesticides/cumulative/rra-op/I_G.pdf [accessed 10 November 2005.
|
16330333
|
PMC1314936
|
CC0
|
2021-01-04 23:41:31
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A804-A805
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.113-1314936
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a00807PerspectivesCorrectionErrata 12 2005 113 12 A807 A807 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
There was an error in Figure 2 of Zeman et al. [
Environ Health Perspect 110:817–822 (2002)]: the y-axis should have been labeled “Nitrite” instead of “Nitrate.” The corrected figure appears below.
In Giusi et al. [
Environ Health Perspect 113:1522–1529 (2005)], the colors were incorrect in the key to Figure 1. The corrected figure appears below.
EHP regrets the errors.
|
0
|
PMC1314937
|
CC0
|
2021-01-04 23:41:30
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A807
|
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-a0080816330336EnvironewsForumSubstance Abuse: Resurgence of Teen Inhalant Use Chepesiuk Ron 12 2005 113 12 A808 A808 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 2004 Monitoring the Future (MTF) survey showed that inhalant use (“huffing”) is rising among American teenage students, particularly 8th graders. The results, released in December 2004, showed that 9.6% of 8th graders used inhalants in 2004, up from 7.7% in 2002 and 8.7% in 2003. Inhalant use was also up slightly among 10th and 12th graders in 2004. Findings from the latest MTF will be released in late December 2005, and researchers are anxious to see if the trend holds.
“These increases are disturbing because they come after a long period of decline in inhalant use by students in all three grades,” says Lloyd D. Johnston, a professor at the University of Michigan Institute for Social Research and principal investigator of the MTF since it began in 1975. “We are concerned that the use of this class of drugs may be about to rebound.”
Each year, the MTF, which is funded under grants from the National Institute on Drug Abuse (NIDA), asks approximately 50,000 8th-, 10th-, and 12th-grade students in some 400 schools nationwide about their use of drugs, alcohol, and cigarettes. The data gathered are used to help government officials and policy makers identify potential drug problem areas so they can target resources to deal with them.
“We know that inhalant use starts early and that long-term abusers are among the most difficult drug abuse patients to treat,” says NIDA director Nora Volkow. “It is critical that research efforts to characterize the behavioral effects of inhalants intensify, so that more effective preventions, interventions, and treatments can be developed.” This year, NIDA announced the continuation of a broad-based research initiative begun in 2002 to address the epidemiologic, social, behavioral, cognitive, and neurobiological consequences of inhalant abuse, as well as treatment and prevention.
More than 1,000 readily available products are used as inhalants, and they can potentially kill, according to the Office of National Drug Control Policy (ONDCP). Such products include glue, shoe polish, gasoline, lighter fluid, and the propellants in spray deodorant, hair sprays, and canned whipped cream.
The ONDCP further reports that glue, shoe polish, and toluene-containing products were the most commonly abused inhalants among users aged 12 to 17. According to the American Association of Poison Control Centers, gasoline accounted for the greatest percentage (44%) of reported inhalant deaths between 1996 and 2001, followed by air fresheners (26%) and propane/butane (11%). Other health effects of inhalant use include headache, nausea, vomiting, slurred speech, loss of motor coordination, and wheezing.
The physical and social environment both play a key role in inhalant use, says Harvey Weiss, executive director of the National Inhalant Prevention Coalition. Treatment sometimes requires removing the abuser from the environment in which he or she is abusing inhalants. “We should not view inhalant abuse [simply] as a substance abuse problem,” Weiss says. “It’s a public health problem, so we need to do more public health outreach to young people.”
Sources believe that education is the key to preventing inhalant use from becoming a dangerous fad. When MTF data from the mid-1990s began showing a long-term gradual increase in inhalant use, the Partnership for a Drug-Free America and NIDA mounted an aggressive media campaign about the dangers of inhalants. The next round of MTF data showed a decline in inhalant use and a concurrent increase in young people viewing inhalants’ use as risky, but use began climbing again after the media campaign ended.
“Of course, the evidence is circumstantial, but we’ve seen the same thing happen for so many other drugs,” Johnston says. “A drug can have a resurgence in use among young people because of what I call ‘generation forgetting’—that is, a new generation of young people comes along that hasn’t heard too much about a drug, so it is naïve about the consequences of its use. That begins to change when a public education campaign is launched.”
Despite the MTF findings, the U.S. government hasn’t yet documented a trend indicating a rise in inhalant use among teenagers, says Terry Zobeck, deputy associate director for policy and budget at the ONDCP. But he adds, “The MTF is respected and well documented. We will be quite concerned if its next survey shows that inhalant use is up for the third year in a row.”
Huffing is up.
A new survey shows the practice of inhaling toxic—often deadly—substances is increasing among American teens.
|
16330336
|
PMC1314938
|
CC0
|
2021-01-04 23:41:30
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A808
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.113-a808
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a00811EnvironewsForumEHPnet: National Eye Institute Dooley Erin E. 12 2005 113 12 A811 A811 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 Eye Institute (NEI) is the primary institute of the NIH for supporting and conducting research on preventing, diagnosing, and treating eye diseases and other vision disorders. Currently the NEI oversees approximately 1,600 research projects at more than 250 institutions in addition to the research ongoing at its own facilities in Bethesda, Maryland. The institute also works to translate research findings into clinical applications and to raise public awareness about eye and vision problems. The NEI uses its website, located at http://www.nei.nih.gov/, to help disseminate information about its many programs.
The What’s New section provides links to newly released NEI-funded research and other topics of interest to those in the field. The more in-depth News and Events section includes press releases, clinical alerts for professionals, information on meetings and special events, and a list of official statements and reports on vision.
The Health Information page links to information on 21 eye diseases and disorders. There is also a section on basic eye anatomy with diagrams of the eye, links to glossaries of eye terminology, a collection of eye care resources, NEI information provided in Spanish, and a way to order NEI materials online. The collection of eye care resources consists of an eye health organizations database; a page of frequently asked questions about clinical trials and how they are conducted; and tips on finding an eye care professional, procuring financial assistance for eye care, and talking to doctors about eye health.
More information on NEI clinical trials is available on the Research Funding page and through the Clinical Studies Database. The Research Funding page has information on grant and funding opportunities for researchers, news on staff appointments, updates on grants and funding policies, and overviews of councils and workshops of interest to NEI researchers, among other resources. The Clinical Studies Database provides a list of all ongoing and completed NEI-supported studies. This section also includes study results and lists of journal articles that have been generated by the research, as well as a list of NEI studies that are currently enrolling participants. Site visitors can search the database for studies under six topic areas or by keyword, study location, age of study participants, patient recruitment status, or study status.
The Education Programs page offers overviews of NEI outreach activities. Through the National Eye Health Education Program, the NEI conducts large-scale public and professional educational activities in partnership with national organizations. Specialty initiatives within this program focus on diabetic eye disease, glaucoma, low vision (when everyday tasks become difficult to do even with corrective lenses, medicine, or surgery), and educating Spanish-speaking Americans about eye and vision problems. VISION is a teaching supplement for grades 4 through 8 that is available for download at no charge. This 16-page guide helps teachers plan lessons about how the eye works, eye problems, and eye safety. The supplement was developed in cooperation with the Association for Research in Vision and Ophthalmology. THE EYE SITE is an NEI-sponsored exhibit that travels to shopping malls around the United States to educate the public about low vision, vision rehabilitation services, and vision adaptive devices, as well as about the NEI itself. The exhibit features five colorful kiosks and an interactive multimedia program. Another exhibit, VISION, educates visitors to science museums about how vision works and about how researchers are working to develop ways to protect our eyes from disease and developmental problems.
|
0
|
PMC1314939
|
CC0
|
2021-01-04 23:41:30
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A811
|
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-a0081216330337EnvironewsNIEHS NewsAbdiction/Addiction Connection Hood Ernie 12 2005 113 12 A812 A814 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
Sometimes in science, as in politics, connections arise that may at first glance appear to be strange bedfellows. That might be the natural first impression of a potential association between chemical intolerance and addiction. But although the conditions are manifested by behaviors that appear to be polar opposites—substance avoidance (or abdiction, as some are beginning to call it) by the chemically intolerant, and compulsive substance use by the addicted—there is evidence to suggest that, biologically, they may actually have much in common.
That was the concept behind “Addiction and Chemical Intolerance: A Shared Etiology?” This conference, held 19–20 September 2005 in Research Triangle Park, North Carolina, was the first scientific meeting to be cosponsored by the NIEHS and the National Institute on Alcohol Abuse and Alcoholism (NIAAA). It was also the first time researchers from the fields of environmental health and addiction convened to explore common ground and potential collaborations.
“The idea of hosting a conference on chemical intolerance and addiction stems from a long history of individual physicians’ reporting observations on patients that looked like addiction to chemicals, foods, caffeine, or alcoholic beverages,” explained conference chair Claudia Miller, a professor and researcher in environmental medicine at The University of Texas Health Science Center at San Antonio. “There is a striking resemblance between the symptoms and responses to substances reported by chemically intolerant patients and individuals addicted to drugs or alcohol.”
Firm numbers on addiction and chemical intolerance are hard to come by, in part because both conditions often go undiagnosed. Approximately 67% of all Americans drink alcohol, yet 90% of the alcohol is consumed by only 30% of the population, said NIAAA director Ting-Kai Li in his keynote address. In the latter half of 2003 (the most recent year for which figures are available), there were 627,923 drug-related emergency room visits in the United States, according to the Drug Abuse Warning Network of the U.S. Substance Abuse and Mental Health Services Administration. As for chemical intolerance, epidemiologic figures compiled and reported at the meeting by William Meggs, a professor of emergency medicine at East Carolina University, suggest the prevalence of the condition (self-reported) to be approximately 12% of the U.S. population, with approximately 4% self-reporting as “seriously affected.”
Miller contends that addiction and chemical intolerance represent divergent physiologic responses to a shared underlying disease mechanism she calls toxicant-induced loss of tolerance (TILT). In TILT, a chemical exposure—either acute or chronic and low-level—initiates sensitization to even small amounts of structurally diverse chemicals found in foods, drugs, alcoholic and caffeinated beverages, pesticides, mold toxins and other elements of indoor air, implanted devices, solvents, cleaning chemicals, and more. Thereafter, when affected individuals are exposed to everyday “triggering” substances such as foods, traffic exhaust, or fragrances, they report multisystem symptoms including headache, nausea, difficulty breathing, muscle spasms, and rashes. The fact that different people exhibit different constellations of symptoms has made it difficult to conduct epidemiologic studies or arrive at a case definition, Miller says. In the past, these difficulties have led some observers to speculate that chemical intolerance is psychogenic in origin.
As she outlined in her presentation to the approximately 120 attendees, Miller postulates that the TILT mechanism can lead to either abdiction or addiction, with both behaviors intended to avoid unpleasant withdrawal symptoms. She further proposes that TILT may underlie a wide variety of chronic diseases that are increasing in prevalence worldwide, such as asthma, autism, chronic fatigue syndrome, fibromyalgia, and depression. (She described these proposals in depth in an article in the January 2001 issue of Addiction.)
Parallel Paths
Whether or not chemical intolerance and addiction are flip sides of the same coin, it is clear that researchers in the two fields have much to learn from each other. Li said, “Some people become alcohol-dependent and then they recover because the environmental risks have been removed; there’s a gene–environment interaction. I think it’s true also for chemical intolerances. It’s an environmentally induced condition, and when you remove the environmental risk, the person may still be genetically high-risk, but without the environmental component they can then recover.”
As things stand today, however, there are no easy answers for the chemically intolerant. Environmental epidemiologist Howard Hu daily perceives the need for more research in his role as a clinician at the Harvard School of Public Health. “Our environmental medicine clinic has several hundred patients who have this disorder, and we have not made any progress in ways to evaluate and manage them that has led to any sustainable improvements in their condition,” he said. “So we really appreciate the need for good research that will shed light on the biology of the disorder and allow us to devise methods to manage and treat it.”
Hu felt that the conference was a good step forward in helping to define a research agenda. “Some of the approaches to chemical addiction and alcoholism [research] have provided a roadmap of where the chemical intolerance research needs to go, in terms of understanding genetic susceptibility and the molecular changes that might be the mechanism of how the intolerance phenotype develops,” he said. One role model for progress described by Miller might be the Japanese government, which has established several environmentally controlled medical units (EMUs) in hospitals for the research, diagnosis, and treatment of chemical intolerance. To date, there is no comparable facility in the United States.
One speaker called attention to “tantalizing morsels” of convergence that have emerged between chemical intolerance and addiction. For example, it appears all but certain that genetic susceptibility plays an important role in both conditions, and one of the most compelling ideas to emerge was the possibility that susceptibility to both conditions may arise from polymorphisms in the same genomic neighborhood—genes including CYP2D6, PON1, and others that are known to regulate the metabolism of exogenous agents such as drugs and pesticides. PON1 is involved in the detoxification of organophosphate pesticides; CYP2D6 functions in the metabolism of structurally diverse substances that affect the central nervous system, including various classes of antidepressants, amphetamines, codeine, and neurotoxicants. The question of whether variant alleles of these genes give rise to the abdiction and addiction phenotypes is a primary target for investigation in the future.
There is a striking resemblance between the symptoms and responses to substances reported by chemically intolerant patients and individuals addicted to drugs or alcohol.
–Claudia Miller
The University of Texas Health Science Center at San Antonio
For now, case–control study results presented by researchers Cornelia Baines and Gail McKeown-Eyssen of the University of Toronto (which were published in the October 2004 International Journal of Epidemiology) clearly show an elevated risk for chemical intolerance associated with variations in the enzymatic metabolism genes CYP2D6, PON1, and NAT2. A gene–gene interaction detected between CYP2D6 and NAT2 suggested that rapid metabolism alleles in both genes may confer as much as an 18-fold elevated risk for chemical intolerance. These findings point toward a biologic basis for the condition.
Brain imaging studies presented at the conference by Hu, Marc Potenza of the Yale University School of Medicine, and Leonid Bunegin of The University of Texas Health Science Center at San Antonio showed striking similarities between chemically intolerant patients and addicted individuals in terms of the neural regions involved and the types of activation detected. Many signs point to the mesolimbic system, where the activity of neurotransmitters such as dopamine is regulated. Among individuals who are genetically susceptible to either chemical intolerance or addiction, the homeostasis of the brain’s reward system may be upset or perhaps changed permanently by exposures to certain drugs or chemicals. Thus, although the outcomes of addiction and abdiction may be polar opposites, the underlying causes and mechanisms may prove to be very similar.
You Say Tomato . . .
Differences in nomenclature often pose a challenge and require reconciliation when two fields begin to work together. As one conference presenter waggishly put it, “Scientists would rather use each other’s toothbrushes than use each other’s terminology.”
Perhaps the best example of varying terminology arose as speakers from both fields presented some of the leading hypotheses in each field. In chemical intolerance, researchers refer to “initiation” (the exposure that leads to the development of intolerance) and “triggering” (subsequent exposures resulting in symptoms); in addiction research, scientists refer to neurologic “sensitization” to a substance leading to “amplification” of its effects.
According to Miller, future research may show that neurologic sensitization also explains initiation and triggering. “Perhaps the processes [underlying addiction and chemical intolerance] are one and the same, but we don’t know that quite yet,” she says. “Eventually, once the biology has been worked out, the terminology may reconcile, clarifying the links between the two fields. It was one of the most striking parallels to emerge from the meeting.”
Another impediment discussed during the proceedings is the longstanding struggle to precisely define phenotypes of chemical intolerance for research purposes. Single-minded focus on this difficulty in the past has been the excuse for doing no research, said Miller, who added that facilities like Japan’s EMUs could be used to assess individual responses in the absence of any consensus on case definitions or phenotypes. “Just as there is no single case definition or phenotype that encompasses all forms of drug and alcohol addiction, there is no single case definition that can be applied to all forms of abdiction, because we are dealing with a general mechanism for new classes of diseases that have varied manifestations,” she explained.
Establishing a chemical intolerance phenotype or case definition is further complicated by a phenomenon called “masking.” Underlying chemical or food triggers may be masked by overlapping symptoms resulting from simultaneous or sequential exposures to other foods or chemicals, from addiction to caffeine, alcohol, or tobacco, and from varying degrees of habituation to triggering substances. For example, Miller wrote in her Addiction paper, “[i]f an individual is sensitive to many different substances, then the effects of everyday exposures to chemicals, foods, or drugs may overlap, producing a confusing array of symptoms. The individual would feel sick most of the time, but the effect of any single exposure would not be apparent to either the individual or his physicians.” Masking therefore confounds diagnosis and treatment because clinicians tend to address patients’ overt symptoms without discovering the underlying intolerances, much less the initiating exposures that led to illness in the first place.
The lack of phenotypes may also hamper the application of systems biology to the study of chemical intolerance. Systems biology integrates tools from genomics, proteomics, metabolomics, and informatics to detect and validate novel biomarkers of disease. “Without a phenotype, it’s difficult to move to the next level,” said William Slikker, Jr., deputy center director for research at the National Center for Toxicological Research. First, he suggested, we still need to define phenotypes in a way in which they can be systematically examined. “Once that is done,” he said, “then I can see setting hypotheses that can be tested using the systems biology approach.”
At the same time, the availability of a research EMU—the equivalent of a detox unit for alcohol or drug withdrawal—would provide a unique tool for examining individuals’ genetic and protein expression before and after removal of chemical and food triggers and before and after specific challenges, said Miller. “Just as systems biology will enable researchers to understand individual responses to complex environments, the EMU is a tool that [would allow] us to identify the responses of individuals to a wide variety of exposures,” she said.
Miller said the approaches are completely compatible and complementary. “A clear advantage of the EMU is that it can be done now—well before sophisticated genomic and proteomic approaches become widely available—and begin to benefit patients with a wide variety of environmentally induced illnesses.”
The Road Ahead
NIEHS deputy director Samuel Wilson, who opened the meeting, agreed that the future of the field depends largely on researchers’ ability to carefully identify researchable questions. “It’s going to be up to the scientists writing the proposals or bringing the problems forward to figure out experimental themes or researchable problems that they can make a case for, and then work up and make solid discoveries on,” he said. “There’s no substitute for having quantitative traits to look at—quantitative biochemical markers or biomarkers that can be related with exposure and with these very complex behavioral phenotypes.”
Wilson added that only when the molecular science embedded in the pathophysiology and biology of chemical intolerance and addiction is uncovered will the extent of overlap between the two conditions be established.
Several attendees expressed great interest in pursuing collaborative projects with colleagues from the other field, and many were optimistic that the conference would ultimately result in cross-institute initiatives between the NIEHS and NIAAA. For environmental health researchers, addiction has long been a blind spot; in addiction research, the same is true for environmental exposures. With greater interactions between the two fields, both may achieve a clearer view of these conditions and the road to health.
|
16330337
|
PMC1314940
|
CC0
|
2021-01-04 23:41:30
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A812-A814
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.113-a812
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0081416353307EnvironewsNIEHS NewsBeyond the Bench: Nurses Adapt to Changing Health Care Climate Tillett Tanya 12 2005 113 12 A814 A815 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
With increased emphasis being placed on the importance of environmental health comes the need for an expanded variety of environmental health care practitioners. Nursing is one of several professions that are augmenting and advancing the capabilities of their practitioners to meet this need. Now the Community Outreach and Education Program (COEP) of the University of New Mexico’s Center for Environmental Health Sciences has partnered with the New Mexico Environment Department to create an environmental health nursing internship as a component of its outreach program.
In cooperation with the university College of Nursing, the partners recruit nursing students who are interested in implementing environmental health care initiatives in the surrounding community. For four years, the program has taken nursing students beyond the traditional curriculum and shown them firsthand how interaction with environmental factors affects human health. The program also gives students valuable experience in taking measures to abate hazardous exposures in communities.
The interns are currently working on a number of projects that will affect different community members. They are helping to develop surveys and compiling data for a project investigating uranium exposure and subsequent kidney damage among the Navajo Nation, whose members live near and work in uranium mines. They are gathering and assembling materials for a community education and survey project on mercury in surface waters and other environmental health concerns among the Cheyenne River Sioux Tribe. They are helping COEP staff write and field-test an integrated environmental health curriculum on diabetes for middle school students. And in a fourth project, they are helping to develop a best-practices manual for applying farm waste fertilizer to croplands in a way that minimizes human exposure to aerosolized waste.
The local ties of some of the nursing interns have enhanced the program’s role as an effective community advocate. Intern Krystyn Yepa is a Native American from the Pueblo of Jemez who became interested in environmental health nursing because she wanted to understand some of the health effects in her tribe resulting from exposure to different pollutants in the environment. “Learning about the different environmental health problems that exist in New Mexico has given me the willpower to finish nursing school to ultimately achieve my goal of improving the health and lifestyles of my people by placing importance on the environment,” she says.
Yepa explains further, “As a Native American, I was raised to respect and appreciate the environment, and working at the COEP has only strengthened my values related to the environment.” She also credits the internship with teaching her important assessment tools that lay emphasis on the environment when completing a health history on patients, something a traditional nursing internship would not likely provide.
“The nursing interns have been an incredible asset to our COEP,” says staff member Stefani Hines. “It is a mutually beneficial situation on many levels—the student nurses gather valuable real-world experiences in environmental health; the nursing school has access to additional, unique placements for their students; we have additional support for projects, which helps make them better; and the communities we work with benefit from the students’ efforts as well.”
Dedicated, enthusiastic environmental health nursing interns will only continue to play an important role in advancing the COEP mission. One project currently in development will help community members get involved in city and county zoning processes, which will both encourage a healthy community mindset and minimize exposures to pollutants.
Effecting change in Southwest communities.
Krystyn Yepa (right) is one of several nursing interns working with Navajo and Sioux communities to reduce exposures and increase knowledge through a program of the University of New Mexico Center for Environmental Health Sciences.
|
16353307
|
PMC1314941
|
CC0
|
2021-01-04 23:41:31
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A814-A815
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.113-a814
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a00815EnvironewsNIEHS NewsHeadliners: Reproduction: Oocyte Generation in Adult Mice Phelps Jerry 12 2005 113 12 A815 A815 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
Johnson J, Bagley J, Skaznik-Wikiel M, Lee H-J, Adams GB, Niikura Y, Tschudy KS, Tilly TC, Cortes ML, Forkert R, Spitzer T, Iacomini J, Scadden DT, Tilly JL. 2005. Oocyte generation in adult mammalian ovaries by putative germ cells in bone marrow and peripheral blood. Cell 122:303–315.
The theory that female mammals are born with a finite number of germ cells (oocytes) has been accepted as an unquestionable truth for over 50 years. Recent research has challenged this accepted dogma by showing that mice and flies can produce oocytes and follicles during puberty and adulthood. Now NIEHS grantee Jonathan L. Tilly and colleagues at the Harvard Medical School have shown that adult mice can produce large numbers of new oocytes in a short period of time, providing additional evidence to challenge the accepted belief of a fixed complement of oocytes at birth. The Harvard researchers also discovered a source of germline stem cells in the bone marrow.
Oocytes are found in the ovaries surrounded by somatic cells in structures known as follicles. Only a small fraction of follicles actually reach ovulation, producing an egg capable of being fertilized. Conventional wisdom hold that in humans, only about 30,000 of an original pool of about 1 million oocytes present at birth are still present at puberty, and this number is thought to gradually decline throughout adulthood until the complete loss of oocytes at around age 50 stimulates menopause. Acceptance of the concept that adult mammals can continue to produce oocytes has been slow, likely due to the lack of direct evidence of the existence of mammalian female germline stem cells.
The Harvard team conducted gene expression analysis and bone marrow transplantation studies on mice that had been sterilized through chemotherapy. Within 24 hours of treatment, follicles were regrowing in the animals’ ovaries. By 2 months after treatment, there was no difference between the treated animals and controls. In other studies, mice whose bone marrow was destroyed with chemotherapeutic agents were injected with peripheral blood from transgenic animals with germline cells expressing green fluorescent protein. Oocytes found in the test animals’ ovaries within 30 hours of treatment also expressed the fluorescent protein.
The researchers have not yet determined whether oocytes derived from germline stem cells can undergo fertilization and subsequently develop into viable offspring. However, the results do prove that bone marrow and peripheral blood are sources of germline stem cells and can sustain oocyte production into adulthood. If adult oocyte production is also possible in humans, it could have major implications for the treatment of infertility and other disorders such as osteoporosis, although much additional research is needed before this potential can be realized.
|
0
|
PMC1314942
|
CC0
|
2021-01-04 23:41:29
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A815
|
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-a0081616330338EnvironewsFocusCritical Care: Applying Genomics to Inflammation Outcomes Schmidt Charles W. 12 2005 113 12 A816 A821 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
What do gunshot wounds, burns, heart attacks, arthritis, asthma, and cancer all share in common? Apart from inflicting misery, these conditions—and others too—involve inflammation, an immune response to injury and infection that normally protects, but sometimes endangers or kills patients. Caused by immune cells accumulating at a site of injury, inflammation typically guards against infection and speeds recovery; it is a critical process and, per se, does not cause disease. But unchecked inflammation that spreads or fails to subside poses chronic and acute health risks for millions of people. Asthma patients, for instance, can’t breathe because inflammatory compounds cause airway linings to swell and mucus to spread in the lungs. Inflammation also exacerbates cancer, scientists believe, by facilitating the proliferation of abnormal cells. An acute condition called sepsis—caused when infection or inflammation spills into the bloodstream—produces organ failure and shock in critically ill patients. Up to 215,000 Americans die from sepsis every year, according to the National Institute of General Medical Sciences. Worldwide, sepsis is estimated to kill 1,400 people each day, according to a consensus document published in the June 1992 issue of Chest.
In light of its implications, inflammation has become one of the hottest areas in biomedical research. J. Perren Cobb, a professor of surgery and genetics at Washington University in St. Louis, says a wide array of medical specialties stand to benefit from these investigations. “Inflammation is a major unifying syndrome, the investigation of which provides opportunities for multidisciplinary convergence,” he explains. “Studies of inflammation cut across all the domains at the NIH; it’s a fundamental process in human biology that ties everything together.”
Growing evidence suggests that genetic factors drive key aspects of an individual’s inflammatory outcome. Scientists studying inflammation are trying to identify the genes that drive inflammation as well as biomarkers from throughout the course of inflammation. Stephen Chanock, who heads the Section on Genomic Variation in the Pediatric Oncology Branch at the National Cancer Institute, emphasizes that the current critical care orientation of this research has broad multidisciplinary implications that extend to environmental health. “Injuries represent the ultimate gene–environment interactions,” he explains. “Usually environmental health focuses on chronic exposures, but in this case we’re studying environmental insults that are more dangerous and intense. So, the ‘environment’ in environmental health isn’t just about pollution, it’s also experiential. We’re developing practical methods for looking at inflammation that will ultimately be applied to larger public health issues.”
Toward Better Knowledge of Inflammation
Today, genomics defines the cutting edge of inflammation research. Genomic studies, in addition to their proteomic and metabolomic cousins, aim to resolve an age-old mystery: namely, why some patients recover readily from inflammation while others suffer and die from it. The current research emphasis focuses on critical care, particularly of trauma and burn patients, who face the lethal dangers of septic complications. Ideally, new gene-based discoveries will provide diagnostic biomarkers to predict who among these patients will react poorly to inflammation and why. If doctors could reliably predict this outcome in advance, they might tailor antibiotics and other treatment options to a patient’s own inflammatory system, potentially saving lives.
Better knowledge of inflammation biology could also spawn new treatment options, Cobb says. The newest drug for sepsis—an Eli Lilly and Company product called Xigris that came on the market in 2001—helps some patients, but its cost is exorbitant: nearly $7,000 per course of treatment. What’s more, the drug reduces the risk of death by just 6% and can produce side effects such as excessive bleeding.
Among the numerous programs moving inflammation research forward is an effort funded by a National Institute of General Medical Sciences “glue grant,” so named because it “glues together” multi-disciplinary efforts to tackle biomedical questions beyond the means of any one research group. This program, called Inflammation and the Host Response to Injury, strives to determine why patients can have dramatically different outcomes after traumatic injuries and burns. Headed by Ronald Tompkins, a professor of surgery at Harvard Medical School and chief of Massachusetts General Hospital’s Burn Service, the program uses genomic and proteomic methods to study inflammation at 22 clinical centers located throughout the country. A total of $37 million was made available for the program’s first five years.
When the Inflammation and Host Response to Injury program was launched in 2001, its leaders decided to create a broad research infrastructure with uniform protocols as a first priority. “One of our first challenges was to develop guidelines, not just for the sample collection and analysis, but also for patient management,” says Lyle Moldawer, a glue grant recipient and professor of surgery at the University of Florida College of Medicine. “We recognized that all the funded centers have different protocols for the immediate care of trauma and burn patients, and we were concerned that those differences in early management might contribute to gene expression changes.”
Tompkins says creating a uniform infrastructure for the program was like building a highway. “We needed the gas stations, the on-ramps, the off-ramps,” he says. “No one had ever tried to introduce this technology into critical care medicine before.” With standard operating procedures in place and the program now in its fourth year, scientists have begun to address a subsequent challenge: extracting useful knowledge from the reams of genomic data flowing out of the program’s 22 clinics.
At the same time the glue grant program was gearing up, Cobb, senior investigator Anthony Suffredini of the NIH Critical Care Medicine Department, and Robert Danner, who heads the Infectious Diseases Section in the same department, created the Consortium for Expression Profile Studies in Sepsis specifically to identify the needs of those applying genomic methods to critical care. The consortium hosted four meetings throughout the country before evolving into the NIH Functional Genomics of Critical Illness and Injury Symposia series, which now provides a forum where glue grant recipients and others discuss research progress and results. The most recent symposium, hosted by the NIH at its Bethesda campus on 21–22 April 2005, was attended by scientists from 10 countries, all seeking to advance genomics in inflammation research.
An Inflammation Primer
Once triggered, inflammation proceeds similarly whether caused by pollutants, pathogens, trauma, radiation, or burns. Localized mast cells in affected tissues produce histamine, a chemical mediator that dilates blood vessels at the site of injury, producing redness and heat. Histamine also renders blood vessels permeable, so leukocytes (white blood cells) can reach the injury. Leukocytes are attracted to the injury site by chemotactic proteins known as chemokines, which are secreted by endothelial cells of the blood vessels.
Leukocytes originate in bone marrow and include diverse cell types, such as neutrophils, eosinophils, basophils, monocytes, lymphocytes, and macrophages. Neutrophils arrive at the affected area first. These remarkable cells roam the body and kill pathogens on demand with a toxic blend of free radicals and protein-chewing enzymes that destroy bacterial cell walls. Monocytes engulf cellular debris and mature into macrophages, which are larger leukocytes that consume entire bacteria. These cells also secrete a variety of cytokines that recruit and activate other cell types. Lymphocytes are divided in two broad classes—B cells and T cells—each with different roles. B cells, once activated, make antibodies that attack foreign substances, while T cells kill infected cells directly.
Chemical mediators released by leukocytes during inflammation come in many varieties. Cytokines, for instance, help to regulate inflammation, whereas inter-leukins regulate T cell activity and produce systemic effects such as fever.
Normally, the whole inflammation process is self-limited and short-lived; leukocytes disperse after dispensing with infectious agents, and inflammation dies down within hours or days. Problems crop up when the response persists or spreads systemically, damaging and killing normal tissues in the process. Chronic inflammation can persist for years, causing illnesses that end with the suffix “-itis,” such as bronchitis, arthritis, and bursitis. Systemic inflammation—sepsis being one variety—occurs when cytokines reach the bloodstream and spread through the body, damaging organs far from the initial injury’s source.
Candidate Genes
No one knows precisely what happens when inflammation goes awry. Years of immunology research have implicated hundreds of genes in abnormal inflammation, but the evidence linking them to particular outcomes is weak. Of these genes, the one coding for C-reactive protein (CRP), an acute-phase molecule whose levels shoot up during systemic inflammation, is perhaps the best known. High CRP levels are prognosticators for heart disease and stroke (which are both linked to inflammation), but its role in these conditions remains unclear. Another well-known gene—tumor necrosis factor–alpha (TNF-α)—codes for a pro-inflammatory cytokine that normally regulates leukocyte and endothelial cell activity, in addition to other functions.
By the 1990s, however, candidate gene studies had yet to produce clinical benefits for inflammation. Suffredini says scientists at the time were extremely frustrated with the lack of progress. “People were throwing up their hands and feeling [painted] into corners,” he says.
A turning point emerged at the turn of the millennium, when a rough draft of the human genome and the advent of microarrays made it possible to assess the expression of thousands of genes simultaneously. “The analogy is that for years, we’d been working on the ground to see how candidate genes interact,” Cobb explains. “But microarrays allowed us to look down at the genome from twenty thousand feet, so to speak, and that has enabled us to model much broader interactions.”
With these tools, scientists could search for entirely new genes and molecular pathways involved in disease processes. Cancer researchers were among the first to exploit the technology for clinical aims, Suffredini says, inspiring their counterparts in critical care to do the same. Thus, inflammation research entered a new phase of gene discovery that drives much of the progress in the field today. Scientists are now investigating a variation in the promoter region of TNF-α(the region that initiates protein production after binding transcription factors) that might contribute to sepsis.
Injuries represent the ultimate gene–environment interactions. Usually environmental health focuses on chronic exposures, but in this case we’re studying environmental insults that are more dangerous and intense.
–Stephen Chanock National Cancer Institute
While cancer genomics inspired similar efforts in critical care, both specialties operate under vastly different research settings. For one thing, cancer patients typically have the time and awareness to provide informed consent for blood and tissue sampling. In addition, the cohorts tend to be large and matched for age, sex, treatment history, and other parameters that can influence genomic profiles. Trauma and burn patients, on the other hand, are rushed—often unconscious—into the emergency room or intensive care unit, where live-saving treatment is the first priority. In this frenetic environment, informed consent is difficult to secure, and research sampling becomes a secondary concern.
Moreover, cancer and trauma induce totally different types of gene expression—whereas tumors typically produce localized, stable expression profiles corresponding to small portions of the genome, critical injuries trigger enormous genomic changes that affect all tissues and shift rapidly over time. Temporal factors are extremely important in critical care sampling because they have a tremendous influence on the gene profile; a sample taken 15 minutes after injury will be vastly different than one taken several hours later.
Into the Data
According to Tompkins, investigators with the glue grant program chose to investigate normal and abnormal inflammation trajectories sequentially, each in five-year increments. Genomic and proteomic data for the normal trajectory—compiled using samples from trauma and burn patients who recovered uneventfully—are now being analyzed.
At the same time, program scientists augmented the clinical research with additional genomewide expression studies of leukocytes sampled from healthy volunteers dosed intravenously with bacterial endotoxin. These studies—which induced low-level systemic inflammation that permitted validation of sample processing protocols—enabled scientists to compare baseline and inflammatory genomic changes at varying time points. Patients weren’t harmed by the experiments, and all responses returned to normal within 24 hours.
The results, published in the 31 August 2005 issue of Nature, showed how complex inflammatory networks really are—between 3,000 and 5,000 genes, up to 20% of the entire genome, were activated, according to Moldawer, one of the study’s authors. “The research revealed that the magnitude of the changes was much larger than we anticipated,” he says. “We expected to see up-regulation of stress-related genes during the acute phase, but much to our surprise, the diversity of the changes was much greater than we thought it would be.”
Many of those changes, Moldawer adds, were seen in genes involved in mitochondrial energy transfer, protein synthesis, and antigen recognition—in short, biological processes that enable leukocytes to become more efficient antimicrobial agents, he says. Preliminary analyses suggested the magnitude and nature of the endotoxin response shared some similarities with the response seen in real patients. At press time, the clinical data from actual patient cohorts were still being assessed.
Although the amounts of genomic data may be computationally daunting, recent evidence from another study suggests efforts to distinguish good inflammatory outcomes from bad might have promise. This study, published in the 29 March 2005 Proceedings of the National Academy of Sciences, made several key discoveries. First, hospitalization and repeated sampling had only a modest effect on gene expression in healthy volunteers. Thus, the experience of being hospitalized (with its enforced bed rest and defined nutritional intake) is unlikely to influence gene expression in ways that undermine the detection of signature profiles for specific inflammatory outcomes. Second, the researchers showed that gene expression differences in whole-blood leukocytes drawn from severe trauma patients could be divided into injury-specific patterns. Taken together, says coauthor Tompkins, the findings indicate that expression profiling may yield “low-hanging fruit” in the form of highly correlated data.
Linking Sepsis-Related Genes to Biology
Meanwhile, researchers in Germany have shown that subsets of genes can be linked directly to sepsis. Among these researchers is Trinad Chakraborty, who directs the Institute of Medical Microbiology at Justus-Liebig University. Chakraborty is completing a study of genomic factors contributing to sepsis in patients with multiple trauma or pneumonia. The study—part of a broader effort to understand why patient outcomes differ after similar injuries and illnesses—involved screening up to 20,000 genes in peripheral blood during a 14-day post-injury period. The effort, conducted in 185 patients, found 690 genes whose expression appears to correlate with sepsis. In future research, Chakraborty plans to look for single-nucleotide polymorphisms within candidate genes that predispose the sepsis phenotype, and to identify protein-based biomarkers for diagnostic use.
But Chakraborty adds that computational challenges are a serious holdup. “When we started the research, getting the microarrays to be sufficiently robust was the bottleneck,” he says. “Now we’ve resolved that problem, and bioinformatics is the bottleneck.” He and his colleagues hope to trim the 690 genes to a lesser population of 25 or so. “Then we could develop an algorithm that recognizes a profile within that smaller set of genes to indicate whether you have a likelihood of sepsis or not.”
Inflammation is a major unifying syndrome, the investigation of which provides opportunities for multidisciplinary convergence. . . .
It’s a fundamental process in human biology that ties everything together.
–J. Perren Cobb Washington University in St. Louis
U.S. scientists have also correlated genes with sepsis and used these findings to suggest a preliminary mechanism for its lethality. Led by Hector Wong, who directs the Division of Critical Care Medicine at Cincinnati Children’s Hospital Medical Center, the scientists used microarrays to compare gene profiles between children who survived sepsis and those who died from it. Children respond uniquely to sepsis in that their fatality rates are much lower than those of adults—roughly 10% compared to 30% among the latter, says Chanock.
Wong suspects that children respond better to sepsis in part because they have fewer comorbidities such as diabetes and heart disease (a status that is changing somewhat with rising childhood obesity). But he further suspects genetic factors underlie important biological differences that improve their outcomes, though at this point he can’t say how.
In recent studies presented at the April symposium, Wong found that among non-surviving children, six genes coding for metallothionein—a protein that binds zinc and removes it from the bloodstream—appeared to be highly expressed. These findings led him to a hypothesis: if severely septic children had high blood metallothionein levels, he proposed, then their blood zinc levels might be correspondingly low. “And in fact, that turned out to be true,” he says.
Another interesting finding was that the profiles showed altered expression patterns for a host of proteins that either depend on zinc or take part in zinc homeostasis. “So there’s a lot of biology there to look at,” Wong says. “We don’t know how or whether zinc is involved; there’s very little information out there about the effects of acute zinc deficiency. I find it hard to believe the foundation for sepsis is zinc, but . . . I think it can be tested.” After considering this position further, Wong adds that this is how high-throughput investigations are useful: they suggest biological mechanisms that scientists can explore further in the laboratory.
Future Needs
Today, a genomic research culture is slowly seeping into the front lines of care for the critically ill and injured. But establishing that culture isn’t easy—emergency room and intensive care unit settings challenge researchers in many ways. Issues like informed consent for study participation and repeated intrusive blood sampling to assess temporal changes in the genome are difficult to manage, Tompkins says. Ideally, new technologies will reduce sample volume requirements, lessen the amount of time required for microarray analysis (which now averages 24 hours), and reduce microarray costs to the extent that they can be used routinely in the clinic.
We need to do a better job of educating people about the importance of this process.
–J. Perren Cobb Washington University in St. Louis
Chakraborty adds that microarray platforms need to accommodate sample degradation too. As it stands now, he says, RNA in blood samples drawn in the emergency room has a higher degradation potential than RNA in samples drawn from the more controlled environment of a research laboratory. “The platforms need to become more robust,” Chakraborty says. “That way, if the quality of the RNA drops to fifty or seventy percent rather than a hundred percent, we would still be able to get meaningful results.” Researchers with the glue grant program are also seeking to set up guidelines for standardized research procedures that will help lessen the potential for sample degradation.
Inflammation genomics also poses enormous computational challenges. Studies that lack sufficient statistical rigor are a persistent problem, Cobb says—emergency room and intensive care unit cohorts tend to be smaller than optimal, and patients come in after the trauma has occurred so they can’t serve as their own controls. At the same time, lists of inflammation-specific genes identified during microarray experiments need to be incorporated into biological models that describe their molecular interactions. Bioinformatic research and associated databases are continually advancing to meet these needs, however, and collaborations among research groups both within the United States and abroad are helping to drive the science forward.
Cobb emphasizes that despite its broad public health impact, inflammation research has yet to achieve the same public awareness as that of cancer or heart disease. “We need to do a better job of educating people about the importance of this process,” he says. This means reaching other scientists as well as the public, whose concerns often drive research funding.
In the meantime, genomic methods have generated incremental advances in our understanding of inflammation. Scientists have barely scratched the surface of its vast complexity, but perhaps in the not-too-distant future, patients will reap the benefits of their efforts.
The Inflammatory Response
Small advances.
Fatality rates from sepsis are much lower in children than adults, so much may be learned from how children’s bodies deal with inflammation.
|
16330338
|
PMC1314943
|
CC0
|
2021-01-04 23:41:29
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A816-A821
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.113-a816
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0082216330339EnvironewsFocusFocusing on Vision Through an Environmental Lens Barrett Julia R. 12 2005 113 12 A822 A827 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
Our eyes are our window to the world, but for many people the view becomes dim or even darkens entirely due to visual impairment. Although the full impact of the environment on sight is unknown and significant gaps remain in our understanding of vision disorders, many reports have shown that low vision and blindness can be directly or indirectly related to environmental exposures.
Vision is described in terms of visual acuity and field of vision. Visual acuity is a measure of how well an individual sees compared with someone with normal sight—for example, a person with 20/60 vision must be within 20 feet of an object to see it as clearly as a normal-sighted person at 60 feet—and a normal field of vision is 160 to 170 degrees.
The World Health Organization (WHO) estimates that approximately 124 million people have low vision, which it defines as visual acuity between 20/60 and 20/400 with the best possible correction or a visual field of 20 degrees or less. Another 37 million people meet the WHO definition for blindness, which is visual acuity that cannot be corrected to better than 20/400 or a visual field of 10 degrees or less. An analysis published in the April 2004 Archives of Ophthalmology by the Eye Disease Prevalence Research Group, a consortium representing several institutions, indicates that low vision or blindness affects 3.3 million Americans over age 40 and predicts that this figure may be as high as 5.5 million by 2020. (In this study, low vision was defined as visual acuity between 20/40 and 20/200 with best correction, while blindness was defined as visual acuity of 20/200 or worse with best correction.)
According to the WHO, the rising trend is likely to be seen globally as well. An expanding population explains some of the increase, but more critically, the fastest-growing population sector comprises people older than 50. Worldwide, more than 80% of people who are blind are 50 or older, although they represent only 19% of the world’s population. Gender is also significantly associated with visual impairment. Women represent two-thirds of those with blinding eye disease, even after controlling for women’s longer life spans.
Risk also varies by race, ethnicity, and world region. Socioeconomic development often predicts regional prevalence of a disorder. Of the approximately 1.4 million children with blindness in the world, about 75% live in high-poverty areas in Asia and Africa. Among children in high- and middle-income countries, optic nerve defects, other neurological problems, and retinopathy of prematurity (a consequence of incomplete eye development) are the most common causes of blindness.
Developing nations are disproportionately affected due in large part to the burdens associated with poverty: lack of clean water and sanitation, limited or nonexistent health care, and malnutrition. Among children in low-income countries, vision problems arise mostly from complications of measles or rubella, nutritional deficiency, improper or inadequate treatment, and eye infections in the first days of life. In Tibet, an area with one of the highest prevalences of cataract, a lack of vitamin A is compounded by exposure to high-altitude ultraviolet (UV) light, soot and pathogens from indoor burning of coal and yak dung, and a dusty, windy environment. As a result, 10.9% of the total Tibetan population suffers visual impairment.
Cataract
The primary insult to the eye is age, and one very common result of aging is cataract, in which the lens acquires color and may also become clouded or opaque. “If you live long enough, you will get cataract,” says Roger Truscott, an associate professor and senior research fellow at the Save Sight Institute at the University of Sydney in Australia. Non–age-related cataract arises from specific mutations in membrane proteins, injury, toxic or infectious exposures, and diabetes. Family studies have shown that genetics has a role in heritable cataract and may influence the development of age-related cataract, although no specific genes have been identified.
Cataract ranks as the leading cause of blindness and low vision worldwide. The WHO estimates that nearly 48% of global blindness arises from cataract. Cataract removal is one of the most common surgeries in the United States, with approximately 1 million operations performed each year. In developing countries, cataract can mean permanent blindness because sight-restoring treatment is unavailable or unaffordable for many people.
Among the suspected environmental contributors to age-related cataract are UV light exposure, cigarette smoking, and a diet low in antioxidants. Natalie Kurinij, program director of the NEI Vision Research Program, cites results published over the years from the Chesapeake Bay Waterman Study and the Salisbury Eye Evaluation Project as supportive evidence of the risk presented by UV exposure. “We’re fairly confident that sunlight exposure plays a role,” says Kurinij.
That role seems to be linked to oxidative damage, which follows the generation of free radicals. The same sort of mechanism is suspected with cataract risk associated with smoking. Smoking generates free radicals throughout the body, and those may be responsible for lens damage. The precise mechanisms by which oxidative damage to the lens occurs are still being investigated.
The relationship between free radicals and lens damage may not be direct, however. “The epidemiological data are surprisingly weak when you think that our eyes are exposed for about half of our lifetime, and it seems to make sense that a transparent tissue, designed to transmit light, should suffer ultraviolet damage,” says Truscott. The lens possesses a good UV filter system, but it decreases with age. Further, the lens’s ability to maintain low oxygen levels within its center also seems to diminish, and the lens thus becomes more susceptible to oxidative damage. Due to these changes in the lens, the compounds that serve as UV filters may bind to lens proteins, which then become more sensitive to UV radiation–induced damage.
A diet rich in free radical–scavenging antioxidants might be protective, but proof is lacking. “Vitamin data are not convincing with regard to cataract, but [vitamins] E and C could be useful. There’s still much more to know,” says Truscott. Kurinij agrees that support for nutrition’s role varies. “There have been conflicting reports from observational studies regarding the role of antioxidant nutrients and the development of cataract,” she says. “Any potential effect of antioxidant nutrients on cataract will probably depend on the nutritional status of the population to begin with.”
To illustrate the complexity of researching dietary effects on vision, Kurinij compares randomized clinical trial results from the NEI’s Age-Related Eye Disease Study to cataract studies conducted in Linxian, China. In the NEI study, high-dose antioxidant and zinc supplements over a six-year period was not associated with lens opacities in a healthy, well-nourished U.S. population. However, in the relatively nutritionally deprived Linxian subjects a multi-vitamin supplement or a supplement with riboflavin and niacin was associated with fewer cases of cataract.
Oxidative damage also figures in research conducted by Debra Schaumberg, a clinical associate scientist at Schepens Eye Research Institute and an assistant professor of medicine and ophthalmology at Harvard Medical School. Schaumberg and her colleagues reported in the April 1999 issue of the Annals of Epidemiology that people with higher blood levels of C-reactive protein, an indicator of systemic inflammation, had a higher incidence of cataract. “This was really the first time that anyone had shown that systemic inflammation, with no clinically detectable inflammation in the eye, increased the risk of cataract,” says Schaumberg, who notes that older people and obese persons tend to have higher levels of inflammatory activity in the body. “Obesity is one of the strongest contributors to the levels of something like C-reactive protein,” she says.
Schaumberg also identifies heavy metal exposure as a risk factor needing more research. According to a study led by Schaumberg and published in the 8 December 2004 issue of JAMA, low-level lead exposure appears linked to cataract formation in men. Of 642 men ranging in age from 48 to 93 years, 122 were diagnosed with cataract. Bone scans determined that the men’s long-term, low-level lead exposure was comparable to that of the general population. Men in the highest exposure group (8.17–35.0 micrograms per deciliter) had 2.5 times the risk of having cataract as the men in the lowest exposure group (1.0–3.0 micrograms per deciliter). “As far as we know, this paper . . . was really the only epidemiological study looking at heavy metals in relation to eye disease. I think it’s really an area that we don’t know much about,” she says.
At the opposite end of the age spectrum, children may have cataract at birth or develop the condition in infancy due to prenatal infection with rubella or toxoplasmosis, among other causes. Such cases need immediate treatment. One consequence of untreated cataract is nystagmus, any of a variety of involuntary movements of the eyes. Amblyopia, or “lazy eye,” is another condition found in children. “In the first weeks up until the first couple of months [of life], if vision is disturbed in both eyes that then will cause poor vision for the rest of life because nystagmus cannot be treated in any way,” says Jill Keeffe, an associate professor at the Centre for Eye Research Australia. “With cataract, it needs to be treated within weeks, whereas with amblyopia, which might develop from strabismus [drifting or crossing of one or both eyes] or from uneven refractive error between the two eyes, the window of opportunity is much longer. Obviously, the earlier, the better.”
Retinal Disorders
Retinal disorders also pose a threat worldwide. Age-related macular degeneration (AMD) refers to damage to the area of the retina responsible for sharp central vision. It is the third most common global cause of blindness, accounting for approximately 8.7% of total blindness, and the primary cause of blindness in developed countries. In the United States, there are approximately 1.8 million people with vision loss due to AMD, and another 7.3 million are at risk. As the average age of the world’s population creeps upward, AMD will become even more significant.
As much as 30% of AMD may be related to smoking. A prospective study published in the 9 October 1996 issue of JAMA linked smoking with AMD, and more recently a study in the 14 April 2005 British Journal of Ophthalmology showed that smokers were twice as likely as nonsmokers to develop AMD. The risk declines if one stops smoking, to the point that after 20 years of not smoking former smokers have about the same level of risk as nonsmokers. Possible mechanisms of damage linked to smoking include depressed levels of antioxidants, reduced oxygen, and altered blood flow.
The effect of diet on AMD risk shares some of the same components as cataract; specifically, low-level antioxidant levels may heighten the chances of developing the disease. Obesity and high blood pressure, fat intake, and cholesterol levels also appear to increase AMD risk, but the specifics are not yet clear.
Family studies imply a genetic link, which is supported by three papers published in the 15 April 2005 issue of Science and a fourth published in the 2 May 2005 issue of Proceedings of the National Academy of Sciences. “Age-related macular degeneration is certainly a disease that’s affected by our genes,” says Timothy Stout, an associate professor of ophthalmology at the Casey Eye Institute in Portland, Oregon. “What’s interesting is that it’s also a disease that’s influenced by our environment. The link between those two has been puzzling in the past, and I think there are new studies that suggest that some of the genes that play a role in the development of macular degeneration are genes that may be involved in inflammation and our immune response.”
In early 2005, the four teams independently associated AMD with a gene coding for complement factor H, an inflammatory component. “The recent research implicating the complement factor H gene in AMD is a major breakthrough,” says Peter Humphries, a professor of medical molecular genetics at Trinity College in Dublin, Ireland. “Many studies have resulted in localizing so-called susceptibility genes to chromosomal regions, but the studies recently reported are the first to identify an actual gene. I expect that we will find out a great deal more about the so-called molecular pathology of AMD as a result of this discovery.”
Further, up to six regions within the genome have been implicated as potentially harboring AMD genes, and a second gene was reported in the November 2005 Human Molecular Genetics. “As yet we have very little information about this most recent gene,” says Humphries. “[However], once more is known about the mechanisms of action of such variants, we stand to know a lot more about the cause of AMD, and hence the prospects for eventual prevention will become more realistic.”
Retinal damage is also a hallmark of diabetic retinopathy, which blinds about 5 million people worldwide. The U.S. Centers for Disease Control and Prevention estimates that 13.8 million Americans have been diagnosed with diabetes and another 5.2 million have it without realizing it. Duration of diabetes and its control affect risk of diabetic retinopathy, and approximately 5.3 million Americans over age 18 have the eye condition. With diabetes rates increasing, in part due to increasing obesity, diabetic retinopathy can be expected to become more prevalent. Dietary and genetic factors may also affect its development, as may high blood pressure and high cholesterol.
“All of these things—hypertension, diabetes, hypercholesterolemia—tend to have bad effects at the level of the small blood vessels, the capillaries,” says Stout. Retinal vein blockages associated with high cholesterol, high triglycerides, and high blood pressure can create capillary-bursting pressure. The tissue then becomes hypoxic, or insufficiently oxygenated. Retinal hypoxia also occurs in diabetes when the capillary network dies through mechanisms that are not completely clear.
Hypoxia triggers production of vascular endothelial growth factor, which promotes formation of new blood vessels, but the process is disorganized. “The body tends to not build the blood vessels in the right place, so they will grow not as nicely formed capillaries in the retina, but at the optic nerve or into the center part of the eye or at the part of the eye where the drainage system is, and that causes all sorts of problems,” says Stout. Further, the poorly built new vessels leak, as may existing blood vessels. Ultimately, the retina detaches from the underlying layers, and vision is lost.
Infections and Nutritional Deficiencies
Slightly more than 5% of global blindness arises from injury- or disease-associated corneal opacity. Distinct from this category, trachoma accounts for an additional 3.6% of global blindness. Trachoma is the most common infectious cause of vision loss and affects approximately 84 million people, primarily in remote rural areas of Africa, Asia, Central and South America, Australia, and the Middle East. This disorder arises from repeated infection by Chlamydia trachomatis bacteria that are spread by close contact with an infected person or by flies. After numerous infections, eyelid scarring turns the eyelashes inward, and they rake against the cornea, a condition called trichiasis. The irritation scars the cornea and eventually renders it opaque.
Infection typically starts in childhood, although the blinding effects do not occur until well into adulthood. “It’s not just one infection—it’s repeated infections,” says Keeffe. “We’ve seen scarring in the lids of preschool-aged children, . . . but it’s usually not until the forties and fifties that vision loss occurs.” In some areas, 60–90% of pre-school-age children carry active infections. Women account for 75% of late-stage blinding trachoma cases, possibly because they have greater contact with children.
In 1997 the WHO launched GET (Global Elimination of Trachoma) 2020 with the goal of eradicating trachoma altogether. A major part of GET 2020 is a primary health care plan known as the SAFE strategy. The SAFE strategy utilizes lid surgery (S), antibiotics (A) to treat active infections, facial cleanliness (F), and environmental changes (E) geared toward improving sanitation and access to clean water. A review of the SAFE strategy published in the June 2003 issue of The Lancet Infectious Diseases found strong support for the use of antibiotics and surgery in warding off infection and blindness, although the evidence for face washing and environmental improvements was weaker.
Onchocerciasis, also known as river blindness, is caused by Onchocerca volvulus, a parasite transmitted by blackflies in riverside areas. The disease is endemic in West and Central Africa, Yemen, and several South American countries. When a blackfly bites, a juvenile form of the parasite enters the body. Once mature, females release high numbers of larvae that migrate to the skin and eyes. Associated lesions form in all eye tissues except for the lens, and lead to inflammation, bleeding, secondary infections, and eventually blindness. More than 17 million people are infected with the parasite, approximately 500,000 people are visually impaired as a result, and another 250,000 are blind. Fear of infection prevents arable riverside land from being used, and local economic growth stagnates.
Global efforts at halting river blindness started with effective vector control efforts in 1974, and ongoing community-based treatment with ivermectin, an antiparasite medication, began in 1996. The disease has been reduced in most areas and may be eradicated from Latin America by 2010.
Among children, cornea-clouding vitamin A deficiency is the most common cause of preventable childhood blindness. Poor night vision is the key symptom of the very early stages of the corneal damage preceding blindness. At this stage, children can retain their vision with repeated doses of vitamin A. In late-stage vitamin A deficiency, the cornea becomes very white and cloudy, and vision loss is irreparable; cornea transplants are impossible because the tissue becomes too damaged. Immunization and good nutrition are key to preventing this form of blindness, but where these interventions are not immediately possible, doses of vitamin A make “lovely primary health interventions,” says Keeffe. In addition to supplying vitamin A supplements, health organizations also strongly promote vitamin A–rich diets. Breastfeeding provides ample vitamin A to babies, and older children and adults receive the nutrient through garden produce and supplemented foods.
An Eye to the Future
The WHO estimates that up to 75% of all blindness is preventable. Vision 2020: The Right to Sight, a program instituted in 1999 by the WHO and the International Agency for the Prevention of Blindness, builds upon previous programs, dovetails with pre-existing efforts of many organizations, identifies remaining regional and national needs, and provides a framework for filling gaps. Through Vision 2020, a coordinated effort is under way to eliminate preventable blindness by 2020 by increasing awareness of eye disease, garnering resources for prevention and treatment, controlling major causes of avoidable blindness, training ophthalmologists and other eye care personnel to diagnose and treat the diseases specific to certain regions, and providing these specialists with necessary technology and infrastructure.
Other organizations also have carried out large-scale international programs. For example, Lions Clubs International, a service organization with a long history of combating low vision and blindness, instituted the worldwide SightFirst program with three major goals: treating and preventing diseases such as river blindness, trachoma, and cataract; providing education and training of health care workers to diagnose and treat eye disease; and constructing and equipping health care facilities. The group is currently building SightFirst II as well as collaborating with the WHO on the Project for the Elimination of Avoidable Childhood Blindness.
On a national level, the U.S. government sponsors prevention and treatment through the Department of Health and Human Services’ Healthy Vision 2010 and the NEI’s National Eye Health Education Program. Healthy Vision 2010 is part of Health People 2010, a national program to improve the health of Americans, and seeks to promote regular eye examinations for adults and children, vision screening for preschoolers, and injury prevention. There’s also a component to educate people with low vision about treatment. The National Eye Health Education Program has a more specific focus, encouraging early detection and treatment of glaucoma and diabetic eye disease, and providing education about treatment for low vision. This program provides materials that communities can use to educate the public about eye disease and the importance of early detection and treatment.
A critical gap in eliminating preventable blindness and low vision is delivery of health care. Without access to health care, opportunities are missed to diagnose problems early when treatment is most likely to be successful. Access to health care is a problem in many nations, including the United States. Stout offers diabetes as an example of the United States falling short in this regard. “That’s a huge population problem for us in the United States, because as people are falling through the health care cracks and aren’t [controlling diabetes] they’re going to get very significant blinding diabetic retinopathy at a relatively early age,” he says. “Vision loss in anybody is not a good thing, but vision loss in a relatively young, healthy person who presumably has a productive career ahead of them—it’s a real issue.”
For all that’s been learned about eye disease, there remain gaps, and new hypotheses continue to be generated. Researchers at the Schepens Eye Research Institute recently embarked on the $2.2 million, three-year Planning Grant for Research on Blinding Eye Diseases. Like other NIH “roadmap” grants, this one is designed to promote interdisciplinary collaborations regarding complex health challenges. “While experts from these areas often collaborate informally on [eye] disease, ophthalmology has remained somewhat specialized and in some ways isolated from other disciplines,” remarked Darlene Dartt, director of scientific affairs at Schepens Eye Research Institute, in announcing the grant. “This is really the first federal program to formalize collaboration.” In learning more about the cascade of events occurring in other parts of the body in diseases ranging from Alzheimer disease to rheumatoid arthritis, researchers might gain some insight into the processes of blinding eye diseases.
Clouding our vision.
Cataract, the leading cause of blindness worldwide, arises from a combination of factors including genetics, age, and environmental exposures. Though largely treatable, poor access to health care leaves many around the world to suffer. (Top to bottom) An acute sudden-onset cortical cataract in a person with type 1 diabetes; a hypermature age-related cataract; a white congenital cataract.
Insight into the problem.
Flies attracted to eye secretions are one way trachoma is transmitted (above). In its bid to eradicate this disease by 2020, the WHO encourages facial cleanliness (right top) and improvements in sanitation such as burying waste (right bottom).
Focusing on onchocerciasis.
When parasites spread by blackflies (above) enter the body, their larvae migrate to the eyes and skin, causing lesions that result in inflammation, bleeding, and severe visual impairment or blindness if left untreated treated (right top). Treatment with ivermectin (right bottom) has reduced the disease in most areas and nearly eradicated it in some.
Seeing a way past trachoma.
A 15-minute procedure performed by a doctor or trained nurse using local anesthetic can reverse the conditions that cause trachoma and reduce the risk of blindness. Globally, however, lack of access to basic health care keeps many patients in the dark.
|
16330339
|
PMC1314944
|
CC0
|
2021-01-04 23:41:29
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A822-A827
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.113-a822
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0082816330340EnvironewsSpheres of InfluenceGAO Sounds Off on Chemical Regulation Black Harvey 12 2005 113 12 A828 A830 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
Since 1976 the Toxic Substances Control Act (TSCA) has given the federal government the power to require that chemicals are properly tested and regulated before they reach the market, and that they don’t pose unreasonable risks to human and environmental health. TSCA is the key piece of legislation governing the way the U.S. Environmental Protection Agency (EPA) reviews and regulates chemicals including solvents and constituents of paints, fuels, and plastics. Yet, concerns persist about chemical safety and the adequacy of regulation.
Now, in a June 2005 report titled Chemical Regulation: Options Exist to Improve EPA’s Ability to Assess Health Risks and Manage Its Chemical Review Program, the Government Accountability Office (GAO) has reviewed the EPA’s efforts to control the risks of new chemicals not yet in commerce, to assess the risks of existing chemicals used in commerce, and to publicly disclose information provided by chemical companies under TSCA. The report points out shortcomings in TSCA and its implementation, and suggests ways to strengthen the law.
HPV Chemicals
TSCA authorized the EPA to both assess new chemicals before they enter the marketplace and to review chemicals already on the market. But when the law was enacted, thousands of chemicals already being used were grandfathered in. “Those sixty-two thousand or so chemicals were just accepted as being okay to be in commerce without any kind of EPA risk analysis,” says David Bennett, the report’s lead analyst.
Even aside from these grandfathered chemicals, the EPA has an enormous number of chemicals to examine, so the agency has narrowed its approach. “We have decided to focus our work on the high-volume chemicals, using volume as a surrogate for [human and environmental] exposure,” says Charles Auer, director of the EPA Office of Pollution Prevention and Toxics. The High Production Volume (HPV) Challenge Program was started in 1998 with the goal of looking at some 2,800 chemicals that were produced in quantities exceeding 1 million pounds per year as of 1990. The voluntary program was established by the EPA, Environmental Defense, the American Chemistry Council (ACC), and the American Petroleum Institute to identify and fill gaps in basic hazard data for these chemicals, and to make those data publicly available by 2005.
The information garnered from the HPV Challenge Program “will allow us to prioritize among the chemicals and then obtain additional information where appropriate or take control actions,” says Auer. Michael P. Walls, managing director of health, products, and science policy at the ACC, adds, “Through the [HPV Challenge] Program we provide a mechanism to assure the agency that there is adequate information on which to base current risk management decisions.”
“The program is a light-some-candles-rather-than-sit-and-curse-the-darkness initiative,” says Karen Florini, a senior attorney with Environmental Defense. “It gathers preliminary basic screening-level information. It’s clearly valuable; it’s just limited.”
Despite progress made to date, the GAO report states there are 300 chemicals in the HPV Challenge Program “for which chemical companies have not agreed to provide the minimal test data that EPA believes are needed to initially assess their risk.” Auer regards that situation as “unfinished business.” He says the EPA is developing rules to require industry to test those chemicals. “We hope to finalize that test rule at the end of this year or early next year,” he says.
Both Auer and Walls note there could be a variety of reasons for the chemical industry not doing this testing voluntarily. For example, in some instances domestic manufacturers of a chemical have said they would be willing to provide information only if foreign competitors who export the chemical to the United States would share the cost of testing—support that was not forthcoming. However, once a test rule is in effect, anyone who produces or imports the chemical must comply and provide the data.
New Chemicals
The GAO report also voices concern about the EPA’s efforts to regulate new chemicals. Though the report noted that the EPA has taken actions to regulate exposure to about 3,500 of 32,000 new chemicals submitted for review since TSCA was enacted, the GAO has qualms about the way in which those chemicals were examined.
The EPA typically does not have enough data on a submitted chemical’s properties to determine its toxicity. Consequently, it may compare a new chemical with closely related model chemicals to predict whether the new compound will pose a safety hazard. “We found evidence that in some cases the models were not entirely predictive. The problem is that in some cases there is just not a lot of data out there to show how predictive the models are,” says Bennett.
Auer counters that the models do what they are supposed to—identify potentially hazardous candidates for further testing. He adds that the models also tend to err on the side of caution—that is, they tend to identify chemicals that appear to be hazardous but prove to be safe upon further examination.
The GAO report also notes that TSCA does not require chemical companies to submit data to the EPA on the toxicity, routes of exposure, or potential extent of exposure of new chemicals. “I think it is scandalous that new chemicals can be brought to market without being accompanied by any actual data,” Florini says. “Eighty-five percent of PMNs [premanufacture notices, which must be submitted to the EPA at least 90 days before production of a new chemical begins] are submitted with no health data, and reliable models aren’t available for many end points, particularly for long-term health effects other than cancer.”
But Walls says chemical manufacturers must be prepared to supply data to the EPA if a chemical has characteristics of persistence, bioaccumulation, and toxicity. “The assumption that new chemical applications are filed in the United States with no information is not right,” he says. He agrees that TSCA does not require safety and exposure data, but adds that if a company is submitting a chemical to which people could be exposed, the company would be “remiss” in not providing that information. Still, the data need be provided only if the agency asks for them.
Auer asserts that requiring toxicity testing before a chemical is actually manufactured, as required by TSCA, could interfere with innovation in the industry. He estimates the cost of providing the information wanted by the EPA to be in the range of a quarter of a million dollars for new chemicals. Auer says the expense of testing before a chemical company knows whether there will be a demand for a chemical could hobble efforts to develop improved chemicals, and maintains that the EPA’s track record of taking action to reduce the risk of new chemicals is a good one.
TSCA, Take Two
Faced with the lack of required data for new chemicals and what the GAO regards as the uncertain effectiveness of the voluntary HPV Challenge Program, the report recommends that Congress give the EPA the authority to require chemical manufacturers to generate and provide test data on HPV chemicals. That recommendation is embodied in legislation introduced in Congress this summer by senators Frank Lautenberg (D–NJ) and James Jeffords (I–VT). The bill also would give the EPA the authority to require those data for all chemicals used, and to prioritize which chemicals the industry would have to test. Auer says the EPA has not yet taken a position on the bill. Environmental Defense supports it, while the ACC opposes it, saying it duplicates the EPA’s existing authority under TSCA.
The GAO offers a number of other recommendations to strengthen the act. Among them are validating and improving the models used by the EPA to assess risks of chemicals; requiring chemical companies to submit testing results of chemicals with PMNs; letting the EPA regulate chemicals if they pose a “significant” risk to health or the environment rather than the more stringent “unreasonable” risk; and setting national goals for reducing the overall use of toxic chemicals.
Auer says the EPA will soon have much better exposure information under amendments to the TSCA Inventory Update Rule, which requires chemical manufacturers to submit basic production data every four years for chemical substances (including imports) manufactured for commercial purposes in amounts of 25,000 pounds or more at a single site. According to the EPA, the 2003 amendments tailor reporting requirements to more closely match the EPA’s information needs, provide a vehicle for the EPA to obtain updated information on the potential human and environmental exposures of chemical substances listed on the TSCA Inventory, and improve the utility of the information reported under the rule.
With the amendments, “we will know the chemicals that are in consumer products,” Auer says. “We will know the number of workers that are exposed to chemicals. We will have a better basic idea of the uses of chemicals. So EPA in sixteen months will have basic exposure information on [HPV] chemicals.”
Moreover, Auer says the EPA will require the information to be updated every five years so the agency can understand how chemical use is shifting and whether safer alternatives have been developed in the meantime. He says the information will allow the EPA to identify HPV chemicals that are candidates for more testing or for enhanced regulation. Further, Walls says the chemical industry on its own initiative will supply the EPA with hazard data on around 500 chemicals that reached the high-volume threshold between 1990 and 2002.
Nevertheless, Bennett reserves judgment on the HPV Challenge Program’s effectiveness. “The process is ongoing, so I would hesitate to say how successful it is, because EPA has not received all the data that industry has promised to deliver. We won’t know for some time whether the program will be successful,” he says.
Florini echoes that view, and also points out that the EPA is “way behind” in making public the information submitted to date. “Six years into the HPV Challenge Program, it’s really regrettable that the database hasn’t yet been released, though it apparently will be out by year-end,” she says.
After three decades of existence, it is appropriate that TSCA is undergoing significant examination, as the EPA, the chemical industry, environmentalists, and legislators all look at ways to revise this major statute. What a revised TSCA will look like after this examination, however, is far from certain.
|
16330340
|
PMC1314945
|
CC0
|
2021-01-04 23:41:29
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A828-A830
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.113-a828
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0083216330341EnvironewsInnovationsMaking Succinate More Successful Potera Carol 12 2005 113 12 A832 A835 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
What does the word “fermentation” bring to mind? Beer? Bread? Ethanol derived from corn and other plant matter? How about succinate? Since 2001, biochemist George Bennett and bioengineer Ka-Yiu San, both professors at Rice University, have been tinkering with Escherichia coli to coax it to convert sugars to succinate, a chemical with multiple industrial uses. Now their efforts are bearing fruit as “green” succinate is starting to become a reality in chemical commerce.
Who uses succinate? By itself, succinate is used as a flavor enhancer in food products and as a stabilizer in pharmaceuticals. It is also used to produce other industrial chemicals, including butanediol, tetrahydrofuran, and pyrrolidone, which become ingredients in solvents, paints, deicers, plastics, fuel additives, fabrics, and carpets.
Succinate is traditionally manufactured from petrochemicals through expensive processes. The Rice team’s goal is to make a more environmentally friendly succinate from renewable starting materials. “We want to use agricultural materials that are renewable to make this useful product, and alleviate the drain of limited oil reserves,” says Bennett.
The Department of Energy (DOE) “sees a future for biorefineries that use biomass as feedstocks to make fuels and chemicals,” says department chemist Gene Petersen. In 1994, the agency’s now-defunct Alternative Feedstocks Program assessed the likelihood of making chemicals from biomass. “The category of compounds that seemed most viable were organic acids like succinic, acetic, and citric,” says Petersen.
That evaluation resulted in the DOE’s funding of fermentation research programs at national laboratories and universities. In 2004, the DOE released volume I of a report titled Top Value Added Chemicals from Biomass, coauthored by Petersen (volume II is expected out in 2006). According to the report, succinate tops the list of 12 “building block” chemicals—molecules with multiple functional groups that possess the potential to be transformed into new families of useful materials—that can be produced from sugars via biological conversion.
In 2001, 10 million pounds of succinate were produced from petrochemicals and sold for an average of $2 per pound. “The market is there if we can make succinate more economically through biofermentation,” says Praveen Vadlani, principal research scientist at AgRenew Incorporated in Manhattan, Kansas. By making green succinate in bulk—a potentially cheaper material with the cachet of environmental friendliness—people may even be inspired to find new applications for it, such as bio-based polymers and composites, predicts Vadlani.
Optimizing Glucose
“It’s not a direct route from glucose to succinate,” says Bennett. Several biochemical pathways can produce succinate from sugar. They all start with the degradation of glucose, which contains six carbon atoms, to pyruvate, which contains three carbons. Then pyruvate can be converted not only into succinate (which contains four carbons), but also lactate, ethanol, acetate, and other chemicals. The trick is to speed up the chemical reactions that lead to succinate production while blocking those that make lactate, ethanol, and other chemicals.
Some pathways operate aerobically (they need oxygen) whereas others run anaerobically (they do not use oxygen). Bennett and San spent four years working out both aerobic and anaerobic methods for E. coli to convert glucose into almost pure succinate in yields high enough to be commercially feasible. Their anaerobic method has proven more efficient, with 1.0 gram of glucose yielding 1.44 grams of sodium succinate. Their aerobic process yields about three-quarters that amount.
Bennett and San have engineered a form of E. coli, dubbed SBS550MG, that contains six genetic alterations that allow it to produce succinate anaerobically from two different routes—the glyoxylate pathway and the fermentation route. To accomplish this, the researchers deleted four E. coli genes, including those for lactate and ethanol production, and activated the glyoxylate pathway in order to speed the conversion of glucose solely into succinate. They also added two genes from other bacteria to boost the amount of succinate generated.
Both routes produce succinate through different biochemical reactions that do not compete or interfere with each another. In fact, Bennett and San designed the routes to be complementary. SBS550MG converts glucose to succinate very efficiently and very rapidly, and gives high yields of nearly pure succinate with few by-products, says San. High-pressure liquid chromatography confirms that more than 90% of the starting glucose ends up as succinate.
To make the leap from the laboratory to the marketplace, the Rice scientists teamed up with bioengineering experts at AgRenew. Under Vadlani’s direction, AgRenew will perfect the methods to manufacture succinate from corn and sorghum rather than from the pure glucose used in the laboratory experiments. “We see great promise in the technology, and once the methods are established, we may even switch to cornstalks or agricultural waste,” says Vadlani.
Up and Running
Kris Berglund, chief science officer at Diversified Natural Products (DNP) in Scottville, Michigan, is experiencing new market demands for green succinate. DNP also uses E. coli to ferment sugars to succinate, but the bacterial strain used was licensed from the DOE, which produced it under its Alternative Feedstocks Program. DNP’s fermentation method differs from that created by Bennett and San in that an aerobic process occurs first, followed by an anaerobic process that requires added carbon dioxide. Says Berglund, “We take six carbons from glucose and add two carbons from carbon dioxide to form two molecules of succinate with four carbons each.”
DNP just started large-scale production of succinate from agricultural materials at Agro-Industrie Recherches et Développements (ARD) in Pomacle, France. The joint venture was announced by French president Jacques Chirac on 30 August 2005. In seeking a partner to manufacture its biosuccinate batches, Berglund searched worldwide and chose ARD because “they shared the same vision as we do to replace petroleum-based chemicals with biomass production,” he says.
The staff at ARD’s manufacturing facility, located in the agricultural Champagne region, will produce up to 200 tons of succinate from wheat and sugar beets in the first year. DNP plans to construct a large plant in the United States, too. “As far as we know, we’re the first company to enter commercial production of succinate from biomaterials,” says Berglund. Although production has just begun, Berglund says “customers already want to buy it,” particularly for use as a flavor enhancer, stabilizer, and acidulant for food production. Some customers desire green succinate because they view it as a “natural” ingredient that would be favored by organic food consumers.
Customers also are lining up to buy DNP’s succinate-based runway and wing deicer. Succinate, which lowers the freezing point of water, replaces the formates and acetates in deicers now on the market. These chemicals not only corrode the metal alloy, plastic, and rubber parts of airplanes, but also destroy the concrete surfaces and plastic and metal components of lighting equipment at airports. Federal Aviation Administration approval of the DNP deicer appears imminent, according to Berglund. Other products in the succinate pipeline at DNP include biodegradable solvents that do not cause air pollution or damage the ozone, a diesel fuel additive to reduce particulate emissions, and biodegradable polyesters for use in fabrics or plastics.
DNP does not disclose information about its yields, but “our methods are good enough to compete with any fossil fuel–related process,” says Berglund. Based on estimates calculated when oil sold at $25 per barrel, DNP forecast a selling price of less than $1 a pound for its biosuccinate. With declining petroleum reserves and rising oil prices, “the economics of our process are even more attractive,” says Berglund.
Other companies are following behind on the same commercialization path. Michigan Biotechnology Institute (MBI) International in Lansing developed a patented process based on Actinobacillus succinogenes, a bacterium isolated from the cow’s rumen (a fermentation chamber in the animal’s stomach). MBI scientists created mutant strains for anaerobic production of succinate from biomass sugars, resulting in yields of approximately 1 gram of succinate from 1 gram of glucose. Different types of biomass, including cornstalks, corn fiber, and sugarcane, can be used to fuel the fermentation.
The MBI method also pipes in carbon dioxide. “It’s a greenhouse-friendly fermentation, because we utilize carbon dioxide instead of generating carbon dioxide,” says microbiologist Bernie Steele, manager of quality assurance at MBI. He foresees his company’s biosuccinate method being linked to ethanol plants, which generate carbon dioxide as a waste product. An overall biorefinery program that uses by-products from one production stream to feed another manufacturing process maximizes economic returns.
After 10 years of research and development efforts, MBI is seeking partners to scale up its process to manufacture large quantities of green succinate. “The technology is maturing for the transition of biomass into energy or chemicals,” says Steele.
Future Uses for Succinate
The future for succinate lies not in utilizing it directly as a food additive, but in creating innovative biopolymers like polybutylene succinate. This biodegradable plastic, already being made with petroleum-based succinate, is found in packaging film, bags, flushable hygienic products, and garden mulch. “We have customers waiting to buy our succinate to make polymers,” says Berglund. Other, stiffer biodegradable plastics, like polylactic acid, are formed into drinking cups, food trays, containers, and planter boxes. These “green” alternatives replace products typically made from petroleum-based plastics.
The commitment of corporate giants like Cargill and DuPont to make products from biomass casts “a bright light on the future of biofermentation,” says Petersen. Cargill produces up to 300 million pounds of polylactic acid, sold as NatureWorks→ PLA, from renewable resources such as corn. DuPont’s Sorona→, a polymer of 1,3-propanediol now made from petrochemicals, adds softness and stretch to fabrics. In 2006, DuPont will switch to a fermentation method to make its 1,3-propanediol from corn sugar. Called Bio-PDO∀, the corn-based polymer will be the first product developed by DuPont’s Bio-Based Materials unit.
Despite this buy-in, the future isn’t here yet. In general, the long journey to find an economic way to convert renewable bio-materials into commodity chemicals takes about 10 years; the basic research behind NatureWorks PLA started in the 1980s. “It’s not that easy to get away from petrochemicals, even though we want to environmentally,” says Petersen.
But the large-scale processes under way at Cargill and DuPont indicate long-term business interest in fermentation, says Bennett. He envisions more companies entering the bioproducts business and the economics of succinate and other bioproducts improving through engineering refinements. And as oil prices rise and fermentation becomes more economically appealing, “companies will find different ways to make the same end product,” says Vadlani.
Sweetening the deal
Researchers are refining techniques for producing succinate from biomass such as sorghum (above) rather than petroleum. One group, MBI International, uses ion exchange to further refine succinate into succinic acid (right).
==== Refs
Suggested Reading
Chatterjee R Millard CS Champion K Clark DP Donnelly MI 2001 Mutation of the pstG gene results in increased production of succinate in fermentation of glucose by Escherichia coli Appl Environ Microbiol 67 1 148 154 11133439
Guettler MV Rumler D Jain MK 1999 Actinobacillus succinogenes sp. nov., a novel succinic-acid-producing strain from the bovine rumen Int J Syst Bacteriol 49 pt 1 207 216 10028265
Sanchez AM Bennett GN San KY 2005 Novel pathway engineering design of the anaerobic central metabolic pathway in Escherichia coli to increase succinate yield and productivity Metab Eng 7 3 229 239 15885621
Werpy T Petersen G eds. 2004. Top Value Added Chemicals from Biomass. Volume I—Results of Screening for Potential Candidates from Sugars and Synthesis Gas. Springfield, VA: National Technical Information Service. Available: http://www.nrel.gov/docs/fy04osti/35523.pdf
|
16330341
|
PMC1314946
|
CC0
|
2021-01-04 23:41:31
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A832-A835
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.113-a832
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a00837EnvironewsScience SelectionsLiver Cancer and Aflatoxin: New Information from the Kenyan Outbreak Barrett Julia R. 12 2005 113 12 A837 A838 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
Millions of people are exposed to aflatoxins, toxic compounds produced by Aspergillus molds. These molds infest staple crops such as maize, peanuts, rice, and wheat throughout the world. Outbreaks of aflatoxicosis affecting up to several hundred people at a time have occurred sporadically, most recently in eastern Kenya in early 2004. An investigation of the Kenyan outbreak now yields new information on the risk factors associated with acute aflatoxin poisoning [EHP 113:1779–1783].
Chronic low-level exposure to aflatoxins, particularly aflatoxin B1, is associated with increased risk of developing liver cancer, impaired immune function, and malnutrition. Acute high-level exposure, which is less common, causes early symptoms of diminished appetite, malaise, and low fever. Later symptoms, including vomiting, abdominal pain, and hepatitis, signal potentially fatal liver failure.
The Kenyan outbreak followed a poor harvest of maize that had been damaged and made susceptible to mold by drought. Furthermore, to guard against theft of the meager harvest, people stored the maize in their homes, which were warmer and moister than the granaries where it was usually stored. From January to June 2004, 317 people sought hospital treatment for symptoms of liver failure, and 125 died. Health officials ruled out viral liver diseases; suspecting acute aflatoxin poisoning, they examined maize samples and found aflatoxin B1 concentrations as high as 4,400 parts per billion (ppb), 220 times the Kenyan limit for food.
Researchers conducted a case–control study using records for 40 patients (cases) who had been hospitalized with acute jaundice during late May and early June and 80 randomly selected controls. Jaundice is a nonspecific symptom of liver damage.
Participants or family members completed questionnaires targeting maize quality, storage, preparation, and consumption. The researchers collected 1-kilogram samples of maize from households that still had grain left over from the time of the outbreak for measurement of aflatoxin concentrations. Blood samples from 29 patients and 62 controls were analyzed for concentrations of aflatoxin B1–lysine albumin adduct, a marker of aflatoxin exposure. The researchers also tested blood from 18 patients and 54 controls for hepatitis B surface antigen, an indicator of hepatitis B infection. In people with chronic low-level aflatoxin exposure, this virus enhances the risk of developing liver cancer.
Maize from patients’ homes contained significantly higher amounts of aflatoxin (with a geometric mean of 354.5 ppb) compared to control households (with a geometric mean of 44.1 ppb). Patients’ serum aflatoxin adduct concentrations, which were comparable to those measured in previous outbreaks, were nearly 10 times higher than those of controls. Further, patients who died had higher blood levels of adducts than those who survived. Forty-four percent of the patients tested positive for hepatitis B, compared to 7% of controls.
These analyses, with their greater level of detail, are the first to quantify the association between concentrations of aflatoxin in food, exposure history, concentrations of serum aflatoxin adducts, and acute aflatoxin poisoning. This study is also the first to quantify the independent association between hepatitis B infection and the effects of acute aflatoxin poisoning. The researchers suggest that monitoring both aflatoxin concentrations in crops and the incidence of acute jaundice could permit earlier recognition of food contamination and help prevent an outbreak from becoming widespread. Further, they suggest that future use of blood tests for aflatoxin B1–lysine albumin adducts could serve to diagnose aflatoxin poisoning and to gauge the success of measures for reducing aflatoxin exposure.
Tiny killer.
Chronic low-level exposure to aflatoxins produced by Aspergillus molds (such as A. flavus, above) is associated with increased risk of liver cancer.
|
0
|
PMC1314947
|
CC0
|
2021-01-04 23:41:30
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A837-A838
|
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-a00838EnvironewsScience SelectionsThe Heavy Load of Lead: Ergonomic Stress Heightens Exposure-Related Neuropathy Sharma Dinesh C. 12 2005 113 12 A838 A838 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
Long-term lead exposure among industrial workers can result in neuropathy (a disorder of the peripheral nervous system), while lower exposure levels cause muscle weakness. Until recently, however, the interaction between lead toxicity and chronic repetitive muscle use had not been investigated. Researchers from the Center for Occupational and Environmental Neurology in Baltimore now report that the impact of chronic lead exposure is augmented by concomitant ergonomic stress [EHP 113:1730–1734].
The study included 80 lead smelter workers who were routinely exposed on the job to inorganic lead dust and (to a lesser extent) lead fumes. Historical blood lead records for all the workers were available from the smelter, which checked all employees’ blood lead at least quarterly. These records showed that workers had high chronic exposure in the distant past, much lower exposure in the more proximate past, and still lower exposure at the time of the study. The researchers also measured current blood and bone lead levels and used the historical records to calculate two metrics of cumulative lead exposure—working-lifetime integrated blood lead (IBL) and working-lifetime weighted-average blood lead (TWA).
The team used the current perception threshold test to examine nerve fiber populations in the workers’ shoulders, arms, wrists, and hands. This test measures the amount of electrical current needed to induce a sensation. The team also created a three-tiered ergonomic stress rating based on all the different jobs the workers had ever performed, cumulated over their employment history. This was used to arrive at a time-weighted average ergonomic stressor. Sensory nerve conduction threshold was measured in large myelinated, small myelinated, and unmyelinated nerve fibers.
The results showed that decrements in nerve function—a precursor to neuropathy—were limited to large and small myelinated sensory nerve fibers, with a threshold effect at a TWA of 28 micrograms per deciliter. At higher levels of lead exposure and presence of ergonomic stress, nerve fibers were more susceptible to increased damage, something that has never before been shown in human studies. The investigators suggest that nerves affected by lead are more susceptible to traction or mechanical compression, as would occur in the carpal tunnel of workers who perform activities such as heavy lifting and shoveling.
Measures of chronic lead exposure may serve as strong predictors of impaired nerve function. In addition, the authors believe they have been able to separate the impact of two components of cumulative blood lead—duration and intensity—with exposure intensity appearing to have a greater influence than duration on the outcome studied. Finally, the authors point out that although TWA and IBL are associated with peripheral nerve damage, bone lead—another measure of chronic exposure—is a weak predictor of lead effects in the nervous system because it reflects only that lead stored in the bone compartment and not necessarily the cumulative blood lead to which peripheral nerves were exposed.
Double jeopardy.
Ergonomic stress may heighten the threat posed by on-the-job lead exposure.
|
0
|
PMC1314948
|
CC0
|
2021-01-04 23:41:31
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A838
|
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-a00839AnnouncementsNIEHS Extramural UpdateEnvironmental Influences on Epigenetic Regulation 12 2005 113 12 A839 A839 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
Epigenetics refers to the study of gene silencing and non-Mendelian inheritance of traits. By controlling the expression of key genes and associated pathways, epigenetics provides an elegant mechanism for time and tissue-specific expression of key biologic molecules across the life span.
The available data support the notion that epigenetic processes can be influenced by both exogenous and endogenous factors. Specific agents in the areas of nutrition, environmental chemicals, stressors, and pharmaceutical agents have been shown to affect the epigenetic state. Few data are available to address the potential relationship between such changes and adverse functional outcomes. The role of epigenetics in the regulation of normal development as well as the impact of exogenous influences (e.g., environmental chemicals, stressors, nutrition, drugs) on epigenetic pathways is an emerging area of interest.
The National Institute of Environmental Health Sciences (NIEHS) is developing a comprehensive program that addresses how gene expression is altered under different forms of environmental stress and how this alters the risk of developing disease.
Two recent events signal the role NIEHS is undertaking to increase support of research to enhance understanding of how environmental perturbations occurring throughout the life span can affect phenotypes at different stages of life and/or transgenerationally by altering gene expression though modification of DNA methylation and chromatin structure. First, on November 2–4, in collaboration with Duke University Medical Center, NIEHS co-sponsored an international symposium titled Environmental Epigenomics, Imprinting and Disease Susceptibility (http://www.geneimprint.com/meetings/2005durham/). Second, as part of a series of solicitations to be released over the next 5 years, NIEHS announced RFA ES-05-007, Environmental Influences on Epigenetic Regulation, to support research on environmental stressor effects on gene silencing in somatic cells and altered methylation profiles that are related to disease outcomes at various stages of the life span.
These solicitations are anticipated to stimulate research directed toward the understanding of the role of epigenetic modulation of gene expression in a variety of diseases and dysfunctions and the modulation of epigenetic regulation by environmental chemicals either alone or in combination with altered nutrition. Other areas of emphasis could include understanding the role of epigenetics in the transgenerational effects of environmental agents, the role of environmental stressors in the suppression of gene expression, the identification of epigenetic targets in the genome sensitive to environmental modification, and the mechanism of interaction of environmental chemicals with potential modifiers of epigenetic processes.
Contact
Frederick L. Tyson, PhD |
[email protected]
Jerrold Heindel, PhD |
[email protected]
|
0
|
PMC1314949
|
CC0
|
2021-01-04 23:41:30
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A839
|
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-a00840AnnouncementsFellowships, Grants, & AwardsFellowships, Grants, & Awards 12 2005 113 12 A840 A841 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
Ecology and Oceanography of Harmful Algal Blooms
The U.S. Environmental Protection Agency (EPA), as part of its Science to Achieve Results (STAR) program, and its interagency partners, the National Oceanographic and Atmospheric Administration (NOAA), the National Aeronautic and Space Administration (NASA), and the Office of Naval Research (ONR), are seeking applications proposing targeted research projects of up to 3 years duration and, depending on appropriations, multidisciplinary regional studies for 3–5 years for the Ecology and Oceanography of Harmful Algal Blooms (ECOHAB) program. This program supports research on algal species whose populations may cause or result in deleterious effects on ecosystems and human health. Studies of the causes of such blooms, their detection, effects, mitigation, and control in U.S. coastal waters (including estuaries and Great Lakes) are solicited.
Harmful algal blooms (HABs) are caused by a diverse group of organisms, including toxic and noxious phytoplankton, some protists, cyanobacteria, benthic algae, and macroalgae. While some HABs occur naturally, others may be stimulated by human activities. Blooms can extend over large geographic areas, be composed of more than one harmful or toxic species, and cause significant impacts on fisheries, recreation, human health, and the ecology of both marine and fresh water bodies. HABs are now a recurrent and serious problem in many areas of the United States, and evidence suggests that the frequency and distribution of HABs is also increasing globally, impacting many countries that have commercial and recreational activities in the coastal ocean.
HAB impacts on public health and local/regional economies are also dramatic and increasing. In a recent study, average annual economic losses in the United States from HABs were approximated at $49 million, with costs attributable to maintenance of toxin monitoring programs; closures of shellfish beds; marine mammal stranding networks; collapse of some fisheries; mortality of fish, shellfish, turtles, birds, and mammals; disruptions in tourism; threats to public and coastal resource health; publication of watershed, health, and seafood advisories; and medical treatments (Anderson et al. 2000, available at http://www.whoi.edu/redtide/pertinentinfo/Economics_report.pdf). Despite greater public awareness and advisories of bloom events, human illnesses and even fatalities continue to be reported. Additionally, some toxins may cause only a few documented illnesses but result in serious public reaction and temporary aversion to local seafood products and activities (e.g., $46 million in lost revenue from; the 1997 Maryland fish health/Pfiesteria events; Anderson et al. 2000). These deleterious impacts have increased public awareness and demand for intervention to reduce or eliminate bloom impacts on coastal resources, local economies, and threats to public health.
Over the course of the last decade, numerous national and agency reports have described the magnitude of the HAB problem and outlined research plans to systematically address the issue. The ECOHAB Program was initiated a decade ago as an interagency, scientific program designed to increase the understanding of the fundamental processes underlying the impacts and population dynamics of HABs (ECOHAB 1995). Three major research themes encompassing the priorities of issues of national importance on the HAB phenomenon were identified: 1) organisms, with a goal towards determining the physiological, biochemical, and behavioral features that influence bloom dynamics; 2) environmental regulation, with a goal toward determining and parameterizing the factors that govern the initiation, growth, and maintenance of these blooms; and 3) food web and community interactions, with a goal toward determining the extent to which food webs and trophic structure affect and are affected by the dynamics of HABs. Information in these areas, in turn, supported a critical goal of the ECOHAB program, the development of reliable models to forecast bloom development, persistence, and toxicity. Since 1997, the ECOHAB Program has sponsored nearly 100 projects with topics ranging from molecular aspects of HAB detection to large-scale, multidisciplinary regional studies of bloom formation, maintenance, and dissipation. Projects cover a wide spatial spectrum along the U.S. coastline and its territories. ECOHAB-sponsored projects also address the detection, prevention, control, and mitigation of HABs and their impacts, as well as economic assessments of these recurring events. Project summaries may be viewed at http://www.whoi.edu/science/B/redtide/nationplan/ecohabprojectsummaries.html
Although several research efforts have been completed or are underway, the understanding of the biological, physical, and chemical processes that regulate HABs remains limited. Toxic blooms can impact virtually all compartments of marine food webs, resulting in adverse effects on metabolism, viability, growth, fecundity, and recruitment of marine organisms. HAB-produced toxins can have immediate, acute impacts on marine populations, including marine mammals, birds, and several protected species. Little is known about the effects of chronic, low-level exposure. Dramatic shifts in ecosystem structure can result from plankton blooms and macroalgal overgrowth in benthic systems. In this context, present knowledge is inadequate to define the scale and complexity of many HAB phenomena.
As a result, an additional focus on the early detection of bloom species, the environmental conditions supporting blooms, and the toxins associated with some HAB species is needed. Further, while there is increasing emphasis on manipulating coastal waters to prevent or control HABs in other nations, it is practically absent from U.S coastal management strategies. Finally, there needs to be greater emphasis on ensuring that coastal managers and the public are provided the most current information available in a manner that will maximize its usefulness in mitigating HAB impacts. This would include use of observing systems and models in the development of HAB forecasts.
The solicitation closing date is January 10, 2006. Funding is contingent upon receipt of fiscal years 2006–2010 federal appropriations. It is anticipated that a total of approximately $7–10 million will be awarded, depending on the availability of funds. The agency partners anticipate awarding approximately 15–20 funding agreements under this solicitation, including two regional projects. Awards for targeted studies are typically on the order of $150,000 per year, total costs, for up to 3 years. Multi-investigator and multi-institutional applications may include correspondingly higher budgets and longer project periods, but may not exceed a 5-year project period.
You may submit either a paper application or an electronic application (but not both) for this solicitation. The necessary forms for submitting a paper application will be found on the National Center for Environmental Research website, http://es.epa.gov/ncer/rfa/forms/. To apply electronically you must use the application package available at https://apply.grants.gov/forms_apps_idx.html (see "Submission Instructions for Electronic Applications") plus some additional forms from http://es.epa.gov/ncer/rfa/forms/
Contact: Bronda Harrison, 202-343-9777, e-mail:
[email protected]
Implications of Tropospheric Air Pollution for Surface UV Exposures
The U.S. Environmental Protection Agency (EPA), as part of its Science to Achieve Results (STAR) program, is seeking applications proposing research to better understand the effect of tropospheric pollution (ozone and particulate matter) on surface ultraviolet radiation levels.
Exposure to ultraviolet radiation (UV) has significant health and ecosystem impacts. Under current conditions, one in five Americans will develop skin cancer in their lifetime and one American dies every hour from this devastating disease. Changes in air quality affect UV exposures in ways that we do not fully understand. The goal of this solicitation is to enable research to better understand how changes in tropospheric ozone and particulate matter will alter surface UV exposures.
Several different U.S. government agencies operate UV monitoring programs that employ instrumentation falling into three distinctly different categories: broad-band sensors of radiation intensity within a spectral region between two specified levels, narrow-band instruments measuring radiation in several well-defined wavelength intervals, and spectral devices providing detailed information on radiation intensity as a function of wavelength. Details about the various UV monitoring networks can be found via the web at: http://www.arl.noaa.gov/research/programs/uv.html
Between 1996 and 2004, EPA, in conjunction with the National Park Service, operated a 21-site monitoring network with a goal of characterizing UV exposures and trends in a range of ecoregions across the United States. This 21-site network used Brewer Mark IV single-monochromator spectrophotometers, which measure UV radiation levels reaching the Earth’s surface at 0.5 nm increments over the wavelength range 286.3–363 nm, providing information on UV-B and UV-A exposures. The data from this network are available via the web at http://www.epa.gov/uvnet/. A review of the data collected by this network was recently completed and is available in draft form at http://www.geecresearch.com/EPAUV.htm
EPA is currently working with the National Oceanographic and Atmospheric Administration’s (NOAA’s) Central UV Calibration Facility (CUCF) to redeploy its Brewer instruments at six sites, some of which are part of networks operated by NOAA and the U.S. Department of Agriculture. The new network will include a triad of Brewers at CUCF’s Table Mountain Test Facility, where the Brewers will be compared to NOAA’s reference spectro-radiometer. The new network configuration is intended to help characterize the performance of the Brewers and to collect information relevant to quantifying the effects of tropospheric ozone and fine particles on surface UV levels and radiative forcing.
The solicitation closing date is January 18, 2006. It is anticipated that a total of approximately $600,000 will be awarded under this announcement, depending on the availability of funds. The EPA anticipates funding approximately two grants under this RFA. The projected award per grant is $100,000 per year total costs, for up to 3 years. Requests for amounts in excess of a total of $300,000, including direct and indirect costs, will not be considered. The total project period for an application submitted in response to this RFA may not exceed 3 years. The EPA reserves the right to reject all applications and make no awards under this RFA. The EPA reserves the right to make additional awards under this RFA if additional funding becomes available. Any additional selections for awards will be made no later than 4 months after the original selection decisions.
You may submit either a paper application or an electronic application (but not both) for this announcement. Instructions for both forms of submission follow. For paper applications, forms can be found on the NCER web site: http://es.epa.gov/ncer/rfa/forms/. For electronic applications, use the application package available at https://apply.grants.gov/forms_apps_idx.html (see “Submission Instructions for Electronic Applications”).
Contact: Bronda Harrison, 202-343-9777; e-mail:
[email protected]
Interdisciplinary Partnerships in Environmental Health Sciences
The mission of the National Institute of Environmental Health Sciences (NIEHS) is to promote research that will ultimately reduce the burden of human illness and dysfunction from environmental causes. The research supported by the NIEHS addresses this mission through a diverse grants portfolio consisting of basic in vitro and animal research, population-based studies, and a limited number of patient-oriented studies that focus on the understanding, detection, prevention, and intervention of environmentally related disease and disease processes. Recent technological advances and a growing appreciation that environmental factors contribute to most complex diseases provide unprecedented opportunities for developing new research paradigms that bring together interdisciplinary teams of scientists to move basic environmental health sciences research into clinical and public health practice.
The objective of this initiative is to foster scientific collaboration between clinical and basic investigators to accelerate the application of basic research results into the clinical setting to improve human health in those areas where environmental factors are known or expected to influence the development or progression of human disease. Scientific knowledge achieved through this research program is expected to move the field of environmental health sciences into new directions and approaches for the identification, treatment, and prevention of environmentally related diseases or disorders. Through this initiative, the NIEHS will support both the development of new collaborations between researchers with basic and clinical expertise and the continued efforts of existing collaborations. Both activities must directly support the integration of clinical and basic science research.
For the purpose of this solicitation, clinical research is defined as: 1) patient-oriented clinical research conducted with human subjects, or research on the causes and consequences of disease in human populations involving material of human origin (such as tissue or specimens) and for which an investigator or colleague directly interacts with human subjects in an outpatient or inpatient setting to clarify a problem in human physiology, pathophysiology, or disease; 2) development of new clinically based technologies, therapeutic interventions, or clinical trials; 3) epidemiologic and behavioral studies in humans; such studies are appropriate in cases where the primary focus of the study is on a specific disease or disorder and the clinical investigator is an essential part of the planning, conduct, and analysis of the study.
Basic science research is defined as mechanistic research using experimental approaches and may include use of cell lines, in vitro or in vivo models. Basic research may include the development of new tools to expand the capacity of clinically oriented research.
Applications to this solicitation may be either exploratory in nature, laying the foundation of long-term collaborations or attaining proof of principle for an innovative collaborative approach, or they may be continued development of established collaborations. In either instance, applications must focus on a specific human disease or disorder where there is evidence or a strong rationale for the involvement of environmental factors in its etiology or phenotypic expression and must involve research at both the basic and clinical levels. Suggested topics and example research projects include, but are not limited to, first, integration of basic mechanism-driven and clinical patient-oriented research to gain new insights into the role of environmental factors in complex human diseases: 1) integration of patient phenotype data with high data content techniques such as transcriptomics, proteomics, and metabolomics to investigate the mechanisms by which exposures lead to disease; 2) comparison of animal/model organism and human responses to toxicants to identify biological alterations contributing to the disease etiology; 3) examination of the impact of current and/or prior environmental exposures on the progression, treatment, and survival of patients with existing disease.
Second are collaborations between basic, patient-oriented, and epidemiologic researchers to identify and validate biomarkers and apply them to the development or progression of human diseases: 1) application of omics technologies and modeling to existing populations to identify predictive marker profiles for known exposures and the genesis of disease; 2) use of mechanistic response data obtained from the conduct of basic science experimentation to identify putative predictive markers of exposure and response and their validation in existing cohorts; 3) coordination of engineering and clinical expertise to develop systems that integrate exposure with individual biological response or phenotypic change; 4) innovative approaches to the identification of the determinants of individual susceptibility and the interaction between genes and environmental stressors in human disease.
Third is use of environmental stressors as a probe(s) to identify phenotypic variation in humans and animal models of disease to inform genetic analyses of disease susceptibility: 1) conduct of genome-wide association studies to study gene, gene–gene and gene–environment interactions in well characterized cases of environmentally induced disease; 2) identification and assessment of the functional relevance of SNPs and haplotypes associated with environmentally induced disease and the mechanistic consequences of those variations at the molecular and cellular levels.
Fourth are interdisciplinary approaches to the development of intervention and prevention strategies to alter the progression of environmentally induced human disease: 1) identification of novel compounds or engineering of biocompatible materials that protect against, inhibit, or reverse toxicant actions and their validation in disease relevant settings; 2) development of mechanistically derived prevention strategies and application of them in populations with known genetic susceptibility to environmental stressors.
Fifth is collaborative development or refinement and application of model systems that faithfully replicate human disease condition or species comparisons that can be used to understand environmentally induced human disease processes. NIEHS encourages applicants to utilize existing biological and or other resources to address the topics described above where applicable, such as: 1) new analyses of data from completed studies; 2) assay of archived biological samples from completed and ongoing studies; 3) collection of new data and samples from ongoing or completed studies; 4) analysis of public and other accessible databases.
Proposals addressing perturbation of biological processes in the absence of extension to human disease will be considered nonresponsive in the context of this solicitation.
This funding opportunity will use the NIH Exploratory/Developmental Grant (R21) and Research Project Grant (R01) award mechanisms. As an applicant, you will be solely responsible for planning, directing, and executing the proposed project.
This funding opportunity uses just-in-time concepts. It also uses the modular as well as the non-modular budget formats (see http://grants.nih.gov/grants/funding/modular/modular.htm). Specifically, if you are submitting an application with direct costs in each year of $250,000 or less, use the modular budget format described in the PHS 398 application instructions. Otherwise follow the instructions for nonmodular research grant applications.
The PHS 398 application instructions are available at http://grants.nih.gov/grants/funding/phs398/phs398.html in an interactive format. Applicants must use the currently approved version of the PHS 398. For further assistance, contact GrantsInfo at 301-435-0714 (telecommunications for the hearing impaired: TTY 301-451-0088) or by e-mail:
[email protected]
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.
The letters of intent receipt dates for this PAR are December 11, 2005, 2006, 2007, with the application receipt dates January 11, 2006, 2007, 2008. The complete version of this PA is available at http://grants/guide/pa-files/PAR-05-168
Contact: Cindy Lawler, Cellular, Organs and Systems Pathobiology Branch, Division of Extramural Research and Training, National Institute of Environmental Health Sciences, P.O. Box 12233, MD EC-23, Research Triangle Park, NC 27709 USA, 919-316-4671, fax: 919-541-5064, e-mail:
[email protected]; Kimberly Gray, Susceptibility and Public Health Branch, Division of Extramural Research and Training, National Institute of Environmental Health Sciences, P.O. Box 12233, MD EC-21, Research Triangle Park, NC 27709 USA, 919-541-0293, fax: 919- 316-4606, e-mail:
[email protected]. Reference PAR-05-168
|
0
|
PMC1314950
|
CC0
|
2021-01-04 23:41:31
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A840-A841
|
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-a0800a16330330PerspectivesCorrespondencePesticides and Neurologic Symptoms Burns Carol The Dow Chemical Company, Midland, Michigan, E-mail:
[email protected] Daniel A. The Monsanto CompanyBoth authors are employed by companies that manufacture pesticides.
12 2005 113 12 A800 A800 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 read with interest the recent study titled “Neurologic Symptoms in Licensed Private Pesticide Applicators in the Agricultural Health Study” (Kamel et al. 2005). Although this was a hypothesis-generating study, the authors speculated regarding moderate exposure and associations with neurologic symptoms. Substantiation of hypotheses requires meaningful metrics of exposure and effect, and depends on exclusion and analysis of competing hypotheses for the observations. In our opinion, the article by Kamel et al. falls seriously short in several regards and requires additional data in order to provide credible and defensible conclusions.
Kamel et al. (2005) analyzed a number of symptoms in those “ever” experiencing one of 23 self-reported symptoms in the preceding 12 months. The biologic significance of the outcome “symptom count” is unknown; also, “multiple symptoms” is not a definable disease or illness. The fact that private applicators report headache, nausea, and fatigue does not establish that each is of neurologic origin, particularly given the physical requirements of farming. Indeed, results of the questions used by Kamel et al. (2005) have been shown to agree poorly with objective tests of neurologic function (Lundberg et al. 1997). Further, Kamel et al. limited the analyses to a single episode rather than symptoms that were reported more than once per year (Kamel et al. (2005); Table 2). As a cross-sectional analysis, the data do not permit assessment of the temporal relationship between exposure and symptom onset, and no consideration was given to the transient nature of the reported symptoms. Thus, although the nature of the analysis implies some sort of persistent neurologic condition underlying the reporting of symptoms, no such condition can be established from intermittent symptoms of indeterminate etiology.
In addition to other potential causes for these symptoms, researchers have warned about the role of psychosocial factors in the reporting of non-specific symptoms. According to Spurgeon et al. (1996),
Many occupational and environmental health hazards present as an increased reporting of non-specific symptoms such as headache, backache, eye and respiratory irritation, tiredness, memory problems, and poor concentration. The pattern and number of such symptoms is surprisingly constant from hazard to hazard suggesting that common psychological and social factors, not directly related to the exposure may be involved.
The role of these factors has been well documented in the psychological literature. Such factors include attitudes and belief systems; current or preexisting stress; workers’ perception of the competence and credibility of management; and involvement of the media, pressure groups, and the legal system (Spurgeon et al. 1997). Further, “[p]revention and control strategies are unlikely to be successful if the real sources of the problems are not correctly identified” (Spurgeon et al. 1997).
Because Kamel et al. (2005) relied on self-reported days of application to infer exposure rather than actual measured dose, their assumption of sufficient exposure to cause a biologic effect has severe limitations. The Agricultural Health Study (AHS) has used lifetime exposure days for specific, individual pesticides in other publications (Alavanja et al. 2003, 2004; Engel et al. 2005), but Kamel et al. (2005) offered no support for their change in approach and the validity of a class-wide, rather than pesticide-specific biologic effect. Furthermore, studies indicate that farmers have much less pesticide exposure than is often assumed from self-reported use and even within this low range; the exposure is variable for a given day. For example, in a study of organophosphate applicators, Stokes et al. (1995) identified differences in urinary metabolite levels based on the number of tanks loaded, acres sprayed, and hours sprayed. Other bio-monitoring studies have identified a large range of exposure for different pesticides, including applicators with no detectable exposure (Arbuckle et al. 2002; Mandel et al. 2005). The exposure metric used by Kamel et al. (2005) of cumulative lifetime days applied most likely overestimates exposur, in light of these exposure studies of farmer applicators.
We believe that the findings of Kamel et al. (2005) may well be the result of evaluating multiple pesticides as groups at a time in conjunction with other physical or emotional stress related to farming or even a common ailment such as influenza (Dunn et al. 1995). In any event, the conclusions are not justified by the data because there is no coherent disease outcome and no meaningful exposure metric. It is our view that even hypotheses generated by such non-specific data do not meet the stated AHS objective, which is to “provide information that agricultural workers can use in making decisions about their health and the health of their families” (AHS 2005).
==== Refs
References
Agricultural Health Study 2005. Homepage. Available: http://www.aghealth.org [accessed 1 November 2005].
Alavanja MCR Dosemeci M Samanic C Lubin J Lynch CF Knott C 2004 Pesticides and lung cancer risk in the Agricultural Health Study cohort Am J Epidemiol 160 876 885 15496540
Alavanja MCR Samanic C Dosemeci M Lubin J Tarone R Lynch CF 2003 Use of agricultural pesticides and prostate cancer risk in the Agricultural Health Study cohort Am J Epidemiol 157 800 114 12727674
Arbuckle TE Burnett R Cole D Teschke K Dosemeci M Bancej C 2002 Predictors of herbicide exposure in farm applicators Int Arch Occup Environ Health 75 406 414 12070637
Dunn JT Lees-Haley PR Brown RS Williams CW English LT 1995 Neurotoxic complaint base rates of personal injury claimants: implications for neuropsychological assessment J Clin Psychol 51 577 584 7593681
Engel LS Hill DA Hoppin JA Lubin JH Lynch CF Pierce J 2005 Pesticide use and breast cancer risk among farmers' wives in the Agricultural Health Study Am J Epidemiol 161 121 135 15632262
Kamel F Engel LS Gladen BC Hoppin JA Alavanja MCR Sandler DP 2005 Neurologic symptoms in licensed private pesticide applicators in the Agricultural Health Study Environ Health Perspect 113 877 882 10.1289/ehp.7645 [Online 15 April 2005].16002376
Lundberg I Hogberg M Michelsen H Nise G Hogstedt C 1997 Evaluation of the Q16 questionnaire on neurotoxic symptoms and a review of its use Occup Environ Med 54 343 350 9196457
Mandel JS Alexander BH Baker BA Acquavella JF Chapman P Honeycutt R 2005 Biomonitoring for farm families in the Farm Family Exposure Study Scand J Work Environ Health 31 suppl 1 98 104 16190155
Spurgeon A Gompertz D Harrington JM 1996 Modifiers of non-specific symptoms in occupational and environmental syndromes Occup Environ Med 53 361 366 8758029
Spurgeon A Gompertz D Harrington JM 1997 Non-specific symptoms in response to hazard exposure in the work-place J Psychosom Res 43 43 49 9263930
Stokes L Stark A Marshall E Narang A 1995 Neurotoxicity among pesticide applicators exposed to organophosphates Occup Environ Med 52 648 653 7489054
|
16330330
|
PMC1314951
|
CC0
|
2021-01-04 23:41:31
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A800a
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.113-1314952
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0800b16330330PerspectivesCorrespondencePesticides and Neurologic Symptoms: Kamel et al. Respond Kamel Freya Gladen Beth C. Hoppin Jane A. Sandler Dale P. National Institute of Environmental Health Sciences, National Institutes of Health, Department of Health and Human Services, Research Triangle Park, North Carolina, E-mail:
[email protected] Lawrence S. Memorial Sloan-Kettering Cancer Center, New York, New YorkAlavanja Michael C.R. National Cancer Institute, National Institutes of Health, Department of Health and Human Services, Bethesda, MarylandThe authors declare they have no competing financial interests.
12 2005 113 12 A800 A801 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
Burns and Goldstein raise several issues regarding our paper (Kamel et al. 2005), in which we reported that applicators chronically exposed to moderate levels of pesticide experience more neurologic symptoms. They assert that our measures of exposure and effect are not “meaningful.” We disagree.
Burns and Goldstein state that “‘multiple symptoms’ is not a definable disease or illness.” Although this is true, symptoms cause many medical visits and so are significant to public health. Further, we made no claim that applicators reporting more symptoms had a particular disease. Indeed, in some of our analyses we purposely excluded individuals reporting neurologic disease in order to evaluate associations of pesticide use specifically with symptoms. We studied a mixed group of symptoms, all sometimes associated with neurologic dysfunction or disease, although with varying specificity. Excluding two relatively nonspecific symptoms (headache and fatigue) did not appreciably change the distribution of the symptom variable. The assertion that we limited our analysis to ”a single episode” is inaccurate: our main analyses evaluated multiple rather than single symptoms, and we took symptom frequency into account in our analysis of individual symptoms (Kamel et al. 2005; Table 4). We acknowledged the limitations of cross-sectional analysis in our article. However, the associations we observed were with cumulative pesticide use; accounting for recent use did not change results.
The issue is not whether the symptoms we studied are diagnostic of neurologic or other disease, but whether experiencing these symptoms is associated with pesticide exposure. Burns and Goldstein cite Lundberg et al. (1997) but omit Lundberg et al.’s conclusion that the exposure-related relationship of symptom reporting to organic solvent exposure makes this approach useful for comparing groups with different exposures. At least 23 previous studies used symptom reporting to evaluate neurologic effects of pesticide exposure, with 19 reporting positive associations (Kamel and Hoppin 2004). We extended this approach to a very large group of applicators who had detailed exposure information available.
Burns and Goldstein discuss potential factors related to symptoms, citing Spurgeon’s biopsychosocial model (Spurgeon et al. 1996). We agree that personal and social factors likely influence both experience and reporting of symptoms. However, Spurgeon (2002) noted that
Discussion of the determinants of symptom reporting does not constitute a dismissal of the farmer’s illness but simply a recognition that it is likely to result from a complex interaction of physical, psychological, and social processes.
She described a study of farmers whose symptoms were associated with five factors, one being handling sheep within 48 hr of pesticide dipping. Thus, pesticide exposure may still be associated with increased symptoms even (or perhaps especially) when psychosocial factors are taken into account. Most of the factors Burns and Goldstein list are unlikely to be related to exposure in licensed applicators and so cannot explain the associations seen. Further, confounding by psychosocial factors would likely produce associations with all types of pesticides, but our findings were specific to insecticides. Finally, we do not understand Burn’s and Goldstein’s comment that our findings are “the result … of a common ailment such as influenza”; are they suggesting that pesticide exposure is associated with increased risk of flu?
Burns and Goldstein call our exposure measures limited, citing biomonitoring studies which show that variations in internal exposure are not completely correlated with external exposure. This point is largely irrelevant because the associations seen depend not on identifying the absolute level of pesticide exposure but rather on ranking applicators as relatively more or less exposed. Variation in the degree to which self-reported days of use represents internal exposure is probably nondifferential with respect to symptom reporting, with resulting misclassification likely to bias associations towards the null; the true relationship may be stronger than we observed. Our findings of associations with insecticides only, and with cumulative but not recent exposure, suggest that recall bias does not fully account for our results. We see no problem in combining pesticides for a class-wide analysis, particularly because many grouped pesticides exert effects through similar or related biologic mechanisms. Using class-wide analyses may minimize confounding because most applicators used multiple pesticides. Ultimately, it will be interesting to evaluate the effects of individual chemicals; we are planning such studies.
Thus, our measures of both exposure and effect are sufficient for their purpose, which is to examine the association of symptom reporting with moderate insecticide exposure. Our study clearly demonstrates such an association. Importantly, it is independent of both recent exposure and a history of high exposure or poisoning, suggesting that lifetime exposure at moderate levels may have health consequences. This finding has implications for farmers’ health and deserves to be reported and evaluated further.
==== Refs
References
Kamel F Engel LS Gladen BC Hoppin JA Alavanja MCR Sandler DP 2005 Neurologic symptoms in licensed private pesticide applicators in the Agricultural Health Study Environ Health Perspect 113 877 882 16002376
Kamel F Hoppin JA 2004 Association of pesticide exposure with neurologic dysfunction and disease Environ Health Perspect 112 950 958 15198914
Lundberg I Hogberg M Michelsen H Nise G Hogstedt C 1997 Evaluation of the Q16 questionnaire on neurotoxic symptoms and a review of its use Occup Environ Medicine 54 343 350
Spurgeon A 2002 Models of unexplained symptoms associated with occupational and environmental exposures Environ Health Perspect 110 suppl 4 601 605 12194893
Spurgeon A Gompertz D Harrington JM 1996 Modifiers of non-specific symptoms in occupational and environmental syndromes Occup Environ Med 53 361 366 8758029
|
0
|
PMC1314952
|
CC0
|
2021-01-04 23:41:31
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A800b-A801
|
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-a0802a16330332PerspectivesCorrespondenceTungsten Alloy and Cancer in Rats: Kalinich Responds Kalinich John F. Armed Forces Radiobiology, Research Institute, Bethesda, Maryland, E-mail:
[email protected] author declares he has no competing financial interests.
12 2005 113 12 A802 A802 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 would like to address Schell’s comments about our article published in EHP (Kalinich et al. 2005). Schell expresses concern about certain statements we made in our article about embedded tungsten alloy fragments, especially our reference to the undisputed Centers for Disease Control and Prevention (CDC) finding that there is an increased incidence of childhood leukemia in areas where there are high levels of environmental tungsten (CDC 2003; Sheppard and Witten 2004). Schell contends that our results showing the carcinogenic potential of embedded tungsten alloy fragments have no bearing on the situation in Fallon, Nevada, and believes that our mentioning them “is both inappropriate and misleading.” We respectfully disagree.
In our article (Kalinich et al. 2005) we report an unexpected response in rats to tungsten alloys that could not have been predicted by looking at tungsten toxicity alone. We suggested that our results support the advisability for further consideration of tungsten compounds or synergistic effects of tungsten with other environmental factors in cases such as Fallon. We cited several reports in support of such a view. Miller et al. (2001, 2002) indicated that the presence of tungsten in an in vitro model system increased the toxicity of both nickel and cobalt in a synergistic manner. Wei et al. (1985, 1987) reported that tungsten exhibited a promoting effect on N-nitroso-N-methylurea–induced mammary carcinogenesis in rats. Other investigators have also suggested the cause for the Fallon cancer cluster might be an as yet uninvestigated factor or the result of simultaneous or sequential exposure to one or more agents (Daughton 2005).
At this time it is not clear whether these similar findings from diverse research are an unrelated coincidence or whether they suggest a toxicologic property of tungsten not yet understood. What is clear, however, is the need for further research in this area, not only a toxicologic assessment of tungsten alone but also potential synergistic interactions with known toxic agents. The currently proposed National Toxicology Program study of tungsten is an important first step in resolving these issues.
==== Refs
References
CDC 2003. Cross-Sectional Exposure Assessment of Environmental Contaminants in Churchill County, Nevada. Final Report. Atlanta, GA:Centers for Disease Control and Prevention. Available: http://www.cdc.gov/nceh/clusters/Fallon/study.htm [accessed 2 November 2005].
Daughton CG 2005 Overlooked in Fallon? [Letter] Environ Health Perspect 113 A224 A225 15811811
Kalinich JF Emond CA Dalton TK Mog SR Coleman GD Kordell JE 2005 Embedded weapons-grade tungsten alloy shrapnel rapidly induces metastatic high-grade rhabdomyosarcomas in F344 rats Environ Health Perspect 113 729 734 15929896
Miller AC Mog S McKinney L Luo L Allen J Xu J 2001 Neoplastic transformation of human osteoblast cells to the tumorigenic phenotype by heavy-metal tungsten-alloy metals: induction of genotoxic effects Carcinogenesis 22 115 125 11159749
Miller AC Xu J Prasanna PGS Page N 2002 Potential late health effects of the heavy metals, depleted uranium and tungsten, used in armor piercing munitions: comparison of neoplastic transformation and genotoxicity using the known carcinogen nickel Mil Med 167 120 122 11873492
Sheppard PR Witten ML 2004. Heavy metal content in airborne dust of childhood leukemia cluster areas: even small towns have air pollutants [Abstract]. Eos Trans American Geophysical Union 85: Fall meeting suppl., abstract A11A-0025. Available: http://www.agu.org/cgi-bin/SFgate/SFgate?&listenv=table&multiple=1&range=1&directget=1&application=fm04&database=%2Fdata%2Fepubs%2Fwais%2Findexes%2Ffm04%2Ffm04&maxhits=200&=%22A11A-0025%22 [accessed 4 November 2005].
Wei HJ Luo XM Yang SP 1985 Effects of molybdenum and tungsten on mammary carcinogenesis in SD rats J Natl Cancer Inst 74 469 473 3856053
Wei HJ Luo XM Yang SP 1987 Effect of molybdenum and tungsten on mammary carcinogenesis in Sprague-Dawley (SD) rats [in Chinese] Zhonghua Zhong Liu Za Zhi 9 204 207 3447862
|
0
|
PMC1314953
|
CC0
|
2021-01-04 23:41:31
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A802a
|
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-a0802b16330332PerspectivesCorrespondenceSynthetic Musk Compounds and Effects on Human Health? Salvito Daniel Research Institute for Fragrance Materials, Inc., Woodcliff Lake, New Jersey, E-mail:
[email protected] author is employed by the Research Institute for Fragrance Materials, a nonprofit organization that publishes its work in peer-reviewed literature under the guidance of an independent scientific panel and receives support from the private sector.
12 2005 113 12 A802 A803 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 recent article by Luckenbach and Epel (2005) on in vitro observations of inhibitory properties exhibited by certain nitromusk and polycyclic musk fragrance ingredients on mussel cells raised some concerns regarding potential environmental risks and safety to humans that may be associated with nitromusk and polycyclic musk compounds. The Research Institute for Fragrance Materials would like to address several points that may help readers more clearly understand the meaning and context of the reported research.
The tonnages of musk compounds reported by Luckenbach and Epel (2005) in their article (7,000–8,000 tons) are higher than the industry-reported global tonnage of these materials. From 1995 to 2000, the total worldwide usage declined from 300 tons to 200 tons for musk xylene and musk ketone combined. The 2000 worldwide use of polycyclic musks is approximately 4,000 tons.
Measured concentrations of these compounds in the environment are less than the effects concentrations reported by Luckenbach and Epel (2005). In a review of measured environmental concentrations, Rimkus (1999) stated that the highest reported measurement of hexahydrohexamethyl-cyclopenta (γ)-2-benzopyran (HHCB) in surface water was 12.5 μg/L (0.048 μM). The IC50 (concentration that inhibits 50%) reported for polycyclic musks was 2.34 μM. Overall, measured environmental concentrations were 2–6 orders of magnitude lower than the effects concentrations reported by Luckenbach and Epel (2005).
The data reported by Luckenbach and Epel (2005) reflect a method under development. There are many steps between the observation of an in vitro effect and effects on whole organisms, ecosystems, and humans. In vivo studies in mussels and studies linking mussel gill tissue to undefined tissues in mammals and humans are some of the research necessary to conclude that these higher level effects may exist. These effects would then need to be placed into a risk-based context by comparing them to exposure concentrations.
The safety of nitromusk and polycyclic musk compounds for humans has been extensively tested and affirmed by numerous regulatory agencies and academic scientists around the world [Scientific Committee on Cosmetic Products and Non-Food Products (SCCNFP) 2002a, 2002b]. The trace environmental levels of the musks continue to be investigated, and environmental safety and monitoring studies are ongoing so that the public can be assured of their safety.
Regarding the environmental effects of synthetic musks, the IC10 (concentration that inhibits 10%) values should be compared to no observed effect concentrations (NOECs). The IC10 values of the synthetic musks are around the level of the lowest in vivo NOECs observed for aquatic organisms.
Table 1 shows that the in vitro multi-xenobiotic resistance (MXR) transporter activity in mussel gill is of the same sensitivity as the effects observed in the standard toxicity tests with aquatic organisms. Thus, at the exposure level where the protective transporter efflux is decreased, rendering the cell more accessible to other potential toxicants, the effects of the synthetic musks are also indicated in other end points, such as development and growth.
The observed effects are not limited to the nitromusk and polycyclic musk compounds. For example, the other chemicals used by Luckenbach and Epel (2005)—verapamil and quinidine—also produced the phenomenon. In the case of verapamil, the IC10 was reported at 1–2 orders of magnitude below the nitromusks and polycyclic musks.
I look forward to continued discussions with the Luckenbach and Epel to determine the relevance of the results of this study.
Table 1 Intensity of in vitro MXR transporter activity in mussel gill.
IC10) (mg/L) Lowest NOEC (mg/L
MK 0.14 mmol = 0.041 0.063
MX 0.09 mmol = 0.027 0.056
AHTN 0.35 mmol = 0.090 0.035
HHCB 0.37 mmol = 0.095 0.068
Abbreviations: MK, musk ketone; MX, musk xylene.
==== Refs
REFERENCES
Luckenbach T Epel D 2005 Nitromusk and polycyclic musk compounds as long-term inhibitors of cellular xenobiotic defense systems mediated by multi-drug transporters Environ Health Perspect 113 17 24 15626642
Rimkus G 1999 Polycyclic musk fragrances in the aquatic environment Toxicol Lett 111 37 56 10630702
SCCNFP 2002a. Opinion of the Scientific Committee on Cosmetic Products and Non-Food Products Intended for Consumers Concerning 6-Acetyl-1,1,2,4,4,7-Hexamethyl-tetraline (AHTN). SCCNFP/0609/02. Available: http://europa.eu.int/comm/health/ph_risk/committees/sccp/documents/out176_en.pdf [accessed 2 November].
SCCNFP 2002b. Opinion of the Scientific Committee on Cosmetic Products and Non-Food Products Intended for Consumers Concerning Hexahydro-Hexamethyl-Cyclopenta (γ)-2-Benzopyran (HHCB). SCCNFP/0610/02. Available: http://europa.eu.int/comm/health/ph_risk/committees/sccp/documents/out179_en.pdf [accessed 2 November].
|
16330332
|
PMC1314954
|
CC0
|
2021-01-04 23:41:30
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A802b-A803
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.113-a802b
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0805a16330334PerspectivesCorrespondencePesticide Testing on Humans: Resnick and Portier Respond Resnik David B. Office of the Scientific Director, National Institute of Environmental Health Sciences, National Institutes of Health, Department of Health and Human Services, E-mail:
[email protected] Christopher National Toxicology Program, National Institute of Environmental Health Sciences, National Institutes of Health, Department of Health and Human ServicesThe authors declare they have no competing financial interests.
12 2005 113 12 A805 A805 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 letter, Needleman et al. suggest that our arguments regarding the ethics of human testing of pesticides (Resnik and Portier 2005) are “vague, tendentious, and essentially incorrect.” However, they offer no effective sustained arguments in support of this conclusion.
They cite the Food Quality Protection Act (FQPA 1996) and note that this act led pesticide companies to sponsor studies in humans, in part, to avoid an additional 10-fold human safety factor. They then chastise our article for saying effectively the same thing. Their next point focuses on what they view as a U.S. Environmental Protection Agency (EPA) failure to enforce the FQPA for organophosphate pesticides, an issue that has no bearing on the ethical question of human testing of pesticides. They claim that we failed to specify exactly how human studies could benefit society, but we clearly stated in our article (Resnik and Portier 2005) that these studies could promote public health by providing knowledge that may be useful in regulating pesticides. It is the responsibility of the party sponsoring or conducting a human testing study to demonstrate the relevance and utility of the proposed study with regard to human toxicity.
Needleman et al. contend that we have missed an important issue—the relevance of adult testing to children’s risk. This argument is not relevant to our article because we focused on human testing of pesticides on adults, not on children. We seriously doubt whether testing pesticides on children, or on other vulnerable populations, could ever be justified on ethical grounds. It is possible that studies of the effects of pesticides on adults could enhance our understanding of how pesticides affect children, but the party(s) sponsoring the study would need to provide some evidence for this supposition. In our article (Resnik and Portier 2005) we argued that there may be some cases where the public health benefits of testing pesticides on adults justify imposing risks on human subjects. We did not argue that one of the benefits of testing pesticides on adults is to improve the health of children, but we acknowledge that some studies might have this potential benefit.
We addressed arguments concerning the statistical power of human studies in our Supplemental Material, which is available online (http://ehp.niehs.nih.gov/members/2005/7720/suppl.pdf). Good statistical design is one of the key principles that must be considered in evaluating the acceptability of human testing, and we acknowledge that some of the disputed pesticide studies have been underpowered. It is surprising to us that Needleman et al. would raise this concern when their key premise is that human testing of pesticides is never ethical. If testing of pesticides is never ethical, then statistical issues, such as sample size, are irrelevant.
Needleman et al. have confused the issues relating to the FQPA, 10X safety factors, and risk assessment with the questions surrounding human testing of pesticides. We wholeheartedly agree that to conduct a clinical study in humans for the sole purpose of keeping a product commercially viable is unethical. They assume that the only reason why anyone would develop and conduct human studies of pesticides is to promote the interests of pesticide manufacturers. Again, we address this issue in our Supplemental Material (http://ehp.niehs.nih.gov/members/2005/7720/suppl.pdf).
We argue that a study that benefits private industry can be ethical, provided that it also offers scientific or social benefits. For example, clinical trials of new drugs benefit pharmaceutical companies, but they also benefit patients, enhance our understanding of human disease, and improve public health.
==== Refs
References
FQPA 1996. Food Quality Protection Act of 1996. Public Law 104–170.
Resnick DB Portier C 2005 Pesticide testing on human subjects: weighing benefits and risks Environ Health Perspect 113 813 817 16002367
|
0
|
PMC1314955
|
CC0
|
2021-01-04 23:41:30
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A805a
|
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-a0805b16330334PerspectivesCorrespondenceInfant Methemoglobinemia: Causative Factors Zeman Catherine Health Division, University of Northern Iowa Cedar Falls, Iowa, E-mail:
[email protected] author declares she has no competing financial interests.
12 2005 113 12 A805 A806 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
That individually and environmentally mediated cofactors function in the development of infant methemoglobinemia (iMHG) is not a new finding. Studies cited by Fewtrell (2004) note these cofactors. In my work on iMHG, using a nested case–control study that was not cited by Fewtrell (2004), I confirmed that cofactors (feeding practices, individual and infant physiology, etc.) played a role in the disease status of populations under study (Zeman 2000; Zeman et al. 2002a).
Cofactor work completed with Ustyogova et al. (2002) indicated that in vitro studies examining exposures below and above the maximum contaminant limit for nitrate show impacts to lymphocyte proliferation and cytokine production with shift in immune response from a Th 1 lymphocyte immune status to a Th 2 lymphocyte, indicating possible decreased resistance to pathological states. Could this be another factor in iMHG? Ustyogova et al. (2002) examined healthy adults, but the study raises the issue of the effects of exposure on the developing immune system of infants. The microbial status of drinking water for participants in the case–control study (Bauer et al. 2003), has been evaluated at the bacterial and parasite levels (Bauer et al. 2003, Zeman et al. 2005). Findings indicated that most water was highly contaminated with fecal coliforms (0–1,000/100 mL) and protozoan oocysts (0–84 cysts/L); when the likelihood of contamination was compared to data on whether or not an iMHG case had occurred in the household, no significant relationship was found.
Fewtrell (2004) claimed that no exposure–response data are available, but two articles (Zeman et al. 2002a, 2002b) reporting on the iMHG case–control study and associated exposure assessment to nitrate/nitrite contradict this. In one of these studies (Zeman et al. 2002a), a bivariate fit of nitrate level in well water and nitrite exposure through water and dietary sources (p = 0.0001) validated the exposure assessment methodology. Table 9 of this article illustrates the relationship strength under bivariate test for a variety of risk factors, and Table 11 provides a multivariate analysis showing the most predictive factors for this study population—exposure to drinking water nitrates, breast-feeding duration, and lack of vitamin use (Zeman et al. 2002a). By stratifying these data for bivariate analysis and comparing the calculated nitrite exposure for each child for low to medium (< 0.1 mg/kg/day to ≥0.1–1.5 mg/kg/day) and low to high (< 0.1 mg/kg/day to ≥1.5 mg/kg/day) exposures, the likelihood (L) and Pearson (P) calculations show a definite gradation in effect and significance in both situations: low to medium (L = 6.574, p = 0.0103; and P = 4.377, p = 0.0364); low to high (L = 20.7474, p = 0.0001; and P = 15.605, p = 0.0001). I agree, however, that no dose–response relationship has been documented comparing calculated exposure to measured blood methemoglobin level at the time of a clinically diagnosed iMHG case. This would be a gold standard that would help us to tease out the causative factors of iMHG and to establish solidly or refute what looks like, to date, the centrality of the role of nitrate exposure in the etiology of iMHG.
==== Refs
References
Bauer RJ Vlad M Sinca A Moga D Mirestean I Zeman CL 2003. Field methodology for the determination of the prevalence of Giardia cysts and Cryptosporidium oocysts in drinking water and its association to the development of diarrheal disease in the Transylvania Region of Romania. In: Proceedings of the Sixth International Symposium and Exhibition on Environmental Contamination in Central and Eastern Europe (Herndon RC, ed). Tallahassee, FL:Institute for International Cooperative Environmental Research, Florida State University, 233–238.
Fewtrell L 2004 Drinking-water nitrate, methemoglobinemia, and global burden of disease: a discussion Environ Health Perspect 112 1371 1374 15471727
Ustyugova IV Zeman C Dhanwada K Beltz LA 2002 Nitrates/nitrites alter human lymphocyte proliferation and cytokine production Arch Environ Contam Toxicol 43 270 276 12202921
Zeman CL 2000. Exposure Assessment Methodology Development in Support of a Pilot Study of the Long-Term Neuropsychological Impact of Methemoglobinemia and High Nitrate Exposure in Infants of Transylvania, Romania, with an added Case Control Study of MGH Risk Factors.. Ann Arbor MI:UMI, Bell & Howell.
Zeman CL Kross B Vlad M 2002a A nested case–control study of methemoglobinemia risk factors in children of Transylvania, Romania Environ Health Perspect 110 817 822 12153765
Zeman CL SeuleanuI I Sinca M Sinca A Moga D Vlad M 2005 Environmental illness may have contributed to the origins of Transylvanian vampire myths Int J Global Health 3 2 29 39
Zeman CL Vlad ML Kross B 2002b Exposure methodology and findings for dietary nitrate exposures in children of Transylvania, Romania J Expos Anal Environ Epidemiol 12 54 63
|
16330334
|
PMC1314956
|
CC0
|
2021-01-04 23:41:30
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A805b-A806
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.113-a805b
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0806a16330335PerspectivesCorrespondenceInfant Methemoglobinemia: Fewtrell Responds Fewtrell Lorna Centre for Research into Environment and Health, Crewe, Cheshire, United Kingdom, E-mail:
[email protected] author declares she has no competing financial interests.
12 2005 113 12 A806 A806 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 her letter, Zeman seems to be objecting to three points relating my article (Fewtrell 2004): that the role of cofactors is not new, that her articles were not cited, and that exposure–response data are available.
First, in my article (Fewtrell 2004) I did not suggest that the role of cofactors was a novel discovery, as evidenced by the selection of articles I cited noting such factors. Rather, I noted the fact that the role of cofactors often seems to be overlooked in some of the literature.
Second, as stated in the conclusion (Fewtrell 2004), “the study did not set out to review the role of nitrates in the causation of methemoglobinemia” nor, by extension, the role of cofactors; thus the literature citation was selective.
Finally, I assessed the article by Zeman et al. (2002) in the literature review for my study (Fewtrell 2004), but I felt it did not provide useful drinking water (i.e., exposure) data related to the level of methemoglobinemia in infants (i.e., response data); therefore, the article by Zeman et al. (2002) was not cited (although I do consider the new data presented in Zeman’s letter to be of interest).
Three points that influenced my decision not to cite the article by Zeman et al. (2002) are worth noting. First, Zeman et al.’s Figure 2 shows an apparent relationship between nitrate level in wells (parts per million) and “nitrate” (this is presumably nitrite, as described in the figure legend and Zeman’s letter) exposure in milligrams per kilogram per day. This was reported to have a correlation of 0.71, presumably resulting in a coefficient of determination of 0.50. However, some reported concentrations are remarkably high, exceeding 1,000 ppm nitrate. The reported relationship appears, visually, to be dependent on a few very high value(s) for the claimed correlation, making inference—or application of the functional relationship—within more usual exposure levels inappropriate. The data points within more “usual” elevated exposure ranges (say < 500 ppm) do not appear to exhibit a clear correlation between nitrate concentration (parts per million) in well water and calculated nitrite intake in milligrams per kilogram per day.
Second, this figure simply claims a correlation between a concentration (i.e., nitrate in water, parts per million) and a precurser of the outcome condition (i.e., calculated nitrite intake, milligrams per kilogram per day), not the “outcome” of interest I discussed (Fewtrell 2004). It is unclear how the boiling of water (which may lead to an increase in nitrate concentrations) is accounted for in the relationship presented by Zeman et al. (2002). The relationship is also likely to be location specific, being dependent upon local feeding habits [e.g., level of formula, tea (chi), vegetables, etc., given to the infant].
Third, Zeman et al. (2002) did not provide a detailed explanation of how the dependent variable numerical values in their Figure 2 were derived, making it difficult to assess the quality of this information.
The selection of appropriate studies to include in any global assessment is difficult and will always be contentious. I hope that these observations explain my decision not to cite the article by Zeman et al. (2002).
==== Refs
References
Fewtrell L 2004 Drinking-water nitrate, methemoglobinemia, and global burden of disease: a discussion Environ Health Perspect 112 1371 1374 15471727
Zeman CL Kross B Vlad M 2002 A nested case-control study of methemoglobinemia risk factors in children of Transylvania, Romania Environ Health Perspect 110 817 822 12153765
|
0
|
PMC1314957
|
CC0
|
2021-01-04 23:41:30
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A806a
|
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-a0806b16330335PerspectivesCorrespondenceAvian Influenza and UV-B Blocked by Biomass Smoke Mims Forrest M. IIIGeronimo Creek Observatory Seguin, Texas, E-mail:
[email protected] author declares he has no competing financial interests.
12 2005 113 12 A806 A807 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
Washam (2005) described various poultry inoculation strategies being considered for controlling the spread of avian influenza in Southeast Asia and China. Longini et al. (2005) proposed that a future avian influenza A pandemic might be contained at the source by targeted prophylaxis, quarantine, and prevaccination.
Washam (2005) correctly noted that “Asian farmers, though, are running out of options.” I propose a new option: Avian influenza might be controlled by a substantial reduction in regional scale biomass smoke in Southeast Asia that will allow natural solar ultraviolet-B radiation (UV-B) to suppress the virus before infection occurs.
Influenza viruses and various non-pigmented bacteria are killed by UV-B wavelengths in sunlight (Hollaender and Oliphant 1944). Biomass smoke significantly suppresses natural levels of UV-B, and severe smoke pollution reduced UV-B by up to 95% during the burning seasons in Brazil in 1995 (Mims 1996) and 1997 (Mims FM III, White B, unpublished data). Reduced UV-B on 6 days in August 1997 was well correlated (r2 = 0.83) with an increase in the ratio of nonpigmented bacteria vulnerable to UV-B to pigmented bacteria that are protected from UV-B (Mims and White 1998). Although airborne influenza viruses were not measured, 1997 hospital admission records at Alta Floresta, Brazil, showed that influenza incidence was highest during the burning season (de Castro GC, personal communication).
Human cases of avian influenza in Thailand and Vietnam peaked during the winter burning seasons of 2003 and 2004 (Thailand Ministry of Public Health 2005). Assuming similar optical properties of biomass smoke in Southeast Asia and Brazil, where UV-B and optical depth are highly correlated, optical depth measurements over Thailand and Vietnam by NASA’s Terra and Aqua satellites suggest highly suppressed UV-B during these avian influenza outbreaks (Mims FM III, unpublished data).
Human cases of avian influenza in Thailand and Vietnam since December 2003 have peaked during both the rainy season and the burning season. Thus, periods of prolonged cloudiness and severe smoke pollution could play a role in initiating avian and other influenza outbreaks by attenuating the solar UV-B that might otherwise suppress influenza viruses in outdoor air exposed to sunlight. The transmission of avian influenza to people during these periods is enhanced by the fact that poultry raised for human consumption are often kept within several meters of where people live (World Health Organization 2004).
==== Refs
References
Hollaender A Oliphant J 1944 The inactivating effect of monochromatic ultraviolet radiation on influenza virus J Bacteriol 48 447 454 16560850
Longini IM Jr Nizam A Xu S Ungchusak K Hanshaoworakul W Cummings DAT 2005 Containing pandemic influenza at the source Science 309 1083 1087 16079251
Mims FM III 1996 Significant reduction in UV-B caused by smoke from biomass burning in Brazil Photochem Photobiol 64 123 125
Thailand Ministry of Public Health 2005. Avian Influenza Surveillance in Human as of July 1, 2005. Available: http://thaigcd.ddc.moph.go.th/AI_case_report_050701.html [accessed 25 August 2005].
Washam C 2005 On hens and needles Environ Health Perspect 113 A370 15929877
World Health Organization 2004. Avian Influenza A(H5) in Rural Areas in Asia: Food Safety Considerations. Available: http://www.who.int/foodsafety/micro/avian2/en/ [accessed 25 August 2005].
|
16330335
|
PMC1314958
|
CC0
|
2021-01-04 23:41:29
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A806b-A807
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.113-a806b
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0809a16353305EnvironewsForumAutoimmune Disease: Phthalate Linked to Lupus in Mice Potera Carol 12 2005 113 12 A809 A809 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
No one knows to what degree genetics or environmental agents cause lupus, an autoimmune disorder that affects the skin, joints, and internal organs including the kidneys. However, researchers at Indiana State University may have strengthened the environmental evidence by discovering that phthalates trigger lupus antibodies in a mouse model.
Phthalates are found in adhesives, cosmetics, fragrances, vinyl flooring, polyvinyl chloride pipe, and certain toys and medical supplies. According to a report out of the Centers for Disease Control and Prevention and the National Toxicology Program, published in the October 2000 issue of EHP, phthalate exposure is more extensive than previously suspected, especially in women aged 20–40 years. Other studies have pointed to possible links with asthma, rhinitis, and eczema in children as well as altered genital development in male infants. The new lupus findings add to a growing list of potential health effects caused by these chemicals.
In lupus, the immune system loses its ability to tell the difference between foreign substances (antigens) and the body’s own cells and tissues. The immune system makes antibodies against the body itself, causing inflammation, tissue injury, and pain. Up to 1.5 million Americans have been diagnosed with lupus, and another 16,000 develop the disease each year, according to the Lupus Foundation of America.
While investigating the gene sequence of a monoclonal antibody used as a marker for tumor growth, biochemist Swapan Ghosh, interim chair of the Life Sciences Department at Indiana State University, noticed that it shared 98% similarity with an antibody protein component (light chain) made by NZB mice, a popular model for autoimmune diseases. In lupus, such antibodies attack DNA in the kidneys, heart, and lungs. The finding, published in the December 2003 issue of Immunology, was a surprise: “I was not studying lupus or autoimmune diseases at all,” says Ghosh. But he took advantage of the unexpected turn and has launched a series of experiments to further explore the phthalate–lupus connection.
In the latest study, Ghosh and graduate student So-Yon Lim injected four types of mice, including NZB mice, with di-(2-ethylhexyl) phthalate, or DEHP. Initially, all the mice generated antiphthalate antibodies, but only the lupus-prone NZB mice developed nephritis, which led to kidney failure and early death. The other mice initially produced antiphthalate antibodies, but the antibodies were counteracted by CD8+ suppressor T cells, which prevented kidney damage. “There’s something different about the immune systems of NZB mice [that makes them more susceptible to phthalates],” says Ghosh. The details of the investigation are reported in the August 2005 issue of the Journal of Autoimmunity.
Although the phthalate–lupus connection has been observed only in mice, “many things found in the mouse immune system have proven to be true in humans,” says Ghosh. On the other hand, “not everything seen in a mouse model reflects what happens in humans,” cautions Betty Diamond, chief of rheumatology at Columbia University.
Although Ghosh’s results are far from applicable to humans, they do raise several questions for future studies on the potential phthalate–lupus link in people. Do lupus patients have high levels of antiphthalate antibodies? Ghosh plans to screen lupus patients and healthy people in the future to find out. Does exposure to phthalates increase the risk for lupus? He plans to explore this, too, by measuring blood levels in workers exposed to phthalates in the plastics manufacturing industry. Lupus is five times more common in women than men. Might this be because women use more phthalate-containing cosmetics and perfumes than men do?
The American Chemistry Council (ACC), an industry trade group, has criticized Ghosh’s study because he combined DEHP to proteins like bovine serum albumin. “The attached proteins may cause autoimmune and allergic responses,” says Marian Stanley, director of the ACC’s Phthalate Esters Panel. Ghosh counters, “We also studied DEHP not complexed to a protein, and it evoked an anti-DNA response.” He explains that he attached a main metabolite of DEHP to proteins because some studies have suggested that phthalate metabolites show an affinity for albumin in the body.
So far, exposure to ultraviolet light is the only environmental factor that has been clearly linked to lupus in genetically susceptible patients. As lupus researchers continue to investigate other environmental causes, “we need to be open-minded, but not jump to conclusions,” Diamond says.
|
16353305
|
PMC1314959
|
CC0
|
2021-01-04 23:41:29
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A809a
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.113-a809a
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0809b16353305EnvironewsForumThe Beat Dooley Erin E. 12 2005 113 12 A809 A811 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
Fly the Environmentally Friendly Skies
In June 2005, the British airline industry unveiled a 15-year initiative to make itself more environmentally friendly. The industry wants to improve its fuel efficiency, reduce perceived external noise, and lower carbon dioxide emissions on new planes by 50% and nitrogen oxide emissions by 80%. Also planned are ways to give travelers information on the amount of fuel used and pollutants emitted on routes that they travel. The industry may also prohibit foreign carriers from flying older, more-polluting aircraft into the United Kingdom.
A Loan for Colombia
In June 2005 the World Bank announced it was granting a $150 million loan to Colombia to help that nation integrate sustainability principles into its environmental programs and policies and meet the UN Millennium Development Goals, including halving the number of people without adequate water and sanitation facilities. The monies are earmarked for three areas: development of a framework for planning and monitoring the progress toward meeting the UN goals; increased interinstitutional cooperation and public participation in environmental decision making; and development of laws and policies related to air and water quality, solid waste management, and environmental licensing. Bank officials hope the work financed by the loan will also decrease child mortality rates related to respiratory and diarrheal diseases.
Wave Power in the Works
Just off the northern coast of Portugal is the site of the world’s first commercial wave-generated electric plant. The contract was signed in May 2005 for the $9.6 million project, under which three wave energy converters will be built at the site. The long, hinged converters move with the flow of tidal currents, pumping fluid to hydraulic motors that drive generators.
The wave power plant is expected to provide electricity for more than 1,500 Portuguese households while displacing more than 6,000 metric tons of carbon dioxide produced each year by conventional power plants. If this first phase proves successful, 30 additional wave converters will be ordered by the end of 2006.
Baytril Gets the Boot from Bird Farms
Amidst calls from doctors and public health advocates, the FDA has banned the use of the antibiotic Baytril in poultry. The FDA is also reviewing requests to ban the use of other drugs given to animals. Although Perdue Farms and other producers stopped using Baytril before the July 2005 ban, an industry spokesman said alternative drugs are not as effective in dealing with respiratory illnesses in mass-produced poultry.
The ban is intended to stop the increase of drug-resistant strains of foodborne Campylobacter. Campylobacter infection causes abdominal symptoms and fever, and is one of the most common bacterial causes of diarrheal illness in the United States. According to the FDA, 20% of human Campylobacter infections involve the resistant strain.
WHO Knows About Radon?
The WHO has launched the International Radon Project to educate the public about the hazards of this chemically inert, radioactive gas that occurs naturally in soils and rocks around the world. The project will include a database of average radon levels in member nations, radon action levels, and mitigation measures, among other information. The WHO has also published a new fact sheet on radon and cancer as part of the project.
Radon may cause 6–15% of lung cancer cases, and moderate exposure may increase the risk of lung cancer in smokers by 25 times. Radon exposure in homes varies according to a home’s location, ventilation, and presence of exterior cracks and openings.
From Carpet to Kilowatts
Each year some 4.7 billion pounds of carpet are taken to U.S. dumps, taking up almost 1% of the country’s landfill space. Now Shaw Industries, the world’s largest carpet maker, has opened a $10 million power plant that is fueled by the 16,000 tons of scrap the company turns out annually as well as by 6,000 tons of sawdust produced by wood flooring manufacturing. The new plant powers one of the company’s main factories, and should save the company $2.5 million in fuel oil each year. The plant engineers say the process emits about the same amount of pollution as natural gas.
|
0
|
PMC1314960
|
CC0
|
2021-01-04 23:41:29
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A809b-A811
|
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-a0810a16353306EnvironewsForumInfectious Disease: New Human Retroviruses Phillips Melissa Lee 12 2005 113 12 A810 A810 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
Retroviruses called human T-lymphotropic viruses (HTLVs) are found in two types—HTLV-1 and HTLV-2—in people all over the world. Genetic evidence suggests that they crossed into humans from simian T-lymphotropic viruses (STLVs) and that each type, plus various subtypes, have crossed independently. Now, two more types of HTLV have been found in humans living in central Africa.
At least 22 million humans are infected with HTLV-1 or HTLV-2, and the viruses are endemic in several areas. About 2–5% of those infected with HTLV-1 develop adult T cell leukemia. HTLV-1 also causes a neurologic disease called tropical spastic paraparesis/HTLV-1 associated myelopathy. HTLV-2 is less pathogenic but is thought to cause similar neurologic illnesses and increase susceptibility to opportunistic infection.
William Switzer, a researcher at the Centers for Disease Control and Prevention, and his colleagues sequenced HTLV strains from a high-risk population: people in Cameroon who reported contact with nonhuman primate tissues through hunting and butchering or keeping primate pets. The study uncovered many previously unknown subtypes of HTLV-1, most with known correlates in nonhuman primates. The team also found that two people carried previously unknown HTLV types. One, HTLV-3, is similar to the nonhuman primate virus STLV-3. The other, HTLV-4, is genetically different from any known virus in humans or other primates. The findings appear in the 31 May 2005 issue of the Proceedings of the National Academy of Sciences.
Because HTLV-4 is so divergent from other HTLVs, this virus may have evolved in humans over quite some time, Switzer says. It’s possible, though, that primates are infected with an equally divergent simian version that just hasn’t been found yet. “We’re screening primates in that same area to see if we can answer that question,” Switzer says.
A group led by Antoine Gessain, head of the Epidemiology and Physiopathology of Oncogenic Viruses Unit at the Pasteur Institute, also recently found a subtype of HTLV-3 in a human, but it’s somewhat different from the subtype Switzer and his colleagues found, which suggests “another example of multiple independent, cross-species transmission events,” Switzer says. The HTLV-3 strain Gessain found is extremely similar to a strain reported in the red-capped mangabey, which suggests that it crossed to humans very recently, Switzer says. Gessain’s findings were published 9 May 2005 in Retrovirology.
The current dogma surrounding retroviruses is that cross-species transmission is rare, but finding so many near-identical strains between humans and nonhuman primates suggests this is not a rare event. Benign retroviruses probably cross from nonhuman primates to humans frequently, but we don’t notice them because we don’t get sick, says Bernard Poiesz, a professor of medicine at SUNY Upstate Medical University. “But every once in a while,” he says, “one of them will jump and we may not handle it so well.”
Cross-species predator.
A T lymphocyte infected with HTLV-1 (green), which causes a type of leukemia. Such viruses are believed to have crossed to humans from simians.
|
16353306
|
PMC1314961
|
CC0
|
2021-01-04 23:41:29
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A810a
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.113-a810a
|
oa_comm
|
==== Front
Environ Health PerspectEnviron. Health PerspectEnvironmental Health Perspectives0091-67651552-9924National Institute of Environmental Health Sciences ehp0113-a0810b16353306EnvironewsForumEcological Change: Life Lessons Bonn Dorothy 12 2005 113 12 A810 A810 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
“All global environmental change eventually ends up as a human health problem,” said Eric Chivian, director of the Harvard Center for Health and the Global Environment, opening the August 2005 First International Conference on Health and Biodiversity in Galway, Ireland. Speaker after speaker showed how careless disregard for the environment and its variety of life forms squanders potential new medicines, endangers our food security, and exposes us to new risks of infectious disease.
Many frequently prescribed drugs are derived from or patterned after compounds in natural sources, Chivian noted. For example, ziconotide—a pain killer 1,000 times more powerful than morphine—comes from marine cone snails that inhabit narrow ranges in coral reefs and thus are increasingly endangered by coral bleaching, mostly from global warming. How many other useful species are lost without our ever recognizing their potential?
Species loss may also mean the loss of valuable models for medical research, said Chivian. Black bears, which hibernate for several months over the winter without losing bone mass, could provide a clue to the cause of osteoporosis, an enormous public health problem. But bear populations in many parts of the world are threatened by habitat destruction and overhunting.
Discussion of sustainable food systems for developing countries focused on promoting the use of indigenous plants. In Lebanon, where diets are high in bread and refined grains but low in fruits, vegetables, and fish, a quarter of the children are overweight and a third of the women of child-bearing age are anemic. Malek Batal, a nutrition professor at the American University of Beirut, is exploring how wild plants such as fennel, mint, and salsify have the potential to increase diversity of nutrient intake and food security in poor communities. He found that wild plants offer antioxidants, flavonoids, fiber, iron, calcium, and many other nutrients. Being easily accessible, easy to use, and palatable, they also contribute to food security.
Interfering with ecosystems can have dire consequences for biodiversity, as conservation biologist Diana Bell of the University of East Anglia explained: when the South American myxoma virus was introduced into Europe in the 1950s to control rabbit populations, it contributed to the collapse of a species-rich ecosystem in which the rabbit was the keystone prey for more than 45 predators. Bell also identified the illegal trade in wildlife (especially small carnivores) in Southeast Asia as a dual threat to human health (as the origin of the SARS coronavirus) and massive species loss in this “biodiversity hot spot.” She believes an interdisciplinary approach involving ecologists, microbiologists, medical specialists, and others will best advance research in the twin fields of human health and species loss.
The time to address biodiversity loss is now, speakers agreed. As Chivian said, “We are in deep, deep trouble with what we are doing to life on Earth. . . . We are tampering with the life support systems of the Earth in ways that we barely understand.”
|
0
|
PMC1314962
|
CC0
|
2021-01-04 23:41:31
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A810b
|
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-a0836aEnvironewsScience SelectionsAsthma in Young Children: Prenatal DDE Exposure May Increase Risk Freeman Kris 12 2005 113 12 A836 A836 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
Most countries have banned the agricultural use of the organochlorine insecticide DDT because of the way this persistent, fat-soluble compound accumulates in the food chain. However, DDT is still widely sprayed in developing countries to combat malaria-bearing mosquitoes. Studies have linked exposure to DDT and its persistent metabolite p,p′-DDE to changes in the immune responses of human cells, and to asthma prevalence in children and adults. A longitudinal study now shows that prenatal exposure may provide the fundamental window for asthma susceptibility linked to DDT [EHP 113:1787–1790].
Investigators collected umbilical cord blood from 482 children born on the Spanish island of Menorca and tested 84% for the presence of organochlorine compounds. DDT is not used on Menorca. However, the parents of the children in the study ate relatively large amounts of fish, which can be a source of exposure to DDT residues. According to self-reports of diet on questionnaires, more than half of the mothers ate fish more than twice a week during pregnancy.
All of the children tested had p,p′-DDE in their cord serum (the median concentration was 1.03 nanograms per milliliter [ng/mL]). Serum levels tended to be higher in children with older mothers, although the mothers’ fish consumption during pregnancy correlated poorly with the children’s DDE levels. Each child’s serum also contained hexachlorobenzene and polychlorinated biphenyls.
The researchers correlated the children’s prenatal exposure to risk of having asthma or atopy at age 4. Asthma was defined as one or more episodes of wheezing in the fourth year alone, one or more episodes of wheezing per year in consecutive years (“persistent wheezing”), or a physician’s diagnosis of asthma. Atopy was defined as having blood levels of specific immunoglobulin E antibodies for dust mites, cats, or grasses. Of the initial participants, 97% provided medical information yearly through age 4 and 75% provided blood samples at age 4; 306 of these samples were tested for antibodies and for peripheral white blood cells, a sign of the underlying inflammation responsible for asthma.
Wheezing was reported at age 4 for 11.6% of the children whose blood was tested for organochlorines. In addition, 12.6% of those who gave blood at age 4 had antibodies for the specified allergens in their blood. The risk of wheezing increased with the concentration of p,p′-DDE in the child’s cord serum. Of the children in the lowest quartile of exposure (less than 0.57 ng/mL), 9% reported wheezing compared to 19% of the children in the highest quartile of exposure (more than 1.90 ng/mL). There were no correlations between wheezing in the children and maternal consumption of fish during pregnancy.
There was no apparent link between atopy and the relationship between DDT and wheezing; children both with and without atopy had a similar increase of wheeze with increasing p,p′-DDE. The researchers speculate that the lack of an association between DDT exposure and atopy in their study could be due to the young age of the children studied, as sensitization to allergens tends to increase during childhood. There was no correlation between the other organochlorine compounds measured and wheezing or atopy.
Further study is needed to determine if the link between DDT and asthma susceptibility is caused by the effect of the insecticide on the immune system or the hormonal system. In addition to its direct impact on immune cells as shown in previous research, p,p′-DDE has also been shown to interfere with hormonal receptors and to mimic estrogen activity, which might indirectly affect immune responses. The researchers suggest that their results be considered when evaluating the risk of spraying DDT in anti-malaria campaigns.
Then and now.
A study of Spanish mother–child pairs shows that DDT exposure in utero may contribute to later asthma in children.
|
0
|
PMC1314963
|
CC0
|
2021-01-04 23:41:30
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A836a
|
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-a0836bEnvironewsScience SelectionsDeath by Particles: The Link Between Air Pollution and Fatal Coronary Heart Disease in Women Kessler Rebecca 12 2005 113 12 A836 A837 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 growing body of evidence links chronic exposure to air pollution––especially particulate matter (PM)––with mortality resulting from a variety of heart, lung, and respiratory diseases. A new study corroborates this association, and indicates that women may be at greater risk than men of fatal coronary heart disease (CHD) as a result of exposure to airborne PM [EHP 113:1723–1729]. When ozone (O3) or sulfur dioxide (SO2) is also present, women’s risk appears even greater.
The study, by a team of epidemiologists at Loma Linda University, is part of the 22-year Adventist Health Study on the Health Effects of Smog. It followed 3,239 nonsmoking, non-Hispanic white adults in several mainly urban areas in California from 1976 to 1998. The researchers associated CHD deaths with prior exposure to various levels of several common air pollutants: PM2.5, PM10–2.5, PM10, O3, SO2, and nitrogen dioxide (NO2).
Participants completed a baseline health and lifestyle questionnaire in 1976, and four subsequent questionnaires covering personal sources of air pollution, such as secondhand tobacco smoke and fumes in the workplace. The researchers used airport visibility measurements (for PM2.5 only) and data from state-run air pollution monitors (for all other pollutants) to estimate pollutant levels over time for the zip code centroids of participants’ work sites and residences. Documented pollutant levels ranged from negligible to above legal limits. California’s death certificate files and the National Death Index provided data on numbers and causes of deaths.
The researchers found that CHD caused 23.7% of all the deaths in the study cohort (155 women and 95 men). Adjusting for past smoking, body mass index, education level, frequency of eating meat, and calendar year (as PM levels declined over the study period), the researchers conducted statistical analyses to determine whether fatal CHD was associated with long-term exposure to the pollutants, either singly or in combinations of single gases and PM.
Women showed a relative risk for fatal CHD of 1.42, 1.38, and 1.22 with each increase of 10 micrograms per cubic meter (μg/m3) of airborne PM2.5, PM10–2.5, and PM10, respectively, in the air pollution they encountered during the four years preceding death. Postmenopausal women showed higher relative risks of 1.49, 1.61, and 1.30 for each 10 μg/m3 increase in PM2.5, PM10–2.5, and PM10, respectively. Neither O3, SO2, nor NO2 was associated with fatal CHD on its own. O3 and to a lesser degree SO2 (but not NO2) increased the effect of all sizes of PM. O3 in conjunction with PM2.5 yielded the most striking results: a relative risk of 2.0 in all women. Contrary to findings from several other studies that found increased risk of cardiopulmonary deaths due to PM in both genders, men showed no response to any of the pollutants.
The researchers highlight several physiological mechanisms that may explain their findings. Short-term exposure to PM is known to increase arrhythmia, inflammation, and blood viscosity, and to decrease heart rate variability, among other adverse effects that could lead to fatal CHD. Other findings show that O3 exposure increases lung permeability, perhaps easing PM’s entry into the bloodstream. Finally, several studies have indicated that PM deposits differently—and perhaps more harmfully—in women’s lungs than in men’s. This may provide a starting point for teasing out the study’s finding of an association between PM and risk of fatal CHD in women, but not in men.
|
0
|
PMC1314964
|
CC0
|
2021-01-04 23:41:31
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A836b-A837
|
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-a0858a16330342AnnouncementsBook ReviewThe Environmental Science of Drinking Water Bourne Shawna Shawna Bourne is a Certified Inspector of Public Health and works for the Ontario Ministry of the Environment where she manages the Drinking Water Program for the London District. She is also a board member of the Stanier Institute (Linking Health to Hygiene) and is an executive on behalf of the Ontario Branch Members of the Canadian Institute of Public Health Inspectors.12 2005 113 12 A858 A858 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
Patrick J. Sullivan, Franklin J. Agardy, and James. J.J. Clark
Burlington, MA:Elsevier, 2005. 368 pp. ISBN: 0-7506-7876-3, $59.95 cloth
Our ever-increasing knowledge of the subtle impacts of contaminants on the ecosystem and the public health of the community demand an evidence-informed and fact-based approach to dealing with the contamination of what has been coined the “lifeblood of the planet.” The authors of The Environmental Science of Drinking Water question the current culture of drinking water management in the United States and ask whether the reasoning behind the current management system is a sound one—much like the Native American proverb “The frog does not drink up the pond in which he lives,” a foolish practice that would lead to its own demise; the authors conclude that we should approach the management of our water supplies as the frog does and not foul the foundation of our, thus far, successful existence.
The authors raise a red flag early, citing the innumerable, unquantified industrial chemicals, pesticides, pharmaceuticals, and treatment byproducts released into our communal waterways. Five chapters cover basic water chemistry, modern chemical contaminants found in water systems, and the treatment technologies available for contaminant control and remediation. This information is supplemented by numerous relevant appendices and a glossary of terms; unfortunately, the work is scattered with abbreviations that perplex the unfamiliar reader and make the text difficult to follow at times. The authors stress the importance of an ecosystem approach to risk management, noting that a paradigm shift is necessary to protect against the long-term synergistic effects of the “chemical soup” that our water resources are becoming.
The Environmental Science of Drinking Water is an excellent compilation of fact, case study, policy analysis, opinion, and good old-fashioned common sense. This enlightening book is an excellent text suitable for technical experts in the field, students honing their skills in environmental health or engineering, or laypersons interested in understanding the origins of policies developed by the U.S. government.
Unfortunately, this text quickly dismisses the significance of microbial contaminants, which is a faulty approach from my perspective. Just as chemical risks due to the contamination of water supplies have changed over time, so too have the biological risks of the water supply and the susceptibilities of the population at large who consume it. The focus on water chemistry and related risks may warrant a review of the title of this textbook, with a more appropriate title being Drinking Water: The Chemical Risks in Modern Water Supplies. Such a title would more accurately reflect the scope and focus of this body of excellent work.
Clearly, these problems are not restricted to the United States. Given the many great examples and case studies used throughout the text, additional benefit would be gained by considering the state of affairs around the world and both the failures and successes in the implementation of water policy; this would add to the weight of the authors’ policy arguments. Some government agencies have begun to look at the broader impact of chemical and other contamination of water supplies; for example, in Ontario, Canada, goals include leading industry beyond minimum standards and toward “environmental improvements” through incentives.
Overall, the thesis is presented methodically and comprehensively. The arguments sway the reader to conclude that waste management and water treatment technology can be used better to reduce the risk to our water supplies. Source protection and drinking water treatment are available to exact a change. The new question is how to share this knowledge with the public, articulate common values and expectations, and mobilize the populace to put its money where its mouth is. The most remarkable achievement of this book is that it has the ability to move these and other important questions to the forefront of the minds of those who read it and begin that paradigm shift through the discourse of reason.
|
0
|
PMC1314965
|
CC0
|
2021-01-04 23:41:30
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A858a
|
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-a0858b16330342AnnouncementsNew BooksNew Books 12 2005 113 12 A858 A858 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
Advances in Air Pollution Modeling for Environmental Security: Proceedings of the NATO Advanced Research Workshop
István Faragó, Krassimir Georgiev, Ágnes Havasi, eds.
New York:Springer-Verlag, 2005. 406 pp. ISBN: 1-4020-3350-8, $84.95
Advances in Arsenic Research: Integration of Experimental and Observational Studies and Implications for Mitigation
Peggy A. O’Day, Dimitrios Vlassopoulos, Xiaoguang Meng, Liane G. Benning, eds.
Cary, NC:Oxford University Press, 2005. 450 pp. ISBN: 0-8412-3913-4, $159.50
Applied Research in Environmental Economics
Christoph Böhringer, Andreas Lange, eds.
New York:Springer-Verlag, 2005. 314 pp. ISBN: 3-7908-1587-X, $84.95
Biomeasurement: Understanding, Analysing, and Communicating Data in the Biosciences
Dawn Hawkins
Cary, NC:Oxford University Press, 2005. 312 pp. ISBN: 0-19-926515-1, $35
Comparative Genomics
Aoife McLysaght, Daniel H. Huson, eds.
New York:Springer-Verlag, 2005. 167 pp. ISBN: 3-540-28932-1, $58
Dimensions in Environmental and Ecological Economics
Nirmal Chandra Sahu, Amita Kumari Choudhury
Andhra Pradesh, India:Universities Press, 2005. 612 pp. ISBN: 81-7371-463-0, $16.17
DNA Conformation and Transcription
Takashi Ohyama, ed.
New York:Springer-Verlag, 2005. 220 pp. ISBN: 0-387-25579-6, $134
Ethical Dimensions of Health Policy
Marion Danis, Carolyn Clancy, Larry R. Churchill, eds.
Cary, NC:Oxford University Press, 2005. 424 pp. ISBN: 0-19-530083-1, $34.50
Global Warming—Myth or Reality?
Marcel Leroux
New York:Springer-Verlag, 2005. 510 pp. ISBN: 3-540-23909-X, $129
Managing Natural Wealth: Environment and Development in Malaysia
Jeffrey R. Vincent, Rozali Mohamed Ali
Washington, DC:RFF Press, 2005. 452 pp. ISBN: 1-933115-20-3, $85
Nuclear Hazards in the World: Field Studies on Affected Populations and Environments
Jun Takada
New York:Springer-Verlag, 2005. 134 pp. ISBN: 3-540-25272-X, $89.95
Plows, Plagues, and Petroleum: How Humans Took Control of Climate
William F. Ruddiman
Princeton, NJ:Princeton University Press, 2005. 272 pp. ISBN: 0-691-12164-8, $24.95
Presenting and Representing Environments
Graham Humphrys, Michael Williams, eds.
New York:Springer-Verlag, 2005. 218 pp. ISBN: 1-4020-3813-5, $129
Progress in Preventing Childhood Obesity: Focus on Schools—Brief Summary: Institute of Medicine Regional Symposium
Committee on Progress in Preventing Childhood Obesity
Washington, DC:National Academies Press, 2005. 24 pp. ISBN: 0-309-10040-2, $10.80
Questions and Answers in Environment Science
S.K. Basu, A.K. De
Andhra Pradesh, India:Universities Press, 2005. 396 pp. ISBN: 81-7371-547-5, $4.59
Strategic Planning in Environmental Regulation: A Policy Approach That Works
Steven Cohen, Sheldon Kamieniecki, Matthew A. Cahn
Cambridge, MA:MIT Press, 2005. 304 pp. ISBN: 0-262-03341-0, $60
The Coupling of Climate and Economic Dynamics: Essays on Integrated Assessment
Alain Haurie, Laurent Viguier, eds.
New York:Springer-Verlag, 2005. 381 pp. ISBN: 1-4020-3424-5, $169
The Laboratory Rat, 2nd ed.
Mark A. Suckow, Steven Weisbroth, Craig L. Franklin, eds.
Burlington, MA:Elsevier, 2005. 928 pp. ISBN: 0-12-074903-3, $139.95
The Mediterranean Sea
Alain Saliot, ed.
New York: Springer-Verlag, 2005. 414 pp. ISBN: 3-540-25018-2, $249
Toxicological Evaluation of Certain Veterinary Drug Residues in Food
Joint FAO/WHO Expert Committee on Food Additives
Geneva:World Health Organization Press, 2005. 189 pp. ISBN: 92-4-166053-8, $36
|
0
|
PMC1314966
|
CC0
|
2021-01-04 23:41:31
|
no
|
Environ Health Perspect. 2005 Dec; 113(12):A858b
|
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 10.1289/ehp.8230ehp0113-00175516330359ResearchAryl Hydrocarbon Receptor–Independent Toxicity of Weathered Crude Oil during Fish Development Incardona John P. 1Carls Mark G. 2Teraoka Hiroki 3Sloan Catherine A. 4Collier Tracy K. 1Scholz Nathaniel L. 11 Ecotoxicology and Environmental Fish Health Program, Environmental Conservation Division, Northwest Fisheries Science Center, National Oceanic and Atmospheric Administration, Seattle, Washington, USA2 Auke Bay Laboratory, Alaska Fisheries Science Center, National Oceanic and Atmospheric Administration, Juneau, Alaska, USA3 Department of Toxicology, School of Veterinary Medicine, Rakuno Gakuen University, Ebetsu, Japan4 Environmental Assessment Program, Environmental Conservation Division, Northwest Fisheries Science Center, National Oceanic and Atmospheric Administration, Seattle, Washington, USAAddress correspondence to J.P. Incardona, Environmental Conservation Division, Northwest Fisheries Science Center, National Oceanic and Atmospheric Administration, 2725 Montlake Blvd. E, Seattle, WA 98112 USA. Telephone: (206) 860-3347. Fax: (206) 860-3335. E-mail:
[email protected] authors declare they have no competing financial interests.
12 2005 10 8 2005 113 12 1755 1762 19 4 2005 10 8 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. Polycyclic aromatic hydrocarbons (PAHs), derived largely from fossil fuels and their combustion, are pervasive contaminants in rivers, lakes, and nearshore marine habitats. Studies after the Exxon Valdez oil spill demonstrated that fish embryos exposed to low levels of PAHs in weathered crude oil develop a syndrome of edema and craniofacial and body axis defects. Although mechanisms leading to these defects are poorly understood, it is widely held that PAH toxicity is linked to aryl hydrocarbon receptor (AhR) binding and cytochrome P450 1A (CYP1A) induction. Using zebrafish embryos, we show that the weathered crude oil syndrome is distinct from the well-characterized AhR-dependent effects of dioxin toxicity. Blockade of AhR pathway components with antisense morpholino oligonucleotides demonstrated that the key developmental defects induced by weathered crude oil exposure are mediated by low-molecular-weight tricyclic PAHs through AhR-independent disruption of cardiovascular function and morphogenesis. These findings have multiple implications for the assessment of PAH impacts on coastal habitats.
cardiovascular functionfish developmentnon-point source pollutionoil spill
==== Body
Every recent assessment of coastal habitats worldwide, from tropical reefs to temperate estuaries, has cited land-based pollution or runoff as a major threat to aquatic ecosystem health (Fabricius 2005; Li and Daler 2004; Pew Oceans Commission 2003; U.S. Commission on Ocean Policy 2004). As pervasive components of runoff from impervious surfaces, polycyclic aromatic hydrocarbons (PAHs) are a part of this problem, and there is very little understanding of their biologic impacts on aquatic resources. Because of urbanization and increased heavy vehicle use, storm water runoff and atmospheric deposition are now the largest sources of aquatic PAH contamination (Li and Daler 2004; Lima et al. 2002; National Research Council 2003; Van Metre and Mahler 2003; Van Metre et al. 2000). An understanding of the effects of PAHs on aquatic organisms is essential to understanding fully the impacts of urbanization and non-point source pollution on coastal habitats.
On a smaller scale, oil spills have provided a more conspicuous view of the impacts of PAH pollution on aquatic resources. Hydrocarbons from oil spills can persist in nearshore sediments for decades or longer and have long-term effects on aquatic ecosystems (Peterson et al. 2003; Reddy et al. 2002; Short et al. 2004). The deleterious effects of PAHs on fish early-life stages were investigated extensively after the 1989 Exxon Valdez oil spill in Prince William Sound, Alaska, which contaminated nearshore and intertidal spawning grounds for Pacific herring (Clupea pallasi) and pink salmon (Oncorhynchus gorbuscha) with Alaska North Slope (ANS) crude oil. Field and laboratory studies in these species and others demonstrated a common syndrome of oil-induced embryolarval toxicity that occurs in a range of teleosts, including marine, freshwater, temperate, and tropical species (Carls et al. 1999; Couillard 2002; Heintz et al. 1999; Marty et al. 1997; Pollino and Holdway 2002). This was characterized by pericardial and yolk sac edema, jaw reductions, and curvature of the body axis. Increased weathering of crude oil enriches the fraction of tricyclic PAHs and their alkylated homologs and increases the frequency of malformations (Carls et al. 1999; Heintz et al. 1999). Additionally, delayed mortality also occurred in the absence of external malformations, as indicated by the reduced oceanic survival of pink salmon exposed to weathered crude oil as embryos and released as smolts (Heintz et al. 2000).
The mechanisms leading to PAH-associated malformations and sublethal effects during fish development are unknown. Most PAHs bind the aryl hydrocarbon receptor (AhR), a ligand-activated basic-helix-loop-helix-Per-Arnt-Sim family transcription factor that controls the expression of a battery of genes encoding enzymes that convert PAHs to water-soluble derivates that are excreted, including mixed-function oxygenases such as cytochrome P450 1A (CYP1A) family members (Nebert et al. 2004). Although CYP1A for decades has been the most widely used biomarker for PAH exposure (Whyte et al. 2000), its role as a bioindicator of PAH toxicity has been debated. Genetic analysis in the mouse has led to a dual model in which the AhR pathway mediates both an adaptive response by which xenobiotic compounds are metabolized and detoxified, and a toxic response whereby receptor activation results in negative impacts in the exposed animal (Nebert et al. 2004; Schmidt and Bradfield 1996). Generally, the toxic response occurs with AhR ligands that are poor substrates for CYP enzymes, in particular, halogenated aromatic hydrocarbons such as dioxins and poly-chlorinated biphenyls. Because these ligands are resistant to metabolism, they accumulate in tissues and persistently activate the AhR. On the other hand, PAHs are classically associated with the adaptive response, by which they are eliminated from tissues. Nevertheless, some high-molecular-weight PAHs such as benzo(a)pyrene are converted to carcinogenic reactive intermediates by CYP1A (Phillips 1983), and it is widely held that much of the acute toxicity of PAHs is due to oxidative stress and cellular damage arising from CYP1A catalytic activity.
In fish embryos, PAH and dioxin toxicities are usually equated because exposure to potent AhR ligands such as 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) induces a super-ficially similar syndrome (Peterson et al. 1993). In zebrafish (Danio rerio), a brief exposure to TCDD shortly after fertilization results in the appearance of vascular dysfunction, pericardial and yolk sac edema, and anemia in hatching-stage larvae at 72–96 hr postfertilization (hpf) (Belair et al. 2001; Henry et al. 1997). These effects of TCDD exposure require a functional AhR. Because of genome duplication, many teleosts have two AhR genes, AhR1 and AhR2 (Hahn 2002). AhR1 protein is most similar in structure to the single mammalian AhRs, whereas AhR2 is divergent. Although TCDD is a ligand for both receptors from several fish species, AhR2 transcripts are more abundant and widely distributed (Hahn 2002; Karchner et al. 1999), and in zebrafish only AhR2 was found to be a functional receptor for TCDD and other common halogenated AhR ligands (Andreasen et al. 2002). Consistent with these findings, targeted knockdown of AhR2 in zebrafish embryos with antisense morpholino oligonucleotides (MOs) prevented all of the toxic effects of TCDD that occurred within the time frame of morpholino efficacy (Prasch et al. 2003; Teraoka et al. 2003). However, AhR-dependent developmental defects were CYP1A independent. A CYP1A morpholino did not alter dioxin toxicity in zebrafish embryos, implicating other AhR target genes in dioxin pathophysiology (Carney et al. 2004).
A previous analysis showed that micro-molar concentrations of individual tricyclic PAHs representing the homologous series most abundant in weathered crude oil (fluorene, dibenzothiophene, and phenanthrene) caused a syndrome of edema and craniofacial and body axis defects after dose-dependent cardiac dysfunction that was first observed at about 36 hpf (Incardona et al. 2004). These compounds caused cardiac arrhythmias that are characteristic of drugs known to block cardiac K+ channels of the human ether-a-go-go–related gene (HERG) family (Langheinrich et al. 2003). Among four-ring compounds, chrysene (9 μM), which is enriched in highly weathered crude oil, was nontoxic, whereas pyrene (1–5 μM), which is generally absent from weathered ANS crude oil, induced a syndrome with features similar to TCDD exposure that occurred between 80 and 96 hpf. Here, we show that a) both toxic and non-toxic PAHs induce CYP1A in zebrafish embryos, acting tissue specifically through AhR1 and AhR2; b) pyrene toxicity is AhR dependent, whereas tricyclic PAH toxicity is not; c) weathered ANS crude oil causes a syndrome in zebrafish embryos that is both clearly distinct from TCDD toxicity and consistent with cardiac dysfunction expected from the most abundant tricyclic compounds; and d ) the cardiovascular toxicity of weathered crude oil is independent of both AhR1 and AhR2. Rather than mediating the embryo-larval toxicity of weathered crude oil, the AhR pathway confers a measure of protection against the pathophysiologic effects of tricyclic PAHs on the developing fish heart.
Materials and Methods
Chemicals.
Dibenzothiophene (> 99%), phenanthrene (> 99.5%), pyrene (> 99%), and chrysene (98%) were obtained from Sigma-Aldrich (St. Louis, MO, USA). Stock PAH solutions were made in dimethyl sulfoxide (DMSO; tissue culture grade; Sigma-Aldrich) at 10 mg/mL, except chrysene (1 mg/mL). DMSO was ≤0.1% in exposure solutions.
Zebrafish exposures.
Wild-type AB strain zebrafish were maintained and fertilized eggs processed as previously described (Incardona et al. 2004). Fish were treated humanely and anesthetized when necessary. Exposures to individual model PAH compounds were carried out in plastic six-well plates (15–25 embryos in 3 mL) with a static renewal protocol at 28.5°C as described previously (Incardona et al. 2004). All exposures used doses that were above the solubility of the compounds and that, for toxic PAHs, produced effects in 100% of the embryos (Incardona et al. 2004). Each experiment was replicated at least three times. For crude oil exposures, embryos were incubated statically in a water-accommodated fraction (WAF) of crude oil, or in the continuously flowing effluent from a gravel column. The WAF was prepared by an overnight high-energy spin of 50 mL ANS crude [partially weathered by heating to 70°C until 20% reduction in mass (Marty et al. 1997)] in 30 L zebrafish system water using a large fiberglass tank and a paint mixer fixed to a fan motor. The WAF contained an estimated 2.8 mg/L total PAH (initial) but has the disadvantage of a high proportion of other oil components such as alkanes because a large fraction is present in particulate or colloidal form. WAF was diluted with system water to 1:2, 1:5, 1:10, and 1:100, and static exposures were performed in 30 mm glass Petri dishes (25 embryos in 4 mL) in an incubator at 28.5°C.
We used oiled gravel columns to achieve a more environmentally relevant exposure. System water was passed by gravity through gravel (4–6 mm grain diameter) coated with partially weathered ANS crude oil (6.0 g oil/kg gravel) to model conditions in oiled intertidal substrate as described previously (Marty et al. 1997; Short and Heintz 1997). Controls were similarly incubated in water passed through clean gravel. Dosing columns were 2-L glass beakers filled with 1.3 kg rock, with water flow directed to the bottom through 6 mm glass tubes. Columns were placed in glass baking dishes set at a slight angle (~ 6°); effluent overflowed from the tops of the columns, filling the baking dishes as a reservoir for exposing embryos in replicate open glass 30 mm Petri dishes (n = 4–5; ~ 25–50 embryos/dish). Temperature was maintained with submersible aquarium heaters. Column flow was initiated 1 day before embryo exposure to further weather the oil and remove any particulates. Exposures were started at 4–8 hpf. Temperature (nearest 0.5°C) and flow rate (nearest milliliter) were recorded as often as hourly during the day and at one or two time points during the night [Supplemental Table 1; Supplemental Material available online (http://ehp.niehs.nih.gov/docs/2005/8230/supplement.pdf)]. Although embryos developed more slowly at the experimental temperatures, they were staged according to the published standard series (Kimmel et al. 1995), and all developmental times are reported as hpf at standard temperature (28.5°C). We performed six oiled gravel experiments; two included uninjected embryos only (n > 300), three included AhR2 morphants (n > 200), two included AhR1/AhR2 double morphants (n = 145), and one included CYP1A morphants (n = 87). Standard statistical analyses were carried out with Microsoft Excel 2004 for Mac (Microsoft Corporation, Redmond, WA, USA).
Morpholino injections.
AhR1 cDNA and genomic sequences are available as GenBank AF258854 (GenBank 2005) and Ensembl ENSDARG00000020046 (Ensembl 2005), respectively, and AhR2 cDNA sequence as GenBank AF063446 (GenBank 2005). All morpholinos were synthesized by GeneTools (Philomath, OR, USA) and are listed below (mismatch nucleotides in control morpholinos are indicated by lowercase letters). The translation-blocking MOs targeting zfAHR2 (5′-TGTACCGATACCCGCCGACATGGTT-3′ for zfAHR2-MO, 5′-TGaACCcATACCCGCCGtCATcGTT-3′ for the negative control 4Mis-AHR2-MO) and zfCYP1A (5′-TGGATACTTTCCAGTTCTCAGCTCT-3′) have been described previously (Teraoka et al. 2003). Splice-blocking morpholinos for zfAhR1 were designed to target the exon 2/intron 2 splice donor site (5′-CTTTTGAAGTGACTTTTGGCCCGCA-3′ for E2I2-MO, 5′-CTTTTcAAcTGAgTTTTGcCCCcCA-3′ for 5Mis-E2I2-MO) and the intron 2/exon 3 splice acceptor site (5′-GTTCAGGGTTACTGCAAAAGAAAT-3′ for I2E3-MO). Morpholinos were injected at the 1–4 cell stage (0.25–1 hpf) as previously described (Incardona et al. 2004; Teraoka et al. 2003), and embryos were allowed to recover in system water at 28.5°C to 50% epiboly (5–6 hr) before use in exposure studies. AhR1 morpholinos were labeled with fluorescein, whereas AhR2 and CYP1A morpholinos were not. For injections involving AhR1, embryos were selected on an epifluorescent stereoscope based on fluorescence intensity and an even distribution in blastomeres.
Imaging of live embryos/larvae, immunofluorescence, and confocal microscopy.
Digital still micrographs were obtained and video-microscopy of live embryos and larvae performed as described previously (Incardona et al. 2004). Antibodies used were monoclonal 1-12-3 against fish CYP1A (Park et al. 1986), anti-myosin heavy chain monoclonal MF20 (Developmental Studies Hybridoma Bank, University of Iowa, Iowa City, IA, USA) (Bader et al. 1982), and anti-atrial myosin heavy chain monoclonal S46 (Berdougo et al. 2003). For CYP1A immunofluorescence, embryos were fixed overnight in 4% phosphate-buffered paraformaldehyde, and MF20/S46 immunofluorescence was assessed in embryos fixed in either paraformaldehyde or methanol plus 10% DMSO. Processing for immunofluorescence was carried out as described previously (Incardona et al. 2004). Secondary antibodies were AlexaFluor488-conjugated goat anti-mouse IgG1 (S46) and AlexaFluor568-conjugated goat anti-mouse IgG2b (MF20), both from Molecular Probes (Eugene, OR, USA). Immunolabeled embryos were mounted in glycerol or 3% methylcellulose and imaged using a Zeiss LSM 5 Pascal confocal system with Ar and HeNe lasers (Carl Zeiss Advanced Imaging Microscopy, Jena, Germany). For semiquantitative comparisons, treated or control embryos were marked by tail clipping and mixed for antibody labeling, mounted together, and imaged with identical settings.
PAH analysis.
Water samples (200 mL) were collected and stored in brown glass bottles with 20 mL dichloromethane at 4°C for up to 7 days before extraction. After addition of deuterated internal standards, samples were extracted twice with dichloromethane (25 mL each time) for 2 min each with 2 min separation times using 1-L separatory funnels. For quality assurance, a known mixture of PAHs was added to 200 mL zebrafish system water (spiked blank) and extracted. The extracts were processed and hydrocarbons analyzed by gas chromatography–mass spectrometry using selected ion monitoring as previously described (Sloan et al. 2004). The accuracy of the hydrocarbon analyses was estimated by recoveries from the spiked blank, which ranged from 69 to 111% for 17 different PAHs (naphthalene, 102%; fluorene, 85%; dibenzothiophene, not determined; phenanthrene, 98%; chrysene, 83%). Total PAH concentrations were calculated by summing concentrations of individual PAHs. Relative PAH concentrations were calculated as the ratio of PAH concentration to the total PAH concentration.
Results
Differential activation of AhR1 and AhR2 by model PAHs.
We first examined the relationship between CYP1A induction via the AhR and the developmental toxicity of individual PAH compounds [structures shown in Supplemental Figure 1; Supplemental Material available online (http://ehp.niehs.nih.gov/docs/2005/8230/supplement.pdf)]. To determine the roles of the two zebrafish AhRs, we used the previously described (Prasch et al. 2003; Teraoka et al. 2003) translation-blocking AhR2 morpholino (AhR2-MO), and designed AhR1 splice-blocking morpholinos targeting either the exon 2/intron 2 junction or the intron 2/exon 3 junction (Supplemental Figure 1; Supplemental Material available online (http://ehp.niehs.nih.gov/docs/2005/8230/supplement.pdf)]. Morpholinos designed to block splicing or translation have similar efficacy (Draper et al. 2001). Coinjection of AhR2-MO and AhR1 exon 2/intron 2 morpholino failed to produce normal embryos, so all studies used intron 2/exon 3 morpholino (AhR1-MO). Nonfunctional control morpholinos included the AhR2 sequence with a four-base mismatch (AhR2-MIS) and the AhR1 exon 2/intron 2 sequence with a five-base mismatch (AhR1-MIS).
At nominal concentrations ranging from 10 to 60 μM, phenanthrene or dibenzothiophene causes dose-dependent changes in cardiac rhythm ranging from bradycardia through partial (2:1) to complete atrio-ventricular (AV) conduction block (Incardona et al. 2004). In embryos exposed to approximately 30–60 μM phenanthrene or dibenzothiophene with AV block and edema [Figure 1C, Supplemental Movie 1; Supplemental Material available online (http://ehp.niehs.nih.gov/docs/2005/8230/supplement.pdf)], relatively weak CYP1A immunofluorescence was observed predominantly in vessels most proximal to the heart, including the first aortic arch and the carotid artery (Figure 1D,G). However, CYP1A immunofluorescence was not observed in endothelial cells lining the heart in embryos exposed to either phenanthrene or dibenzothiophene (Figure 1D,G). Injection of AhR2-MO largely blocked CYP1A induction by dibenzothiophene and phenanthrene (Figure 1F,H; Table 1). However, despite this effect on CYP1A induction, AhR2 morphants (i.e., MO injected) were not protected from phenanthrene- or dibenzothiophene-induced cardiac dysfunction (Figure 1E, Table 1). The same types of cardiac arrhythmia were observed in AhR2 morphants exposed to phenanthrene or dibenzothiophene at the same developmental stage as controls [~ 36 hpf; data not shown, Supplemental Movie 1; Supplemental Material available online (http://ehp.niehs.nih.gov/docs/2005/8230/supplement.pdf)]. Although we did not quantify changes in the dose response, AhR2 morphants generally had a higher degree of AV block than did controls at a given phenanthrene or dibenzothiophene concentration, consistent with a protective effect of CYP1A induction. Injection of AhR1-MO alone did not prevent CYP1A induction by phenanthrene (data not shown) and, in combination with AhR2-MO, did not prevent tricyclic PAH toxicity (Supplemental Movie 1; Supplemental Material available online (http://ehp.niehs.nih.gov/docs/2005/8230/supplement.pdf)]. Therefore, although the tricyclic PAHs phenanthrene and dibenzothiophene induce CYP1A weakly in some blood vessels through AhR2, the primary toxicity of these PAHs in fish embryos is AhR independent, and their cardiac effects are not associated with AhR activation or CYP1A induction in the endocardium.
In contrast to the tricyclic PAHs, the effects of pyrene during zebrafish development overlap considerably with those previously reported for TCDD exposure. Despite widespread induction of CYP1A throughout the vascular endothelium observed as early as 36 hpf (Table 1 and data not shown), the overt signs of pyrene toxicity do not appear until after 80 hpf (Figure 2). By 100 hpf, uninjected or AhR2-MIS–injected embryos exposed to pyrene showed edema (Figure 2A,C and data not shown), cell death in the neural tube (Figure 2E), and anemia (Figure 2G), as well as CYP1A immunofluorescence in the liver (Figure 2K) and throughout the vascular endothelium of the trunk and head (Figure 2I,K). Most pyrene-exposed larvae die later in the fifth day of development (Table 1). Injection of AhR2-MO largely prevented the morphologic and lethal effects of pyrene exposure (Figure 2B,D,F,H; Table 1) and markedly reduced the levels of CYP1A immunofluorescence assayed at both 48 hpf (Table 1) and 100 hpf (Figure 2J,L). At comparable time points, CYP1A morphants exposed to pyrene generally had defects that were less severe than those of controls. For example, pericardial edema was more severe at 96 hpf in AhR-MIS–injected larvae than in CYP1A morphants, indicated by pericardial cross-sectional areas of 0.034 ± 0.008 mm2 and 0.021 ± 002 mm2, respectively (n = 10 for each, t-test p < 0.001). However, CYP1A-MO injection did not ultimately protect from pyrene toxicity because all of the defects associated with pyrene exposure (including lethality) appeared in CYP1A morphants 12–18 hr later than in controls, possibly due to loss of MO efficacy at later stages (Table 1). These findings indicate that, like phenanthrene and dibenzothiophene, pyrene selectively activates AhR2, but unlike these tricyclic PAHs, pyrene toxicity is AhR dependent. Moreover, delay of pyrene toxicity in CYP1A morphants suggests the involvement of a toxic CYP1A-derived pyrene metabolite.
A trivial explanation for the absence of embryonic toxicity for some compounds [e.g. chrysene (Incardona et al. 2004)] could be a lack of tissue uptake during the course of the exposure. For example, because of differences in water solubility of PAHs with different ring arrangements but similar molecular weight (e.g., 0.67 μM pyrene vs. 0.009 μM chrysene), the effective exposure levels would be very different. However, exposure to 9 μM chrysene, which produces no apparent toxic effects, resulted in robust CYP1A immunofluorescence throughout the epidermis and the vascular endothelium (Figure 3A,G). Most cranial (Figure 3G and data not shown) and trunk vessels (data not shown), as well as the endothelial cells lining the cardiac ventricle (Figure 3H), expressed CYP1A after chrysene exposure. Remarkably, injection of AhR2-MO blocked induction of CYP1A in the epidermis by chrysene while leaving the vascular induction intact and generally increased (Figure 3B). In contrast, injection of AhR1-MIS (Figure 3C) or AhR1-MO (Figure 3D) had no effect on CYP1A induction by chrysene, but coinjection of both AhR1-MO and AhR2-MO markedly reduced both epidermal and endothelial CYP1A immunofluorescence (Figure 3E). Injection of CYP1A-MO eliminated virtually all CYP1A immunofluorescence associated with chrysene exposure (Figure 3F). These findings indicate that chrysene can activate both AhR1 and AhR2 in a tissue-specific manner. Although either AhR1 or AhR2 can mediate the vascular induction of CYP1A by chrysene, the epidermal induction of CYP1A by chrysene is AhR2 specific. However, AhR activation and CYP1A induction by chrysene are not associated with any overt developmental toxicity or cardiac dysfunction (despite CYP1A induction in cardiac endothelial cells).
Embryonic cardiac dysfunction and the weathered crude oil syndrome.
We exposed zebrafish embryos to ANS crude oil weathered using two methods: generation of a WAF by an overnight high-energy spin of an oil–water mixture and incubation in the effluent from a column loaded with oiled gravel (OGE) or control (clean) gravel (CGE). Although chemical analysis demonstrated that these two methods produced different degrees of weathering [Table 2, Supplemental Figures 2 and 3; Supplemental Material available online (http://ehp.niehs.nih.gov/docs/2005/8230/supplement.pdf)], exposure of embryos to both preparations (WAF exposed and OGE exposed) produced identical types of biologic effects (Figure 4 and data not shown).
In general, the defects in WAF-exposed embryos were more severe, consistent with its higher degree of weathering (i.e., larger tricyclic PAH fraction). By 63 hpf, WAF- or OGE-exposed embryos showed a suite of defects that overlapped considerably with exposure to individual model tricyclic PAHs, but with additional features not observed with the model compounds. Grossly, all embryos exposed to weathered crude oil showed dorsal curvature of the body axis; mild to moderate pericardial edema was seen in OGE-exposed embryos (Figure 4B,D), and yolk sac edema was seen in WAF-exposed embryos (data not shown). In both types of exposures, changes in cardiac function were the earliest observed defects. Mild pericardial edema and reduced blood flow associated with poor cardiac contractility and bradycardia were apparent in WAF-exposed embryos at 33 hpf (data not shown) and by 36–39 hpf in OGE-exposed embryos [Supplemental Movie 2; Supplemental Material available online (http://ehp.niehs.nih.gov/docs/2005/8230/supplement.pdf)]. Usually, the first and mildest sign of cardiac dysfunction was regurgitation of erythrocytes from the atrium into the yolk sac [Supplemental Movie 3; Supplemental Material available online (http://ehp.niehs.nih.gov/docs/2005/8230/supplement.pdf)]. Staining with the chamber-specific antibodies MF20 and S46 demonstrated that subtle delay or disruption of cardiac looping was associated with the earliest observed defects in cardiac function (Figure 4E,F). Cardiac looping was assessed quantitatively by measuring the angle between the cardiac chambers relative to the left–right body axis (Figure 4E,F). CGE-exposed embryos fixed at 39 hpf had a mean inter-chamber angle of 28 ± 5° (n = 6), whereas OGE-exposed embryos with weak contractility had a mean interchamber angle of 50 ± 2° (n = 7, t-test p < 0.01). All WAF- or OGE-exposed embryos showed severely abnormal cardiac looping at subsequent stages (typically by 54–64 hpf), and exposures were terminated at 72–80 hpf. The late cardiac morphology typically showed chambers that were stretched along the anterior–posterior axis, with the ventricle stiff and reduced in diameter and the atrium relatively dilated (Figure 4D). Atypical movement was often observed in the wall of the cardiac outflow tract or bulbus arteriosus [Supplemental Movie 4; Supplemental Material available online (http://ehp.niehs.nih.gov/docs/2005/8230/supplement.pdf)]. In most cases, and in particular in WAF-exposed embryos, both cardiac chambers ultimately collapsed into a stringlike structure when embryos were exposed for longer durations (data not shown).
Intracranial hemorrhage and doming of the head due to ventricular edema was very common among embryos exposed to either preparation (85% of OGE-exposed embryos in one experiment, n = 66) and most often involved the mesencephalon/third ventricle and hindbrain/fourth ventricle (Figure 4H). Hemorrhage occasionally involved the branchial arches or eye but was never observed in tissues outside the head (data not shown). The sensitive period for intracranial hemorrhage was late during the second day of development to early in the third day; although not observed at 30 hpf, hemorrhage first occurred between 30 and 39 hpf, with the number of affected embryos maximal by 58–63 hpf.
Finfold defects involving all fins and consisting of irregular edges or blisters were also common by 63–72 hpf (94% of OGE-exposed embryos, n = 66; Figure 4J,L). No abnormalities were observed in CGE-exposed embryos.
Although we did not observe arrhythmias typical of AV conduction block in OGE-exposed embryos, the total PAH levels were well below the levels at which either phenanthrene or dibenzothiophene alone blocks AV conduction. The key finding is that the complex PAH mixture that comprises weathered crude oil caused early cardiac dysfunction similar to effects of model tricyclic PAHs, rather than a syndrome that arises later during the larval period, such as that associated with pyrene or TCDD exposure. Weathered crude oil also produced additional defects in zebrafish embryos (intracranial hemorrhage and finfold defects) that either are not observed with exposure to individual model PAHs or dioxins, or arise during an earlier developmental stage than AhR-mediated TCDD toxicity.
A protective role for the AhR pathway.
Embryos exposed to WAF or OGE showed identical robust patterns of CYP1A induction by 36 hpf. In OGE-exposed embryos, CYP1A immunofluorescence (Figure 5, green) was strong in the epidermis (Figure 5B), cranial vasculature (Figure 5C), trunk vasculature (data not shown), and the endothelium lining both the atrium (Figure 5D) and ventricle (Figure 5E), but was absent in the myocardium (Figure 5D,E, red fluorescence). CGE-exposed embryos showed only background immunofluorescence (Figure 5A). As we observed for chrysene, injection of AhR2-MO blocked the epidermal induction of CYP1A by weathered crude oil, leaving the vascular induction intact (Figure 5I). Confocal imaging showed that the vascular CYP1A immunofluorescence was actually higher in AhR2 morphants than in uninjected or control morphants (data not shown). Nevertheless, AhR2-MO injection, in three separate experiments with maximum morpholino levels allowing viability, did not influence the cardiac dysfunction or intracranial hemorrhage caused by weathered crude oil exposure (n > 200, data not shown). In contrast, coinjection of AhR1-MO and AhR2-MO markedly reduced CYP1A immunofluorescence in both the epidermis and vasculature in OGE-exposed embryos (Figure 5J). CYP1A immunofluorescence was virtually eliminated in OGE-exposed CYP1A morphants (Figure 5K). In contrast to what we observed with pyrene, defects were more severe in OGE-exposed morphants with an inactive AhR pathway (Figure 5F). At the earliest time point assayed (39 hpf), both AhR1/AhR2 double morphants and CYP1A morphants showed a significantly larger fraction with no circulating erythrocytes due to severe reductions in contractility (~ 50%, chi-square p < 0.001), and weak circulation with atrial regurgitation (~150% of controls, chi-square p < 0.001). Although a very small but statistically significant increase in cardiac dysfunction was observed in either control or antisense morpholino-injected embryos incubated in CGE, the nature of the cardiac defects was different from that seen with weathered crude oil exposure and consistent with injection artifact (data not shown). Overall, these data indicate that the AhR pathway actually provides some degree of protection from PAH-induced cardiac toxicity.
Because cardiac output and intracranial hemorrhage are likely to be functionally related, it was difficult to determine whether AhR activation or CYP1A induction played a causal role in intracranial hemorrhage. Both CYP1A morphants and AhR1/AhR2 double morphants had a reduced overall occurrence of intracranial hemorrhage, but this was most likely due to the preponderance of embryos with no circulation. In these animals, erythrocytes would pool in the yolk sac. Intracranial hemorrhage was still present among most embryos that had circulation and a lack of CYP1A immunofluorescence (data not shown), which strongly suggests that the underlying pathophysiology is AhR independent. Because morpholinos are most effective at earlier developmental stages, we focused primarily on cardiac function. We did not assess whether the finfold defects resulting from weathered crude oil exposure required a functional AhR pathway.
Discussion
PAHs of different molecular weight and structure generally are thought to share a common toxic mechanism mediated by the AhR pathway. The data presented here demonstrate that different PAH classes act via distinct toxic mechanisms during fish development. Moreover, our data reveal the complexity of different PAH classes with respect to their distribution in tissues and interactions with AhR family members. The very low abundance of zebrafish AhR1 mRNA coupled with a lack of TCDD binding and transactivation ability suggested that it encodes a nonclassical receptor (Andreasen et al. 2002). However, non-halogenated PAHs were not tested in those studies, and it is clear from our results that AhR1 can be activated by some ligands. Although generally nontoxic to zebrafish embryos, chrysene appears to activate AhR1 and AhR2 in vivo in different tissues. Although the tricyclic PAHs and pyrene cause primarily vascular endothelial CYP1A induction, exposure to chrysene results in extensive CYP1A expression in the epidermis, as well. Although either AhR1 or AhR2 can mediate the vascular induction of CYP1A by chrysene, the epidermal induction of CYP1A by chrysene is AhR2 specific. The simplest interpretation is that chrysene accumulates in or binds to epidermal cells, possibly due to its higher hydrophobicity, whereas pyrene and the tricyclic PAHs do not, and that epidermal cells do not express AhR1. In contrast, based on the present results, we would predict that phenanthrene, dibenzothiophene, and pyrene selectively bind zebrafish AhR2 over AhR1.
Of the four model PAHs examined here, only one—pyrene—has dioxin-like (i.e., AhR-dependent) toxicity. Although there is considerable overlap between the effects of pyrene and TCDD on zebrafish larvae, there are distinct features to both syndromes. For example, exposure to pyrene during embryogenesis does not result in the jaw malformations observed with TCDD exposure. Similarly, although there is increased apoptotic cell death in the brain of TCDD-exposed larvae (Dong et al. 2002), the developing spinal cord does not show the widespread cell death as observed with pyrene exposure. Most important, although CYP1A knockdown failed to modulate TCDD toxicity (Carney et al. 2004), the onset of pyrene toxicity was markedly delayed in CYP1A morphants. Most evidence suggests that TCDD toxicity is related to its poor metabolism by CYP enzymes and concomitant bioaccumulation with persistent, futile AhR activation. Moreover, the toxicity of halogenated aromatic compounds generally correlates with affinity for the AhR (Hestermann et al. 2000) and, by extension, the degree of CYP1A induction. Of the model PAHs tested here, chrysene is the most potent CYP1A inducer [two orders of magnitude greater than pyrene (Barron et al. 2004)] but the least toxic. Thus, even though metabolites of some individual PAHs (e.g., pyrene) may be embryotoxic, CYP1A induction or catalytic activity in itself appears to play little role in the developmental toxicity of petrogenic PAHs. Similarly, because chrysene had no overt toxicity in AhR1/AhR2 double or CYP1A morphants, neither unmetabolized chrysene nor CYP1A-derived metabolites are likely to contribute to the embryotoxicity of weathered crude oil.
Compared with urbanized aquatic systems, where PAHs are almost always mixed with other contaminants (e.g., metals, persistent organochlorines, pesticides), the weathered ANS crude oil model provides a fairly simple system for dissecting the effects of an environmentally relevant PAH mixture. The effects of weathered crude oil exposure have little in common with pyrene or TCDD toxicity and instead overlap to a surprising degree with those of individual tricyclic PAHs. The earliest and most pronounced effect of weathered crude oil is on cardiac function during looping stages and subsequent cardiac morphogenesis. These findings, coupled with our previous work (Incardona et al. 2004), indicate that the tricyclic components of weathered crude oil are the major toxic fraction and that the abundant alkylated PAHs probably have cardiac-specific activities similar to the nonalkylated homologs. Moreover, toxicity can be attributed to the parent compounds because metabolism of PAHs via the AhR pathway confers a degree of protection. These findings have widespread implications for ecological assessment and natural resource management.
First, PAH-contaminated environments are often characterized by mixtures derived from both petrogenic (e.g., oil spills) and pyrogenic (e.g., vehicle exhaust) sources. Within a mixture, the composition of individual PAHs can vary considerably from low- to high-molecular-weight compounds. Because different PAHs act on fish embryos via independent toxic mechanisms, understanding the cumulative toxicity of PAH mixtures will be more challenging than previously appreciated.
Second, these findings implicate cardiovascular dysfunction as a major proximal cause of deformities associated with exposure to petrogenic PAHs. This raises the possibility that lower exposure concentrations cause subtle cardiovascular effects in fish that otherwise appear normal. This might explain, for example, the reduced marine survival of pink salmon exposed as embryos to lower levels of weathered ANS crude oil (Heintz et al. 2000).
Third, the variety of cardiac function abnormalities we observed with both model tricyclic PAHs and weathered crude oil suggests several potential targets in the myocardium, which, in addition to HERG potassium channels, includes sarcoplasmic or plasma membrane calcium channels, as well as gap junctions. In this light, it is notable that among PAHs with two to five rings, tricyclic compounds were the most potent in blocking dye coupling via gap junctions in an in vitro assay (Blaha et al. 2002). The identification of direct cardiovascular pathophysiology as a key process underlying petrogenic PAH toxicity provides a conceptual framework for the development of better tools for assessing PAH effects in wild freshwater and marine fish.
Fourth, our findings emphasize the current limitations to assessing the effects of oil spills and other sources of aquatic PAH contamination. For many years, measuring CYP1A in field-collected samples has been the basis for assessing ecological damage and recovery after oil spills or remediation efforts in urbanized watersheds. However, CYP1A appears to play a protective rather than a causal role in petrogenic PAH toxicity. This greatly reduces the significance of CYP1A as a biomarker of PAH effect.
Finally, these results highlight the relative toxicity of low-molecular-weight tricyclic PAHs, which previously were considered to be weakly toxic based on the AhR agonist model. Phenanthrene inputs into the environment have remained constant in the last two decades, even in areas receiving inputs of predominantly pyrogenic PAHs (Lima et al. 2002). An understanding of the impacts of tricyclic PAHs on at-risk fish species thus deserves greater emphasis.
Supplementary Material
Supplemental Table 1 Supplemental Movie 1 Supplemental Movie 2 Supplemental Movie 3 Supplemental Movie 4 Supplemental Material is available online at http://ehp.niehs.nih.gov/docs/2005/8230/supplement.pdf
We thank J. Stegeman and B. Woodin for providing monoclonal antibody 1-12-3; T. Linbo for fish husbandry; and P. Swanson and K. Peck-Miller for manuscript reviews.
This work was supported by the National Oceanic and Atmospheric Administration Coastal Storms Program; by grants to J.P.I. and N.L.S. from the California Department of Fish and Game’s Oil Spill Response Trust Fund through the Oiled Wildlife Care Network, University of California, Davis; and by a grant to H.T. from the Japanese Ministry of Education, Culture, Sports, Science and Technology, and Cooperative Research from Active Research in Rakuno Gakuen University. J.P.I. was supported by the National Academies/National Research Council Research Associateships Program.
Figure 1 AhR2 knockdown prevents CYP1A induction by phenanthrene and dibenzothiophene, but not cardiac dysfunction. (A–F) Lateral light microscopic views of live embryos at 48 hpf (anterior at left) are paired with corresponding ventral confocal images (anterior at top) of CYP1A (green) and myosin heavy chain (red) immunofluorescence. (A, B) Embryo exposed to solvent (DMSO). (C, D) AhR2-MIS–injected embryo exposed to 28 μM phenanthrene. (E, F) AhR2 morphant exposed to 28 μM phenanthrene. Black arrowheads and arrows (C, E ) indicate pericardial and yolk sac edema, respectively. CYP1A immunofluorescence induced by phenanthrene (D) in the cranial division of the internal carotid artery (CrDI) and optic artery (OA) was blocked by AhR2-MO injection (F). Solid white arrowheads indicate cross-reactive immunofluorescence in the jaw cartilage, and unfilled white arrowheads indicate the ventricular myocardium. (G, H) Higher magnification confocal images showing CYP1A (green) and myocardial myosin heavy chain (red) immunofluorescence at 48 hpf in embryos with cardiac dysfunction after exposure to 28 μM dibenzothiophene (lateral views with anterior at left). In an uninjected embryo (G), the proximal portion of the mandibular arch (AA1, arrow) is CYP1A+, whereas the ventricular endothelium is CYP1A− (asterisk). Only cross-reactive signal is seen in the jaw cartilage (arrowhead) in an AhR2 morphant (H). Bars = 100 μm (A–F) and 50 μm (G, H).
Figure 2 AhR2 morphants are resistant to pyrene toxicity. Control (uninjected or AhR2-MIS injected) and AhR2 morphant embryos were exposed to 5 μM pyrene through 100 hpf. Lateral (A, B) and dorsal (C, D) views showing edema (arrows) in uninjected larvae. Higher magnification light micrographs of the trunk region of uninjected (E, G) and AhR2 morphant (F, H) larvae showing cell death (E, granular appearance) in the neural tube (nt, neural tube; nc, notochord) and a reduction of erythrocytes (G, arrows) in the ventral aorta (VA) and caudal vein (CV) of uninjected larvae. (I–L) CYP1A immunofluorescence in the trunk (I, J) and head regions (K, L) of pyrene-exposed larvae. In AhR2-MIS–injected larvae (I, K) the vasculature (DLAV, dorsal longitudinal anastomotic vessel; Se, intersegmental vessels; AA, branchial arches) and liver are CYP1A+, whereas only weak signal is seen in the liver of the AhR morphant (J, L). In (E–L,) anterior is to the left and dorsal at top. Bars = 200 μm (A–D) and 50 μm (E–L).
Figure 3 Chrysene induces CYP1A through both AhR1 and AhR2. All images show CYP1A immunofluorescence at 72 hpf (A, B, G, H) or 48 hpf (C–F) after exposure to 9 μM chrysene from 6 hpf. (A–F) Lateral epifluorescent images with anterior to the left in uninjected (A), AhR2 morphant (B), AhR1-MIS–injected (C), AhR1 morphant (D), AhR1/AhR2 double morphant (E), and CYP1A morphant (F ) embryos. Epidermal CYP1A is seen as punctate fluorescence on the surface of the embryos. Immunofluorescent signal in the otic capsule and jaw cartilage was often observed in unexposed embryos. This signal was resistant to CYP1A morpholino (F) and is therefore likely to represent a cross-reactive protein. (G, H) Confocal images of uninjected chrysene-exposed embryos. (G) Three-dimensional confocal projection (180 μm series of optical sections) of CYP1A immunofluorescence, ventral view with anterior at top. Arrows indicate CYP1A+ blood vessels; AA1, mandibular arch; CrDI, cranial division of the internal carotid artery; OA, optic artery; ORA, opercular artery. (H) Confocal optical section through the cardiac chambers (anterior at top) with CYP1A (green) and myosin heavy chain (red) immunofluorescence. The asterisk (*) indicates CYP1A+ endothelial cells lining the ventricle. Bars = 250 μm (A–F) and 50 μm (G, H).
Figure 4 Defects resulting from embryonic exposure to OGE. PAH levels are shown in Table 2 (results for column 1) and Supplemental Figure 2 [Supplemental Material available online (http://ehp.niehs.nih.gov/docs/2005/8230/supplement.pdf)]. (A, B) Gross appearance at 64 hpf of CGE-exposed (A) or OGE-exposed (B) larvae. (C, D) Cardiac morphology at 64 hpf in CGE-exposed (C) and OGE-exposed (D) larvae. (E, F) Cardiac chamber–specific immunofluorescence (red, ventricle; green, atrium) in CGE-exposed (E) and OGE-exposed (F) larvae at 39 hpf; dashed white lines indicate the angles measured to assess looping. (G, H) High-magnification views of the midbrain–hindbrain junction in CGE-exposed (G) and OGE-exposed (H) larvae. Arrows indicate red tinge from extracellular hemoglobin; arrowheads mark extravascular erythrocytes, and the floor of the brain ventricles is marked with unfilled arrowheads. (I–L) High-magnification views of pectoral (I, J) and caudal (K, L) fins in CGE-exposed (I, K) and OGE-exposed (J, L) larvae. The finfolds of OGE-exposed larvae have irregular margins and blisters (arrows). Bars = 500 μm (A, B) and 50 μm (C–L).
Figure 5 AhR1/AhR2 or CYP1A morphants are more sensitive to weathered crude oil toxicity. (A–E) Confocal immunofluorescence images of CYP1A (green) and myosin heavy chain marking myocardium (red). CGE-exposed embryos showed no CYP1A immunofluorescence at 39 hpf (A), whereas OGE-exposed embryos showed intense immunofluorescence in the epidermis (B; 180 μm series of optical sections, vasculature of the head (C, optical section; PICA, primitive internal carotid artery; AA1, mandibular arch), and endocardium (asterisks in D, E) in both the atrium (D) and ventricle (E). (F) Cardiac function (good: strong forward flow; weak: forward flow with atrial regurgitation; or none: erythrocytes pooled in the yolk sac) at 39–40 hpf in CGE- and OGE-exposed embryos that were uninjected, control morpholino injected, AhR1/AhR2 double morphant, or CYP1A morphant [Supplemental Movie 3; Supplemental Material available online (http://ehp.niehs.nih.gov/docs/2005/8230/supplement.pdf)] Bars represent the percentage of embryos in the three classifications (numbers within each bar indicate n). The data represent sets of embryos that were exposed sequentially in the effluent of a single pair of columns (control and oiled) run over 8 days. PAH levels are shown in Table 2 (results for column 6) and Supplemental Figure 3 [Supplemental Material available online (http://ehp.niehs.nih.gov/docs/2005/8230/supplement.pdf)]. Because weathering increases with duration of column flow, AhR1/AhR2 double morphants (AhR1+2-MO) and corresponding controls (bars under OGE days 0–2) were exposed to less weathered oil than were the CYP1A morphants and controls (bars under OGE days 5–7; see also Table 2). The increased severity of cardiac dysfunction in control embryos exposed during days 5–7 is consistent with the higher tricyclic component. (G–K) CYP1A immunofluorescence (epifluorescent images) of representative OGE-exposed embryos from experiments shown in (F) and similar experiments with AhR2 single morphants, fixed at 42 hpf. (G) Uninjected control with intense epidermal CYP1A signal. (H) AhR2-MIS–injected control. (I) AhR2 morphant with loss of epidermal CYP1A signal and robust vascular immunofluorescence. (J) AhR1/AhR2 double morphant with markedly reduced overall immunofluorescence. (K) CYP1A morphant with background staining only. Bars= 100 μm (A, B), 50 μm (C–E), and 500 μm (G–K).
Table 1 AhR2 morpholino prevents pyrene toxicity but not tricyclic PAH toxicity (%).
Treatment CYP1A+ at 48 hpf Cardiac dysfunction at 48 hpf Viability at 106 hpf
AhR2-MIS + 28 μM phenanthrene 100 (n = 17) 100 (n = 22) —
AhR2-MO + 28 μM phenanthrene 7 (n = 27) 100 (n = 72) —
Uninjected + 28 μM dibenzothiophene 100 (n = 20) 100 (n = 20) —
AhR2-MO + 28 μM dibenzothiophene 0 (n = 17) 100 (n = 17) —
Uninjected + DMSO 0 (n = 43) — 100 (n = 90)
Uninjected + 5 μM pyrene 100 (n = 39) — 9 (n = 93)
AhR2-MIS + 5 μM pyrene 94 (n = 16) — 0 (n = 70)
AhR2-MO + 5 μM pyrene 3 (n = 31) — 86 (n = 56)
CYP1A-MO + 5 μM pyrene 7 (n = 59) — 92 (n = 39)a
—, not applicable.
a Mortality delayed 18 hr relative to control; 0% viable at 124 hpf.
Table 2 Summary of PAH levels in weathered crude oil exposures.
Weathered oil preparation (experiment) Total PAH (μg/L) Tricyclic PAHs (%)
WAF day 1 1,549 43.7
WAF day 4 264 55.8
OGE day 1 (column 1) 78.0 16.9
OGE day 4 (column 1) 53.5 24.1
OGE day 0 (column 6) 111.1 17.6
OGE day 2 (column 6) 53.2 25.8
OGE day 7 (column 6) 52.7 28.9
==== Refs
References
Andreasen EA Hahn ME Heideman W Peterson RE Tanguay RL 2002 The zebrafish (Danio rerio ) aryl hydrocarbon receptor type 1 is a novel vertebrate receptor Mol Pharmacol 62 234 249 12130674
Bader D Masaki T Fischman DA 1982 Immunochemical analysis of myosin heavy chain during avian myogenesis in vivo and in vitro J Cell Biol 95 763 770 6185504
Barron MG Heintz RA Rice SD 2004 Relative potency of PAHs and heterocycles as aryl hydrocarbon receptor agonists in fish Mar Environ Res 58 95 100 15178019
Belair CD Peterson RE Heideman W 2001 Disruption of erythropoiesis by dioxin in the zebrafish Dev Dyn 222 581 594 11748828
Berdougo E Coleman H Lee DH Stainier DY Yelon D 2003 Mutation of weak atrium/atrial myosin heavy chain disrupts atrial function and influences ventricular morphogenesis in zebrafish Development 130 6121 6129 14573521
Blaha L Kapplova P Vondracek J Upham B Machala M 2002 Inhibition of gap-junctional intercellular communication by environmentally occurring polycyclic aromatic hydrocarbons Toxicol Sci 65 43 51 11752684
Carls MG Rice SD Hose JE 1999 Sensitivity of fish embryos to weathered crude oil: Part I. Low-level exposure during incubation causes malformations, genetic damage, and mortality in larval Pacific herring (Clupea pallasi ) Environ Toxicol Chem 18 481 493
Carney SA Peterson RE Heideman W 2004 2,3,7,8-Tetrachlorodibenzo-p -dioxin activation of the aryl hydrocarbon receptor/aryl hydrocarbon receptor nuclear translocator pathway causes developmental toxicity through a CYP1A-independent mechanism in zebrafish Mol Pharmacol 66 512 521 15322242
Couillard CM 2002 A microscale test to measure petroleum oil toxicity to mummichog embryos Environ Toxicol 17 195 202 12112627
Dong W Teraoka H Yamazaki K Tsukiyama S Imani S Imagawa T 2002 2,3,7,8-Tetrachlorodibenzo-p -dioxin toxicity in the zebrafish embryo: local circulation failure in the dorsal midbrain is associated with increased apoptosis Toxicol Sci 69 191 201 12215674
Draper BW Morcos PA Kimmel CB 2001 Inhibition of zebrafish fgf8 pre-mRNA splicing with morpholino oligos: a quantifiable method for gene knockdown Genesis 30 154 156 11477696
Ensembl Zebrafish Genome Browser 2005. Ensembl Homepage. Available: http://www.ensembl.org/Danio_rerio/index.html [accessed 27 October 2005].
Fabricius KE 2005 Effects of terrestrial runoff on the ecology of corals and coral reefs: review and synthesis Mar Pollut Bull 50 125 146 15737355
GenBank 2005. GenBank Overview. Available: http://www.ncbi.nlm.nih.gov/Genbank/index.html [accessed 27 October 2005].
Hahn ME 2002 Aryl hydrocarbon receptors: diversity and evolution Chem Biol Interact 141 131 160 12213389
Heintz RA Rice SD Wertheimer AC Bradshaw RF Thrower FP Joyce JE 2000 Delayed effects on growth and marine survival of pink salmon Oncorhynchus gorbuscha after exposure to crude oil during embryonic development Mar Ecol Prog Ser 208 205 216
Heintz RA Short JW Rice SD 1999 Sensitivity of fish embryos to weathered crude oil: Part II. Increased mortality of pink salmon (Oncorhynchus gorbuscha ) embryos incubating downstream from weathered Exxon Valdez crude oil Environ Toxicol Chem 18 494 503
Henry TR Spitsbergen JM Hornung MW Abnet CC Peterson RE 1997 Early life stage toxicity of 2,3,7,8-tetrachlorodibenzo-p -dioxin in zebrafish (Danio rerio ) Toxicol Appl Pharmacol 142 56 68 9007034
Hestermann EV Stegeman JJ Hahn ME 2000 Relative contributions of affinity and intrinsic efficacy to aryl hydrocarbon receptor ligand potency Toxicol Appl Pharmacol 168 160 172 11032772
Incardona JP Collier TK Scholz NL 2004 Defects in cardiac function precede morphological abnormalities in fish embryos exposed to polycyclic aromatic hydrocarbons Toxicol Appl Pharmacol 196 191 205 15081266
Karchner SI Powell WH Hahn ME 1999 Identification and functional characterization of two highly divergent aryl hydrocarbon receptors (AHR1 and AHR2) in the teleost Fundulus heteroclitus . Evidence for a novel subfamily of ligand-binding basic helix loop helix-Per-ARNT-Sim (bHLH-PAS) factors J Biol Chem 274 33814 33824 10559277
Kimmel CB Ballard WW Kimmel SR Ullmann B Schilling TF 1995 Stages of embryonic development of the zebrafish Dev Dyn 203 253 310 8589427
Langheinrich U Vacun G Wagner T 2003 Zebrafish embryos express an orthologue of HERG and are sensitive toward a range of QT-prolonging drugs inducing severe arrhythmia Toxicol Appl Pharmacol 193 370 382 14678746
Li D Daler D 2004 Ocean pollution from land-based sources: East China Sea, China Ambio 33 107 113 15083656
Lima ALC Eglinton TI Reddy CM 2002 High-resolution record of pyrogenic polycyclic aromatic hydrocarbon deposition during the 20th century Environ Sci Technol 37 53 61 12542290
Marty GD Short JW Dambach DM Willits NH Heintz RA Rice SD 1997 Ascites, premature emergence, increased gonadal cell apoptosis, and cytochrome P4501A induction in pink salmon larvae continuously exposed to oil-contaminated gravel during development Can J Zool-Rev Can Zool 75 989 1007
National Research Council 2003. Oil in the Sea III: Inputs, Fates, and Effects. Washington, DC:National Academies Press.
Nebert DW Dalton TP Okey AB Gonzalez FJ 2004 Role of aryl hydrocarbon receptor-mediated induction of the CYP1 enzymes in environmental toxicity and cancer J Biol Chem 279 23847 23850 15028720
Park SS Miller H Klotz AV Kloepper-Sams PJ Stegeman JJ Gelboin HV 1986 Monoclonal antibodies to liver microsomal cytochrome P-450E of the marine fish Stenotomus chrysops (scup): cross reactivity with 3-methylcholanthrene induced rat cytochrome P-450 Arch Biochem Biophys 249 339 350 2428309
Peterson CH Rice SD Short JW Esler D Bodkin JL Ballachey BE 2003 Long-term ecosystem response to the Exxon Valdez oil spill Science 302 2082 2086 14684812
Peterson RE Theobald HM Kimmel GL 1993 Developmental and reproductive toxicity of dioxins and related compounds: cross-species comparisons Crit Rev Toxicol 23 283 335 8260069
Pew Oceans Commission 2003. America’s Living Oceans: Charting a Course for Sea Change. Arlington, VA:Pew Oceans Commission.
Phillips DH 1983 Fifty years of benzo(a )pyrene Nature 303 468 472 6304528
Pollino CA Holdway DA 2002 Toxicity testing of crude oil and related compounds using early life stages of the crimson-spotted rainbowfish (Melanotaenia fluviatilis ) Ecotox Environ Saf 52 180 189
Prasch AL Teraoka H Carney SA Dong W Hiraga T Stegeman JJ 2003 Aryl hydrocarbon receptor 2 mediates 2,3,7,8-tetrachlorodibenzo-p -dioxin developmental toxicity in zebrafish Toxicol Sci 76 138 150 12883077
Reddy CM Eglinton TI Hounshell A White HK Xu L Gaines RB 2002 The West Falmouth oil spill after thirty years: the persistence of petroleum hydrocarbons in marsh sediments Environ Sci Technol 36 4754 4760 12487296
Schmidt JV Bradfield CA 1996 Ah receptor signaling pathways Annu Rev Cell Dev Biol 12 55 89 8970722
Short JW Heintz RA 1997 Identification of Exxon Valdez oil in sediments and tissues from Prince William Sound and the northwestern Gulf of Alaska based on a PAH weathering model Environ Sci Technol 31 2375 2384
Short JW Lindeberg MR Harris PM Maselko JM Pella JJ Rice SD 2004 Estimate of oil persisting on the beaches of Prince William Sound 12 years after the Exxon Valdez oil spill Environ Sci Technol 38 19 25 14740712
Sloan CA Brown DW Pearce RW Boyer RH Bolton JL Burrows DG 2004. Extraction, Cleanup and Gas Chromatography/Mass Spectrometry Analysis of Sediments and Tissues for Organic Contaminants. NMFS-NWFSC-59. Seattle, WA:U.S. Department of Commerce, National Oceanic and Atmospheric Administration, National Marine Fisheries Service.
Teraoka H Dong W Tsujimoto Y Iwasa H Endoh D Ueno N 2003 Induction of cytochrome P450 1A is required for circulation failure and edema by 2,3,7,8-tetrachlorodibenzo-p -dioxin in zebrafish Biochem Biophys Res Commun 304 223 228 12711302
U.S. Commission on Ocean Policy 2004. An Ocean Blueprint for the 21st Century: Final Report of the U.S. Commission on Ocean Policy—Pre-Publication Copy. Washington, DC:U.S. Commission on Ocean Policy. Available: http://www.oceancommission.gov/documents/full_color_rpt/wel-come.html [accessed 26 October 2005].
Van Metre PC Mahler BJ 2003 The contribution of particles washed from rooftops to contaminant loading to urban streams Chemosphere 52 1727 1741 12871740
Van Metre PC Mahler BJ Furlong ET 2000 Urban sprawl leaves its PAH signature Environ Sci Technol 34 4064 4070
Whyte JJ Jung RE Schmitt CJ Tillitt DE 2000 Ethoxyresorufin-O -deethylase (EROD) activity in fish as a biomarker of chemical exposure Crit Rev Toxicol 30 347 570 10955715
|
16330359
|
PMC1315066
|
CC0
|
2021-01-04 23:41:30
|
no
|
Environ Health Perspect. 2005 Dec 10; 113(12):1755-1762
|
utf-8
|
Environ Health Perspect
| 2,005 |
10.1289/ehp.8230
|
oa_comm
|
==== Front
PLoS PathogPLoS PathogppatplpaplospathPLoS Pathogens1553-73661553-7374Public Library of Science San Francisco, USA 1636207510.1371/journal.ppat.001003705-PLPA-RA-0108R2plpa-01-04-04Research ArticleParasitologyPlasmodiumEukaryotesInvasion by P. falciparum Merozoites Suggests a Hierarchy of Molecular Interactions Molecular Hierarchy in Malaria InvasionBaum Jake Maier Alexander G Good Robert T Simpson Ken M Cowman Alan F *Division of Infection and Immunity, The Walter and Eliza Hall Institute of Medical Research, Parkville, Victoria, AustraliaYoung John EditorSalk Institute for Biological Studies, United States of America* To whom correspondence should be addressed. E-mail: [email protected] 2005 16 12 2005 1 4 e3727 7 2005 27 10 2005 Copyright: © 2005 Baum 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.Central to the pathology of malaria disease are the repeated cycles of parasite invasion and destruction of human erythrocytes. In Plasmodium falciparum, the most virulent species causing malaria, erythrocyte invasion involves several specific receptor–ligand interactions that direct the pathway used to invade the host cell, with parasites varying in their dependency on these different pathways. Gene disruption of a key invasion ligand in the 3D7 parasite strain, the P. falciparum reticulocyte binding-like homolog 2b (PfRh2b), resulted in the parasite invading via a novel pathway. Here, we show results that suggest the molecular basis for this novel pathway is not due to a molecular switch but is instead mediated by the redeployment of machinery already present in the parent parasite but masked by the dominant role of PfRh2b. This would suggest that interactions directing invasion are organized hierarchically, where silencing of dominant invasion ligands reveal underlying alternative pathways. This provides wild parasites with the ability to adapt to immune-mediated selection or polymorphism in erythrocyte receptors and has implications for the use of invasion-related molecules in candidate vaccines.
Synopsis
The repeated cycles of parasite invasion and destruction of human red cells is central to malaria disease. In Plasmodium falciparum, the most virulent species that causes malaria, invasion involves the interaction of several parasite ligands with receptors that line the red cell surface. Central to the success of the P. falciparum parasite is its ability not only to utilize a number of these receptors but also to vary the primary route used. Here we show that in some parasite strains when you remove their key invasion ligands, rather than activating an alternative molecular machinery to compensate, the parasite has at its disposal a secondary means of invading that was present in the parent parasite but whose role was masked by the dominant invasion route. This suggests that the interactions that direct invasion are organized hierarchically, where silencing of the dominant ligand reveals underlying alternative means to invade. Such a mechanism gives the parasite population the ability to avoid host immune-mediated selection or to adapt to variation in red cell surface receptors and still successfully infect the human host. This stresses the importance of directing a blood-stage malaria vaccine at multiple receptor–ligand interactions to prevent parasite adaptation to invade via alternative routes.
Citation:Baum J, Maier AG, Good RT, Simpson KM, Cowman AF (2005) Invasion by P. falciparum merozoites suggests a hierarchy of molecular interactions. PLoS Pathog 1(4): e37.
==== Body
Introduction
Unlike many other members of the phylum Apicomplexa, malaria parasites limit their infection of host cells to the restricted population of erythrocytes in the bloodstream. The invasion of erythrocytes and the subsequent cycles of growth, replication, and rupturing of infected cells are responsible for the majority of symptoms relating to malaria disease, with severe parasite infections giving rise to rapid hemolysis and metabolic acidosis [1]. This makes the blood stage of the parasite life cycle a primary target for novel interventions to prevent invasion and combat malaria disease.
Erythrocyte invasion is a rapid process governed by molecular interactions between the invading blood-stage parasite, the merozoite, and the host cell surface [2]. The merozoites bind to the erythrocyte surface, reorientate to their apical pole, and then, following formation of a tight junction between host and parasite apical tip (Figure 1A), invade, forming an isolated parasitophorous vacuole [2]. In Plasmodium falciparum, the most virulent of malaria species infecting humans, studies with erythrocytes modified by enzyme treatment [3–11], or from human donors lacking surface antigens [5,6,12–16], have identified a number of erythrocyte receptors used by merozoites for attachment (Figure 1B). These include glycophorins A [17], B [12], and C [14] as well as unknown receptors referred to as X [12], Y [7], Z [8], and E [10]. The dependency on these receptors for invasion is known to vary among parasites both in laboratory strains [12,15,18,19] and in isolates from the field [20–22]. Furthermore, in certain laboratory strains, such as W2mef (and its clone Dd2), the dependency on different receptors for invasion can change following either disruption of the host receptor [9] or targeted modification of the parasite ligand to which it binds [8,23].
Figure 1 A Number of Receptor–Ligand Interactions Mediate Merozoite Invasion of the Erythrocyte
(A) Estimates for the dimensions of the interaction between the apical merozoite tip and the erythrocyte surface suggest a finite number of receptors may, together with parasite ligands, direct invasion (dimensions based on [55]; receptor figures based on [56]).
(B) Known receptor–ligand interactions between the invading merozoite and erythrocyte surface receptors and their enzyme sensitivities to neuraminidase (Nm), trypsin, and chymotrypsin (Chymo). S indicates sensitivity of invasion to this enzyme (inhibits invasion/ligand-binding). R indicates resistance (does not inhibit invasion/ligand-binding). Unknown ligands or receptors are indicated by a “?,” with unknown receptors E, X, Y, and Z known only from their enzyme sensitivities. Dots in table indicate that no data are available. Data are taken from [7,8,10–12,14,17,27,33,45].
The specialized apical complex that defines all Apicomplexan parasites contains two main vesicular bodies, termed micronemes and rhoptries (Figure 1A). Two families of parasite proteins that localize to these organelles are thought to underlie the receptor dependency of different parasite strains. The first family shares homology with the Duffy binding protein from Plasmodium vivax [24] and includes erythrocyte-binding antigen 175 (EBA-175) [17], EBA-140 [11,14] (also known as BAEBL [24]), and EBA-181 [10] (also known as JESEBL [24]). The EBA proteins bind sialic acid (SA) residues of glycophorins A, C, and the unknown receptor E, respectively, on the erythrocyte surface and are therefore sensitive to treatment by the sialidase neuraminidase (Figure 1B) [4,10,14,25]. The second group of proteins is a high-molecular-weight, family-sharing homology with the reticulocyte-binding proteins of P. vivax [26]. These include the P. falciparum reticulocyte binding-like homolog 1 (PfRh1) [7,27], PfRh2a, and PfRh2b [8,28], and PfRh4 proteins [29] (also known as NBP proteins −1, −2a, −2b, and −4 [7,28,29]). PfRh1 binding to the erythrocyte surface via an unknown receptor Y is SA-dependent, whereas the inferred binding characteristic of PfRh2b (to its unknown receptor Z) and PfRh4 are SA-independent (Figure 1B). Binding characteristics of PfRh2a are not known. There is also evidence for variability in the expression of the functional EBA and PfRh proteins in different parasite lines, a factor that may underlie the use of alternative invasion pathways [8,27]. Both families also include proposed pseudogenes EBA-165 (PFD1155w) [30] and PfRh3 (PFL2520w) [31], respectively, which are transcribed but do not appear to form protein products.
The 3D7 strain of P. falciparum has been fully sequenced [32], and can invade erythrocytes in both a SA-dependent and independent manner [8]. Previous studies have successfully disrupted the expression of EBA-140 [14], EBA-175 [33], PfRh1, PfRh2a, and PfRh2b [8] in 3D7 by gene knockout strategies. In each case this did not result in any measurable change in efficiency of invasion into normal erythrocytes, demonstrating that there is considerable redundancy in invasion pathways and that the role of each protein is either expendable or can be readily compensated [8,14,33]. The 3D7ΔEBA-175 parasites invade chymotrypsin-treated erythrocytes at a greatly reduced rate when compared to the parental 3D7 strain. This suggests that EBA-175 is functional in the parent, and as a result of the gene disruption, the mutant parasite has a reduced repertoire of potential invasion receptors [14,33], although on which alternative receptor(s) it is relying is not known. Disruption of PfRh1, PfRh2a, or EBA-140 in 3D7 does not result in any noticeable alteration in the sensitivity of invasion, even in enzyme-treated erythrocytes, suggesting these proteins play a less important role in the parent [8,14,27]. Unlike the other knockout lines and 3D7 parent, the 3D7ΔRh2b parasite uses receptors that are chymotrypsin-resistant [8]. This suggests that an invasion pathway, not normally used by the parent strain, has been utilized in the mutant parasite. This mirrors the phenotype observed when EBA-175 function is removed in W2mef (a parasite usually sensitive to neuraminidase), resulting in a dramatic shift to SA-independent invasion [23].
Here we have analyzed the novel invasion pathway used by the 3D7ΔRh2b parasite. Our results suggest that changes in the receptor–ligand dependency of parasites are directed by a hierarchy of molecular interactions and not, at least in the case of 3D7, by the activation of new ligands. This mechanism would provide parasites with the ability to rapidly adapt to host immune selection directed at primary invasion ligands or to polymorphism in erythrocyte receptors.
Results
PfRh3 Is Activated in 3D7ΔRh2b Parasites
A number of genes that encode invasion-related proteins have been disrupted in the P. falciparum parasite strain 3D7 [8,14,33]. However, so far only the 3D7ΔRh2b parasite, lacking the PfRh2b gene, alters its invasion receptor dependency, utilizing erythrocyte receptors that are more resistant to chymotrypsin than the parent [8,11]. To investigate the molecular basis for this novel invasion pathway, transcription of key invasion-related genes in late-stage (40–48 h postinvasion) 3D7 and 3D7ΔRh2b parasites was measured. The unrelated isolate D10 was included as it naturally lacks the PfRh2b gene, as well as the EBA-140 gene, and invades in a chymotrypsin-resistant manner similar to 3D7ΔRh2b [8]. The relative amount of RNA transcribed (relative to either the actin and histone 2b genes, or the late-stage expressed msp2 gene) for the EBA (EBA-140, EBA-165, EBA-175, and EBA-181) and PfRh (PfRh1, PfRh2a, PfRh2b, PfRh3, and PfRh4) families of genes is shown in Figure 2. As predicted, transcription of the PfRh2b in both 3D7ΔRh2b and D10 was greatly reduced (or absent) when compared to wild-type 3D7 (Figure 2A and 2B). The absence of EBA-140 was also confirmed for D10 (Figure 2B). Among the PfRh and EBA genes, the only gene that showed significant increases in relative levels of RNA transcript, in comparison to 3D7, was PfRh3, a putative pseudogene. This gene appeared to be transcribed approximately 10-fold more in both 3D7ΔRh2b and D10 relative to 3D7 (Figure 2A and 2B).
Figure 2 Comparison of Transcription of Key Invasion Genes in 3D7, 3D7ΔRh2b, and D10
(A and B). Expression levels of EBA and PfRh genes controlled by the relative expression of either actin and histone 2b (two constitutively expressed genes) or msp2 (a gene expressed later on in the life cycle). Error bars represent the 95% confidence interval (CI) of values from two to three independent amplifications.
(C and D) Affymetrix microarray comparison of log gene expression intensity (as measured by MOID) between cRNA isolated from late-schizont 3D7 and (C) 3D7ΔRh2b or (D) D10 parasites. Data shown are restricted to 548 genes whose expression profile matches clusters 4, 13, 14, 15 [34], including genes that encode proteins known to be involved in invasion. Color represents significance of the change in the level of gene expression as measured by the t-statistic (<2 black, >2 red). The control genes PfRh2b (knocked out) and EBA-140 (deleted in D10) and upregulation of PfRh3 are shown.
To confirm the apparent activation of PfRh3 in 3D7ΔRh2b and D10 and investigate other genes that may have changed with the loss of the PfRh2b gene, we probed Affymetrix gene-chips representing 95% of the predicted genes in the 3D7 P. falciparum genome [34]. Global analysis of expression carried out using the MOID [35], RMA [36], and GCRMA algorithms [37] showed a number of genes that have changed their transcription levels between parasite lines (Dataset S1). Measurements from replicate chips were used to calculate a moderated t-statistic. The t-statistic is used to rank genes in order of confidence of differential expression [38]. Analysis was restricted to those genes that are expressed at similar times to genes whose products are known to be involved in invasion (clusters 4, 13, 14, and 15 [34]) but excluding the known virulence genes (vars, rifins, and stevors [39]), leaving a total of 548 genes (Datasets S1–S4). As expected, PfRh2b is downregulated in both 3D7ΔRh2b and D10, as is EBA-140 in D10. To investigate which gene might have been upregulated to compensate for loss of PfRh2b, a low cutoff point of t greater than or equal to 2 was chosen to define those genes that have increased their transcription between the 3D7ΔRh2b or D10 parasite and 3D7 wild type. By this cutoff, and comparison between the algorithms, seven genes showed significant increase in transcriptional activity in the 3D7ΔRh2b (Figure 2C; Table 1). PfRh3 is activated in both 3D7ΔRh2b and D10 parasites, by approximately 60-fold and approximately 15-fold, respectively (values from the MOID calculation, Figure 2D; Table 1), supporting the result observed using RT-PCR. Of the other genes that have increased in transcriptional activity (PlasmoDB ID: PFE1465w, PFB0680w, PFF0670w, PF08_0036, PF13_0173, and PF14_0567), all are uncharacterized genes with no obvious adhesive domain and lack a signal peptide and transmembrane domain. PF08_0036 encodes a putative protein transporter, which is unlikely to function directly in invasion. Additionally, and in contrast to PfRh3, even by the most generous estimates these genes have only increased by 4-fold in 3D7ΔRh2b and do not show up regulation in D10. Therefore, although their protein products may play an indirect role in the new invasion pathway of 3D7ΔRh2b, it is unlikely that they directly compensate for the loss of function of PfRh2b.
Table 1 Genes That Have Altered Their Transcription Levels >2-Fold in 3D7ΔRh2b Relative to 3D7 Wild Type with a t-Statistic ≥2
Since PfRh3 transcription was very low in 3D7, its elevated level in 3D7ΔRh2b and D10, both of which invade more efficiently into chymotrypsin-treated erythrocytes, may indicate a role for Rh3 in compensating for the loss of function of Rh2b, and as such define the molecular basis for the novel chymotrypsin-insensitive pathway. However, PfRh3 may also be activated in 3D7ΔEBA-175 and 3D7ΔEBA-40, although these parasites show no change in the phenotype or efficiency of invasion. (Table 1, Datasets S1–S5).
PfRh3 Is Not Essential for the Novel Invasion Pathway Used by 3D7ΔRh2b
The presence of a conserved frame-shift mutation(s) in the PfRh3 gene from multiple parasite lines [40], and the inability to detect a protein product for the gene using rabbit serum raised against recombinant portions of the gene [31] suggest that PfRh3 is a transcribed pseudogene. To investigate whether the increase in levels of PfRh3 transcript in 3D7ΔRh2b and D10 was associated with the production of a functional PfRh3 protein, genomic DNA and cDNA from the three strains were used as template to amplify a 5′ region of PfRh3. Sequences of this region, from genomic or cDNA, all retained the 5′ frame-shift mutation (data not shown), suggesting no protein product could be produced. The possibility that posttranscriptional modification might remove the frame-shift mutation from PfRh3 mRNA, and therefore produce a functional protein product, was excluded by failure of rabbit anti-Rh3 antisera to detect any protein product from either 3D7ΔRh2b or D10 (data not shown).
To finally rule out the role of PfRh3 in the 3D7ΔRh2b parasite, we constructed a plasmid (pCC4-Rh3) (Figure 3A) that would integrate into the PfRh3 locus in the 3D7ΔRh2b parasite by double recombination crossover [41]. Because the original 3D7ΔRh2b knockout had been achieved using a vector containing the human dihydrofolate reductase (hdhfr) gene [8], the blasticidin-S deaminase gene (BSD, conferring resistance to blasticidin) was used in pCC4-Rh3 to select for parasites carrying the plasmid. As a control, knockout of PfRh3 was also performed in parental 3D7 since, in the absence of other modifications, this gene has been easily disrupted previously [31,41]. To confirm integration of pCC4-Rh3 into both 3D7 and 3D7ΔRh2b, genomic DNA from parental and transfected parasites was analyzed by Southern hybridization. This revealed DNA fragments consistent with double-crossover recombination into the PfRh3 locus (Figure 3B). This represents the first time two genes have been sequentially disrupted in P. falciparum. To ensure that PfRh3 transcript had been lost from 3D7ΔRh3 and 3D7ΔRh2bΔRh3 we used RT-PCR with late-stage RNA from these parasite lines as well as 3D7 and 3D7ΔRh2b. Primers targeted to a specific region of PfRh2a showed consistent expression in all parasites lines (Figure 3C), while those specific for the region of PfRh3 deleted by the gene-disruption strategy showed no expression in the PfRh3 knockout lines, confirming that the gene has been disrupted (Figure 3C).
Figure 3 Targeted Disruption of PfRh3 Shows It Is Not Essential for the 3D7ΔRh2b Chymotrypsin-Resistant Pathway
(A) Disruption of the PfRh3 gene in 3D7. The pCC4-Rh3 plasmid contains the blasticidin-S deaminase selectable marker, a negative selectable marker (A. G. Maier and A. F. Cowman, unpublished data), and 5′ and 3′ Rh3 regions. PfRh3 is shown with homologous target sequences (shaded regions). The double-crossover integration events are shown for 3D7, resulting in the deletion of a 5′ region of the gene. Restriction enzymes are C (ClaI) and X (XbaI), with fragment sizes shown for a C/X digestion. RT-PCR amplification target is shown as a black bar.
(B) Southern blot of genomic DNA from parasites shown digested with ClaI and XbaI and probed with the 5′ Rh3 flank from the pCC4-Rh3 vector.
(C) RT-PCR of PfRh2a and PfRh3 from 3D7 and three knockout lines.
(D) Invasion into enzyme-treated erythrocytes expressed as a percentage of that into untreated erythrocytes for 3D7 and three knockout lines. Values above each column indicate the mean % invasion, with error bars representing the 95% CI from three independent assays.
Finally, to confirm that PfRh3 plays no role in the novel invasion pathway used by the 3D7ΔRh2b parasite, 3D7, 3D7ΔRh2b, 3D7ΔRh3, and 3D7ΔRh2bΔRh3 were grown in erythrocytes treated with neuraminidase and chymotrypsin. Invasion rates into untreated cells were similar for all parasite lines (data not shown). Invasion into neuraminidase-treated cells showed similar rates for the 3D7 and 3D7ΔRh3 parasites of approximately 75% to that into untreated cells and approximately 60% in the 3D7ΔRh2b and 3D7ΔRh2bΔRh3 parasites (Figure 3D). Chymotrypsin treatment similarly grouped the parasites into those in which PfRh2b had been disrupted with 3D7 and 3D7ΔRh3 parasites, invading at approximately 35% to that into untreated cells and between 70%–100% in the 3D7ΔRh2b and 3D7ΔRh2bΔRh3 parasites. These figures are similar to those seen in the original analyses of 3D7ΔRh2b [41] and demonstrate that PfRh3 plays no role in the novel chymotrypsin-resistant pathway.
Therefore, despite the consistent finding of PfRh3 up-regulation in both the 3D7ΔRh2b and D10 parasites, these data clearly demonstrate that it plays no role in determining the novel invasion pathway used by 3D7ΔRh2b. Indeed, PfRh3 activation in both 3D7ΔEBA-140 and 3D7ΔEBA-175 (which do not appear to use this novel pathway) further argues against a role for PfRh3 in compensating for the loss of PfRh2b function (Dataset S5). The lack of obvious candidate genes that have properties consistent with their role in invasion suggests that proteins that are already present in 3D7 mediate the previously unused pathway.
Relative Levels of Key Invasion Proteins Have Not Changed between Wild-Type 3D7 and the 3D7ΔRh2b Parasite
A number of parasite proteins have been implicated in mediating the alternative invasion pathways used by P. falciparum to invade erythrocytes [4,7,8,10,11,14,17]. Although transcript levels of these invasion-related genes do not appear to have changed with loss of the PfRh2b gene in 3D7, posttranscriptional modification resulting in changes in the absolute amounts of protein product may underlie the novel invasion pathway used in the 3D7ΔRh2b parasite line. Semiquantitative Western blots, where loading was controlled for expression of the SERA5 protein, were undertaken to determine if changes in absolute levels of invasion-related proteins had occurred (Figure 4A). Other than the absence of detectable protein in knockout lines (as expected), no discernable differences are present between the different 3D7 parasite lines for the PfRh1, PfRh2a, PfRh2b, EBA-140, EBA-175, or EBA-181 proteins (Figure 4A), suggesting that the absolute levels of the invasion proteins have not changed. Proteomic analysis of late schizonts from both 3D7 and 3D7ΔRh2b also failed to show any dramatic changes in the presence of any particular proteins (unpublished data). However, as with previous studies [42], invasion genes were poorly represented in the analysis, with only a few peptide fragments from one or two proteins showing up. This suggests that the absolute amount of most invasion proteins is not sufficient for detection, using current proteomic methods. This demonstrates that the absolute levels of key invasion proteins in 3D7, following the disruption of PfRh2b, appear to remain constant despite a shift in the invasion receptor dependency.
Figure 4 Changes in Protein Levels Do Not Underlie the 3D7ΔRh2b Chymotrypsin-Resistant Pathway
(A) Western blot of parasite culture supernatant material probed with rabbit polyclonal antibodies against the functional EBA and PfRh proteins. SERA5 is used as a loading control. Antibodies marked with an asterisk (*) indicate the same gel stripped and reprobed with a different antibody.
(B) Invasion into chymotrypsin-treated erythrocytes for five parasite lines represented as the percentage of invasion to that into untreated erythrocytes.
(C) Invasion into either untreated or chymotrypsin-treated erythrocytes in the presence of protein-G purified polyclonal rabbit antiserum raised against recombinant EBA-140 and EBA-175 (or both together). Values are represented as the percentage of invasion in the presence of NRS. Error bars represent the 95% CI from three independent assays.
Invasion Inhibitory Antibodies against EBA-140 and EBA-175 Inhibit 3D7ΔRh2b and the Parental 3D7 Strain
The inhibition caused by antibodies raised against recombinant EBA-140 and EBA-175 relate directly to the function of these proteins in invasion, as shown by the failure of specific antibodies to inhibit parasites where the respective genes have been disrupted [14,33]. The binding of EBA-140 and EBA-175 to the erythrocyte surface is insensitive to chymotrypsin treatment [11] since their receptors, glycophorins A and C, are resistant to digestion by this enzyme (although some glycophorin A is sensitive to chymotrypsin, as much as 60% remains following enzyme treatment because of a variable glycosylation site near the enzyme cleavage point [43]). Given that the pathway being used by 3D7ΔRh2b is relatively chymotrypsin-resistant (Figure 4B), alterations in the importance or reliance of 3D7ΔRh2b on EBA-140 or EBA-175 might underlie this unidentified pathway. To address this, invasion assays in the presence of αEBA-140 and αEBA-175 were undertaken with the 3D7 mutant parasite lines, 3D7 and D10.
As found previously [14], invasion of 3D7 into untreated erythrocytes in the presence of αEBA-140 was approximately 80% compared to that in the presence of normal (preimmunization) rabbit serum (NRS) (Table 2; Figure 4C), demonstrating a role for the EBA-140 protein despite there being no altered or observable phenotype following the disruption of its gene in 3D7 [14]. Invasion of 3D7ΔRh2b in the presence of αEBA-140 was approximately 70%, the range of which is not significantly different from that in 3D7 (Table 2; Figure 4C). Both 3D7ΔEBA-140 and D10 were unaffected by αEBA-140, confirming the specificity of the antibody [14]. Interestingly, in the presence of αEBA-140, 3D7ΔEBA-175 invasion was significantly less inhibited (approximately 90% invasion) than wild-type 3D7 or 3D7ΔRh2b (Table 2; Figure 4C). This suggests that in the absence of EBA-175, invasion by the 3D7 parasite is entirely SA-independent and no longer relies on any residual function of EBA-140.
Table 2 Percentage Invasion of Parasites into Untreated and Chymotrypsin-Treated Erythrocytes in the Presence of IgG-Purified Rabbit Antiserum Raised against Recombinant EBA-140 and EBA-175
The 3D7, 3D7ΔRh2b, and 3D7ΔEBA-140 parasites invaded erythrocytes at approximately 40% in the presence of αEBA-175. This adds strong support to the observation that 3D7 primarily uses the EBA-175/GYPA receptor–ligand interaction to mediate invasion despite being able to invade via alternative (SA-independent) routes when this pathway is blocked (for example following neuraminidase treatment) [33]. Inhibition of 3D7ΔEBA-140 by EBA-175 antibodies demonstrates a more dominant role for EBA-175 over EBA-140. Invasion of 3D7ΔEBA-175 was not inhibited by αEBA-175, confirming the specificity of the antibody [23].
Invasion of 3D7 into erythrocytes pretreated with chymotrypsin for all parasite lines confirmed results found previously (Figure 4B). This enzyme digests many surface receptors but does not entirely remove glycophorins C or A, the receptors of EBA-140 and EBA-175, respectively [11]. As such, by removing a number of other putative receptors, but leaving glycophorins C and A, chymotrypsin treatment will accentuate the inhibitory effects of αEBA-140 and αEBA-175. The relative inhibition of 3D7 with αEBA-140 (compared to that in the presence of NRS) was reduced to approximately 40% invasion (Table 2; Figure 4C), confirming that reported previously [14]. However, unlike 3D7, chymotrypsin treatment did not significantly reduce the relative inhibition of 3D7ΔRh2b with αEBA-140, suggesting that EBA-140 plays a less important role in the PfRh2b knockout parasites. Following chymotrypsin treatment, 3D7ΔEBA-140 and D10 showed no inhibition from αEBA-140. This was also true for 3D7ΔEBA-175, which further supports its shift to reliance on SA-independent invasion. The lack of reduction in relative invasion of 3D7ΔRh2b with αEBA-175 and chymotrypsin treatment, like that with αEBA-140, demonstrates that the PfRh2b knockout appears to be less reliant on EBA-175 than the parent strain; 3D7, the EBA-140 knockout line, and D10, in the presence of αEBA-175, all showed only a moderate to no increase in the relative inhibition compared to NRS. Finally, it is worth noting that inhibition in the presence of both antibodies together (both with and without chymotrypsin treatment) is not dramatically different from that seen in the presence of αEBA-175 alone (Table 2; Figure 4C). A caveat to the observations following chymotrypsin treatment is that our ability to detect subtle differences in invasion is limited with chymotrypsin treatment since, in both parasites, it reduces the efficiency of invasion and, as such, means that the absolute numbers of parasites are less in each assay.
Decreased inhibition of invasion in the presence of αEBA-140 and αEBA-175 demonstrates that without PfRh2b, the 3D7 parasite utilizes receptors other than glycophorins C and A. Furthermore, it supports the general assertion that in wild-type 3D7, EBA-175 plays a more critical role than EBA-140 in mediating invasion.
Long-Term Culturing of 3D7 in Chymotrypsin and Low-Trypsin-Treated Cells Does Not Lead to a Switch in Invasion Phenotype
Following selection on enzyme-treated cells, certain parasite lines (notably W2mef/Dd2) are able to switch their route of invasion to a previously unused pathway [9]. Having demonstrated that neither a change in transcription nor a change in protein levels, or the function (as determined by inhibition) of alternative invasion ligands, underlies the novel invasion pathway used by the 3D7ΔRh2b parasite line, long-term selection of 3D7 was undertaken to attempt to select for the chymotrypsin-resistant pathway (thereby mimicking knockout of the PfRh2b gene), as has been observed with selection on neuraminidase-treated cells and the knockout of EBA-175 in W2mef [23]. After selection on chymotrypsin- and trypsin-treated cells (the most marked phenotype of the new pathway in 3D7ΔRh2b) for more than 30 d, no significant change in enzyme sensitivity occurred in 3D7 to make it more like the 3D7ΔRh2b parasite (Figure 5A and 5B).
Figure 5 Long-Term Selection of (A) 3D7 and (B) 3D7ΔRh2b in Erythrocytes Pretreated with Chymotrypsin (1.5 mg/ml) and Trypsin (0.1 mg/ml) Shows No Significant Change in Receptor Dependency
Parasitemia was measured by microscopy; then cultures were diluted down to 0.5% parasitemia at 48-h intervals. Cultures were maintained for greater than 30 d. Secondary y-axis represents invasion into enzyme-treated erythrocytes as a percentage of invasion into untreated erythrocytes, with a regression line plotted across the values.
Therefore, even though the underlying mechanism for invasion into chymotrypsin- and trypsin-treated erythrocytes appears to be already present in the 3D7 parasite, it cannot be upregulated when PfRh2b is present. This suggests, that PfRh2b is the a priori ligand used by 3D7 and that the physical presence of functional PfRh2b relegates any potential role of a secondary invasion ligand(s) (and its resulting chymotrypsin-resistant pathway), reducing the invasion efficiency of 3D7 into certain enzyme-treated cells.
Discussion
The availability of the completed P. falciparum genome has led to the identification of two families of proteins that are thought to underlie the variable receptor–ligand interactions used by malaria parasites to invade human erythrocytes. Gene disruption of each of these EBA and PfRh proteins have been generated in various parasite lines, and in 3D7, the genome reference malaria parasite, null mutants for EBA-140, EBA-165, EBA-175, PfRh1, PfRh2a, PfRh2b, and PfRh3 are available [8,14,30,33,41]. Of these lines only the 3D7ΔRh2b parasite apparently uses an invasion pathway that is qualitatively different to that of the parent 3D7 strain [8]. Here we have used specific gene disruptions and a comparative analysis of gene and protein expression and function to demonstrate that the receptor-mediated invasion pathway of 3D7ΔRh2b likely uses ligands that are already present in wild-type 3D7, but our ability to measure their function was masked by the presence of PfRh2b.
Compensation for the Function of PfRh2b in 3D7
Comparison of gene expression between 3D7 and 3D7ΔRh2b showed that PfRh3, a related PfRh pseudogene, does increase its mRNA expression levels significantly following PfRh2b disruption. However, no protein product is produced (it being a pseudogene [31]). Furthermore, the possibility that PfRh3 might function at the RNA level is ruled out by gene disruption on a 3D7ΔRh2b genetic background. This is the first time two genes have been disrupted sequentially in the malaria parasite. Previous work analyzing PfRh transcription in the parasite strains FCB1, T996, and 3D7 showed that PfRh3 levels are naturally much lower in 3D7 than the other two strains [44]. Therefore the explanation for PfRh3's anomalous increase in transcription in 3D7ΔRh2b might lie with its low levels of transcription in the 3D7 parent strain, which changes in the knockout parasite. This is supported by a similar activation of PfRh3, albeit to a smaller extent, in the 3D7ΔEBA-140 and 3D7ΔEBA-175 parasites. The rigors of transfection and drug selection may induce changes in gene regulation across the parasite genome that nonspecifically activates the expression of PfRh3. This anomalous upregulation would suggest a degree of caution in the interpretation of comparative microarray data and stress the importance of following up the functionality of up- or downregulated gene products following any selection regimen.
In addition to PfRh3, a small number of other genes do change their transcriptional activity (although not to the same extent as PfRh3). However, lack of obvious domains that might have an adhesive or binding function, and the absence of a signal peptide (necessary for export to secretory organelles or the merozoite surface) or transmembrane domain would suggest that none of these genes are likely to play a direct role in invasion. This apparent absence of change also appears to be true at the protein level for the known invasion proteins. Taken together, this strongly suggests that in 3D7ΔRh2b, the chymotrypsin-resistant invasion pathway revealed in the absence of PfRh2b is the result of a redeployment of the existing ligand repertoire of the parent parasite and not any transcriptional or posttranscriptional change in other invasion ligands.
The Role of Other EBA or Rh Proteins in the Rh2b-Independent Pathway
A plausible explanation for the 3D7ΔRh2b parasites' increased ability to invade chymotrypsin-pretreated erythrocytes is that the parasite has become more reliant on EBA-175 or EBA-140, both of which bind red cells in a chymotrypsin-insensitive manner (mediating invasion via glycophorins A or C) [11,14,23]). However, the reverse seems to be true since 3D7ΔRh2b parasites are less inhibited than wild-type 3D7 in the presence of αEBA-175 or EBA-140 antibodies (see Figure 4C). Although inhibitory antibodies against EBA-181 are not available [10], it is clear that this ligand is not involved since its unknown receptor, receptor E, is chymotrypsin-sensitive [10]. Other possible ligands that may compensate for loss of PfRh2b are the other three functional PfRh proteins. PfRh1 binds to the erythrocyte via an unknown receptor, Y, that is chymotrypsin-resistant but neuraminidase-sensitive [7,27]. However, it is present at relatively low levels in 3D7 and cannot define the entire phenotype, given that 3D7ΔRh2b invasion is only moderately neuraminidase-sensitive [8]. This leaves either PfRh2a, PfRh4 (the receptors of which are unknown or not fully characterized [45]), or another as yet uncharacterized invasion protein. Therefore, at present it is not possible to identify the invasion ligand(s) that compensate for the loss of PfRh2b.
The Absence of the Rh2b-Independent Pathway in the Wild-Type Parent Suggests a Molecular Hierarchy Mediates Invasion
In the absence of any differences in mRNA or protein level, the inability of the wild-type parent to invade using the same pathway as the 3D7ΔRh2b parasite suggests that the presence of PfRh2b obscures other potential receptor-mediated pathways. This is supported by our inability to adapt 3D7 parasites to chymotrypsin-treated erythrocytes following long-term selection. Since 3D7 does not become more resistant to enzyme treatment, this shows that the shift toward using an alternative receptor-mediated pathway is only possible following the removal of PfRh2b but crucially not its target receptor. The lack of changes in gene expression or protein levels and inability to alter invasion in the presence of PfRh2b suggests that a molecular hierarchy operates to direct which receptor-mediated pathway is used by the 3D7 parasite to invade.
When the parasite begins to invade the erythrocyte (see Figure 1), the tight junction is likely to contain micronemal proteins such as EBA-175, EBA-181, and EBA-140 and rhoptry proteins such as the PfRhs [8,27,29], which together will direct the receptor-mediated pathway used to invade. In 3D7, EBA-175 and PfRh2b are likely to dominate these interactions, with EBA-140, EBA-181, and PfRh1 playing only minor roles as evidenced by the lack of phenotype following their functional disruption [8,14,27] (J. Thompson and A. F. Cowman, unpublished data).
For invasion by wild-type 3D7 parasites, removal of the receptors for EBA-140, EBA-175, EBA-181, and PfRh1 by neuraminidase treatment [4,7,10,11,14,27,33], results in utilization of the PfRh2b pathway. When erythrocytes are treated with chymotrypsin, the receptors for PfRh2b [8] and EBA-181 [10] are removed, but invasion proceeds via the interaction of the remaining EBA proteins and any residual PfRh1 function. It is of note that although EBA-175 and EBA-140 both function in invasion (as demonstrated by the antibody inhibition data; see Figure 4C), it is clear that EBA-175 is dominant to the role of EBA-140 [14,33], a feature that is supported by the evidence for diversifying selection operating on EBA-175 but not (detectably) on EBA-140 [46]. It is possible that the EBA proteins may compensate for and function simultaneously with Rh (as previously suggested) [47], although they are structurally distinct protein families.
For 3D7ΔEBA-175 parasites, chymotrypsin treatment dramatically reduces the efficiency of invasion in 3D7ΔEBA-175 (<10% [33]) and can be explained by the absence of the PfRh2b pathway, which is chymotrypsin-sensitive. Critically, like the parental 3D7, no invasion ligand is present that can rescue invasion following chymotrypsin treatment, suggesting PfRh2b still dominates SA-independent invasion.
Finally, for 3D7ΔRh2b parasites, pretreatment with neuraminidase removes the receptors for EBA (and PfRh1), but the chymotrypsin-resistant pathway of 3D7ΔRh2b can compensate for their absence, it being SA-independent.
The presence of a secondary receptor-mediated pathway for invasion being redeployed from a subordinate position also appears to explain the uncovering of an unrelated invasion pathway in the T994 parasite strain following disruption of the PfRh1 ligand [27]. Here, in the absence of any noticeable changes in other invasion ligands, the knockout parasite becomes more neuraminidase- and trypsin-resistant than in wild-type T994 [27]. Following the proposed model, PfRh1 function dominates invasion in T994, but with its removal (by Rh1 disruption) a secondary ligand compensates for its lost function in the tight junction, again via a new receptor-mediated pathway.
While these data appear to explain 3D7 and T994 invasion, they clearly do not fit all the invasion phenotypes [9] associated with these ligand families. In particular, recent work with W2mef, either following selection on neuraminidase-treated cells or after knockout of EBA-175, has shown activation of another ligand PfRh4; this activation is critical for switching to the new SA-independent invasion phenotype [45].
Functional Consequences of Utilization of the PfRh and EBA Families
Variation and expression of PfRh and EBA proteins in both laboratory and field strains is likely to underlie invasion phenotypes [8,20–22,27]. Evidence from the mouse malaria parasite, Plasmodium yoelii, for the differential transcription of members of the Py235 gene family (a family closely related to the PfRh family) across life cycle stages [48], suggests that variable transcription may be a general feature of Rh-related proteins in Plasmodia. The placement of Rh and EBA genes in subtelomeric regions of the genome may allow them to be selectively switched on or off [49], as has been recently demonstrated for Rh4 in W2mef [45]. It is therefore possible that the invasion pathways used (as determined by expression of invasion ligand) by individual clones of P. falciparum may be selected to change through time. This might be evidenced by changes in the invasion phenotype of infecting parasites across the lifetime of an exposed human host, akin to the situation described for virulence genes [50], arbitrated by immune-mediated selection where the most commonly used invasion ligand is constantly selected against. Thus at the population level, successful strategies for preventing malaria disease that are directed against invasion will need to include both dominant and secondary ligands to reduce the likelihood of selection for alternative pathways.
Materials and Methods
Parasite cultures and transfection.
P. falciparum asexual parasites were maintained in human erythrocytes (blood group O+) at a hematocrit of 4% with 10% Albumax II (GIBCO, San Diego, California, United States) [51]. The 3D7 was originally obtained from David Walliker at Edinburgh University. Parasites 3D7ΔRh2b, 3D7ΔEBA-175, and 3D7ΔEBA-140 were generated in previous studies [8,14,33]. D10 is cloned from Papua New Guinea isolate FC27. Cultures were synchronized as described [52].
For disruption of the PfRh3 gene in 3D7 and 3D7ΔRh2b, parasites were transfected with 80–100 μg of plasmid (QIAGEN) as described [41]. Positive selection for transfectants was achieved using 2.5 μg/ml of blasticidin-S [53] with negative selection against the plasmid backbone using 5 fluoro-cytosine (A. G. Maier and A. F. Cowman, unpublished data).
The PCC4-Rh3 vector for disruption of the PfRh3 gene was constructed using pCC4, a derivative of the pHTK plasmid [41], modified to include the BSD gene [53]) under control of histidine-rich protein (hrp) 2 promoter and the cytosine deaminase gene (A. G. Maier and A. F. Cowman, unpublished data). The 5′ and 3′ flanks (both approximately 1 kb) for homologous recombination into the PfRh3 gene were amplified from 3D7 genomic DNA with the primer pairs 5′-GATCccgcggGGAAGGAGTAAAGTTTCGAAGG-3′/5′-GATCtctagaCTTATCTCCCAATATTCTC-3′ (inserts a 5′ SacII site and 3′ XbaI site) and 5′-GATCgaattcGACGGATTAGTTGAAAATAAATCC-3′/5′-GATCccatggCCATCAACTAAGGTTTCATC-3′ (inserts a 5′ EcoRI site and 3′ NcoI site), respectively. The 5′ and 3′ flanks were inserted into the vector to flank the BSD cassette. Plasmid integration was confirmed by Southern blot, using a 5′ Rh3 flank as a probe, following standard protocols.
Long-term growth assays were undertaken for cultures of 3D7 and 3D7ΔRh2b maintained in erythrocytes treated with 1.5 mg/ml chymotrypsin (Worthington Biochemical, Lakewood, New Jersey, United States) and 0.1 mg/ml trypsin (Sigma, St. Louis, Missouri, United States). Percentage parasitemia was calculated per 1,000 red cells by microscopy every 48 h at early to midtrophozoite stage. Parasitemia was adjusted to 0.5% for the next round of invasion.
RT-PCR and microarray analysis.
Total RNA was isolated from synchronized parasites 40–48 h postinvasion, using TRIzol (Invitrogen, Carlsbad, California, United States). RNA was further purified using DNaseI digestion by passage over an RNAeasy column (Qiagen, Valencia, California, United States).
Total RNA (5 μg) was reverse transcribed either with or without SuperScript II reverse transcriptase, using random hexamers (Invitrogen). A LightCycler (Roche, Basel, Switzerland) was used to quantify cDNA using the QuantiTect SYBR Green PCR kit (Qiagen) with PfRh or EBA gene-specific primers. Serial dilutions of 3D7 genomic DNA were used as standard controls. Relative expression ratios of EBA and PfRh genes compared to three reference genes, actin and histone2b (constitutively expressed [34]) and msp2 (transcribed late in the erythrocytic cycle [34]) were calculated for each strain. Each gene expression level was calculated as the average of three independent amplifications for 3D7 and 3D7ΔRh2b, although only two rounds were undertaken for D10. Samples with no reverse transcriptase were used to control for possible genomic DNA contamination of cDNA preparations.
RT-PCR for verification of the PfRh3 gene deletion was as described above, except reactions were separated on 1% TBE agarose gels to determine presence or absence of cDNA product; RT-negative reactions were included to ensure amplification was from cDNA and not genomic DNA. Primers used were Rh2aRTfwd 5′-GATGAGGTCATAAAAGATAATGAG-3′, Rh2aRTrev 5′-GAACATCATCATTCGGTTCAAAAGC-3′, Rh3RTfwd 5′-CAACGAATCAAGCACGTTTACC-3′, and Rh3RTfwd 5′-CTTATCATTTCTAAGGTAGAACC-3′.
cRNA was produced from cDNA using a T7-in vitro transcription kit (MEGAscript Kit, Ambion, Austin, Texas, United States) and used to hybridize to an Affymetrix (Santa Clara, California, United States) oligonucleotide DNA microarray as previously described [34]. Signal intensity and background noise correction were measured using the MOID [35], RMA [36], or GCRMA [37] algorithms. Differential expression was assessed for each parasite strain compared to 3D7 wild type by calculating a moderated t-statistic for each gene using the Limma package [54]. Features of genes that changed their transcriptional activity were investigated through publicly available databases held at http://www.PlasmoDB.org and http://www.geneDB.org.
SDS/PAGE and immunoblot analysis.
Parasite culture supernatants were obtained from cultures following schizont rupture. Proteins were separated by SDS-polyacrylamide gels and transferred to membranes (Schleicher and Schuell Bioscience, Dassel, Germany) and probed with affinity-purified rabbit polyclonal antibodies raised against PfRh1 [8], PfRh2a and PfRh2b [7], EBA-140 [11], EBA-175 [23], and EBA-181 [10]. Supernatant loading was controlled using antibodies raised against SERA5 (a generous gift from S. Miller and B. Crabb). Blots were processed by enhanced chemiluminescence (ECL; Amersham Biosciences, Little Chalfont, United Kingdom).
Invasion inhibition assays.
Erythrocyte invasion was assayed using normal and chymotrypsin-treated cells in the presence or absence of protein G purified rabbit anti-EBA-140 (raised against the F2 domain) [14] and rabbit anti-EBA-175 (raised against the region between the F2 and the 3′ cysteine-rich domain) [23]. Ring-stage cultures were split, half of which was treated with 1.5mg/ml chymotrypsin (Worthington Biochemical), washed, and resuspended in complete medium. Invasion assays were prepared in triplicate, using 100-μl aliquots of culture at a final hematocrit of 4% and parasitemia of 0.5% in 96-well flat-bottom microtiter plates (Becton Dickinson, Franklin Lakes, New Jersey, United States). Rabbit anti-EBA-140, anti-EBA-175, or NRS was added at a final concentration of 1 mg/ml in PBS. A PBS nonserum control was included. Control wells were smeared 72 h postassay preparation to determine life cycle stage (mid- to late-trophozoites postreinvasion); 10 μl of each resuspended assay well was added to 190 μl of filtered 10 μg/ml ethidium bromide in PBS. This was mixed and incubated for 5 min at room temperature. Parasitemia was measured using a FACScan (Becton Dickinson). Assays were performed on at least three separate occasions to calculate the mean and standard error of invasion inhibition relative to the NRS control.
Supporting Information
Dataset S1 Comparative Gene Expression
Comparative gene expression levels for 3D7, 3D7ΔRh2b, 3D7ΔEBA-175, 3D7ΔEBA-140, and D10 late-stage parasites using Affymetrix gene chips. Signal intensity and background noise correction were measured using the MOID [35], RMA [36], or GCRMA [37] algorithms.
(5.1 MB XLS)
Click here for additional data file.
Dataset S2 Invasion Genes: MOID
As in Dataset S1, but analysis is restricted to those genes that are expressed at similar times to genes whose products are known to be involved in invasion (clusters 4, 13, 14, and 15 [34]), excluding the known virulence genes (vars, rifins, and stevors [39]). Signal intensity and background noise correction were measured using the MOID [35] algorithm.
(134 KB XLS)
Click here for additional data file.
Dataset S3 Invasion Genes: RMA
As in Dataset S2, but signal intensity and background noise correction were measured using the RMA [36] algorithm.
(134 KB XLS)
Click here for additional data file.
Dataset S4 Invasion Genes: GCRMA
As in Dataset S2, but signal intensity and background noise correction were measured using the GCRMA [37] algorithm.
(134 KB XLS)
Click here for additional data file.
Dataset S5 Summary of PfRh3 Activation in 3D7ΔEBA-140 and 3D7ΔEBA-175 Parasite Lines
(32 KB DOC)
Click here for additional data file.
Accession Numbers
The PlasmoDB (http://plasmodb.org/PlasmoDB.shtml) accession numbers for the genes discussed in this paper are EBA-140 (MAL13P1.60), EBA-165 (PFD1155w), EBA-175 (PF07_0128), EBA-181 (PFA0125c), PfRh1 (PFD0110w), PfRh2a (PF13_0198), PfRh2b (MAL13P1.176), PfRh3 (PFL2520w), PfRh4 (PFD1150c), and other genes (PFE1465w, PFB0680w, PFF0670w, PF08_0036, PF13_0173, and PF14_0567).
Many thanks to Elizabeth Winzeler from the Scripps Research Institute, California, for generously providing the Affymetrix gene chips and to Jeff Johnson and John Yates III for their help with proteomics. Thanks to Susanne Miller and Brendan Crabb for providing the SERA5 antibody. Thanks to Tony Triglia and Manoj T. Duraisingh for their invaluable input into the project. This work was supported by the Australian National Health and Medical Research Council and the Wellcome Trust. JB is funded by a Peter Doherty Fellowship. AFC is a Howard Hughes International Scholar.
Competing interests. The authors have declared that no competing interests exist.
Author contributions. JB, AGM, KMS, and AFC conceived and designed the experiments. JB, AGM, RTG, and KMS performed the experiments. JB, AGM, RTG, KMS, and AFC analyzed the data. JB and AFC wrote the paper.
Abbreviations
BSD
blasticidin-S deaminase
CIconfidence interval
EBAerythrocyte-binding antigen
PfRh2b
P. falciparum reticulocyte binding-like homolog 2b
NRSnormal (preimmunization) rabbit serum
Rhreticulocyte binding-like homolog
SAsialic acid
==== Refs
References
Miller LH Baruch DI Marsh K Doumbo OK 2002 The pathogenic basis of malaria Nature 415 673 679 11832955
Gaur D Mayer DC Miller LH 2004 Parasite ligand-host receptor interactions during invasion of erythrocytes by Plasmodium merozoites Int J Parasitol 34 1413 1429 15582519
Friedman MJ Blankenberg T Sensabaugh G Tenforde TS 1984 Recognition and invasion of human erythrocytes by malarial parasites: Contribution of sialoglycoproteins to attachment and host specificity J Cell Biol 98 1672 1677 6373782
Mayer DC Kaneko O Hudson-Taylor DE Reid ME Miller LH 2001 Characterization of a Plasmodium falciparum erythrocyte-binding protein paralogous to EBA-175 Proc Natl Acad Sci U S A 98 5222 5227 11309486
Miller LH Haynes JD McAuliffe FM Shiroishi T Durocher JR 1977 Evidence for differences in erythrocyte surface receptors for the malarial parasites, Plasmodium falciparum and Plasmodium knowlesi
J Exp Med 146 277 281 327014
Perkins M 1981 Inhibitory effects of erythrocyte membrane proteins on the in vitro invasion of the human malarial parasite (Plasmodium falciparum) into its host cell J Cell Biol 90 563 567 7026577
Rayner JC Vargas-Serrato E Huber CS Galinski MR Barnwell JW 2001 A Plasmodium falciparum homologue of Plasmodium vivax reticulocyte binding protein (PvRBP1) defines a trypsin-resistant erythrocyte invasion pathway J Exp Med 194 1571 1581 11733572
Duraisingh MT Triglia T Ralph SA Rayner JC Barnwell JW 2003 Phenotypic variation of Plasmodium falciparum merozoite proteins directs receptor targeting for invasion of human erythrocytes EMBO J 22 1047 1057 12606570
Dolan SA Miller LH Wellems TE 1990 Evidence for a switching mechanism in the invasion of erythrocytes by Plasmodium falciparum
J Clin Invest 86 618 624 2200806
Gilberger TW Thompson JK Triglia T Good RT Duraisingh MT 2003 A novel erythrocyte binding antigen-175 paralogue from Plasmodium falciparum defines a new trypsin-resistant receptor on human erythrocytes J Biol Chem 278 14480 14486 12556470
Thompson JK Triglia T Reed MB Cowman AF 2001 A novel ligand from Plasmodium falciparum that binds to a sialic acid-containing receptor on the surface of human erythrocytes Mol Microbiol 41 47 58 11454199
Dolan SA Proctor JL Alling DW Okubo Y Wellems TE 1994 Glycophorin B as an EBA-175 independent Plasmodium falciparum receptor of human erythrocytes Mol Biochem Parasitol 64 55 63 8078523
Hadley TJ Klotz FW Pasvol G Haynes JD McGinniss MH 1987
Falciparum malaria parasites invade erythrocytes that lack glycophorin A and B (MkMk): Strain differences indicate receptor heterogeneity and two pathways for invasion J Clin Invest 80 1190 1193 3308959
Maier AG Duraisingh MT Reeder JC Patel SS Kazura JW 2003
Plasmodium falciparum erythrocyte invasion through glycophorin C and selection for Gerbich negativity in human populations Nat Med 9 87 92 12469115
Mitchell GH Hadley TJ McGinniss MH Klotz FW Miller LH 1986 Invasion of erythrocytes by Plasmodium falciparum malaria parasites: Evidence for receptor heterogeneity and two receptors Blood 67 1519 1521 3516259
Pasvol G Wainscoat JS Weatherall DJ 1982 Erythrocytes deficiency in glycophorin resist invasion by the malarial parasite Plasmodium falciparum
Nature 297 64 66 7040988
Sim BK Chitnis CE Wasniowska K Hadley TJ Miller LH 1994 Receptor and ligand domains for invasion of erythrocytes by Plasmodium falciparum
Science 264 1941 1944 8009226
Binks RH Conway DJ 1999 The major allelic dimorphisms in four Plasmodium falciparum merozoite proteins are not associated with alternative pathways of erythrocyte invasion Mol Biochem Parasitol 103 123 127 10514089
Perkins ME Holt EH 1988 Erythrocyte receptor recognition varies in Plasmodium falciparum isolates Mol Biochem Parasitol 27 23 34 2830508
Baum J Pinder M Conway DJ 2003 Erythrocyte invasion phenotypes of Plasmodium falciparum in the Gambia Infect Immun 71 1856 1863 12654801
Lobo CA de Frazao K Rodriguez M Reid M Zalis M 2004 Invasion profiles of Brazilian field isolates of Plasmodium falciparum: Phenotypic and genotypic analyses Infect Immun 72 5886 5891 15385490
Okoyeh JN Pillai CR Chitnis CE 1999
Plasmodium falciparum field isolates commonly used erythrocyte invasion pathways that are independent of sialic acid residues of glycophorin A Infect Immun 67 5784 5791 10531229
Reed MB Caruana SR Batchelor AH Thompson JK Crabb BS 2000 Targeted disruption of an erythrocyte binding antigen in Plasmodium falciparum is associated with a switch toward a sialic acid-independent pathway of invasion Proc Natl Acad Sci U S A 97 7509 7514 10861015
Adams JH Blair PL Kaneko O Peterson DS 2001 An expanding ebl family of Plasmodium falciparum
Trends Parasitol 17 297 299 11378038
Sim BK Orlandi PA Haynes JD Klotz FW Carter JM 1990 Primary structure of the 175K Plasmodium falciparum erythrocyte binding antigen and identification of a peptide which elicits antibodies that inhibit malaria merozoite invasion J Cell Biol 111 1877 1884 2229177
Galinski MR Medina CC Ingravallo P Barnwell JW 1992 A reticulocyte-binding protein complex of Plasmodium vivax merozoites Cell 69 1213 1226 1617731
Triglia T Duraisingh MT Good RT Cowman AF 2005 Reticulocyte-binding protein homologue 1 is required for sialic acid-dependent invasion into human erythrocytes by Plasmodium falciparum
Mol Microbiol 55 162 174 15612925
Rayner JC Galinski MR Ingravallo P Barnwell JW 2000 Two Plasmodium falciparum genes express merozoite proteins that are related to Plasmodium vivax and Plasmodium yoelii adhesive proteins involved in host cell selection and invasion Proc Natl Acad Sci U S A 97 9648 9653 10920203
Kaneko O Mu J Tsuboi T Su X Torii M 2002 Gene structure and expression of a Plasmodium falciparum 220-kDa protein homologous to the Plasmodium vivax reticulocyte binding proteins Mol Biochem Parasitol 121 275 278 12034462
Triglia T Thompson JK Cowman AF 2001 An EBA175 homologue which is transcribed but not translated in erythrocytic stages of Plasmodium falciparum
Mol Biochem Parasitol 116 55 63 11463466
Taylor HM Triglia T Thompson J Sajid M Fowler R 2001
Plasmodium falciparum homologue of the genes for Plasmodium vivax and Plasmodium yoelii adhesive proteins, which is transcribed but not translated Infect Immun 69 3635 3645 11349024
Gardner MJ Hall N Fung E White O Berriman M 2002 Genome sequence of the human malaria parasite Plasmodium falciparum
Nature 419 498 511 12368864
Duraisingh MT Maier AG Triglia T Cowman AF 2003 Erythrocyte-binding antigen 175 mediates invasion in Plasmodium falciparum utilizing sialic acid-dependent and -independent pathways Proc Natl Acad Sci U S A 100 4796 4801 12672957
Le Roch KG Zhou Y Blair PL Grainger M Moch JK 2003 Discovery of gene function by expression profiling of the malaria parasite life cycle Science 301 1503 1508 12893887
Zhou Y Abagyan R 2002 Match-only integral distribution (MOID) algorithm for high-density oligonucleotide array analysis BMC Bioinformatics 3 3 11835687
Irizarry RA Bolstad BM Collin F Cope LM Hobbs B 2003 Summaries of Affymetrix genechip probe level data Nucleic Acids Res 31 e15 12582260
Wu Z Irizarry RA 2004 Preprocessing of oligonucleotide array data Nat Biotechnol 22 656 658
Smyth GK 2004 Linear models and empirical Bayes methods for assessing differential expression in microarray experiments Stat Appl Genet Mol Biol 3 3
Kyes S Horrocks P Newbold C 2001 Antigenic variation at the infected red cell surface in malaria Annu Rev Microbiol 55 673 707 11544371
Rayner JC Huber CS Galinski MR Barnwell JW 2004 Rapid evolution of an erythrocyte invasion gene family: The Plasmodium reichenowi reticulocyte binding like (RBL) genes Mol Biochem Parasitol 133 287 296 14698440
Duraisingh MT Triglia T Cowman AF 2002 Negative selection of Plasmodium falciparum reveals targeted gene deletion by double crossover recombination Int J Parasitol 32 81 89 11796125
Florens L Washburn MP Raine JD Anthony RM Grainger M 2002 A proteomic view of the Plasmodium falciparum life cycle Nature 419 520 526 12368866
Issit PD Anstee DJ 1998 Applied blood group serology. 4th Ed Durham (North Carolina) Montgomery Scientific Publications 1208 p.
Taylor HM Grainger M Holder AA 2002 Variation in the expression of a Plasmodium falciparum protein family implicated in erythrocyte invasion Infect Immun 70 5779 5789 12228308
Stubbs J Simpson KM Triglia T Plouffe D Tonkin CJ 2005 Molecular mechanism for switching of P. falciparum invasion pathways into human erythrocytes Science 309 1384 1387 16123303
Baum J Thomas AW Conway DJ 2003 Evidence for diversifying selection on erythrocyte-binding antigens of Plasmodium falciparum and P. vivax
Genetics 163 1327 1336 12702678
Cowman AF Crabb BS 2002 The Plasmodium falciparum genome—A blueprint for erythrocyte invasion Science 298 126 128 12364790
Preiser PR Jarra W Capiod T Snounou G 1999 A rhoptry-protein-associated mechanism of clonal phenotypic variation in rodent malaria Nature 398 618 622 10217144
Duraisingh MT Voss TS Marty AJ Duffy MF Good RT 2005 Heterochromatin silencing and locus repositioning linked to regulation of virulence genes in Plasmodium falciparum
Cell 121 13 24 15820675
Peters J Fowler E Gatton M Chen N Saul A 2002 High diversity and rapid changeover of expressed var genes during the acute phase of Plasmodium falciparum infections in human volunteers Proc Natl Acad Sci U S A 99 10689 10694 12142467
Trager W Jensen JB 1976 Human malaria parasites in continuous culture Science 193 673 675 781840
Lambros C Vanderberg JP 1979 Synchronization of Plasmodium falciparum erythrocytic stages in culture J Parasitol 65 418 420 383936
Mamoun CB Gluzman IY Goyard S Beverley SM Goldberg DE 1999 A set of independent selectable markers for transfection of the human malaria parasite Plasmodium falciparum
Proc Natl Acad Sci U S A 96 8716 8720 10411941
Smyth GK 2005 Limma: Linear models for microarray data Gentleman R Carey V Dudoit S Irizarry RA Huber W Bioinformatics and computational biology solutions using R and bioconductor New York Springer 397 420
Torii M Aikawa M 1998 Ultra structure of asexual stages Sherman IW Malaria: Parasite biology, pathogenesis, and protection Washington, DC ASM Press
Anstee DJ 1990 The nature and abundance of human red cell surface glycoproteins J Immunogenet 17 219 225 2093725
|
16362075
|
PMC1315277
|
CC BY
|
2021-01-05 12:11:25
|
no
|
PLoS Pathog. 2005 Dec 16; 1(4):e37
|
utf-8
|
PLoS Pathog
| 2,005 |
10.1371/journal.ppat.0010037
|
oa_comm
|
==== Front
PLoS PathogPLoS PathogppatplpaplospathPLoS Pathogens1553-73661553-7374Public Library of Science San Francisco, USA 1636207610.1371/journal.ppat.001004105-PLPA-RA-0131R1plpa-01-04-06Research ArticleBioinformatics - Computational BiologyImmunologyInfectious DiseasesNeurology - NeurosurgeryPathologyVirologyVirusesHomo (Human)Recognition of Conserved Amino Acid Motifs of Common Viruses and Its Role in Autoimmunity T Cell Recognition of Viruses in MSSospedra Mireia 1Zhao Yingdong 2zur Hausen Harald 3Muraro Paolo A 1Hamashin Christa 4de Villiers Ethel-Michele 3Pinilla Clemencia 45Martin Roland 1¤*
1 Cellular Immunology Section, Neuroimmunology Branch, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, Maryland, United States of America
2 Computational and System Biology Group, Biometric Research Branch, National Cancer Institute, National Institutes of Health, Bethesda, Maryland, United States of America
3 Division for the Characterization of Tumorviruses, Deutsches Krebsforschungszentrum (German Cancer Research Center), Heidelberg, Germany
4 Mixture Sciences, San Diego, California, United States of America
5 Torrey Pines Institute for Molecular Studies and Mixture Sciences, San Diego, California, United States of America
Virgin Herbert EditorWashington University School of Medicine, United States of America* To whom correspondence should be addressed. E-mail: [email protected], [email protected]¤ Current address: Institucio Catalana de Recerca i Estudis Avançats, Fundacio per a la Recerca, Hospital Universitari Vall D'Hebron, Barcelona, Spain
12 2005 16 12 2005 1 4 e4116 8 2005 3 11 2005 Copyright: © 2005 Sospedra et al.2005This is an open-access article distributed under the terms of the Creative Commons Public Domain declaration, which stipulates that, once placed in the public domain, this work may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose.The triggers of autoimmune diseases such as multiple sclerosis (MS) remain elusive. Epidemiological studies suggest that common pathogens can exacerbate and also induce MS, but it has been difficult to pinpoint individual organisms. Here we demonstrate that in vivo clonally expanded CD4+ T cells isolated from the cerebrospinal fluid of a MS patient during disease exacerbation respond to a poly-arginine motif of the nonpathogenic and ubiquitous Torque Teno virus. These T cell clones also can be stimulated by arginine-enriched protein domains from other common viruses and recognize multiple autoantigens. Our data suggest that repeated infections with common pathogenic and even nonpathogenic viruses could expand T cells specific for conserved protein domains that are able to cross-react with tissue-derived and ubiquitous autoantigens.
Synopsis
Infectious agents have been discussed as possible triggers for multiple sclerosis (MS). Molecular mimicry, meaning an antigenic similarity between pathogen proteins and self-proteins (also called autoantigens), is one mechanism that can activate autoreactive T cells. To identify potential triggers and autoantigens in MS, the authors of this study determined the specificity of T cell clones (TCCs) from cerebrospinal fluid (CSF) of an MS patient, which were clonally expanded during disease exacerbation. The CSF is in intimate contact with the central nervous system, which is damaged by autoreactive T cells in MS. The authors observed that these TCCs recognize amino acid motifs from functional protein domains that are evolutionarily conserved between viruses, prokaryotes, and eukaryotes. This phenomenon is reminiscent of pattern recognition by the innate immune system via Toll-like receptors, and represents an interesting bridge as to how immune responses against foreign agents may be misdirected against autoantigens. Three TCCs recognize arginine-rich motifs and respond to peptides from the ubiquitous, nonpathogenic Torque Teno virus (TTV), but also from other common viruses and autoantigens. TTV recognition by clonally expanded CSF TCCs, and the demonstration of viral infection in brains of people with MS, suggest that this virus may participate in triggering or sustaining autoimmune diseases such as MS.
Citation:Sospedra M, Zhao Y, zur Hausen H, Muraro PA, Hamashin C, et al. (2005) Recognition of conserved amino acid motifs of common viruses and its role in autoimmunity. PLoS Pathog 1(4): e41.
==== Body
Introduction
Multiple sclerosis (MS) is considered a CD4+ T helper-1-mediated autoimmune disease that affects the central nervous system (CNS). The etiology of MS remains unclear, but the disease develops in genetically susceptible individuals and likely requires environmental triggers. Epidemiological studies have shown that MS relapses often follow common viral infections, and the viral etiology of certain human demyelinating diseases, and studies of virus-induced disease models [1–3], all point to a role of viruses in MS. Numerous agents have been linked with MS based on serology, pathology, or virus isolation, but none of the associations has been conclusive. The difficulty in identifying a single microorganism as the cause of MS or other autoimmune diseases could indicate that Koch's paradigm, “one organism - one disease,” does not apply to such complex diseases, although we should not discard the possibility that “the” MS agent may still be discovered. However, the bulk of current data suggests that MS is induced and/or exacerbated by several different agents, and that these are most likely ubiquitous pathogens and highly prevalent in the population.
Molecular mimicry, i.e., cross-recognition of foreign agents and self-proteins, is one mechanism by which infectious agents can induce autoimmune diseases [4]. In this context, the traditional search for triggers for MS has been based on choosing a likely target autoantigen (e.g., myelin), establishing CD4+ T cells specific for immunodominant peptides, and then searching for molecular mimics in viral and bacterial databases [5]. The main drawbacks of previous studies are that (a) autoreactive T cells have been established almost universally from the peripheral blood instead of an affected tissue compartment and without knowledge about whether they are related to disease activity; and (b) they have been established based on the consideration that some or all autoreactive T cells in MS are reactive to myelin [6]. This last reasoning is probably too simplistic, since autoimmune diseases can be exquisitely organ- or tissue-specific, and nevertheless autoimmune T cells can be directed against ubiquitous autoantigens, e.g., pyruvate dehydrogenase in primary biliary cirrhosis [7]. To overcome the above problems, we decided to isolate T cells from the cerebrospinal fluid (CSF), a compartment in intimate contact with the affected brain tissue, and focus on T cells that are clonally expanded in vivo during active disease, and hence likely relevant to the autoimmune disease process. Our methodological approach combines unbiased expansion of virtually every T cell using a universal T cell stimulus (phytohemagglutinin [PHA]) [8] with determination of in vivo clonal expansion by T cell receptor (TCR) complementarity-determining region 3 (CDR3) spectratyping, and the unbiased identification of stimulatory peptides by integration of data from screening of positional scanning synthetic combinatorial peptide libraries (PS-SCLs) with protein database analyses [9,10].
Results
Generation of In Vivo-Expanded, CSF-Infiltrating T Cell Clones
We isolated CSF-infiltrating cells from a patient with relapsing-remitting MS during disease exacerbation and cloned them by limiting dilution with PHA as an unbiased stimulus [11]. Growing colonies were characterized for CD4/CD8 and TCRVβ expression, and clonality was confirmed by TCR-variable (TCR VB) chain sequencing (Figure 1A and Table 1). In vivo clonal expansion was assessed by TCR CDR3 spectratyping [12], by comparing the CDR3 length of individual T cell clones (TCCs) with the CDR3 spectrum of CSF mononuclear cells at the time of the lumbar puncture during disease exacerbation (CSF1), and again 14 mo later (CSF2), during remission. Our analysis focused on five CD4+ TCCs (MN10, MN19, MN27, MN36, and MN47), because they were clonally expanded in vivo at exacerbation and reduced 14 mo later, suggesting involvement in disease exacerbation (Figure 1A). Each of these TCCs produced T helper-1 cytokines, and their HLA-class II restriction was also characterized (Table 1).
Figure 1 In Vivo-Expanded CSF-Infiltrating TCCs and Their Response to PS-SCL
(A) TCR BV rearrangement (*Arden's nomenclature) of selected TCCs. Histograms of the relative CDR3 length distributions of each TCC (top histogram) and CSF T cells (middle and bottom histograms). Fluorescence intensity is listed on the y-axis, and on the x-axis, the electrophoresis time resolving in-frame rearrangements of TCRB CDR3 at 3-nt intervals. Red boxes identify the correct alignment. The bottom graph represents the percent contribution (expressed as AUC) of the TCCs' CDR3 to all CDR3s with the same BV chain in the CSF samples.
(B) Proliferative response of MN19 to a complete decapeptide PS-SCL. Single-letter aa codes are listed on the x-axes, and proliferation (cpm) is shown on the y-axis. Data represent one experiment of three. The mixtures with R as the defined aa inducing the highest response are shown in red.
(C) Score matrix for MN19. Each number represents the SIPS-SCL (mean of three independent experiments) of each of the 200 mixtures of a decapeptide PS-SCL (rows, aa; columns, positions). The last column represents the optimal composition of stimulatory peptides “peptidome.” The aa contributing the most to “peptidome” is shown in red.
(D) Peptidomes of the five in vivo-expanded TCCs.
Table 1 Characterization of TCCs
CD4+ T Cell Clones Show Preferential Recognition of Specific Amino Acids
Subsequently, we tested the five TCC with a decapeptide PS-SCL as described [9,10], which allows the identification of stimulatory peptides. The results of a representative experiment for TCC MN19 are presented in Figure 1B (for the other four TCC, see Figure S1). Most remarkably, each of the five TCCs responded strongest to one defined amino acid (aa) in multiple positions of the libraries (marked in red in Figures 1B and S1). Based on the stimulatory indices (SIs) from testing the PS-SCL mixtures (SIPS-SCL), we generated a scoring matrix for each TCC as described [10]. The matrix for MN19 is shown in Figure 1C. We used the matrix to estimate for each TCC the preference of a given aa in the composition of the predicted stimulatory peptides (“peptidome”) by calculating the sum of the SIPS-SCL in each of the ten positions for each of the 20 L-amino acids (L-aa) and expressing it as a fraction of 100% (Figure 1C). If we assume a uniform distribution of the 20 aa in each position, the probability of each aa in each position would be 5%. The “peptidome,” or composition of the predicted stimulatory peptides, for each TCC shows a substantial preference for one aa (Figure 1D, boxed in red): in three TCCs the preference is for R (MN19 [53.2%], MN27 [25.4%], and MN36 [25.1%]), in one TCC for V (MN10 [35.4%]), and in one TCC for K (MN47 [44.6%]).
CSF-Infiltrating T Cell Clones Recognize Torque Teno Virus
Next, we predicted stimulatory peptides for these CD4+ TCCs by PS-SCL biometrical analysis [10], and then selected peptides from human infectious agents (see Materials and Methods). Table 2 shows the actual number of predicted bacterial and viral peptides (column labeled #) and this number normalized for the total number of decamer entries in the database multiplied by one million to facilitate analysis (column labeled NPP). The normalization step was performed to avoid biased representation of agents with large numbers of sequence entries such as HIV or Escherichia coli.
Table 2 Human Infectious Agents Predicted to Be Recognized by the TCCs
Table 2 Continued
Although the prediction included a few peptides from bacteria that had previously been related to MS or CNS infections, the NPP values for these agents were very low. Furthermore, no specific bacterium was preferentially recognized by the three TCCs with R-enriched peptidomes (Table 2). In contrast, and exclusively for these three TCCs, much higher values (NPP > 100) were found for two related small DNA viruses, Torque Teno virus (TTV) and TTV-like mini virus (TLMV) (Table 2 in bold). TTV was identified in 1997 [13] and thought to mediate hepatitis infection, although further follow-up did not confirm this suspicion [14,15]. TLMV was identified in 1999 in a study of TTV prevalence in blood donors [16]. These two viruses are now recognized as ubiquitous agents in the human population and are considered orphan viruses, i.e., they have not been related to a known disease. Interestingly, in the context of MS, TTV DNA has been detected in CSF and brain samples [17,18]. We have confirmed this neurotropism by detecting TTV DNA in five of 11 brain samples from MS patients (Tables 3 and S1) and also in 32 of 41 brain tumors (astrocytoma, medulloblastoma, and ependymoma) (E. M. de Villiers and W. Scheurlen, unpublished data). Interestingly, an identical sequence was detected in three different patients.
Table 3 Detection of TTV-DNA in MS Brain
Next, we tested a total of 228 TTV and TLMV peptides predicted to be recognized by MN19, MN27, or MN36 with higher scores, particularly those with scores higher than 0.7 of the maximal theoretical score. Assuming that the shared preference for R-enriched peptides by these three TCCs reflects similarities in specificity, we tested the peptides predicted for one TCC also with the other two TCCs. The number of stimulatory peptides identified for each TCC is summarized in Figure 2A. Detailed information about all stimulatory peptides is available in Table S2. Fewer peptides are stimulatory for MN36, although in contrast to MN19 and MN27, MN36 preferentially recognizes peptides longer than decamers. Since the current search strategy is based on decamer libraries, our prediction efficacy was therefore much lower for MN36. Despite this, it is interesting to note that all three TCCs recognize peptides not only from one organism, TTV, but even the same peptides. The few TTV peptides that stimulate MN36 are recognized by all three TCCs, and 48 peptides are recognized by MN19 and MN27 (Figure 2B). However, as shown in Figure S2 by dose titration experiments, the different clones recognize these peptides with different affinities, i.e., at different concentrations.
Figure 2 Stimulatory Peptides from TTV and TLMV
(A) Number and SIs of the stimulatory peptides from TTV and TLMV identified for TCCs MN19, MN27, and MN36.
(B) Number and SIs of the peptides co-recognized by different TCCs. Peptides have been tested in proliferation assays at 10 μg/ml and using PBMCs as antigen-presenting cells.
MS Patient Is Infected with TTV at Disease Exacerbation
In order to determine whether the MS patient was infected with TTV, serum samples from the time of the first CSF isolation (time 0) and 1, 3, 8, 14, and 24 mo later, were tested for TTV-DNA by PCR amplification using two primer combinations with the respective nested primers. Cloning and sequencing of all amplicons confirmed the presence of TTV DNA in the first serum sample obtained during exacerbation and also 1 mo later, during a second exacerbation (Figure 3A and Table S1). The next three serial samples during remission were negative, but TTV-DNA was again detected 24 mo later, suggesting a second infection or reactivation. The highly conserved region of the TTV upstream regulatory region (URR) amplified here did not allow for determination of the specific TTV types involved. However, sequence alignment of the samples with the closest related known TTV strains indicates that different isolates were present in the positive serum samples (Figure 3B). Seven different isolates were identified at month 0, three at month 1, and two at month 24; and only one isolate was shared by two isolates (time 0 and 1 mo). Interestingly, this isolate was identical to the one shared by the three brain samples (see above). The CSF could not be checked for TTV infection, as CSF became unavailable.
Figure 3 TTV-DNA in Patient's Serum Samples
(A) Detection of TTV-DNA by PCR amplification using two primer combinations with the respective nested primers. Six different serum samples from the same patient obtained at different time points were analyzed. The first two samples were obtained during relapse. Time points with simultaneous CSF are indicated. Plus symbol indicates presence of TTV DNA; minus symbol indicates absence.
(B) The sequences obtained by cloning and sequencing of all amplicons and the alignment with the closer related known TT viruses are shown. One isolate that was present in two different serum samples is shown in red.
The absence of TTV-DNA in the serum during remission does not exclude its persistence in infected cells, since presence of TTV DNA in tissue samples has been reported despite its absence in corresponding sera [17].
Cross-Recognition of Conserved TTV Epitopes
Next we examined the location of each individual stimulatory peptide (Table S2). TTV has five open reading frames (ORFs) and, interestingly, 96.2% (152/158) of the stimulatory peptides were from a narrow region in the 74-aa N-terminal sequence of ORF1 (Figure 4A). ORF1 and especially its N-terminal region are highly enriched in positively charged aa (mainly R) when compared to the proteomes of all organisms (Figure 4B). The N-terminal R-rich domain of TTV ORF1 corresponds to a potential nuclear localization signal (NLS). A significant proportion of the stimulatory peptides identified are shared by different TTV isolates, indicating that this N-terminal sequence is conserved (Table S2).
Figure 4 Characterization TTV Peptides
(A) Configuration of ORF1 from TTV showing the hypervariable region (HVR) and the 74-aa N-terminal domain. Sequence of the first 74 aa for a prototype TTV is shown. Distribution of all TTV stimulatory peptides identified between the first 74 aa is shown.
(B) AA composition of all proteins, ORF1 from TTV, and the 74-aa N-terminal region of ORF1.
Since multigenotype and persistent infections with TTV are frequent, an expansion of TCCs recognizing these conserved domains is expected. We therefore addressed whether the precursor frequency of T cells responding to R-enriched peptides was increased in the patient in our study. Using IL-7 primary proliferation assays, we stimulated peripheral blood cells with mixtures of ten peptides enriched in R, ten peptides enriched in K, ten peptides not enriched in any specific aa, and a decamer peptide mixture in which all 20 L-aa are present in randomized order in each of the ten positions (X10). Detailed information about these mixtures is available in Table S3. We demonstrated a higher frequency of peripheral T cells specific for TTV R-enriched peptides up to 3 y after the last exacerbation (Figure 5A). The majority of these T cells originated from the memory T cell pool (Figure 5B).
Figure 5 TCL Response to R-Enriched Peptides in Peripheral Blood
(A) Proliferative response of peripheral TCL to the following: a mixture of ten peptides enriched in R, a mixture of ten peptides enriched in K, a mixture of ten peptides not enriched in any specific aa, and a mixture of randomized decapeptides. Negative control is medium without peptide mixture. Responses higher that the mean of the negative control plus 4 standard deviations have been considered positive. * Detailed information of mixtures used in IL-7 primary proliferation assay are available in Table S3.
(B) Comparison of the origin (naïve versus memory) of all TCLs with confirmed reactivity against R-enriched peptides.
Cross-Recognition of Conserved R-Rich Protein Domains in Common Viruses
We also identified stimulatory peptides for these three TCCs from other common human viruses, particularly adenovirus and papillomavirus (Table 4). Interestingly, and similar to what we observed for TTV, various peptides are located in the same protein region and shared by different, but related, strains of one virus. The four stimulatory peptides identified from different types of human papillomavirus are located between aa 447–461 in the minor capsid protein L2 (Table 4). The C terminus of this protein is enriched in basic aa, and the region between aa 456 and 461 is described as a putative NLS [19]. Regarding human adenovirus, we identified three different stimulatory peptides from pVII protein shared by different strains (Table 4). PVII protein is a core protein enriched in R with a histone-like function and tightly bound to the viral DNA. Previous observations indicate that this close association is likely preserved during transport of the viral genome to the nucleus [20]. Interestingly, both proteins share important characteristics with the 74-aa N-terminal sequence of ORF1 from TTV.
Table 4 Stimulatory Peptides from Common Viruses
TTV-Specific T Cells Cross-React with R-Rich Domains in Autoantigens
R-enriched domains are frequent in eukaryotes, prokaryotes, and viruses as part of DNA-binding regions, NLSs, and other functional sequences, providing a bridge for cross-reactivity between pathogens and autoantigens. In support of this notion, three of the stimulatory TTV peptides recognized by TCC MN19 are identical to peptides from two human proteins (Table 5). The first peptide stems from a DEAD box protein. Although the function of most DEAD box proteins is unknown, helicase activity and interactions with DNA or RNA have been associated with several of them [21]. The other two peptides are part of the α-1B adrenergic receptor (α1-AR). They are located between aa 367–380, an R-rich motif that has been reported to interact with the multifunctional protein gC1qR, controlling their expression and subcellular localization [22].
Table 5 Peptides Derived from TTV and Human Proteins with Identical Peptide Sequences
TCC MN19 is the most expanded clone, and it shows the highest preference for R-enriched peptides and, consequently, for R-rich protein domains. For this TCC we have identified five overlapping decapeptides from α1-AR (Table 6), two of these identical to two TTV peptides, that completely cover the R-rich motif between aa 367–380 (see above). MN19 also cross-recognizes two peptides from other adrenergic receptors and three overlapping peptides from the N-terminal R-enriched region of ARP (i.e., arginine-rich protein). Although the expression and function of this protein are unknown, starting at aa 56 the translated DNA shares 100% sequence identity with the secreted human protein MANF (mesencephalic astrocyte-derived neurotrophic factor), which protects dopaminergic neurons in the substantia nigra of the brain [23].
TCCs MN27 and MN36 showed a lower preference for R-enriched peptides than did MN19, which is reflected in the corresponding “peptidomes” and in the composition of the stimulatory peptides. Despite this relatively lower recognition of R-rich domains, we nevertheless identified several stimulatory decapeptides with at least three arginines and potential biological relevance in the context of the CNS or the immune system (Table 6). Among these CNS-related molecules is the purinergic receptor PX2A, which is expressed on oligodendrocytes and astrocyte; several other neurotransmitter receptors; ion channels; and molecules that play a role in brain function and metabolism. Among the immunologically relevant autoantigens stimulatory for MN27 and MN36 are immunoglobulin chains and the pattern recognition receptor TLR9. These antigens are of particular interest in the context of perpetuating the autoimmune response once it has started in the CNS.
Table 6 Cross-Reactive Autoantigens
Discussion
In order to identify putative triggers in MS, we focused in the current study on TCCs that are clonally expanded in vivo at the time of disease exacerbation and in a tissue that is in intimate contact with the affected CNS. Taking T cells directly from the brain or reinjecting potentially autoreactive T cells back into a patient would provide more direct evidence for the relation to disease, but for obvious reasons these steps are either very difficult to justify, i.e., brain biopsies for nondiagnostic reasons, or impossible in humans. Thus, as a next step, we applied methods that included the unbiased expansion of T cells and a search strategy using combinatorial peptide chemistry and bioinformatics. This approach allowed, to our knowledge for the first time, the identification of target epitopes of T cells for which nothing was known in terms of their antigen specificity a priori.
The first interesting and novel point of our data is that the recognition pattern of the five TCCs we studied showed a preference for one specific aa at several positions of the PS-SCL that translates into recognition of peptides enriched in this aa. The most stimulatory peptides for MN19, which shows the strongest bias for R (53.2%), contain on average 7.8 R residues within a decamer. The preference for R-enriched peptides shared by the clones MN19, MN27, and MN36, together with their marked clonal expansion in the CSF, suggest that these three TCCs could have been activated in the periphery in response to the same foreign agent(s) before migrating to the CSF. The analysis of predicted peptides from human infectious agents confirmed this hypothesis. A large number of peptides from TTV and TMLV were predicted exclusively for the three TCCs with the preference for R-enriched peptides. At this stage, no other pathogen was predicted at a comparable level. We therefore assumed that TTV and related viruses are the most likely candidate targets for MN19, MN27, and MN36. We synthesized a large number of these predicted peptides and found not only multiple stimulatory TTV peptides for each of the three TCCs, but many that were recognized by two or even all three TCCs. With few exceptions all peptides are located in an R-rich area of TTV ORF1, a putative capsid protein that is thought to mediate binding to viral DNA and transport to the nucleus. The chromatin association of this R-enriched region of TTV has been demonstrated in in vitro experiments (R. Kellner and E. M. de Villiers, unpublished data). This region is shared by different TTV isolates, suggesting that it represents a conserved functional domain. Human T cell recognition of epitopes conserved between different but related subtypes of viruses has been described for enteroviruses [24,25], flaviviruses [26], influenza A virus [27], and adenoviruses [28]. It has been proposed that exposure to consecutive infections with different strains results in repeated cycles of stimulation and expansion of T cells specific for shared epitopes [27]. The frequent multigenotype infections that characterize TTV and occurred in the patient in this study, together with the unusually large size of the R-rich domain (74 aa) in this virus, are likely relevant in amplifying T cell expansion, a notion supported by the high frequency of peripheral memory T cells specific for R-enriched peptides.
The fact that basic aa such as R play an important role in the interaction of proteins with DNA and with other proteins such as shuttle proteins implies that R-enriched domains are frequent in all organisms as part of DNA/RNA-binding regions, NLSs, or other functional domains. The recognition of such evolutionarily conserved domains by adaptive immune cells such as the TCCs examined here is reminiscent of pattern recognition by innate immune receptors such as Toll-like receptors and could facilitate the cross-recognition of different organisms and human proteins. The fact that several stimulatory peptides from common human viruses other than TTV, such as adenovirus and papillomavirus, are also from R-enriched protein domains with characteristics similar to the 74-aa N-terminal region of ORF1 supports this hypothesis. Important additional evidence includes the identity between three stimulatory TTV peptides and three peptides from human proteins with functional R-enriched domains. A series of other interesting R-enriched peptides from autoantigens have been identified, including several overlapping peptides from a R-rich motif of α1-AR. The density of α1-ARs in the CNS is among the highest of any tissue in the body [29], and although the specific functional roles of this receptor remain uncertain, it has been implicated in motor control by the CNS [30–33]. Interestingly, no myelin autoantigens were among the autoantigenic peptides with highest predicted stimulatory scores for these TCCs. This could be explained by the fact that axonal damage, gliosis, and inflammation also play a role besides demyelination, and that nonmyelin autoantigens, such as alpha-B crystalline, S-100, and others, have already been implicated in MS or experimental allergic encephalomyelitis.
One important component that remains unidentified is the initial activator of these T cells. Both TTV infections and autoantigens are plausible. However, since TTV infections occur frequently and probably also early in life, and since activation of T cells probably starts in the periphery and transmigration into the CNS/CSF and damage of tissue are subsequent events, we believe that the most likely scenario may be that repetitive TTV infections lead to expansion of these T cells.
T cell recognition of R-enriched peptides may not be critical for clearance of the infectious agent, since most of these peptides are recognized at only moderate concentrations, suggesting low functional avidity of the T cells. Furthermore, deletion of high-avidity cells specific to these common, conserved domains in the thymus is expected. However, the frequent occurrence of these peptides in nature implies that T cells specific for R-enriched areas may be activated repeatedly, resulting in lower requirements for costimulation and expansion despite relatively low functional avidity. The resulting reduction in the activation threshold, together with the repetitive expansion of these cells, may facilitate responses to suboptimal autoantigens in the target tissue acting as an “acquired susceptibility trait” in MS. Since TTV infections are also frequent in normal donors, additional predisposing factors, such as human leukocyte antigen and other susceptibility genes, compromised CNS repair processes, increased tissue vulnerability, or variations in central tolerance, are probably necessary for MS development.
Despite this demonstration that several CSF-infiltrating and in vivo-expanded TCCs from an MS patient in exacerbation recognize large numbers of peptides from TTV, we do not suggest this virus as the latest “MS agent.” The fact that these likely disease-related TCCs recognize R-enriched conserved domains shared between different viruses and human autoantigens suggests that the specificity of these T cells results in recognition of specific types of protein domains rather than a specific organism. This kind of specificity or recognition of evolutionarily conserved domains could be involved in inducing and perpetuating autoreactive T cells. It will be important to examine whether the proposed mechanism applies to other MS patients and to other autoimmune diseases, and whether the recognition of conserved protein domains by adaptive immune cells plays a role during protective immune responses.
Materials and Methods
TCCs and antigen-presenting cells.
TCCs were established from CSF of an untreated patient with relapsing-remitting MS during exacerbation by limiting dilution at 0.3 and 3 cells/well with 2 × 105 autologous, irradiated PBMCs and 2.5 μg ml−1 of PHA-P (Sigma, St Louis, Missouri, United States) in IMDM containing 100 U ml−1 penicillin/streptomycin, 50 μg ml−1 gentamicin, 2 mM L-glutamine (BioWhittaker, Gaithersburg, Maryland, United States), and 5% human serum (Gemini Bio-products, Woodland, California, United States). After 24 h, 20 U ml−1 of human recombinant IL-2 (hrIL-2, National Cancer Institute, National Institutes of Health, Bethesda, Maryland, United States) were added. Cells were restimulated every 2 wk with 2.5 μg ml−1 PHA, 20 U ml−1 hrIL-2, and autologous irradiated PBMCs, and hrIL2 was added every 3–4 d.
RT-PCR and sequencing of TCR rearrangements.
TCC TCR VB gene usage was analyzed by PCR using 21 TCRAV and 23 TCRBV family-specific oligonucleotide primers. Nucleotide sequencing of PCR products was performed as described [34]. TCR gene designations are in accord with Arden's nomenclature [35].
Cytokine production.
TCCs were stimulated with coated anti-CD3 antibody, and supernatants were collected after 48 h from cultures with/without antibody. IFN-γ, GM-CSF, IL-4, and IL-10 levels were determined by ELISA following the manufacturer's protocol (Biosource, Camarillo, California). For TCCs MN19, MN27, and MN36 the cytokine production was confirmed with several stimulatory peptides.
CDR3 spectratyping.
For high-resolution TCR β-chain CDR3 spectratyping, 2.5 μl of PCR product from each TCR-BV were used as template in a 12.5 μl primer-extension (“runoff”) reaction containing 1.25 μl of 5′FAM-labeled BV primer, 0.25 μl of 10 mM dNTPs, 0.06 μl of Pfu DNA polymerase, 1.25 μl of Pfu reaction buffer, and 7.2 μl H2O. After thermal cycling (95 °C for 2 min; followed by ten cycles of 94 °C for 20 s, 55 °C for 45 s, and 72 °C for 45 s; and a final extension of 72 °C for 10 min), 2 μl of runoff product was mixed with loading buffer containing four Cy-5-labeled DNA size markers, heat-treated at 80 °C for 2 min, and run on a 6% polyacrylamide gel on an OpenGene (Visible Genetics, Toronto, Ontario, Canada) sequencer. Electropherograms were analyzed for peak size (bp), peak height, and area under the curve (AUC). The percentage represented by each CDR3 peak in a BV spectrum (corresponding to the representation of clonal populations with a given CDR3 length) was calculated according to the formula %AUC BVn = (AUC BVn/AUC all BV) × 100. TCR CDR1, CDR2, and CDR3 boundaries were defined according to the IMGT [36].
Peptide combinatorial libraries and individual peptides.
A synthetic N-acetylated, C-amide L-aa decapeptide combinatorial library in a positional scanning format (PS-SCL; 200 mixtures) was prepared as described [37]. Each OX9 mixture consists of 3.2 × 1011 (199) different decamer peptides at approximately equimolar concentration. Individual decapeptides were synthesized with a custom multiple peptide synthesizer using solid-phase Fmoc chemistry. The purity and identity of each peptide were characterized by mass spectrometry.
Proliferative assays.
TCC proliferation responses to PS-SCL mixtures or individual decapeptides were tested by seeding in duplicate 2 × 104 T cells and 1 × 105 irradiated PBMCs (3,000 rad) with or without PS-SCL mixtures or individual decapeptides. Because the specificity of TCCs was unknown, PHA-P stimulation served as positive control. Proliferation was measured by methyl-3H-thymidine (Amersham Biosciences, Little Chalfont, United Kingdom) incorporation. The stimulatory index for a PS-SCL mixture (SIPS-SCL) with an aa defined at one position was calculated as SIPS-SCL =SI′/mean all SI′ in the library, where SI′ = (mean of duplicate cpm, mixture) − (mean cpm, background). Responses to mixtures were considered positive when SI > 2. The SI for individual peptides was calculated as SI = (mean of duplicate cpm, peptide)/(mean cpm, background). Responses to individual peptides were considered positive when SI > 3, cpm > 1,000, and at least three standard deviations above average background cpm in at least three independent experiments.
Biometric analysis and database searches.
Responses to PS-SCLs were analyzed as described [9,10]. A positional scoring matrix was generated by assigning a value of the stimulatory potential to each of the 20 defined aa in each of the ten positions. Based on a model of independent contribution of individual aa to peptide antigen recognition, the predicted stimulatory score of a given peptide is the sum of the stimulatory potential of all aa contained in the peptide in each position. Using a web-based search tool [38], the scoring matrix was applied to rank, according to their stimulatory score, of all the naturally overlapping 10-mer peptides in the protein sequences within the GenPept database (version 136) (ftp://ftp.ncifcrf.gov/pub/genpept), and for viral peptides, within RefSeq (http://www.ncbi.nlm.nih.gov/RefSeq). We analyzed viral and bacterial peptides with scores higher than 0.7 of the predicted maximal theoretical score (Smax). The cut-off of 0.7 of Smax is based on prior experience of the sensitivity and accuracy of the approach [10,39,40]. We then selected peptides from human infectious agents, counted the predicted peptides from each individual organism, and normalized the value, taking into account the total number of decamers for each organism in the GenPept database, in order to avoid a bias related to the number of database entries. This problem would have otherwise significantly skewed the data for organisms with large numbers of reported sequences, such as HIV.
Precursor frequency in peripheral blood.
Primary proliferation assays were performed as described [41]. Briefly, PBMCs were seeded in 96-well plates at 1 × 105 cells/well on day 0 in the presence of antigen and IL-7. A mixture of ten R-enriched peptides was used as antigen. Three different controls were included, a mixture of ten K-enriched peptides, a mixture of ten peptides not enriched in any specific aa, and a decamer peptide mixture in which all 20 L-aa are present in randomized order in each of the 10 positions (X10) without any defined aa. After 7 d, cell cultures were divided in half, and positive wells were identified by comparing the amount of actively proliferating cells in split cultures with the proliferation of PBMCs seeded without antigen. The remaining half of the positive wells were restimulated, and a confirmation assay was performed at days 17–19. Confirmed positive cultures were used to determine the naïve versus memory origin of precursors T cells by flow cytometry using anti-CD45RA and anti-CD45RO antibodies (Pharmingen, BD, Palo Alto, California, United States).
TTV detection.
Total DNA was extracted from serum samples by the High Pure Viral Nucleic Acid Kit (Roche Diagnostics, Penzberg, Germany). DNA from brain samples was extracted using phenol and chloroform-isoamyl alcohol. PCR amplification of each sample was performed twice. The primer combinations NG133-NG352 with nested NG249-NG351 [42] were used to amplify a 134-bp fragment of the TTV URR. The latter overlaps the highly conserved region of 71 bp, which is amplified by primers NG472-NG352 and nested NG473-NG351 [43] used in the second PCR amplification. All amplicons were cloned, and at least 12 clones per sample sequenced. Sequences were compared to all available TTV sequences.
Supporting Information
Figure S1 Proliferative Response of TCCs MN36, MN27, MN10, and MN47 to 200 Mixtures of a Decapeptide PS-SCL in Which Each Mixture Has One Defined aa in One Position and the Other Positions Contain All L-aa Except Cysteine
Horizontal axes, single-letter aa code; vertical axes, proliferation as counts per minute. Data represent one experiment of three. The defined aa inducing the highest response at several positions for each TCC is shown in red.
(61 KB PDF)
Click here for additional data file.
Figure S2 Proliferation of TCCs MN19 and MN27 in Response to Different Concentrations of Five Decapeptides
Proliferation is shown as SI. Peptide sequences are listed on top.
(32 KB PDF)
Click here for additional data file.
Table S1 Proportion of TTV Clones Identified After Single and Nested PCR
(55 KB PDF)
Click here for additional data file.
Table S2 Stimulatory Peptides from TTV and TLMV
(71 KB PDF)
Click here for additional data file.
Table S3 Mixtures Used in IL-7 Primary Proliferation Assay
(42 KB PDF)
Click here for additional data file.
Accession Numbers
The GenBank (http://www.ncbi.nlm.nih.gov/) accession numbers of the proteins discussed in this paper are α1-AR (AAB60352), ARP (AAB60352), DEAD box protein (AAH40185), minor capsid protein L2 (AAA79427), pVII protein (AAP49207 and AAA92212), PX2A (AAD42947), and TLR9 (BAB19259).
We thank Joan Ohayon, Gregg Blevins, Carlos Mora, Riccardo Cassiani Ingonis, Azita Kashani, and Amy Packer (all Neuroimmunology Branch [NIB], National Institute of Neurological Disorders and Stroke [NINDS], National Institutes of Health [NIH]) for their assistance; Romana Kimmel (Deutsches Krebsforschungszentrum [DKFZ]) for excellent technical assistance; and Andreas Hunziker (DKFZ) for sequencing. We also thank Jan Lünemann (NIB, NINDS, NIH) and Barbara Rehermann (National Institute of Diabetes and Digestive and Kidney Diseases, NIH) for critical comments. We thank Richard Simon (Computational and Systems Biology Group, National Cancer Institute, NIH) for his advice, and Cedric S. Raine (Albert Einstein College, Yeshiva University, New York) and Sergio Baranzini (University of California, San Francisco) for brain samples. M. Sospedra was supported by a NINDS Competitive Fellowship. This work was supported in part by Multiple Sclerosis National Research Institute (CP).
Competing interests. The authors have declared that no competing interests exist.
Author contributions. MS and RM conceived and designed the experiments. MS and PAM performed the experiments. MS, YZ, HH, PAM, EV, CP, and RM analyzed the data. YZ, HH, CH, and CP contributed reagents/materials/analysis tools. MS and RM wrote the paper.
Abbreviations
α1-ARα-1B adrenergic receptor
aaamino acid(s)
AUCarea under the curve
CDR3complementarity-determining region 3
CNScentral nervous system
CSFcerebrospinal fluid
L-aaL-amino acid(s)
MSmultiple sclerosis
NLSnuclear localization signal
ORFopen reading frame
PHAphytohemagglutinin
PS-SCLpositional scanning synthetic combinatorial peptide library
SIstimulatory index
TCCT cell clone
TCRT cell receptor
TLMVTT-like mini virus
TTVTorque Teno virus
URRupstream regulatory region
==== Refs
References
Sibley WA Bamford CR Clark K 1985 Clinical viral infections and multiple sclerosis Lancet 1 1313 1315 2860501
Johnson RT 1994 The virology of demyelinating diseases Ann Neurol 36 (Suppl) S54 60 8017889
Soldan SS Jacobson S 2001 Role of viruses in etiology and pathogenesis of multiple sclerosis Adv Virus Res 56 517 555 11450311
Fujinami RS Oldstone MB 1985 Amino acid homology between the encephalitogenic site of myelin basic protein and virus: Mechanism for autoimmunity Science 230 1043 1045 2414848
Wucherpfennig KW Strominger JL 1995 Molecular mimicry in T cell-mediated autoimmunity: Viral peptides activate human T cell clones specific for myelin basic protein Cell 80 695 705 7534214
Sospedra M Martin R 2005 Immunology of multiple sclerosis Annu Rev Immunol 23 683 747 15771584
Shimoda S Nakamura M Ishibashi H Hayashida K Niho Y 1995 HLA DRB4 0101-restricted immunodominant T cell autoepitope of pyruvate dehydrogenase complex in primary biliary cirrhosis: Evidence of molecular mimicry in human autoimmune diseases J Exp Med 181 1835 1845 7536796
Moretta A Pantaleo G Moretta L Cerottini JC Mingari MC 1983 Direct demonstration of the clonogenic potential of every human peripheral blood T cell. Clonal analysis of HLA-DR expression and cytolytic activity J Exp Med 157 743 754 6600491
Hemmer B Gran B Zhao Y Marques A Pascal J 1999 Identification of candidate T-cell epitopes and molecular mimics in chronic Lyme disease Nat Med 5 1375 1382 10581079
Zhao Y Gran B Pinilla C Markovic-Plese S Hemmer B 2001 Combinatorial peptide libraries and biometric score matrices permit the quantitative analysis of specific and degenerate interactions between clonotypic TCR and MHC peptide ligands J Immunol 167 2130 2141 11489997
Arenz M Herzog-Hauff S Meyer zum Buschenfelde KH Lohr HF 1997 Antigen-independent in vitro expansion of T cells does not affect the T cell receptor V beta repertoire J Mol Med 75 678 686 9351707
Gorski J Yassai M Zhu X Kissela B Kissella B 1994 Circulating T cell repertoire complexity in normal individuals and bone marrow recipients analyzed by CDR3 size spectratyping. Correlation with immune status J Immunol 152 5109 5119 8176227
Nishizawa T Okamoto H Konishi K Yoshizawa H Miyakawa Y 1997 A novel DNA virus (TTV) associated with elevated transaminase levels in posttransfusion hepatitis of unknown etiology Biochem Biophys Res Commun 241 92 97 9405239
Simmonds P Davidson F Lycett C Prescott LE MacDonald DM 1998 Detection of a novel DNA virus (TTV) in blood donors and blood products Lancet 352 191 195 9683208
Prati D Lin YH De Mattei C Liu JK Farma E 1999 A prospective study on TT virus infection in transfusion-dependent patients with beta-thalassemia Blood 93 1502 1505 10029577
Takahashi K Iwasa Y Hijikata M Mishiro S 2000 Identification of a new human DNA virus (TTV-like mini virus, TLMV) intermediately related to TT virus and chicken anemia virus Arch Virol 145 979 993 10881684
Pollicino T Raffa G Squadrito G Costantino L Cacciola I 2003 TT virus has a ubiquitous diffusion in human body tissues: Analyses of paired serum and tissue samples J Viral Hepat 10 95 102 12614465
Maggi F Fornai C Vatteroni ML Siciliano G Menichetti F 2001 Low prevalence of TT virus in the cerebrospinal fluid of viremic patients with central nervous system disorders J Med Virol 65 418 422 11536254
Zhou J Sun XY Louis K Frazer IH 1994 Interaction of human papillomavirus (HPV) type 16 capsid proteins with HPV DNA requires an intact L2 N-terminal sequence J Virol 68 619 625 8289365
Burg JL Schweitzer J Daniell E 1983 Introduction of superhelical turns into DNA by adenoviral core proteins and chromatin assembly factors J Virol 46 749 755 6304339
Rocak S Linder P 2004 DEAD-box proteins: the driving forces behind RNA metabolism Nat Rev Mol Cell Biol 5 232 241 14991003
Xu Z Hirasawa A Shinoura H Tsujimoto G 1999 Interaction of the alpha(1B)-adrenergic receptor with gC1q-R, a multifunctional protein J Biol Chem 274 21149 21154 10409668
Petrova P Raibekas A Pevsner J Vigo N Anafi M 2003 MANF: A new mesencephalic, astrocyte-derived neurotrophic factor with selectivity for dopaminergic neurons J Mol Neurosci 20 173 188 12794311
Marttila J Hyoty H Vilja P Harkonen T Alho A 2002 T cell epitopes in coxsackievirus B4 structural proteins concentrate in regions conserved between enteroviruses Virology 293 217 224 11886241
Graham S Wang EC Jenkins O Borysiewicz LK 1993 Analysis of the human T-cell response to picornaviruses: Identification of T-cell epitopes close to B-cell epitopes in poliovirus J Virol 67 1627 1637 7679749
Aihara H Takasaki T Matsutani T Suzuki R Kurane I 1998 Establishment and characterization of Japanese encephalitis virus-specific, human CD4+ T-cell clones: Flavivirus cross-reactivity, protein recognition, and cytotoxic activity J Virol 72 8032 8036 9733842
Haanen JB Wolkers MC Kruisbeek AM Schumacher TN 1999 Selective expansion of cross-reactive CD8+ memory T cells by viral variants J Exp Med 190 1319 1328 10544203
Heemskerk B Veltrop-Duits LA van Vreeswijk T ten Dam MM Heidt S 2003 Extensive cross-reactivity of CD4+ adenovirus-specific T cells: Implications for immunotherapy and gene therapy J Virol 77 6562 6566 12743315
Wilson KM Minneman KP 1989 Regional variations in alpha 1-adrenergic receptor subtypes in rat brain J Neurochem 53 1782 1786 2553865
Stone EA Lin Y Itteera A Quartermain D 2001 Pharmacological evidence for the role of central alpha 1B-adrenoceptors in the motor activity and spontaneous movement of mice Neuropharmacology 40 254 261 11114404
Stone EA Grunewald GL Lin Y Ahsan R Rosengarten H 2003 Role of epinephrine stimulation of CNS alpha1-adrenoceptors in motor activity in mice Synapse 49 67 76 12710017
Zuscik MJ Sands S Ross SA Waugh DJ Gaivin RJ 2000 Overexpression of the alpha1B-adrenergic receptor causes apoptotic neurodegeneration: Multiple system atrophy Nat Med 6 1388 1394 11100125
Drouin C Darracq L Trovero F Blanc G Glowinski J 2002 Alpha1b-adrenergic receptors control locomotor and rewarding effects of psychostimulants and opiates J Neurosci 22 2873 2884 11923452
Muraro PA Jacobsen M Necker A Nagle JW Gaber R 2000 Rapid identification of local T cell expansion in inflammatory organ diseases by flow cytometric T cell receptor Vbeta analysis J Immunol Methods 246 131 143 11121554
Arden B Clark SP Kabelitz D Mak TW 1995 Human T-cell receptor variable gene segment families Immunogenetics 42 455 500 8550092
Lefranc MP 2003 IMGT, the international ImMunoGeneTics database Nucleic Acids Res 31 307 310 12520009
Pinilla C Appel JR Houghten RA 1994 Investigation of antigen-antibody interactions using a soluble, non-support-bound synthetic decapeptide library composed of four trillion (4 × 1012 ) sequences Biochem J 301 847 853 7519851
Zhao Y Grover L Simon R 2001 TEST: A web-based T-cell epitope search tool. Proceedings of the 14th IEEE Symposium on Computer-Based Medical Systems. 26–27 July 2001 Bethesda IEEE Computer Society 493 497
Nino-Vasquez JJ Allicotti G Borras E Wilson DB Valmori D 2004 A powerful combination: The use of positional scanning libraries and biometrical analysis to identify cross-reactive T cell epitopes Mol Immunol 40 1063 1074 15036911
Rubio-Godoy V Dutoit V Zhao Y Simon R Guillaume P 2002 Positional scanning-synthetic peptide library-based analysis of self- and pathogen-derived peptide cross-reactivity with tumor-reactive Melan-A-specific CTL J Immunol 169 5696 5707 12421949
Bielekova B Muraro PA Golestaneh L Pascal J McFarland HF 1999 Preferential expansion of autoreactive T lymphocytes from the memory T-cell pool by IL-7 J Neuroimmunol 100 115 123 10695722
Okamoto H Takahashi M Kato N Fukuda M Tawara A 2000 Sequestration of TT virus of restricted genotypes in peripheral blood mononuclear cells J Virol 74 10236 10239 11024155
Peng YH Nishizawa T Takahashi M Ishikawa T Yoshikawa A 2002 Analysis of the entire genomes of thirteen TT virus variants classifiable into the fourth and fifth genetic groups, isolated from viremic infants Arch Virol 147 21 41 11855633
|
16362076
|
PMC1315278
|
CC0
|
2021-01-05 12:11:25
|
no
|
PLoS Pathog. 2005 Dec 16; 1(4):e41
|
utf-8
|
PLoS Pathog
| 2,005 |
10.1371/journal.ppat.0010041
|
oa_comm
|
==== Front
PLoS GenetPLoS GenetpgenplgeplosgenPLoS Genetics1553-73901553-7404Public Library of Science San Francisco, USA 1636207710.1371/journal.pgen.001007405-PLGE-RA-0122R2plge-01-06-03Research ArticleGenetics/Gene FunctionGenetics/Genetics of DiseaseGenetics/Disease ModelsGenetics/Complex TraitsEukaryotesAnimalsMus (Mouse)Ablation of the Sam68 RNA Binding Protein Protects Mice from Age-Related Bone Loss Sam68 Regulates Osteogenesis and Bone MassRichard Stéphane 1*Torabi Nazi 1Franco Gladys Valverde 2Tremblay Guy A 1¤aChen Taiping 1¤bVogel Gillian 1Morel Mélanie 1Cléroux Patrick 1Forget-Richard Alexandre 1Komarova Svetlana 3Tremblay Michel L 4Li Wei 2Li Ailian 2Gao Yun Jing 2Henderson Janet E 21 Terry Fox Molecular Oncology Group and the Bloomfield Center for Research on Aging, Lady Davis Institute for Medical Research, Sir Mortimer B. Davis Jewish General Hospital, and Departments of Medicine and Oncology, McGill University, Montréal, Québec, Canada
2 Department of Medicine and Centre for Bone and Periodontal Research, McGill University, Montréal, Québec, Canada
3 Faculty of Dentistry, McGill University, Montréal, Québec, Canada
4 McGill Cancer Centre and Department of Biochemistry, McGill University, Montréal, Québec, Canada
Barsh Gregory EditorStanford University School of Medicine, United States of America* To whom correspondence should be addressed. E-mail: stephane.richard@ mcgill.ca¤a Current address: INRS-Institut Armand-Frappier, Laval, Canada
¤b Current address: Novartis Institutes for BioMedical Research, Cambridge, Massachusetts, United States of America
12 2005 16 12 2005 11 11 2005 1 6 e743 6 2005 11 11 2005 Copyright: © 2005 Richard 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 Src substrate associated in mitosis of 68 kDa (Sam68) is a KH-type RNA binding protein that has been shown to regulate several aspects of RNA metabolism; however, its physiologic role has remained elusive. Herein we report the generation of Sam68-null mice by homologous recombination. Aged Sam68−/− mice preserved their bone mass, in sharp contrast with 12-month-old wild-type littermates in which bone mass was decreased up to approximately 75%. In fact, the bone volume of the 12-month-old Sam68−/− mice was virtually indistinguishable from that of 4-month-old wild-type or Sam68−/− mice. Sam68−/− bone marrow stromal cells had a differentiation advantage for the osteogenic pathway. Moreover, the knockdown of Sam68 using short hairpin RNA in the embryonic mesenchymal multipotential progenitor C3H10T1/2 cells resulted in more pronounced expression of the mature osteoblast marker osteocalcin when differentiation was induced with bone morphogenetic protein-2. Cultures of mouse embryo fibroblasts generated from Sam68+/+ and Sam68−/− littermates were induced to differentiate into adipocytes with culture medium containing pioglitazone and the Sam68−/− mouse embryo fibroblasts shown to have impaired adipocyte differentiation. Furthermore, in vivo it was shown that sections of bone from 12-month-old Sam68−/− mice had few marrow adipocytes compared with their age-matched wild-type littermate controls, which exhibited fatty bone marrow. Our findings identify endogenous Sam68 as a positive regulator of adipocyte differentiation and a negative regulator of osteoblast differentiation, which is consistent with Sam68 being a modulator of bone marrow mesenchymal cell differentiation, and hence bone metabolism, in aged mice.
Synopsis
Osteoporosis is a debilitating bone disease that is characterized by reduced bone mass and microarchitectural damage, which result in increased bone fragility and susceptibility to fracture. Peak bone mass, which is achieved by the age of 30 in humans, has been identified as a major determinant of resistance or susceptibility to osteoporosis. The authors generated mice deficient for the Sam68 RNA binding protein, a protein of unknown physiologic function. The mice develop normally and are protected against bone loss during aging. Age-related bone loss has long been associated with an increase in marrow adipocytes, which are derived from the same mesenchymal lineage as osteoblasts in bone marrow. The authors showed that Sam68 regulates the differentiation of this mesenchymal lineage, such that in its absence, osteoblasts continued to be generated in aging bone, leading to preservation of bone mass. This study identifies a physiologic role for Sam68 as a modulator of the bone marrow stem cell niche and hence of bone metabolism. The data identify Sam68 as a potential therapeutic target for the prevention and treatment of age-related bone loss.
Citation:Richard S, Torabi N, Franco GV, Tremblay GA, Chen T, et al. (2005) Ablation of the Sam68 RNA binding protein protects mice from age-related bone loss. PLoS Genet 1(6): e74.
==== Body
Introduction
During skeletal development, the anabolic activity of osteoblasts [1] is favored over the catabolic activity of osteoclasts [2], which results in a net gain in bone mass. At skeletal maturity, bone mass is maintained through the balanced activity of osteoblasts and osteoclasts during the remodeling cycle. During skeletal aging, there is a shift in the balance that favors osteoclast over osteoblast activity, which results in net bone loss [3]. The amount and rate at which bone is gained during development and lost during aging are determined in large part by genetics [4–6] but also by physical activity and by alterations in the availability and response of bone cells to circulating hormones [7–9] and locally derived growth factors [10,11]. Whereas genetic-based studies have provided novel insights into the pathways that regulate bone development [12–14], relatively little is known about the etiology of age-related bone loss. Increased bone resorption in elderly men and women is associated with a reduction in bone mass and an increase in circulating levels of bone biomarkers [15]. These changes have been attributed primarily to nutritional deficits resulting in alterations in the parathyroid hormone–vitamin D axis [16], to gonadal hormone deficiency [17], to leptin levels and the sympathetic nervous system [8,18–20], and to alterations in bone cell apoptosis [21]. Bone loss in the elderly has also been attributed to alterations in the response of bone marrow stromal cells to their microenvironment that favors differentiation down the adipocyte lineage rather than the osteoblast lineage [22].
Aging has long been associated with an increase in marrow fat, where the generation of adipocytes is favored over osteoblasts [23]. Osteoblasts and adipocytes are derived from a common mesenchymal precursor cell present in bone marrow, and the factors that control this age-induced switch toward adipogenic differentiation is not well understood [24]. While several transcriptional regulatory proteins have been associated with cell fate determination of bone marrow mesenchymal cells, including peroxisome proliferator–activated receptor γ (PPARγ) and KLF5 [25–30], the role of RNA binding proteins in this process remains unknown. RNA binding proteins of the KH type are known regulators of cellular differentiation. For example, expression defects in the KH domain proteins NOVA and FMRP are known to cause paraneoplastic neurologic disorders [31] and the fragile X syndrome, respectively, in humans [32]. The phenotype of the quaking viable mice suggests a role for the QUAKING RNA binding protein in oligodendrocytes and myelination [33]. Indeed, the ectopic expression of the QKI-6/7 isoforms in vivo led to the formation of glial cells rather than neurons from neural progenitor, demonstrating its role in cell fate determination [34]. The loss of GLD-1 protein in Caenorhabditis elegans prevents the appearance of stem cells [35], and the absence of the KH domain protein HOW in Drosophila prevents muscle differentiation and results in flies with held-out-wings [36,37]. The Src substrate associated in mitosis of 68 kDa (Sam68) is also a member of the family of KH domain RNA binding proteins [38]; however, its physiologic role has remained undefined.
Sam68 was identified as an SH3 and SH2 domain interacting protein for Src family kinases and is also a known substrate of Src kinases [39–42] and of the breast tumor kinase [43]. Sam68 has been shown to facilitate the export of unspliced HIV RNA [44] and to regulate pre-mRNA processing [45]. In the present paper, we report the generation of Sam68−/− mice and analysis of their skeletal phenotype. Our data indicate that the absence of Sam68 confers resistance to age-related bone loss in mice such that old Sam68 have a higher bone mass than their wild-type littermates. We provide evidence that Sam68 regulates the differentiation of bone marrow stromal cells by showing that cells isolated from Sam68−/− animals had enhanced osteogenic activity and decreased adipogenic activity than those harvested from wild-type littermates. Furthermore, Sam68−/− mouse embryo fibroblasts (MEFs) were impaired in their capacity to differentiate into adipocytes, consistent with Sam68 being a regulator of bone marrow mesenchymal cell differentiation. These results also characterize a new animal model to study bone metabolism, regeneration, and repair during aging.
Results
Sam68 Is Expressed in the Developing Skeleton of Embryonic Mice
The Sam68 mRNA is known to be widely expressed [38], whereas its pattern of protein expression in vivo remains to be defined. Sections of paraffin-embedded wild-type E14.5 and E16.5 embryonic mice were immunostained with the well-characterized AD1 anti-Sam68 antibody, raised in rabbits against an immunizing peptide that corresponds to amino acids 330 to 348 of the mouse Sam68 protein [46]. Sam68 immunoreactivity was observed in the skeleton and soft tissues of developing wild-type mice at E14.5 and E16.5 (Figure 1). Intense staining was seen in the nucleus of cells in the developing brain, heart, and small intestine (Figure 1B), as well as in chondrocytes in the nasal septum and the glandular tissue adjacent to the nasal cartilage (Figure 1C, panels A–D), in the vertebra and intervertebral discs (panels E–H) and in the epiphysis (panels I–K) and metaphysis (panel L) of long bones. Nuclear staining was observed in proliferating chondrocytes and hypertrophic chondrocytes (Figure 1C, panel H), in osteoblasts (panel L; Dataset S1), and in osteoclasts (Dataset S1). No staining was observed with preimmune serum or when the antiserum was preadsorbed with the immunizing peptide (Figure 1C, panels B, F, and J). Taken together, these data demonstrate that Sam68 protein is selectively expressed in the developing mouse embryo, with particularly elevated expression in cartilage and bone.
Figure 1 Immunohistochemical Localization of Sam68 in Embryonic Mice
(A) Embryonic mice were removed from pregnant dams at E14.5 and E16.5, fixed in 4% paraformaldehyde, and embedded in paraffin. The entire embryo was immunostained with the AD1 anti-Sam68 antibody and counterstained with methyl green, and the image was captured at ×1.2 magnification.
(B) Embryonic soft tissues from the brain, heart and gut were stained with hematoxylin (left) and immunostained with anti-Sam68 antibody (right), and images were captured at ×20 magnification.
(C) Intense anti-Sam68 immunoreactivity was seen in chondrocytes in the nasal septum (panels A–D), in developing vertebra (panels E–H), and in the femoral epiphysis (panels I–K), as well as in diaphyseal osteoblasts (panel L). Adjacent sections were stained with hematoxylin and eosin (panels A, E, and I) or with antibody preadsorbed with the immunizing peptide (panels B, F, and J). Sam68 was localized primarily in the nucleus of cells in a variety of tissues but was also found occasionally in the cytoplasm. Magnification at source ×20, except for panels D, H, and L, which were ×40. Staining patterns are representative of three to five embryos.
Sam68 is Not Essential for Mouse Development
To define the physiologic role of Sam68, we generated Sam68-deficient mice by targeted disruption of exons 4 and 5 of the sam68 gene, which encode the functional region of the KH-type RNA binding domain (Figure 2A). The integrity of the targeted allele was verified by Southern blot analysis (Figure 2B) and by PCR of genomic DNA (unpublished data). Sam68 transcripts encoded by exons 1 to 5 were absent as evidenced by RT-PCR (Dataset S2), and Sam68−/− mice were devoid of Sam68 protein, as visualized by immunoblotting with anti-Sam68 AD1 and SC-333 antibodies or control normal rabbit serum or anti-Sam68 AD1 preabsorbed with peptide (Figure 2C). SC-333 is a rabbit anti-Sam68 antibody that was raised against the C-terminal 20 amino acids of Sam68 [47]. These data confirmed the generation of a mouse deficient in Sam68. The genotypes of offspring from heterozygote intercrosses exhibited a Mendelian segregation at E18.5 (Table 1). Despite the lack of visible deformity, many of the Sam68−/− pups died at birth of unknown causes. Sam68+/- mice were phenotypically normal and Sam68−/− pups that survived the perinatal period invariably lived to old age. Despite evidence that Sam68 mRNA is widely expressed and phosphorylated in mitosis [41,42], the Sam68−/− mice did not develop tumors and showed no immunologic or other major illnesses. Sam68−/− mice did, however, have difficulty breeding due to male infertility and the females rarely provided adequate care to their young.
Figure 2 Generation of Sam68-Deficient Mice
(A) The genomic organizations of the wild-type and targeted sam68 alleles after homologous recombination are depicted. The location of the DNA fragment used as a probe for the Southern blot analysis is shown, as well as the sizes of the two BglII fragments detected for wild-type and targeted sam68 alleles. The targeted allele replaces exon 4 and part of exon 5 of sam68 with a PGK-neomycin cassette.
(B) Southern-blot analysis of genomic DNA from wild-type (+/+), heterozygous (+/−), and homozygous (−/−) mice. DNA fragments corresponding to wild-type (4.5 kb) and the targeted (5.5 kb) alleles are illustrated.
(C) Western blot analysis of Sam68 expression. Protein extracts from wild-type, heterozygous, and homozygous cells subjected to immunoblot analyses using normal rabbit serum, anti-Sam68 AD1 antibody, the peptide antibody AD1 preabsorbed with the immunogenic peptide corresponding to amino acids 330–348 of mouse Sam68, anti-Sam68 Sc333 antibody that recognizes the C-terminal 20 amino acids of Sam68, and anti-actin antibodies as loading control. The migration of the molecular mass markers is known on the left in kDa.
Table 1 Progeny of Sam68 Heterozygote Breeding
Sam68 Deficiency Protects Mice from Age-Related Bone Loss
Src null animals are known to have bone metabolism defects [48], and since Sam68 is a substrate of Src, we decided to analyze the Sam68 mice for skeletal abnormalities. Cohorts of Sam68+/+ and Sam68−/− mice were euthanized at 4 and 12 months of age for skeletal phenotyping. To minimize differences in the bone phenotype that might arise secondary to gender or weight differences, we selected age-matched female mice for these analyses. The female mice demonstrated similar increases in body weight, although 4-month-old Sam68−/− mice weighed less than Sam68+/+ mice, and similar changes in bone lengths in the axial and appendicular skeleton between 4 and 12 months of age (Table 2). Faxitron radiography (Faxitron X-ray Corporation, Wheeling, Illinois, United States) (Figure 3A) and micro–computed tomography (CT) (Figure 3B) revealed significant cortical thinning (arrow) and a reduction in metaphyseal bone (asterisk) in the distal femora of 12-month-old Sam68+/+ mice compared with 4-month-old mice of either genotype and 12-month-old Sam68−/− mice. Similar reductions in trabecular bone were shown by Faxitron and micro-CT in the fifth lumbar vertebrae of the 12-month-old Sam68+/+ mice but not in the Sam68−/− mice (unpublished data). Total body bone mineral content (BMC), quantified with a Lunar PIXImus mouse densitometer (GE-Lunar, Madison, Wisconsin, United States), increased in both Sam68+/+ and Sam68−/− mice between 4 months and 12 months of age, but only reached significance in the Sam68−/− mice (Table 2). Similarly, a greater increase in BMC was seen in the femur and vertebra in the Sam68−/− mice. Bone mineral density (BMD) remained constant or decreased in the wild type mice but increased significantly in the total body and in the femoral and vertebral regions of interest in the Sam68−/− (Table 2). These data showed that Sam68−/− mice continued to thrive and accrue bone in the axial and appendicular skeleton for longer than 12 months. This was in contrast to the situation in age-matched, littermate controls in which a significant amount of bone was lost over the same timeframe.
Table 2 Morphology and Bone Mineral Density in 4- and 12-Month-Old Female Mice
Figure 3 Radiologic Assessment of the Femur of Young and Old Sam68+/+ and Sam68−/− Mice
(A) Mice were given a lethal dose of anesthetic at the indicated times, and contact radiographs of the distal femora were obtained on a Faxitron MX20 equipped with an FPX-2 Imaging system. Representative radiographs of the distal femur of Sam68+/+ (+/+) and Sam68−/− (−/−) mice revealed comparable radiopacity at 4 months (left). At 12 months (right), cortical thinning (arrow) and radiolucency (asterisk) were apparent in the distal femur of +/+ mice but not −/− mice.
(B) Bones were dissected free of soft tissue and fixed overnight in 4% paraformaldehyde before scanning on a Skyscan 1072 static instrument equipped with 3D Creator analytical software. Representative three-dimensional re-constructions and two-dimensional cross-sectional scans demonstrated similar architecture in the distal femur of Sam68+/+ (+/+) and Sam68−/− (−/−) mice. In keeping with the results from Faxitron x-ray, trabecular bone (asterisk) and cortical thickness (arrow) were reduced in the femur of 12-month-old +/+ mice compared with all other groups. The images are representative of those from five to seven animals in each group.
Three-Dimensional Architecture of Bone Is Preserved in Aged Sam68−/− Mice
Bone loss and compromised architecture are characteristic features of the skeletons of aged C57BL/6 mice and resemble the clinical features of age-related bone loss in humans that can predispose an individual to fracture. Quantification of the micro-CT data shown in Figure 3B confirmed the reduction in bone volume compared with tissue volume (BV/TV; Figure 4A, left) in the 12-month-old Sam68+/+ mice (hatched bars) compared with 4-month-old mice of both genotypes (solid bars) and 12-month-old Sam68−/− mice (stippled bars). This was associated with a significant increase (p < 0.01; denoted by the asterisks) in the structure model index, which measures the ratio of plate-like to rod-like structures (Figure 4A, right). A quantifiable increase in the mean trabecular separation (Tb.Sp) (unpublished data) in 12-month-old Sam68+/+ mice was due to an increase in the percentage of spaces falling in the range of 350 to 700 μm (Figure 4B, red hatched line). Trabecular thickness remained constant among the different groups of mice (unpublished data). In effect, this meant that there were fewer trabeculae, rather than equivalent numbers of thin trabeculae, in the 12-month-old Sam68+/+ mice compared with any of the other groups of mice.
Figure 4 Quantitative Micro-CT of Trabecular Bone Composition and Architecture
(A) Bone volume/tissue volume (BV/TV) and structure model index (SMI) were calculated on the femur and fourth lumbar vertebra of six or seven mice in each group using 3D Creator software supplied with the Skyscan instrument. Results expressed as the mean ± SD showed significant differences (p < 0.01) between 4-month-old Sam68+/+ mice (solid black) and 12-month-old Sam68+/+ mice (hatched black) but not between 4-month-old Sam68−/− mice (solid white) and 12-month-old Sam68−/− mice (stippled white).
(B) The distance between trabeculae was reflected in a shift to the right of the distribution curves for 12-month-old Sam68+/+. Solid black = 4-month-old Sam68+/+; hatched black = 12-month-old Sam68+/+; solid red = 4-month-old Sam68−/−; stippled red = 12-month-old Sam68−/−. The asterisks denote p < 0.01.
Bone Remodeling Is Preserved in Aged Sam68−/− Mice
To further define the mechanisms involved in the preservation of bone mass in aged Sam68−/− mice, we prepared sections from plastic-embedded femora and tibia to evaluate osteoblast and osteoclast activity (Figure 5). In situ enzyme histochemical staining for alkaline phosphatase (ALP; brown stain) activity was used as a biomarker for osteoblasts and tartrate-resistant ALP (tartrate-resistant acid phosphatase [TRAP]; red stain) activity as a marker for osteoclasts. Little difference was seen between Sam68+/+ (Figure 5A and 5B) and Sam68−/− (Figure 5C and 5D) mice at 4 months of age. ALP- and TRAP-positive cells were reduced in the 12-month-old Sam68+/+ mice (Figure 5E and 5F) and remained unchanged in 12-month-old Sam68−/− mice (Figure 5G and 5H). The reduction in both osteoblast and osteoclast activity in the 12-month-old Sam68+/+ mice argued against bone being lost primarily due to a relative increase in osteoclast over osteoblast activity, as seen in high turnover disease [49].
Figure 5 Histologic Analysis of Undecalcified Bone from Sam68+/+ and Sam68−/− Mice
Sections of tibia fixed in 4% paraformaldehyde and embedded in plastic were stained for ALP (A–C, E–G) activity to identify osteoblasts or for TRAP (B–D, F–H) activity to identify osteoclasts. Staining patterns were similar in 4-month-old Sam68+/+ (A and B), 4-month-old Sam68−/− (C and D), and 12-month-old Sam68−/− (G and H) mice compared with 12-month-old Sam68+/+ mice (E and F). Magnification at source, left panels ×10 and right panels ×40. Micrographs are representative of those taken from five to seven sections in each group of animals.
Histomorphometric analyses [50] of the long bones (Table 3) corroborated the radiologic evidence of age-related bone loss. Bone volume (BV/TV), newly formed osteoid (OV/TV), and mineral apposition rate (MAR) were all significantly reduced in 12-month-old Sam68+/+ mice compared with 4-month-old Sam68+/+ mice. These reductions were associated with a significant increase in marrow fat (FV/TV) and decreased numbers of osteoblasts (nOB/TV) and osteoclasts (nOC/TV) per tissue volume (Table 3). These results were in contrast to those of the 12-month-old Sam68−/− mice in which all histomorphometric parameters, including marrow fat, resembled those of young mice of either genotype (Table 3). When cells were expressed as a function of the bone perimeter (nOB/BP, nOC/BP), there were no statistical differences and the ratio of osteoblasts to osteoclasts was similar in all groups of mice (Table 3).
Table 3 Histomorphometric Analysis of Long Bones from 4- and 12-Month-Old Mice
Circulating levels of serum C-telopeptide (sCTX; Roche Diagnostics, Mannheim, Germany) and ALP showed little difference among the groups, except for a small decrease in 12-month-old Sam68−/− mice and (Dataset S3). Serum estrogen was significantly decreased in all 12-month-old mice regardless of genotype with a reciprocal increase in interleukin-6 levels (Dataset S3). Given the involvement of leptin in regulating body and bone mass, serum leptin was measured in Sam68+/+ and Sam68−/− mice. Twelve-month-old Sam68−/− mice had significantly lower levels than young and old Sam68+/+ mice and young Sam68−/− mice (Dataset S3). Taken together, these observations suggested that preservation of bone mass in the Sam68−/− mice was not due primarily to altered estrogen status but could have been influenced by differences in circulating leptin levels.
Osteoblast, but Not Osteoclast, Activity Is Altered in Sam68−/− Mice Ex Vivo
Maintenance of bone mass during the adult remodeling cycle is dependent on the coupling of osteoblast to osteoclast activity, such that there is no net gain or loss of bone [49]. To further explore the age-related advantage of Sam68−/− mice with respect to bone preservation, we examined the functional activity of Sam68−/− osteoblasts and osteoclasts ex vivo (Figure 6A and 6B). Cultures of bone marrow stromal cells harvested from 4-week-old Sam68−/− mice and maintained for 18 days in osteoblast differentiation medium demonstrated more intense staining for ALP at 6 days (unpublished data) and 18 days (Figure 6A), and more mineralized nodules at 18 days, than the Sam68+/+ mice, even though similar amounts of fibroblast colony-forming units (CFU-F) were observed (Figure 6A and unpublished data). RT-PCR analysis of molecular markers of osteoblast differentiation (Dataset S2) revealed similar increases over time in RNA from Sam68+/+ and Sam68−/− mice, although type I collagen appeared to be up-regulated at both timepoints in the Sam68−/− mice. Short-term cultures of mature osteoclasts released from the crushed long bones of Sam68+/+ and Sam68−/− mice stained equally well for TRAP and excavated approximately equal numbers of pits of equal size in dentin slices, as visualized by scanning electron microscopy (Figure 6B). These findings suggest that the osteoclasts may not be the primary defect in Sam68−/− mice, as they are in Src null mice [48].
Figure 6 Ex Vivo Activity of Sam68+/+ and Sam68−/− Osteoblasts and Osteoclasts
Marrow stromal cells were isolated from the long bones of juvenile mice and maintained under conditions that promote osteoblast differentiation.
(A) Cultures were fixed in 4% paraformaldehyde after 6 or 18 days and stained in situ for ALP activity and with silver nitrate (von Kossa) to detect mineralized nodules. Sam68−/− cultures stained more intensely for ALP at early and late time points and produced significantly more mineralized nodules after 18 days. Asterisks represent p < 0.01.
(B) Primary osteoclasts were isolated from the crushed long bones of the same mice and plated on glass coverslips or on dentin slices to quantify numbers and activity, respectively. Osteoclasts were identified as cells with three or more nuclei that stained positive for TRAP activity (upper) and excavated pits in dentin slices, as demonstrated by SEM (lower, bar = 20 μm). No statistical differences were observed either in the number of TRAP-positive cells or in their resorptive activity.
The Absence of Sam68 Prevents Adipocyte Differentiation and Promotes Osteoblast Differentiation
It is well recognized that bone marrow stromal cells give rise to both osteoblasts and adipocytes and that age-related bone loss is accompanied by an increase in differentiation down the adipocyte lineage [22]. Therefore, the loss of Sam68 could influence the bone marrow stromal cells to differentiate along the osteogenic versus the adipogenic pathway. Alternatively, the loss of Sam68 could indirectly regulate bone mass as a general disturbance of neuroendocrine control as was shown when leptin and the sympathetic nervous system axis were shown to negatively regulate bone mass [8]. To confirm a role for Sam68 in the regulation of adipocyte differentiation, we isolated primary MEFs from 14.5-day-old Sam68+/+ and Sam68−/− embryos. The primary Sam68−/− MEFs were differentiated into adipocytes in vitro in culture medium containing 5 μM pioglitazone to induce adipogenesis. Cells at days 0, 4, 6, and 12 were stained with Oil red O to monitor adipogenesis. Adipogenesis was more pronounced in the wild-type MEFs cultures than in the Sam68−/− MEFs, consistent with the positive role of endogenous Sam68 in adipocyte differentiation (Figure 7). The expression of key transcription factors including the PPARγ and KLF5 was impaired in Sam68−/− differentiated MEFs compared with Sam68+/+ MEFs, consistent with impaired adipogenesis in the absence of Sam68 (Figure 7). These data, together with data confirming a lean phenotype in Sam68−/− mice (N. Torabi and S. Richard, unpublished data), support the hypothesis that Sam68 modulates the differentiation of mesenchymal cells.
Figure 7 Ex Vivo Adipogenesis Analysis of Sam68−/− Mouse Embryonic Fibroblasts
MEFs were isolated from mouse embryos at embryonic day 14.5. Equal number of MEFs from Sam68+/+ and Sam68−/− was plated on glass cover slips in 24 well-plates. Adipocyte differentiation was carried out at indicated times by the addition of complete media containing the pioglitazone.
(A) Cultures were fixed in 4% paraformaldehyde and stained with Oil Red O to detect the fat droplets stored in adipocytes and photographed (top). The cell images were magnified ×10 and ×20 as indicated.
(B) RT-PCR was carried out on total cellular RNA isolated after differentiation of the MEFs for day 0, 2, 4, 6, and 12. The DNA fragments were visualized on agarose gels stained with ethidium bromide. The expression of adipogenic markers C/EBPβ, C/EBPδ, PPARα, and KLF5 was examined as well as the expression of controls including Sam68, β-actin, and GAPDH.
To further examine this phenotype in a cell autonomous system, we chose the embryonic mesenchymal multipotential progenitor cells C3H10T1/2. The addition of BMP-2 induced osteoblast differentiation, as evidenced by an approximately 400-fold increase in expression of the osteocalcin (OCN) gene [51]. Populations of C3H10T1/2 cells stably transfected with either pSuper-retro (control) and pSuper-retro harboring a short hairpin against Sam68 (Sam68sh) were selected with puromycin. The expression of Sam68 was reduced by approximately 80% as evidenced by immunoblot analyses using β-actin as a loading control (Figure 8A). Osteoblast differentiation was induced in cultures expressing pSuper-retro or pSuper-retro Sam68sh by addition of BMP-2 to the culture medium. . The expression of OCN and β-actin mRNAs was examined by semiquantitative RT-PCR and the Sam68sh-expressing cells displayed a more pronounced osteoblast phenotype compared with control cells, as assessed by the expression of OCN (Figure 8B).
Figure 8 Enhanced Osteogenic Differentiation of the C3HT101/2 Embryonic Cell Line Depleted of Endogenous Sam68
(A) C3HT101/2 cells transfected with an empty vector (pSuper-retro) or a vector containing an shRNA (Sam68 shRNA) were selected with puromycin, and knockdown populations depleted of Sam68 were identified. The reduction in Sam68 protein was analyzed by immunoblotting with anti-Sam68 (AD1) antibody and anti–β-actin antibodies as loading controls.
(B) Osteogenic differentiation was carried out with conditioned medium containing BMP-2 for the indicated times. To assess the level of osteogenic differentiation in these cells, expression of late osteoblast marker, osteocalcin (OCN), was analyzed by RT-PCR and compared with β-actin and GAPDH controls. The DNA fragments were visualized by agarose gel stained with ethidium bromide.
Given the apparent enhancement of mineralized nodule formation by Sam68−/− bone marrow stromal cells ex vivo and the phenotype observed with short hairpin RNA (shRNA)-treated C3H10T1/2, we stained sections of bone from 4- and 12-month-old mice for evidence of changes in marrow adiposity. Figure 9 shows sections of undecalcified bone from 4- and 12-month-old Sam68+/+ and Sam68−/− mice stained with von Kossa and toluidine blue to show mineralized tissue (Figure 9A, black) and marrow adipocytes (Figure 9B, white) and left unstained to show fluorochrome labeling of the mineralization fronts (Figure 9C). Twelve-month-old Sam68+/+ mice showed a noticeable decrease in trabecular bone (Figure 9A), which was associated with a significant increase in marrow adiposity (Figure 9B) and with the two fluorochrome labels superimposed upon one another (Figure 9C). In contrast, the bones of 12-month-old Sam68−/− mice appeared similar to those of the 4-month-old mice of either genotype. These data demonstrate that Sam68 regulates the differentiation of bone marrow mesenchymal cells to promote adipocyte differentiation and inhibit osteoblast differentiation in aging bone.
Figure 9 Old Sam68−/− Mice Are Protected from the Development of Fatty Bone Marrow
Sections of undecalcified bone were stained with von Kossa and toluidine blue and images captured at original magnifications of ×2 (A), ×40 (B and C) to evaluate mineralized tissue (A, black), marrow adipocytes (B, white), and the mineralization fronts (C, yellow and green). The 12-month-old Sam68+/+ bone demonstrated a significant reduction in bone (A) and increase in marrow adipocytes (B) and a decrease in the distance between two consecutive fluorochrome labels (C). Magnification at source was ×40. Micrographs are representative of four to six screened in each group of animals.
Discussion
The present study provides evidence that a physiologic role of Sam68 is to modulate bone marrow mesenchymal stem cells. Young Sam68−/− mice developed normally and contained similar bone mass compared with wild-type littermates. Aged (12 months) Sam68−/− mice displayed a high bone mass phenotype compared with Sam68+/+ littermates. The wild-type littermates underwent age-related bone loss that occurs naturally in mammals, while the Sam68−/− animals preserved their bone mass with aging. The differentiation of bone marrow stem cells isolated from Sam68−/− mice and embryonic mesenchymal multipotential progenitor cells C3H10T1/2 treated with Sam68 shRNA resulted in a more pronounced osteoblast differentiation. These findings demonstrate that the loss of Sam68 enhances osteoblast differentiation. The converse was also true, as MEFs isolated from Sam68−/− animals were impaired in their ability to undergo adipocyte differentiation compared with Sam68+/+ MEFs. These findings suggest that a physiologic role for Sam68 is to regulate the balance between adipogenic and osteogenic differentiation of the bone marrow mesenchymal.
Sam68 protein expression was observed throughout the developing mouse embryo, in keeping with previous reports that identified the Sam68 mRNA to be widely expressed [38]. We observe Sam68 staining in the brain, heart, and intestine (see Figure 1) as well as liver, skin, and kidney (unpublished data) of late-stage embryos. This expression pattern is consistent with previous work that has observed Sam68 in neurons [52] and as a substrate of the intestinal SIK/breast tumor kinase [43]. The expression of Sam68 was not altered in aging bone marrow stromal cells or in senescencing WI-38 cells (unpublished data). The presence of Sam68 in chondrocytes, osteoblasts and osteoclasts in developing cartilage and bone predicted a pivotal role for the protein in skeletal development.
The Sam68−/− mice were generated using a traditional targeting approach where the functional KH domain was deleted and approximately one third of the Sam68−/− mice survived to adulthood with no apparent defects. One explanation for this phenomenon is that Sam68 function is sub-served by one or more of its family members during development such as SLM-1 and SLM-2 [47]. Alternatively, Sam68 is not required during embryonic development. However, two thirds of the Sam68−/−, but not the Sam68+/- pups, were killed by their Sam68+/- mothers. These findings suggest that the mothers are able to detect a subtle defect/difference that we cannot. Adult Sam68−/− mice live a normal life span of approximately two years. The Sam68−/− males were sterile and the Sam68−/− females provided inadequate care to their young, but the pups were not scattered and neglected as observed with FosB−/− mice [53]. Serum levels of estrogen decreased in aging Sam68−/− females as expected; however, the leptin levels decreased in aged Sam68−/− females. The aged Sam68−/− females were not obese and actually weighed less than the littermate controls (see Table 2). Moreover, their appetite was not altered with aging (N. Torabi and S. Richard, unpublished data), suggesting that the observed leptin reduction was not recreating ob/ob-like phenotypes related to weight, appetite and female sterility [54].
The skeletal phenotyping of cohorts of Sam68+/+ and Sam68−/− mice showed that bone mass was preserved in aged Sam68−/− mice. Traditional histology and histomorphometry suggested that the mechanism involved preservation of osteoblast and osteoclast activity. The documented role of the Sam68 regulatory protein, Src, in osteopetrosis led us to investigate the morphology and activity of Sam68−/− osteoclasts ex vivo. The Src tyrosine kinase was shown to play a role in bone remodeling when Src−/− mice died at 6 months of age with an osteopetrotic phenotype [48] and the defect was attributed to defective osteoclast function [55−59]. We therefore cultured mature osteoclasts harvested from Sam68+/+ and Sam68−/− mice ex vivo on dentin slices to quantify their resorptive capacity. The fact that Sam68−/− osteoclasts looked and acted like Sam68+/+ osteoclasts ex vivo and in vivo made it unlikely that this was the primary source of the difference in bone metabolism in 12-month-old Sam68−/− mice. The fact that the CTX levels were lower in young and old Sam68−/− mice suggested that there is reduced bone resorption compared with wild-type littermate controls. However, this reduction in bone resorption occurred with normal osteoclast activity, as assessed by in vitro culturing. These observations are consistent with the Sam68−/− mice having a youth-like bone phenotype. However, it is still possible, that a mild impairment in Sam68−/− mice osteoclast function may manifest itself later in life in overall accumulation of bone and this will require further detailed studies.
Ex vivo differentiation of primary bone marrow stromal cells, harvested from Sam68+/+ and Sam68−/− mice, down the osteoblast lineage revealed an osteogenic advantage in the cultures of cells derived from the Sam68−/− mice. It will be important to demonstrate that a similar effect is observed using primary bone marrow stromal cells from aged Sam68−/− mice. Similar findings were observed when embryonic mesenchymal multipotential progenitor cells C3H10T1/2 treated with Sam68 shRNA, were differentiated into osteoblasts with BMP2. Our findings that aged Sam68−/− mice do not develop fatty bone marrow and that MEFs derived from Sam68 −/− mice have impaired adipocyte differentiation, indicate that Sam68 regulates both adipocyte and osteoblast differentiation. These findings identify Sam68 as the first RNA binding protein to regulate mesenchymal cell differentiation and the challenge will be to identify the specific RNA targets that it regulates during this process. The fact that osteoblast function is altered in Src−/− mice [60,61] raises the possibility that preservation of bone mass in the Sam68−/− mice could be linked with and regulated by Src.
The pathway by which leptin regulates bone resorption was identified to involve the sympathetic nervous system relaying to the osteoblasts via the β-adrenergic pathway leading to the release of growth factors including RANKL that causes the osteoclasts to thrive [8,18–20]. The lowering of leptin levels in aged Sam68−/− mice is consistent with these mice having a high bone mass compared with their aged littermates. These data would suggest that the leptin-sympathetic pathway is unaltered in Sam68−/− mice and that the lowering of leptin may explain the lower levels of CTX in the serum of Sam68−/− mice. These findings suggest that Sam68 may be regulating bone metabolism at two different levels: (1) the absence of Sam68 results in lower leptin levels that may reduce bone resorption via the sympathetic nervous system and (2) the absence of Sam68 favors osteoblast, rather than adipocyte, differentiation.
In conclusion, our data define a physiologic role for Sam68 in bone metabolism and bone marrow mesenchymal stem cell differentiation. The bone phenotype observed in Sam68−/− mice imply that inhibitors of Sam68 could prevent age-related bone loss. Furthermore, the results also suggest that Sam68 expression levels, hypomorphism, and mutations in humans may influence susceptibility to marrow adipocyte accumulation and osteoporosis. Our findings also identify a new animal model to study aging bone loss.
Materials and Methods
Histologic, immunohistochemical, and histomorphometric analyses.
All analyses were performed essentially as described previously [62,63]. Briefly, embryonic mice were removed from timed pregnant dams at E14.5 and E16.5 and fixed intact for 36 h in 4% paraformaldehyde, rinsed thoroughly in PBS, and processed for paraffin embedding. Serial 4-μm sections were cut on a modified Leica RM 2155 rotary microtome (Leica Microsystems, Richmond Hill, Ontario, Canada), stained with a 1:600 dilution of the AD1 anti-Sam68 antibody [46] and counterstained with either methyl green or hematoxylin.
Adult mice were given an intraperitoneal injection of 30 mg/kg calcein at 7 days and 30 mg/kg tetracycline at 2 days prior to sacrifice to label actively mineralizing surfaces. After overnight fixation in 4% paraformaldehyde and rinsing in PBS, the left femur and tibia were embedded in polymethylmethacrylate (MMA) or a mixture of 50% MMA and 50% glycolmethacrylate (GMA). Serial 4- to 6-μm sections of MMA-embedded tissues were left unstained or stained with von Kossa and toluidine blue or with toluidine blue alone, while 4-μm MMA-GMA sections were stained for TRAP and ALP activity. Images were captured using a Leica DMR microscope (Leica Microsystems) equipped with a Retiga 1300 camera (Qimaging, Burnaby, British Columbia, Canada) and the primary histomorphometric data obtained using Bioquant Nova Prime image analysis software (Bioquant Image Analysis Corp, Nashville, Tennessee, United States). Nomenclature and abbreviations conform to those recommended by the American Society for Bone and Mineral Research [50].
Generation of mice with targeted disruption of the sam68 gene.
A λ bacteriophage clone encompassing Sam68 exons 3 to 9 was isolated from a 129/SvJ genomic library using full-length Sam68 cDNA as a probe. XbaI-digested Sam68 genomic DNA fragments of 4 kb (encompassing exon 4 and part of exon 5) and 3kb (spanning part of exon 5 and exon 6) were subcloned in Bluescript SK resulting in pBS4 and pBS3, respectively. A DNA fragment was amplified from pBS4 with the following oligonucleotides (5′-AAT GTC TAG AAA CAA CTC ATA TAC AGA C-3′) and the universal primer. The XbaI-digested 1-kb DNA fragment was subcloned in the XbaI site of pPNT (a gift from Andrew Karaplis, McGill University, Montréal, Quebec, Canada). The 3-kb fragment from pBS3 was amplified by PCR with (5′-GGG ATG CGG CCG CTC TAG AAT TGT CCT ACT TGA ACG G-3') and (5′-CGG TGG CGG CCG CTG TCG ACC TGA GTA ACA TTT CTT A-3′) and subcloned in the NotI site of pPNT. The targeting vector pPNT-Sam68 replaces exon 4 and part of exon 5 with a neomycin-resistant gene cassette. An SalI site was introduced at the 3′ end of the 3-kb DNA fragment and was used to linearize the plasmid for electroporation into embryonic stem (ES) cells. Approximately 1,000 ES colonies were screened and two clones were identified that contained the Sam68 mutant allele, as determined by Southern blotting. Targeted ES cells were injected into 3.5-day-old BALB/c blastocysts and were transferred into CD-1 foster mothers, and animals classified as chimeras by coat color were mated with BALB/c mice. Germ line transmission was achieved and the mice were maintained in C57BL/6 background. The mice used for this study represent mice that were backcrossed in C57BL/6 between three and eight generations; in addition, we maintained the mice in the 129/SvJ strain and observed a similar phenotype.
Genotyping and immunoblot analyses.
All mouse procedures were performed in accordance with McGill University guidelines, which are set by the Canadian Council on Animal Care. Genomic DNA was isolated from tail biopsies and analyzed by Southern blotting and genomic PCR analysis. The DNA fragment utilized as the probe for the Southern blotting analysis was amplified with the following two oligonucleotides (5′-AAG CCT TTA CTG GTT GTG T-3′) and (5′-CTT GAA ACG CAC CGT AGG CT-3′). The wild-type sam68 allele was identified by genomic PCR using the following oligonucleotides 5′-AAA TCC TAA CCC TCC TCA GTC AG-3′ and 5′-GAT ATG ATG GAT GAT ATC TGT CAG-3′. The sam68-targeted allele was identified by genomic PCR using the following oligonucleotides 5′-CTT GGG TGG AGA GGC TAT TCG-3′ and 5′-GTC GGG CAT GCG CGC CTT GAG C-3′.
Radiology and serum biochemistry on Sam68+/+ and Sam68−/− mice.
Radiography, BMD, and micro-CT were performed essentially as described previously [63]. Mice were administered a lethal dose of anesthetic at the indicated times, exsanguinated, and imaged using a Faxitron MX20 equipped with an FPX-2 Imaging system (Dalsa Medoptics, Waterloo, Ontario, Canada). Body fat, BMC, and BMD were evaluated using a Lunar PixiMUS 1.46 (GE-Lunar, Madison, Wisconsin, United States). Morphometric parameters were determined on anesthetized mice at the time of sacrifice by direct measurement or from the Faxitron radiograph.
Micro-CT was performed on the left femur and fourth lumbar vertebra after removal of soft tissues and overnight fixation in 4% paraformaldehyde. The distal metaphysis was scanned with a Skyscan 1072 micro-CT instrument (Skyscan, Antwerp, Belgium). Image acquisition was performed at 100 kV and 98 μA, with a 0.9° rotation between frames. The two-dimensional images were used to generate three-dimensional reconstructions to obtain quantitative data with the 3D Creator software supplied with the instrument.
Serum biochemistry (ALP, Ca, PO4, Mg) was determined at the Rodent Diagnostics Lab (McGill University, Montréal, Quebec, Canada) using routine automated techniques. Commercial assays were used to determine serum levels of CTX (RatLaps ELISA, Nordic Bioscience), 17β estradiol (IBL Immuno-Biological, Hamburg, Germany), leptin (R & D Systems, Minneapolis, Minnesota, United States), and interleukin-6 (R & D Systems).
Ex vivo assessment of osteoblast and osteoclast activity.
Bone marrow was flushed from the tibia and femora of juvenile 4-week-old Sam68+/+ and Sam68−/− mice to obtain stromal cells that were maintained in differentiation medium for 6 or 18 days as described [63]. Cultures were stained in situ for ALP activity and with von Kossa stain to detect mineralized nodules. Total RNA was harvested from parallel cultures for RT-PCR analysis of osteoblast-related gene expression over time as described previously [63].
Osteoclasts were isolated from the crushed long bones of juvenile Sam68+/+ and Sam68−/− mice and used for quantitative studies as described [64]. Briefly, cells were plated on glass coverslips or on dentin slices in 24-well cluster plates for assessment of cell number and pit number, respectively. Cover slips were immersed in 4% paraformaldehyde and stained for TRAP activity after 2 h. Dentin slices were left for 28 h before removing the cells with 1 M ammonium hydroxide and air drying before coating with sputter Au-Pd and examination with scanning electron microscopy (McGill Electron Microscopy Centre). Light microscope images were captured using a Leica DMR microscope equipped with a Retiga 1300 camera. Quantitative analyses were performed using Adobe Photoshop and the data presented as the mean ± SD of two or three independent experiments. Statistical comparisons were made using the Student's t test.
Preparation of mouse embryonic fibroblast and induction of adipocyte differentiation.
MEFs were prepared from 14.5-day-old embryos. Only early-passage MEFs were used for the experimental studies, and induction of adipocyte differentiation was carried out according to the methods previously described [30]. Cultured cells were stained with Oil Red O as described [29].
RNA preparation and RT-PCR to assess adipogenesis in Sam68−/− MEFs.
Total cellular RNA was prepared by TRIzol reagent according to the manufacturer's protocol (Invitrogen, Carlsbad, California, United States). Total RNA (1 μg) was reverse-transcribed, and cDNA samples were subjected to PCR. RT-PCR was normalized by the transcriptional level of GAPDH. The following 5′ and 3′ primers were used to evaluate adipogenic differentiation: KLF5: 5'-AGA CAA GCT GAG ATG CTG C-3′, 5′-GGC AAA CCT CCA GTC GC-3′; PPARγ: 5′-GTG CGA TCA AAG TAG AAC CTG C-3′, 5′-CCT ATC ATA AAT AAG CTT CAA TCG-3′; β-Actin: 5′-TAG GCG GAC TGT TAC TGA GC-3′, 5′-AGC CTT CAT ACA TCA AGT TGG-3′; and Sam68: 5′- GTG GAG ACC CCA AAT ATG CCC A-3′ and 5′-AAA CTG CTC CTG ACA GAT ATC A-3′.
Sam68 down-regulation using Sam68sh.
To generate an shRNA, 64-base duplex DNA nucleotides were purchased from Invitrogen. This fragment comprised 19 base residues specific to mouse Sam68 (GATCCCC AAGATGACGAGGAGAATTATTCAAGAGA TAATTCTCCTCGTCATCTT TTTTTGGAAA). This fragment was cloned into pSUPER retro (OligoEngine, Seattle, Washington, United States). Transfections were performed using LipofectAMINE Plus (Invitrogen) with cloned DNA fragment or empty vector. At 48 h after transfection, populations were selected for 2.5 μg/ml puromycin and the knockdown observed by immunoblotting with anti-Sam68 (AD1) antibody.
Osteogenic differentiation analysis of C3HT101/2 cells.
For osteogenic differentiation, C3HT101/2 (ATCC) cells were incubated in 48-well tissue culture plates at 1.25 × 104 cells/cm2 . After 24-h incubation, the culture media were changed to fresh media containing 300 ng/ml BMP-2 (Sigma, St. Louis, Missouri, United States). Total cellular RNA was prepared as described above, and osteocalcin, β-actin, and GAPDH gene expression at indicated times after BMP-2 treatment was quantified by RT-PCR.
Supporting Information
Dataset S1 Localization of Sam68 in Primary Mouse Osteoblasts and Osteoclasts
(124 KB PPT)
Click here for additional data file.
Dataset S2 Gene Expression Profile of Cultured Stromal Cells at Days 6 and 18 of Culture Isolated from Sam68+/+ and Sam68−/− Mice
(204 KB PPT)
Click here for additional data file.
Dataset S3 Serum Biochemistry and Bone Biomarkers in 4- and 12-Month-Old Mice
(46 KB DOC)
Click here for additional data file.
We acknowledge Jean-Sebastien Binette (Centre for Bone and Periodontal Research) and Jacinthe Sirois (Québec Transgenic Research Network) for excellent technical assistance. We are grateful to Erique Lukong, Daniel Larocque, and Mark Bedford for helpful discussions. This work was supported by grant MT13377 to SR and grant MOP-13419 to JEH from the Canadian Institutes of Health Research (CIHR). SR holds an Investigator Award from the CIHR and JEH is a Chercheur Boursier, Senior of the FRSQ.
Competing interests. The authors have declared that no competing interests exist.
Author contributions. SR and JEH conceived and designed the experiments. SR, NT, GVF, GAT, TC, GV, MM, PC, AFR, SK, WL, AL, and YJG performed the experiments. SR and JEH analyzed the data. MLT generated the mice. SR and JEH wrote the paper.
A previous version of this article appeared as an Early Online Release on November 11, 2005 (DOI: 10.1371/journal.pgen.0010074.eor).
Abbreviations
ALPalkaline phosphatase
BMCbone mineral content
BMDbone mineral density
BMP-2bone morphogenetic protein-2
CTcomputed tomography
CTXC-telopeptide
MEFmouse embryo fibroblast
PPARγperoxisome proliferator–activated receptor γ
shRNAshort hairpin RNA
TRAPtartrate-resistant acid phosphatase
==== Refs
References
Harada S Rodan GA 2003 Control of osteoblast function and regulation of bone mass Nature 423 349 355 12748654
Teitelbaum SL 2000 Bone resorption by osteoclasts Science 289 1504 1508 10968780
Hannan MT Felson DT Dawson-Hughes B Tucker KL Cupples LA 2000 Risk factors for longitudinal bone loss in elderly men and women J Bone Miner Res 15 710 720 10780863
Ralston SH 2003 Genetic determinants of susceptibility to osteoporosis Curr Opin Pharmacol 3 286 290 12810194
Baldock PA Eisman JA 2004 Genetic determinants of bone mass Curr Opin Rheumatol 16 450 456 15201610
Huang QY Recker RR Deng HW 2003 Searching for osteoporosis genes in the post genome era: Progress and challenges Osteoporosis Int 14 701 715
Riggs BL Khosla S Melton LJ 2002 Sex steroids and the construction and conservation of the adult skeleton Endocrine Rev 23 279 302 12050121
Ducy P Amling M Takeda S Priemel M Schilling AF 2000 Leptin inhibits bone formation through a hypothalamic relay: A central control of bone mass Cell 100 197 207 10660043
Elefteriou F Takeda S Ebihara K Magre J Patano N 2004 Serum leptin level is a regulator of bone mass Proc Natl Acad Sci U S A 101 3258 3263 14978271
Erlebacher A Filvaroff EH Ye JQ Derynck R 1998 Osteoblastic responses to TGF-beta during bone remodeling Mol Biol Cell 9 1903 1918 9658179
Goltzman D 2002 Discoveries, drugs and skeletal disorders Nat Rev Drug Discovery 1 784 796 12360256
Karsenty G Wagner EF 2002 Reaching a genetic and molecular understanding of skeletal development Dev Cell 2 389 406 11970890
Kronenberg H 2003 Developmental regulation of the growth plate Nature 423 332 336 12748651
Ornitz DM Marie PJ 2002 FGF signaling pathways in endochondral and intramembranous bone development and human genetic disease Genes Dev 16 1446 1465 12080084
Ross PD Knowlton W 1998 Rapid bone loss is associated with increased levels of biochemical markers J Bone Miner Res 13 297 302 9495524
Tucker KL Chen H Hannan MT Cupples LA Wilson PW 2002 Bone mineral density and dietary patterns in older adults: The Framingham Osteoporosis Study Am J Clin Nutr 76 245 252 12081842
Falahati-Nini A Riggs BL Atkinson EJ O'Fallon WM Eastell R 2000 Relative contribution of testosterone and estrogen in regulating bone resorption and formation in elderly men J Clin Invest 106 1553 1560 11120762
Takeda S Elefteriou F Levasseur R Liu X Zhao L 2002 Leptin regulates bone formation via the sympathetic nervous system Cell 111 305 317 12419242
Elefteriou F Ahn JD Takeda S Starbuck M Yang X 2005 Leptin regulation of bone resorption by the sympathetic nervous system and CART Nature 434 514 520 15724149
Fu L Patel MS Bradley A Wagner EF Karsenty G 2005 The molecular clock mediates leptin-regulated bone formation Cell 122 803 815 16143109
Manolagos SC 2000 Birth and death of bone cells: Basic regulatory mechanisms and implications for the pathogenesis and treatment of osteoporosis Endocrine Rev 21 115 137 10782361
Nuttall ME Gimble JM 2004 Controlling the balance between osteoblastogenesis and adipogenesis and the consequent therapeutic implications Curr Opin Pharmacol 4 290 294 15140422
Meunier P Aaron J Edouard C Vignon G 1971 Osteoporosis and the replacement of cell populations of the marrow by adipose tissue: A quantitative study of 84 iliac bone biopsies Clin Orthop 80 147 154 5133320
Pittenger MF Mackay AM Beck SC Jaiswal RK Douglas R 1999 Multilineage potential of human mesenchymal stem cells Science 284 143 147 10102814
Nakashima K Zhou X Kunkel G Zhang Z Deng JM 2002 The novel zinc finger containing transcription factor osterix is required for osteoblast differentiation and bone formation Cell 108 17 29 11792318
Ducy P Zhang R Geoffroy V Ridall AL Karsenty G 1997 Osf2/Cbfa1: A transcriptional activator of osteoblast differentiation Cell 89 747 754 9182762
Robledo RF Rajan L Li X Lufkin T 2002 The Dlx5 and Dlx6 homeobox genes are essential for craniofacial, axial and appendicular skeletal development Genes Dev 16 1089 1101 12000792
Oishi Y Manabe I Tobe K Tsushima K Shindo T 2005 Kruppel-like transcription factor KLF5 is a key regulator of adipocyte differentiation Cell Metab 1 27 39 16054042
Rosen ED Sarraf P Troy AE Bradwin G Moore K 1999 PPARγ is required for the differentiation of adipose tissue in vivo and in vitro Mol Cell 4 611 617 10549292
Kubota N Terauchi Y Miki H Tamemoto H Yamauchi T 1999 PPARγ mediates high-fat diet–induced adipocyte hypertrophy and insulin resistance Mol Cell 4 597 609 10549291
Darnell RB 2004 Paraneoplastic neurologic disorders: Windows into neuronal function and tumor immunity Arch Neurol 61 30 32 14732616
Jin P Warren ST 2003 New insights into fragile X syndrome: From molecules to neurobehaviors Trends Biochem Sci 28 152 158 12633995
Larocque D Pilotte J Chen T Cloutier F Massie B 2002 Nuclear retention of MBP mRNAs in the Quaking viable mice Neuron 36 815 829 12467586
Larocque D Galarneau A Liu HN Scott M Almazan G 2005 Protection of the p27KIP1 mRNA by quaking RNA binding proteins promotes oligodendrocyte differentiation Nat Neurosci 8 27 33 15568022
Crittenden SL Bernstein DS Bachorik JL Thompson BE Gallegos M 2002 A conserved RNA-binding protein controls germline stem cells in Caenorhabditis elegans Nature 417 660 663 12050669
Baehrecke EH 1997 who encodes a KH RNA binding protein that functions in muscle development Development 124 1323 1332 9118803
Zaffran S Astier M Gratecos D Semeriva M 1997 The held out wings (how) Drosophila gene encodes a putative RNA binding protein involved in the control of muscular and cardiac activity Development 124 2087 2098 9169854
Lukong KE Richard S 2003 Sam68, the KH domain superSTAR Biochim Biophys Acta 1653 73 86 14643926
Wong G Muller O Clark R Conroy L Moran MF 1992 Molecular cloning and nucleic acid binding properties of the GAP-aasociated tyrosine phosphoprotein p62 Cell 69 551 558 1374686
Richard S Yu D Blumer KJ Hausladen D Olszowy MW 1995 Association of p62, a multi-functional SH2- and SH3-binding protein, with src-family tyrosine kinases, Grb2, and phospholipase Cγ-1 Mol Cell Biol 15 186 197 7799925
Fumagalli S Totty NF Hsuan JJ Courtneidge SA 1994 A target for Src in mitosis Nature 368 871 874 7512695
Taylor SJ Shalloway D 1994 An RNA binding protein associated with src through its SH2 and SH3 domains in mitosis Nature 368 867 871 7512694
Lukong KE Larocque D Tyner AL Richard S 2005 Tyrosine phosphorylation of Sam68 by BRK regulates intranuclear localization and cell cycle progression J Biol Chem 280 38639 38647 16179349
Reddy TR Xu W Mau JKL Goodwin CD Suhasini M 1999 Inhibition of HIV replication by dominant negative mutants of Sam68, a functional homolog of HIV-1 Rev Nat Med 5 635 642 10371501
Matter N Herrlich P Konig H 2002 Signal-dependent regulation of splicing via phosphorylation of Sam68 Nature 420 691 695 12478298
Chen T Boisvert FM Bazett-Jones DP Richard S 1999 A role for the GSG domain in localizing Sam68 to novel nuclear structures in cancer cell lines Mol Cell Biol 10 3015 3033
Di Fruscio M Chen T Richard S 1999 Two novel Sam68-like mammalian proteins SLM-1 and SLM-2: SLM-1 is a Src substrate during mitosis Proc Natl Acad Sci U S A 96 2710 2715 10077576
Soriano P Montgomery C Geske R Bradley A 1991 Targeted disruption of the c-src proto-oncogene leads to osteopetrosis in mice Cell 64 693 702 1997203
Manolagas SC Jilka RJ 1995 Bone marrow cytokines and bone remodeling N Engl J Med 332 305 311 7816067
Parfitt AM Drezner MR Glorieux FH Kanis JA Malluche H 1987 Bone histomorphometry: Standardization of nomenclature, symbols and units J Bone Miner Res 2 595 610 3455637
Katagiri T Yamaguchi A Komaki M Abe E Takahashi N 1994 Bone morphogenetic protein-2 converts the differentiation pathway of C2C12 myoblasts into the osteoblast lineage J Cell Biol 127 1755 1766 7798324
Ben Fredj N Grange J Sadoul R Richard S Goldberg Y 2004 Depolarization-induced translocation of the RNA-binding protein Sam68 to the dendrites of hippocampal neurons J Cell Sci 117 1079 1090 14996936
Brown JR Ye H Bronson RT Dikkes P Greenberg ME 1996 A defect in nurturing in mice lacking the immediate early gene fosB Cell 86 297 309 8706134
Ingalls AM Dickie MM Snell GD 1950 Obese, a new mutation in the house mouse J Hered 41 317 318 14824537
Horne WC Neff L Chaterjee D Lomri A Levy JB 1992 Osteoclasts express high levels of pp60c-src in association with intracellular membranes J Cell Biol 119 1003 1013 1385441
Lowe C Yoneda T Boyce BF Chen H Mundy GR 1993 Osteopetrosis in Src-deficient mice is due to an autonomous defect of osteoclasts Proc Natl Acad Sci U S A 90 4485 4489 7685105
Scwartzberg PL Xing L Hoffmann O Lowell CA Garrett L 1997 Rescue of osteoclast function by transgenic expression of kinase-deficient Src in src–/– mutant mice Genes Dev 11 2835 2844 9353253
Tanaka S Amling M Neff L Peymann A Uhlmann E 1996 c-cbl is downstream of c-src in a signaling pathway necessary for bone resorption Nature 383 528 531 8849724
Miyazaki T Sanjay A Neff L Tanaka S Horne WC 2004 Src kinase activity is essential for osteoclast function J Biol Chem 279 17660 17666 14739300
Marzia M Sims NA Voit S Migliaccio S Taranta A 2000 Decreased c-Src expression enhances osteoblast differentiation and bone formation J Cell Biol 151 311 320 11038178
Amling M Neff L Priemel M Schilling AF Rueger JM 2000 Progressive increase in bone mass and development of odontomas in aging osteopetrotic c-src -deficient mice Bone 27 603 610 11062345
Miao D Liu H Plut P Niu M Huo R 2003 Impaired endochondral bone development and osteopenia in Gli2 deficient mice Exp Cell Res 294 210 222
Valverde-Franco G Liu H Davidson D Chai S Carvajal HV 2004 Defective bone mineralization and osteopenia in young adult FGFR3–/– mice Hum Mol Genet 13 271 284 14681299
Komarova SV Pereverzev A Shum JW Sims SM Dixon SJ 2005 Convergent signaling by acidosis and receptor activator of NF-kappaB ligand (RANKL) on the calcium/calcineurin/NFAT pathway in osteoclasts Proc Natl Acad Sci U S A 102 2643 2648 15695591
|
16362077
|
PMC1315279
|
CC BY
|
2021-01-05 07:59:38
|
no
|
PLoS Genet. 2005 Dec 16; 1(6):e74
|
utf-8
|
PLoS Genet
| 2,005 |
10.1371/journal.pgen.0010074
|
oa_comm
|
==== Front
PLoS GenetPLoS GenetpgenplgeplosgenPLoS Genetics1553-73901553-7404Public Library of Science San Francisco, USA 1636207810.1371/journal.pgen.001007705-PLGE-RA-0305R2plge-01-06-02Research ArticleToxicology - Environmental HealthGenetics/EpigeneticsSaccharomycesIn VitroThe Flavoring Agent Dihydrocoumarin Reverses Epigenetic Silencing and Inhibits Sirtuin Deacetylases Dihydrocoumarin Is a Deacetylase InhibitorOlaharski Andrew J 1*Rine Jasper 2Marshall Brett L 3Babiarz Joshua 2Zhang Luoping 1Verdin Eric 34Smith Martyn T 11 School of Public Health, University of California Berkeley, Berkeley, California, United States of America
2 Department of Molecular and Cellular Biology, University of California Berkeley, Berkeley, California, United States of America
3 Gladstone Institute of Virology and Immunology, University of California San Francisco, San Francisco, California, United States of America
4 Department of Medicine, University of California San Francisco, San Francisco, California, United States of America
Dutcher Susan EditorWashington University in St. Louis, United States of America* To whom correspondence should be addressed. E-mail: [email protected] 2005 16 12 2005 1 6 e7728 9 2005 15 11 2005 Copyright: © 2005 Olaharski 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.Sirtuins are a family of phylogenetically conserved nicotinamide adenine dinucleotide-dependent deacetylases that have a firmly established role in aging. Using a simple Saccharomyces cerevisiae yeast heterochromatic derepression assay, we tested a number of environmental chemicals to address the possibility that humans are exposed to sirtuin inhibitors. Here we show that dihydrocoumarin (DHC), a compound found in Melilotus officinalis (sweet clover) that is commonly added to food and cosmetics, disrupted heterochromatic silencing and inhibited yeast Sir2p as well as human SIRT1 deacetylase activity. DHC exposure in the human TK6 lymphoblastoid cell line also caused concentration-dependent increases in p53 acetylation and cytotoxicity. Flow cytometric analysis to detect annexin V binding to phosphatidylserine demonstrated that DHC increased apoptosis more than 3-fold over controls. Thus, DHC inhibits both yeast Sir2p and human SIRT1 deacetylases and increases p53 acetylation and apoptosis, a phenotype associated with senescence and aging. These findings demonstrate that humans are potentially exposed to epigenetic toxicants that inhibit sirtuin deacetylases.
Synopsis
The effects of chronic low-dose human exposure to environmental chemicals are difficult to study and poorly understood. Chemicals are routinely tested for the ability to induce DNA mutations, cause chromosome damage, or produce cell death, but are rarely tested for their ability to cause epigenetic changes, which can influence the behavior of a cell without directly changing the DNA sequence. Epigenetic changes have become the focus of intense research in an attempt to understand the mechanisms by which they function. The Sir2 family of deacetylases is one class of proteins that controls some epigenetic processes and, interestingly, has been implicated in extending the longevity of several organisms. Here the authors describe a novel assay based upon yeast Sir2p function to screen environmental chemicals for their ability to alter epigenetic silencing. From screening a relatively small number of agents, the authors found that dihydrocoumarin, a natural compound found in Melilotus officinalis (sweet clover) that is synthetically manufactured and frequently added to both food and cosmetics, disrupted epigenetic processes in the yeast Saccharomyces cerevisiae. Dihydrocoumarin also inhibited several human Sir2 family deacetylases (SIRT1 and SIRT2) and, when added to cells in culture, increased p53 tumor suppressor protein acetylation and caused elevated levels of apoptosis. The present study suggests that humans are exposed to a number of environmental chemicals that may be classified as epigenetic toxicants.
Citation:Olaharski AJ, Rine J, Marshall BL, Babiarz J, Zhang L, et al. (2005) The flavoring agent dihydrocoumarin reverses epigenetic silencing and inhibits sirtuin deacetylases. PLoS Genet 1(6): e77.
==== Body
Introduction
Members of the silent information regulator 2 (SIR2) family of genes encode highly conserved nicotinamide adenine dinucleotide (NAD+)-dependent deacetylases present in organisms from archaebacteria to eukaryotes [1]. In the yeast Saccharomyces cerevisiae, Sir2p is a histone deacetylase required for heterochromatic silencing at telomeres, ribosomal DNA, and mating type loci [2]. The sirtuin family of deacetylases has a firmly established role in aging [3]. Increased SIR2 activity, mediated either by overexpression or through sirtuin-activating compounds, increases longevity in S. cerevisiae [4], Drosophila melanogaster [5], and Caenorhabditis elegans [5,6]. Conversely, SIR2 deletion reduces life span in S. cerevisiae by 30% [7], and Sir2p inhibition by nicotinamide mimics this effect [8]. Seven apparent homologs of SIR2 (SIRT1–7) exist in humans, with SIRT1 being the presumed Sir2p ortholog due to sequence similarity [1]. Although the human sirtuin deacetylases have a role in heterochromatin modification, they have mainly been identified to have nonhistone protein targets [9].
SIRT1 has been identified to deacetylate p53 [10] as well as a number of other proteins involved with the apoptotic response [10–13]. The p53 tumor suppressor protein is often referred to as the “guardian of the genome” due to its role in cell cycle arrest, senescence, and apoptosis [14]. Lysine acetylation (K320, K373, K382) increases p53 stability [15,16], leading to the transcriptional activation of DNA repair, cell cycle arrest, and proapoptotic genes. Because SIRT1-mediated p53 deacetylation reverses these effects, inhibition of this deacetylation step is hypothesized to promote p53 stability and increase apoptosis levels. Apoptosis and p53 stabilization accompany SIRT1 down-regulation [17], and SIRT1 inhibition by nicotinamide results in p53 hyperacetylation following DNA damage [10]. In addition, SIRT1−/− cells have been identified to be more susceptible to killing by the genotoxic agents cisplatin and staurosporine [18], indicating that SIRT1 abrogation may enhance p53 function. SIRT1-deficient mice are observed to have developmental defects that are likely due to enhanced tumor suppression by a hyperacetylated and stable p53 [19]. p53 activity appears to control a fine balance between appropriate tumor suppression leading to cancer avoidance and stem cell depletion leading to tissue senescence [20,21]. SIRT1 balances these processes.
The discovery that resveratrol, a chemical found in red wine and other foods, increases life span in multiple organisms through a mechanism that may involve the activation of Sir2p [4,5] suggests that the diet and environment might also contain inhibitors of sirtuin enzymes. We devised a pair of haploid strains of the yeast S. cerevisiae that, if exposed to an agent that disrupts one kind of epigenetically heritable chromatin state, would mate and exhibit diploid growth on minimal medium. We tested over 100 chemicals to which humans are commonly exposed, including coumarins, bioflavonoids, benzene metabolites, essential oils, and arsenic, to address the possibility that these compounds could disrupt heterochromatin repression. Here we show that dihydrocoumarin (DHC), a compound found in Melilotus officinalis (sweet clover) that is widely used in the food and cosmetic industries [22,23], tested positive in the yeast mating assay and was identified to inhibit both Sir2p and SIRT1. Concentration-dependent increases in cytotoxicity, apoptosis, and p53 acetylation were observed following DHC exposure in the human TK6 lymphoblastoid cell line. These findings demonstrate that humans are exposed to epigenetic toxicants that inhibit sirtuin deacetylases.
Results
Development of a Screening Assay to Detect Environmental Chemicals That Disrupt Heterochromatin Silencing in S. cerevisiae
Haploid S. cerevisiae exists in two distinct mating types, a and α, controlled by a single locus on Chromosome III, MAT. Additionally, both a and α information are stored at the cryptic mating loci HML
α and HMRa. These cryptic mating loci are used for mating-type interconversion and are silenced in part by Sir2p. Using a strain in which both HML and HMR contain α information and a recessive mata1–1 allele, disruption of any protein required for silencing, including Sir2p, either by chemical inhibition or mutation, would cause a phenotypic switch from an a to an α mating type. Mixing these cells with a strain of yeast that is MATa HMRa HMLa would result in mating only if heterochromatin silencing were inhibited. Mixing strains of these genotypes with complementary auxotrophic markers provided a simple test for agents that disrupt heterochromatic silencing, leading to mating and subsequent growth on minimal medium containing uracil. Using this technically simple indicator assay, we tested over 100 environmental chemicals to which humans are exposed (e.g., arsenic, benzene metabolites, coumarins, bioflavonoids, and essential oils, listed in Table S1, Supporting Information).
DHC and M. officinalis Oil Extract Derepresses Heterochromatic Silencing in S. cerevisiae
DHC (CAS# 119–84–6) from two independent sources derepressed heterochromatic gene silencing in the yeast S. cerevisiae mating assay (Figure 1). Oil extracts from M. officinalis, a common flowering plant that contains small amounts of DHC, also tested as a weak positive in the assay (data not shown). Colony growth following DHC exposure demonstrated a dose-response relationship at micromolar levels that was similar to the level of heterochromatin derepression observed with the established Sir2p inhibitor splitomicin (Figures 1B and 1C).
Figure 1 Effect of DHC on Heterochromatin Silencing in S. cerevisiae at the HMR or HML Loci
(A) DMSO negative control, 1 M nicotinamide positive control, and 50 mM DHC.
(B and C) DHC and splitomicin, respectively, cause dose-dependent increases in heterochromatin silencing and colony formation.
DHC Is a Sir2p Inhibitor
The inhibition of heterochromatic silencing by DHC at levels similar to those observed with splitomicin suggested that it may be a Sir2p inhibitor; further experiments, however, were necessary to test whether Sir2p was the target. To this end, we employed a plasmid under a gal promoter that overexpresses SIR2 and produces a lethal phenotype in the presence of galactose [24]. Addition of 750 μM DHC partially suppressed the Sir2p-induced lethality in galactose, but had no deleterious effect in glucose (Figure 2), thus pinpointing Sir2p as the target of DHC. Addition of a lower concentration of DHC (500 μM DHC) also reversed the lethal SIR2 phenotype (data not shown). A recent report using a yeast telomeric reporter assay to analyze splitomicin analogs [25] support our finding that DHC is a yeast Sir2p inhibitor.
Figure 2 DHC Reverses the Lethal Phenotype of Galactose-Induced Plasmid SIR2 Overexpression
Two-fold yeast serial dilutions were used for this assay. Growth is observed on 0.1% DMSO and 750 μM DHC in dextrose medium. The lethal SIR2 phenotype is observed on the 0.1% DMSO galactose medium but is partially reversed on the 750 μM DHC galactose medium.
DHC Inhibits Human SIRT1 and SIRT2
We tested whether DHC could inhibit the human NAD+-dependent deacetylase SIRT1 and observed that DHC induced a concentration-dependent inhibition of SIRT1 (IC50 of 208 μM) in an in vitro enzymatic assay (Figure 3A). A decrease in SIRT1 deacetylase activity was observed even at micromolar doses (85 ± 5.8 and 73 ± 13.7% activity at 1.6 μM and 8 μM, respectively). The microtubule SIRT2 [26] deacetylase was also inhibited with a similar dose dependency (Figure 3B).
Figure 3 DHC Inhibits Human NAD+-Dependent Deacetylases In Vitro
(A) Dose-dependent inhibition of SIRT1 (y = −19.98x + 96.431).
(B) Dose-dependent inhibition of SIRT2 (y = −20.79x + 101.34). Averages ± standard error from two separate experiments are shown.
DHC Increases p53 Acetylation, Cytotoxicity, and Apoptosis Levels
Experiments with the human TK6 lymphoblastoid cell line further addressed DHC-mediated SIRT1 inhibition. Immunoblot analysis demonstrated that both p53 acetylation and cytotoxicity increased in a dose-dependent manner following DHC exposure (Figure 4A and 4B). Flow cytometric analysis measuring annexin V binding to phosphatidylserine demonstrated that apoptosis levels increased more than 3-fold following DHC exposure (Figure 4C and 4D).
Figure 4 DHC Increases p53 Acetylation, Cytotoxicity, and Apoptosis in the TK6 Cell Line
(A) DHC increases p53 lysine 373 and 382 acetylation in a dose-dependent manner in the TK6 cell line following a 24-h exposure period. The immunoblot is representative of three separate experiments.
(B) DHC increases cytotoxicity in a dose-dependent manner following a 24-h exposure to DHC. The average ± standard error for each of three experiments is shown.
(C) A 5-mM dose of DHC increases apoptosis at the 6-h time point in the TK6 cell line. A combination of annexin V fluorescein isothiocyanate and propidium iodide staining was used to discriminate among apoptotic (lower-right box) and necrotic (upper-right box) cells. Figures are representative of three separate experiments.
(D) DHC increases apoptosis in a dose-dependent manner in the TK6 cell line at the 6-h time point. The average ± standard error for three experiments is shown.
Discussion
Epigenetic drift is hypothesized to occur, in part, through exposure to environmental and dietary compounds that affect the processes responsible for the proper maintenance of the proteome and epigenome [27,28]. The recent discovery that significant differences in epigenetic markers exist in monozygotic twins who lived different lifestyles and spent little time together supports this environmentally driven epigenetic drift hypothesis [29]. Additionally, the finding that methoxychlor, a common DDT substitute used to control mosquitoes, and vinclozolin, a broadly used fungicide applied in the wine industry [30], caused transgenerational epigenetic effects up to the F4 generation [31] indicates that humans are likely exposed to epigenetic toxicants causing epigenetic drift on a regular basis. These are intriguing findings that raise the question of whether the diet and environment contain epigenetic toxicants that may also inhibit the sirtuin deacetylases.
We have designed a yeast mating assay to identify environmental chemicals that inhibit heterochromatic repression in S. cerevisiae to address the possibility that humans are exposed to epigenetic toxicants that inhibit sirtuin deacetylases. After screening more than 100 environmental chemicals to which humans are exposed, including coumarins, bioflavonoids, benzene metabolites, and arsenic, we identified that DHC disrupted heterochromatic repression (see Figure 1A). Further analyses demonstrated that DHC-mediated heterochromatic derepression caused yeast colony formation in a concentration-dependent manner that was similar to splitomicin (Figure 1B and 1C). Splitomicin is an established Sir2p inhibitor [32] with a structure similar to DHC, thus it was likely that DHC-mediated heterochromatic derepression in the mating assay was due to Sir2p inhibition. Further experiments with an overexpressing SIR2-induced death phenotype were conducted to identify if Sir2p was the target of DHC. DHC-mediated reversal of the SIR2 overexpressed death phenotype (Figure 2) indicated that DHC is a Sir2p inhibitor and that this inhibition was responsible for the heterochromatic derepression observed in the mating assay. DHC thus joins a short list of established Sir2p inhibitors that includes nicotinamide, splitomicin, and sirtinol [33].
The identification that DHC is a Sir2p inhibitor is of particular interest because it is a natural component of M. officinalis and is synthetically manufactured for use as a common fragrance in perfumes, cosmetics, lotions, soaps, and as a flavoring agent in beverages and chewing gum [23,34]. DHC is present at concentrations above 100 ppm (670 μM) in gelatins, puddings, soft candy, frozen dairy products, and baked goods [22]. Due to the potential DHC exposure to humans, we tested whether it could also inhibit human sirtuin deacetylases and identified that DHC inhibited both SIRT1 and SIRT2 deacetylases in vitro (Figure 3A and 3B). We also measured SIRT3 deacetylase activity but did not identify it to be affected by DHC (data not shown). It is only recently that the cellular function of the human sirtuin deacetylases has begun to be revealed. Because there are at least seven human sirtuins [1], it is expected that the substrates and functions of each will vary. SIRT2 is a cytoplasmic protein identified to be a microtubule deacetylase [26], an observation that may help explain its requirement for proper progression through mitosis [35]. SIRT1 has been identified to deacetylate a variety of nuclear substrates [9], including several that are transcription factors regulating differentiation and development or proteins involved in the apoptotic response, including FOXO [12,36], ku70 [11], p53 [10], and MEF2 [37]. Although SIRT1 has been identified to have mostly nonhistone targets, there is growing evidence to indicate that SIRT1 interacts with a number of transcription factors with results that may mimic epigenetic modifications. How SIRT1 inhibitors can affect these processes is just beginning to be understood.
Because SIRT1 has been identified to deactivate p53 through deacetylation, experiments in the human TK6 lymphoblastoid cell line were conducted to identify if DHC could induce a phenotype consistent with SIRT1 inhibition and p53 activation. DHC was identified to increase p53 acetylation, cytotoxicity, and apoptosis levels in vitro (Figure 4). In addition, DHC enhanced cell killing to etoposide in both the TK6 human lymphoblastoid and human embryonic kidney 293 cell lines (data not shown), data compatible with the observation that SIRT1−/− cells are more susceptible to killing by genotoxic agents [18]. Previous in vivo and in vitro studies have demonstrated that DHC is not a mutagen, clastogen, or aneugen [34], suggesting that it is unlikely that DHC is stabilizing p53 and increasing cytotoxicity and apoptosis through a genotoxic mechanism. Rather, these data are consistent with the phenotypes of SIRT1 compromised cells [10,17] and support the hypothesis that DHC-mediated SIRT1 inhibition enhances p53 stabilization and increases apoptosis. Increasing tumor suppression and the susceptibility of a cell to undergo apoptosis, however, cannot solely be viewed in a beneficial chemotherapeutic light, particularly in those cases where exposure to an apoptosis-inducing agent may be widespread. The deleterious effects of a hyperactive p53 have been demonstrated in a transgenic mouse model that displayed a shortened life span and a variety of early aging-associated phenotypes [38]. It is hypothesized that increased levels of apoptosis cause a rapid depletion of stem cells, leading to premature tissue senescence [20,21], possibly decreasing longevity. Increasing p53 tumor suppression through inhibition of SIRT1 may tip the cell-survival balance toward apoptosis in a cell that would otherwise not undergo programmed cell death. It is interesting to speculate whether sirtuin deacetylases, which have a role in regulating longevity in a variety of organisms, may also regulate longevity in mammals through the control of apoptosis. Further work is clearly needed to address the possibility that enhanced apoptosis through chemical-mediated inhibition of SIRT1 can increase tissue senescence and affect longevity.
Our finding that the common flavoring agent DHC inhibits sirtuin deacetylases linked to aging is potentially worrisome. SIRT1 inhibition may lead to epigenetic alterations as well as possible stem cell depletion and early tissue senescence due to increased levels of apoptosis. While no compounds to our knowledge are currently classified as “senescegens,” it is possible that a number of environmental chemicals can increase tissue senescence and aging. The present study demonstrates that a more extensive screening of dietary and environmental chemicals is needed to identify other epigenetic toxicants to which humans are commonly exposed.
Materials and Methods
Yeast mating assay.
JRY075 (MATa, HMLa, HMRa, ade6, met, his4, leu2, and ura3) and JRY80 (mata1–1, HMLα, HMRα, leu1, ura3, and ade2) yeast strains [39] were combined in liquid rich-medium (YPD), and 50 μl of the mixture was spread onto YM plates containing 30 μg/ml uracil. Chemicals (Sigma, St. Louis, Missouri, United States; Indofine, Hillsborough, New Jersey, United States) were dissolved in DMSO and added to the YM uracil+ plates at 50 mM concentrations, whereas essential oils and plant oil extracts (Yerba Buena, Woodside, California, United States; Lhasa Kharnak, Berkeley, California, United States; and Nature's Apothecary, Bloomingdale, Illinois, United States) were added directly to the plates. Plates were incubated for at least 72 h at room temperature to allow for diploid growth on minimal medium containing uracil. Nicotinamide (1 M) and DMSO were used as the positive and solvent controls, respectively, for all experiments. Yeast colony DHC and splitomicin dose-response experiments were conducted to compare the ability of DHC to disrupt heterochromatin repression with an established Sir2p inhibitor. Varying doses (1 through 25 μM) of DHC and splitomicin were added directly to the YM and uracil medium.
Inhibition of the SIR2-induced death phenotype.
The yeast strain YSH278 and the galactose-induced SIR2 pSS3 plasmid were the kind gifts of Scott Holmes. The yeast strain YSH278 was transformed with the pSS3 plasmid containing galactose-inducible alleles and grown overnight in raffinose medium lacking leucine. Sets of 2-fold serial dilutions from these cultures were spotted onto YM glucose medium lacking leucine (−LEU) and on galactose medium lacking leucine (−LEU GAL). Both the dextrose and galactose plates contained 750 μM DHC or 0.1% DMSO as a solvent control. Plates were incubated for at least 72 h at room temperature.
SIRT1 and SIRT2 deacetylase assays.
The histone deacetylation assay with recombinant SIRT1 and SIRT2 was performed, as described previously [40], using a [3H] acetylated histone H4 peptide.
Cell culture and immunoblot analysis.
The human TK6 lymphoblastoid cell line (ATCC) was maintained in RPMI1640 medium (GIBCO, San Diego, California, United States) containing 10% FBS (Omega Scientific, Tarzana, California, United States) and 1% penicillin and streptomycin (Omega Scientific) under standard conditions. TK6 cells were exposed to 0.1% DMSO and 1, 2, 3, 4, and 5 mM DHC for 24 h at a concentration of 3.5 × 105–5 × 105. Total cell lysates were collected from 1.5 × 106–2 × 106 cells using 100 μl of radioimmunoprecipitation assay (RIPA) lysis buffer. Protein concentrations were determined by the Bradford assay (Bio-Rad, Hercules, California, United States). Equal protein amounts were resolved by PAGE, transferred onto nitrocellulose membranes, and imunoblotted for lysine 372 and 383 acetylated p53 (Upstate, Charlottesville, Virginia, United States). Protein was visualized using the ECL method per manufacturer's protocol (Amersham Biosciences, Little Chalfont, United Kingdom). Film (Kodak BioMax XAR; Eastman Kodak Corporation, Rochester, New York, United States) was exposed and developed using the Konica SRX-101 developer (Konica Minolta Medical Imaging USA, Wayne, New Jersey, United States).
Cellular cytotoxicity and apoptosis assays.
Cellular cytotoxicity for the aforementioned TK6 DHC experiments was conducted using the Trypan blue exclusion assay. A minimum of 200 cells were analyzed for each dose and experiment. Annexin V apoptosis assays were performed in three additional experiments in which TK6 cells were exposed to 0.1% DMSO and 1, 2, 3, 4, and 5 mM DHC for 6 h. TK6 cells were prepared for flow cytometric annexin V analysis following the manufacturer's protocol (BD Biosciences Pharmingen, San Diego, California, United States), except that 5 μl of 1 mg/ml propidium iodide (PI) was used. At least 10,000 TK6 cells were analyzed on a Beckman Coulter EPICS XL-MCL flow cytometer (Fullerton, California, United States) using System II software for each experiment. Live cells were gated and analyzed for annexin V FITC and PI staining.
Supporting Information
Table S1 List of Chemicals and Essential Oils Tested for Heterochromatic Repression in the Yeast Mating Assay
(32 KB DOC)
Click here for additional data file.
We thank Scott Holmes for his generous donation of the YSH278 yeast strain and the pSS3 plasmid. We thank Kripa Krishnan and Sophia Lim for technical assistance. Supported by National Institutes of Health grants P30 ES01896 (to MTS) and RO1 GM31105 (to JR).
Competing interests. The authors have declared that no competing interests exist.
Author contributions. AJO, JR, LZ, and MTS conceived and designed the experiments. AJO and BLM performed the experiments. AJO and JR analyzed the data. AJO, JR, JB, EV, and MTS contributed reagents/materials/analysis tools. AJO wrote the paper.
Abbreviations
DHCdihydrocoumarin
NAD+nicotinamide adenine dinucleotide
SIR2silent information regulator 2
==== Refs
References
Frye RA 2000 Phylogenetic classification of prokaryotic and eukaryotic Sir2-like proteins Biochem Biophys Res Commun 273 793 798 10873683
Rusche LN Kirchmaier AL Rine J 2003 The establishment, inheritance, and function of silenced chromatin in Saccharomyces cerevisiae
Annu Rev Biochem 72 481 516 12676793
Masoro EJ 2004 Role of sirtuin proteins in life extension by caloric restriction Mech Ageing Dev 125 591 594 15491676
Howitz KT Bitterman KJ Cohen HY Lamming DW Lavu S 2003 Small molecule activators of sirtuins extend Saccharomyces cerevisiae lifespan Nature 425 191 196 12939617
Wood JG Rogina B Lavu S Howitz K Helfand SL 2004 Sirtuin activators mimic caloric restriction and delay ageing in metazoans Nature 430 686 689 15254550
Tissenbaum HA Guarente L 2001 Increased dosage of a sir-2 gene extends lifespan in Caenorhabditis elegans
Nature 410 227 230 11242085
Anderson RM Bitterman KJ Wood JG Medvedik O Cohen H 2002 Manipulation of a nuclear NAD+ salvage pathway delays aging without altering steady-state NAD+ levels J Biol Chem 277 18881 18890 11884393
Bitterman KJ Anderson RM Cohen HY Latorre-Esteves M Sinclair DA 2002 Inhibition of silencing and accelerated aging by nicotinamide, a putative negative regulator of yeast sir2 and human SIRT1 J Biol Chem 277 45099 45107 12297502
North BJ Verdin E 2004 Sirtuins: Sir2-related NAD-dependent protein deacetylases Genome Biol 5 224 15128440
Luo J Nikolaev AY Imai S Chen D Su F 2001 Negative control of p53 by Sir2alpha promotes cell survival under stress Cell 107 137 148 11672522
Cohen HY Miller C Bitterman KJ Wall NR Hekking B 2004 Calorie restriction promotes mammalian cell survival by inducing the SIRT1 deacetylase Science 305 390 392 15205477
van der Horst A Tertoolen LG de Vries-Smits LM Frye RA Medema RH 2004 FOXO4 is acetylated upon peroxide stress and deacetylated by the longevity protein hSir2(SIRT1) J Biol Chem 279 28873 28879 15126506
Langley E Pearson M Faretta M Bauer UM Frye RA 2002 Human SIR2 deacetylates p53 and antagonizes PML/p53-induced cellular senescence EMBO J 21 2383 2396 12006491
Lane DP 1992 Cancer: p53, guardian of the genome Nature 358 15 16 1614522
Ito A Lai CH Zhao X Saito S Hamilton MH 2001 p300/CBP-mediated p53 acetylation is commonly induced by p53-activating agents and inhibited by MDM2 EMBO J 20 1331 1340 11250899
Luo J Li M Tang Y Laszkowska M Roeder RG 2004 Acetylation of p53 augments its site-specific DNA binding both in vitro and in vivo Proc Natl Acad Sci U S A 101 2259 2264 14982997
Wang C Wang MW Tashiro S Onodera S Ikejima T 2005 Roles of SIRT1 and phosphoinositide 3-OH kinase/protein kinase C pathways in evodiamine-induced human melanoma A375-S2 cell death J Pharmacol Sci 97 494 500 15821341
Matsushita N Takami Y Kimura M Tachiiri S Ishiai M 2005 Role of NAD-dependent deacetylases SIRT1 and SIRT2 in radiation and cisplatin-induced cell death in vertebrate cells Genes Cells 10 321 332 15773895
Cheng HL Mostoslavsky R Saito S Manis JP Gu Y 2003 Developmental defects and p53 hyperacetylation in Sir2 homolog (SIRT1)-deficient mice Proc Natl Acad Sci U S A 100 10794 10799 12960381
Campisi J 2005 Senescent cells, tumor suppression, and organismal aging: Good citizens, bad neighbors Cell 120 513 522 15734683
Lombard DB Chua KF Mostoslavsky R Franco S Gostissa M 2005 DNA repair, genome stability, and aging Cell 120 497 512 15734682
Fenaroli G 2004 Fenaroli's handbook of flavor ingredients. Burdock G, editor. 5th ed Boca Raton (Florida) CRC Press 2040 p.
Rastogi SC Johansen JD Menne T 1996 Natural ingredients based cosmetics. Content of selected fragrance sensitizers Contact Dermatitis 34 423 426 8879930
Matecic M Stuart S Holmes SG 2002 SIR2-induced inviability is suppressed by histone H4 overexpression Genetics 162 973 976 12399404
Posakony J Hirao M Stevens S Simon JA Bedalov A 2004 Inhibitors of Sir2: Evaluation of splitomicin analogues J Med Chem 47 2635 2644 15115404
North BJ Marshall BL Borra MT Denu JM Verdin E 2003 The human Sir2 ortholog, SIRT2, is an NAD+ -dependent tubulin deacetylase Mol Cell 11 437 444 12620231
Jaenisch R Bird A 2003 Epigenetic regulation of gene expression: How the genome integrates intrinsic and environmental signals Nat Genet 33 (Suppl) 245 254 12610534
Bjornsson HT Fallin MD Feinberg AP 2004 An integrated epigenetic and genetic approach to common human disease Trends Genet 20 350 358 15262407
Fraga MF Ballestar E Paz MF Ropero S Setien F 2005 From the cover: Epigenetic differences arise during the lifetime of monozygotic twins Proc Natl Acad Sci U S A 102 10604 10609 16009939
Kaiser J 2005 Developmental biology. Endocrine disrupters trigger fertility problems in multiple generations Science 308 1391 1392 15933166
Anway MD Cupp AS Uzumcu M Skinner MK 2005 Epigenetic transgenerational actions of endocrine disruptors and male fertility Science 308 1466 1469 15933200
Bedalov A Gatbonton T Irvine WP Gottschling DE Simon JA 2001 Identification of a small molecule inhibitor of Sir2p Proc Natl Acad Sci U S A 98 15113 15118 11752457
Grubisha O Smith BC Denu JM 2005 Small molecule regulation of Sir2 protein deacetylases FEBS J 272 4607 4616 16156783
National Toxicology Program 1993 NTP toxicology and carcinogenesis studies of 3,4-dihydrocoumarin (CAS No. 119–84–6) in F344/N rats and B6C3F1 mice (Gavage Studies) Natl Toxicol Program Tech Rep Ser 423 1 336 12616288
Dryden SC Nahhas FA Nowak JE Goustin AS Tainsky MA 2003 Role for human SIRT2 NAD-dependent deacetylase activity in control of mitotic exit in the cell cycle Mol Cell Biol 23 3173 3185 12697818
Giannakou ME Partridge L 2004 The interaction between FOXO and SIRT1: Tipping the balance towards survival Trends Cell Biol 14 408 412 15308206
Zhao X Sternsdorf T Bolger TA Evans RM Yao TP 2005 Regulation of MEF2 by histone deacetylase 4- and SIRT1 deacetylase-mediated lysine modifications Mol Cell Biol 25 8456 8464 16166628
Tyner SD Venkatachalam S Choi J Jones S Ghebranious N 2002 p53 mutant mice that display early ageing-associated phenotypes Nature 415 45 53 11780111
Rine J Herskowitz I 1987 Four genes responsible for a position effect on expression from HML and HMR in Saccharomyces cerevisiae
Genetics 116 9 22 3297920
North BJ Schwer B Ahuja N Marshall B Verdin E 2005 Preparation of enzymatically active recombinant class III protein deacetylases Methods 36 338 345 16091304
|
16362078
|
PMC1315280
|
CC BY
|
2021-01-05 08:00:26
|
no
|
PLoS Genet. 2005 Dec 16; 1(6):e77
|
utf-8
|
PLoS Genet
| 2,005 |
10.1371/journal.pgen.0010077
|
oa_comm
|
==== Front
PLoS GenetPLoS GenetpgenplgeplosgenPLoS Genetics1553-73901553-7404Public Library of Science San Francisco, USA 1636207910.1371/journal.pgen.001007805-PLGE-RA-0301R3plge-01-06-08Research ArticleGenetics/GenomicsGenetics/Gene FunctionGenetics/Gene ExpressionGenetics/Complex TraitsHomo (Human)Genome-Wide Associations of Gene Expression Variation in Humans Association of Gene Expression VariationStranger Barbara E 1Forrest Matthew S 1Clark Andrew G 2Minichiello Mark J 1Deutsch Samuel 3Lyle Robert 3Hunt Sarah 1Kahl Brenda 4Antonarakis Stylianos E 3Tavaré Simon 56Deloukas Panagiotis 1*Dermitzakis Emmanouil T 1*1 Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, United Kingdom
2 Department of Molecular Biology and Genetics, Cornell University, Ithaca, New York, United States of America
3 Department of Genetic Medicine and Development, University of Geneva Medical School, Geneva, Switzerland
4 Illumina, Inc., San Diego, California, United States of America
5 Department of Oncology, University of Cambridge, Hutchison/MRC Research Centre, Cambridge, United Kingdom
6 Program in Molecular and Computational Biology, University of Southern California, Los Angeles, California, United States of America
Allison David EditorUniversity of Alabama at Birmingham, United States of America* To whom correspondence should be addressed. E-mail: [email protected] (PD); [email protected] (ETD)12 2005 16 12 2005 1 6 e7813 10 2005 16 11 2005 Copyright: © 2005 Stranger 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 exploration of quantitative variation in human populations has become one of the major priorities for medical genetics. The successful identification of variants that contribute to complex traits is highly dependent on reliable assays and genetic maps. We have performed a genome-wide quantitative trait analysis of 630 genes in 60 unrelated Utah residents with ancestry from Northern and Western Europe using the publicly available phase I data of the International HapMap project. The genes are located in regions of the human genome with elevated functional annotation and disease interest including the ENCODE regions spanning 1% of the genome, Chromosome 21 and Chromosome 20q12–13.2. We apply three different methods of multiple test correction, including Bonferroni, false discovery rate, and permutations. For the 374 expressed genes, we find many regions with statistically significant association of single nucleotide polymorphisms (SNPs) with expression variation in lymphoblastoid cell lines after correcting for multiple tests. Based on our analyses, the signal proximal (cis-) to the genes of interest is more abundant and more stable than distal and trans across statistical methodologies. Our results suggest that regulatory polymorphism is widespread in the human genome and show that the 5-kb (phase I) HapMap has sufficient density to enable linkage disequilibrium mapping in humans. Such studies will significantly enhance our ability to annotate the non-coding part of the genome and interpret functional variation. In addition, we demonstrate that the HapMap cell lines themselves may serve as a useful resource for quantitative measurements at the cellular level.
Synopsis
With the finished reference sequence of the human genome now available, focus has shifted towards trying to identify all of the functional elements within the sequence. Although quite a lot of progress has been made towards identifying some classes of genomic elements, in particular protein-coding sequences, the characterization of regulatory elements remains a challenge. The authors describe the genetic mapping of regions of the genome that have functional effects on quantitative levels of gene expression. Gene expression of 630 genes was measured in cell lines derived from 60 unrelated human individuals, the same Utah residents of Northern and Western European ancestry that have been genetically well-characterized by The International HapMap Project. This paper reports significant variation among individuals with respect to levels of gene expression, and demonstrates that this quantitative trait has a genetic basis. For some genes, the genetic signal was localized to specific locations in the human genome sequence; in most cases the genomic region associated with expression variation was physically close to the gene whose expression it regulated. The authors demonstrate the feasibility of performing whole-genome association scans to map quantitative traits, and highlight statistical issues that are increasingly important for whole-genome disease mapping studies.
Citation:Stranger BE, Forrest MS, Clark AG, Minichiello MJ, Deutsch S, et al (2005) Genome-wide associations of gene expression variation in humans. PLoS Genet 1(6): e78.
==== Body
Introduction
Mapping genetic factors that underlie quantitative traits in humans has been a challenging task in the absence of large samples with accurate phenotypic measures and dense genotypic data. In particular, one of the gaps in our knowledge of human biology is the structure of genetic variation affecting gene regulation and how it contributes to phenotypic variation and disease [1]. Recent studies in model organisms including yeast [2–5], mouse [6–8], maize [8], and rat [9], have attempted to address this issue by testing for linkage and/or associations of gene expression variation among individuals with nucleotide variation. As a result, extensive functional genetic variation has been discovered, suggesting that the overall contribution of regulatory variation to phenotypic variation has been underestimated. In humans, three studies have followed a two-stage approach: first performing linkage analysis to identify regions in which gene expression variation segregates in pedigrees, and then linkage disequilibrium (LD) mapping those regions in a larger sample of unrelated individuals with additional markers [10–12]. Linkage analysis may miss weaker signals in the first stage, as it relies on sufficient differences in phenotypic means among recombinant and non-recombinant genotypes [13]. On the other hand, allele-specific expression experiments can identify signals nearby the gene and in LD with the coding single nucleotide polymorphism (SNP) that is used for the measurement, but cannot provide an unbiased view of regulatory variation in the human genome [14,15]. Association studies have generally more power to detect such signals [16], and the availability of high throughput methods for genotyping and gene expression profiling make genome-wide scans an appealing alternative. But genome-wide studies pose their own challenge in the form of statistical inference in the face of so many simultaneous statistical tests. Obtaining sufficient power using a test with a given rate of false positives in this setting remains a serious challenge [17].
In this study we employed the densely genotyped (>1 million SNPs) HapMap panel of 60 unrelated US residents of Northern and Western European descent (labeled CEU; see [18,19]) to perform a genome-wide association study of gene expression in 630 genes (excluding control genes). We show that even with this limited sample size, we are able to detect strong and highly significant SNP-to-expression associations, most of which are cis (proximal 1-Mb regions) to the gene of interest. This analysis provides the first unbiased and genome-wide view of regulatory associations in samples of unrelated individuals in humans.
Results
Sample and Experimental Design
We selected a set of 630 protein coding genes (see Figure S1 for chromosomal distribution) consisting at the time (January 2004) of all Refseq genes in the ENCODE regions [20] (321 genes, excluding the HSA21 ENCODE regions), all Refseq genes on human Chromosome 21 (191 genes), and all Refseq and manually curated genes from a 10-Mb region at 20q12–13.2 (118 genes), and surveyed their transcript levels in a sample of lymphoblastoid cell lines generated from 60 unrelated CEU individuals. Transcript levels were measured with the Illumina BeadArray technology using two or more unique oligonucleotide probes per gene (1,433 probes including controls; see Materials and Methods) [21]. We performed a minimum of six replicate hybridizations for each of the 60 samples and obtained expression values for all the probes and samples. We normalized the raw intensity values with a quantile normalization method and averaged the replicate values (see Materials and Methods) per probe per sample to obtain a single expression level per individual per probe. We found excellent correlation between measurements of replicates within and between arrays (r2 = 0.96–0.99).
Initial Association Analysis
To perform association tests between gene expression variation and SNP variation, we selected 374 of the 630 tested genes that had probe hybridization signals significantly above the background and were among the most variable (we excluded probes with phenotypic outliers; see Materials and Methods), corresponding to 688 unique probes, and the public release of 753,712 SNPs with minor allele frequency above 5% made by the HapMap project (HapMap version 16b; [19]). The actual determinants of gene expression levels are likely to be associated with many interacting factors across the genome, but in light of the small sample size, we consider here only the simplest models testing for single-SNP effects. We employed a main effects, linear regression model separately for each probe, where the additive effect of a SNP genotype is tested by coding the genotypes of each SNP as 0, 1, and 2 (corresponding to the counts of the minor allele in each genotype) and performing a linear regression of the new variable, “allele counts”, with the normalized gene expression values; this test has one degree of freedom. Note that we treated each probe separately to account for potential differences between the transcript levels inferred by the one or the other due to SNPs or other sequence dependent effects, though in most cases the two probes of each gene were highly correlated (see Figure S2) and produced almost identical results. This makes it unlikely that the associations described below are artifacts of segregation of SNPs within the 50-mer probe sequence but we describe a more detailed analysis below to account for such effects. Because the model considers the effects of each SNP singly, it cannot test the role of potential SNP–SNP interaction effects.
Methods for Multiple-Test Correction
One of the main problems of performing genome-wide analysis with such a large number of phenotypes and SNPs is correcting for multiple testing. Despite the fact that there have been a number of approaches to handle this problem, it is not yet clear which is the optimal way to perform such correction [22]. Thus we applied three methodologies routinely employed for multiple test correction when using a random set of markers from the genome, namely the Bonferroni correction for multiple tests, generating a null distribution of p-values by permuting the phenotypes relative to the genotypes [23], and the false discovery rate (FDR) [24]. In all cases of multiple correction we assigned a significance threshold of p = 0.05 after the correction.
We considered both the genome-wide distribution of p-values as well as a subset of SNPs within 1 Mb from the genes tested. The rationale for the 1-Mb subset is that most of the cis regulatory regions of a gene are located within a small distance from the gene (though known enhancers can be as far as 1 Mb away from the gene; [25]) and therefore the density of relevant sequences around each gene is high. This can be viewed as a “candidate region” approach similar to the candidate gene approach used in disease studies. By restricting our analysis to these candidate regions we are more likely to detect relevant associations. This was evident when we contrasted the distribution of p-values between cis and trans (Figure 1). For the purposes of our analysis we have defined cis to be 1 Mb from the midpoint of the genomic region of the corresponding gene.
Figure 1 QQ Plot of cis versus trans HSA20 −log10
p-Values
The figure shows the contrast of −log10
p-values deriving from associations of SNPs and genes within the 10-Mb region of HSA20 with −log10
p-values deriving from associations between genes on the 10-Mb region HSA20 with SNPs in one of ten ENCODE regions. Note that the distribution falls off the diagonal around −log10
p = 4, which we consider the borderline for the high enrichment of cis significant effects. A similar pattern is observed with any set of trans −log10
p-values on HSA20 or any other cis vs. trans contrast in any region we tested.
Bonferroni correction was applied to the genome-wide analysis as well as to a subset of the associations where SNP−gene distance was less than 1 Mb. The p-values generated from the individual SNP−gene tests were evaluated according to an adjusted significance threshold generated by dividing the 0.05 threshold by the total number of tests (number of genes multiplied by number of SNPs) performed in each case (whole genome or cis). The purpose of applying the Bonferroni correction to the 1-Mb subset of SNPs was to assess significance when only the cis- signal is considered. The whole genome Bonferroni correction was based on all 753,712 SNPs. If we had performed a whole genome analysis using a subset of “tag” SNPs, for example using 500,000 SNPs, the Bonferroni significance threshold would have differed little (p-value equal to 2.67 × 10−10 instead of 1.77 × 10−10; corresponding to −log10
p equal to 9.57 and 9.75, respectively).
In order to obtain significance values using an empirical null distribution of p-values from the observed data, we performed 12,500 permutations, wherein each permutation shuffled the expression values relative to the genotypes and for each permutation we retained the minimum p-value for each gene separately [26]. Empirical p-values were obtained by comparing the observed (non-permuted) p-values for each gene to the distribution of permuted p-values for the same gene. We subsequently corrected for the number of genes tested by adjusting the significance threshold by dividing by the number of genes tested.
The permutation analysis was first performed using genotypes for all 753,712 SNPs, and then again using genotypes for only those SNPs within 1 Mb of the gene of interest to test the significance of cis effects. The permutations were performed separately for each of the two sets of SNPs according to the description above, and significance was assessed for each using the same implementation of multiple test correction as described.
For the third test, we employed an FDR analysis [24], considering only those p-values from associations of all tests with SNP−gene distance less than 1 Mb. The distribution of all of these p-values together (i.e., p-values from all genes and their cis SNPs) was used to calculate the FDR and to assess significance of each individual p-value in the distribution. A whole-genome FDR calculation was not undertaken as the number of genome-wide p-values made it computationally prohibitive to run the analysis. Signals were considered significant if a p-value had a corresponding q-value of less than 0.05.
Assigning Significance to Associations
We detected highly significant (−log10
p up to 13.6) and multiple associations between gene expression level and SNPs in close proximity (less than 400 kb) to the corresponding gene. The signal from the additive model decayed rapidly with distance from the gene (Figure 2A and 2B). In particular, based on the Bonferroni genome-wide threshold (Bonf-gw) we detected a total of 99 SNPs (183 SNP-probe associations) that significantly explained transcript levels of six genes, half of which (three out of six) are in cis and half (three out of six) are in trans. These 99 SNPs cluster into seven genomic regions. When we applied the Bonferroni correction only for SNP−gene associations with distances less than 1 Mb (Bonf-cis) the number increased to 171 SNPs (310 SNP-probe associations) for ten genes, with these SNPs clustering into ten genomic regions.
Figure 2
Cis- Signals of SNP−Gene Associations in the Human Genome
(A) The relationship between statistical significance and distance from gene. Each data point represents the maximum −log10
p for a single gene and SNPs located cis- to its coding locus. The −log10
p-values from the additive model are plotted as a function of distance between the center of the genomic span of the gene and cis- located SNPs (cis- < 4 Mb). Only those gene-SNP associations that have −log10
p > 4 are shown. SNPs are from the 5-kb HapMap. This plot includes data for 101 genes (129 probes). (B) Cis- SNPs with −log10
p ≥ 4 from the 688 probes analyzed are plotted against their chromosomal location on NCBI34 coordinates of the human genome.
Permutation-based assignment of significance was also applied to our analysis. When we performed genome-wide correction (Perm-gw) only three genes had significant associations, and these were in cis. The same three cis associations were identified by the genome-wide Bonferroni correction. Interestingly, no trans associations were significant with the permutation-based analysis; notably absent were the three trans associations identified by the genome-wide Bonferroni correction. This suggests that although Bonferroni is generally a conservative correction method, it might still identify spurious associations if the assumptions of the original parametric model are violated (see below). Permutation-based correction of p-values for SNP−gene distances of less than 1 Mb (Perm-cis) detected ten genes with significant associations, which were the same ten genes identified by the 1-Mb Bonf-cis correction. To further explore the signal cis to the genes tested, and uncover associations that might be biologically relevant but do not cross the significance threshold due to our small sample size, we considered all the genes that were individually significant at the 0.05 significance threshold based on the permutation test (Perm-cis-enrich). A total of 63 genes were determined to be individually significant at this level, while we expect only 19 by chance. This pool of 63 genes is therefore enriched by 44 genes that appear to have significant signals within 1 Mb of the gene.
Finally, we assigned significance based on a FDR of q = 0.05. As mentioned above, it was computationally prohibitive to perform the FDR for all p-values of the genome-wide analysis, so we only used p-values where the SNP–gene distance was less than 1 Mb (FDR-cis) to interrogate the cis signal. We identified 40 genes that had at least one SNP with q < 0.05. Of these, 38 were overlapping with the 63 genes that were individually called significant in the permutation analysis (Perm-cis-enrich). The dependence of SNPs due to linkage disequilibrium may be considered a problem, but as described in Remark D in [24], this dependence may actually lead to conservative estimates of significant genes. However, to account for the effects of dependence of SNPs in associated regions, and be even more conservative, we removed from the FDR analysis all p-values of SNP–gene tests within 100 kb from the lowest p-value for each of the 40 genes above (FDR-cis-trim). When new FDR thresholds were assigned, only 20 genes remained significant, although the remaining 20 were very close to the new threshold. All 20 genes were in the pool of 63 genes identified by the Perm-cis-enrich method. The reason that only 20 genes remained significant is because although we removed the dependence in regions that had a strong signal, we did not remove the dependence in regions with null signal and these regions are more common. So although only 20 genes were left after the FDR-cis-trim analysis, the 40 genes detected with the original FDR-cis analysis should be considered as a set of genes with significant cis signal.
Overlap between Multiple-Test Correction Methods
Table 1 summarizes the numbers of genes identified by each method and the overlaps of associated genes. The biggest discordance was that between Bonf-gw and Perm-gw. Bonf-gw identified three trans associations that were not considered significant by Perm-gw. This is due to the presence at each of these three expression phenotypes of one or two individuals with phenotypic outliers (very slightly exceeding the mean + three standard deviations) that caused extreme p-values with a few SNPs. The presence of these phenotypic outliers violates the normality assumptions of the linear regression model, thus the p-values obtained are not valid. Such problems are not accounted for by the Bonferroni correction if phenotypic distributions are not filtered in advance but can be accounted for by permutation-based thresholds. All analyses in cis showed that ten associated genes were considered significant across all methods that focused on the cis signal only (Perm-cis, Perm-cis-enrich, FDR-cis, FDR-cis-trim). Two of these ten genes, CPNE1 and CSTB, were also shown to have significant cis signals in a previous study [10,12], corroborating the validity of the observed signals. Interestingly, the FDR-cis and Perm-cis-enrich analyses showed that it is likely that more than ten genes have a significant cis signal. The overlap of these methods was almost 100% (38 out of 40 genes identified in the FDR-cis analysis were included in the 63 genes significant by Perm-cis-enrich). It is worth noting that the FDR-cis analysis with q = 0.05 predicts that 95% of the 40 genes (i.e., 38 genes) will be real signals, which is exactly the number that overlaps with Perm-cis-enrich. Overall, we observed a very good concordance between the different statistical methods that were used to assign significance based on multiple test correction.
Table 1 Comparison of Multiple-Test Correction Methods
Power and the Magnitude of Effects
The small sample size and the large number of SNPs and phenotypes bring challenges to the interpretation of the significant effects we detect. Our rationale to focus on cis effect stems from the fact that cis regions are candidate regions to contain regulatory elements that influence the proximal genes, while it is harder to select a priori such regions from the rest of the genome. Therefore, our ability to detect more cis effects is not necessarily because larger effects occur more frequently in cis regions, but because we had more power when performing the analysis in cis regions. In order to estimate the power given the experimental design, one can perform power calculations. However, such calculations require many assumptions about the data. An alternative, given the multiple test correction we have performed, is to describe the magnitude of the effect that each of our analyses can detect. The adjusted r2 values from the linear regression provide an estimate of the fraction of variation that is explained by an individual SNP. Based on the adjusted r2 values from the linear regression performed for the SNPs with the lowest p-values (cis or trans), we have more than 80% power to detect effects of r2 ≥ 0.50 when genome-wide correction (Bonf-gw or Perm-gw) is applied and the type I error is set to 0.05. When we perform the correction based on the 1-Mb distance from the corresponding gene, we have more than 80% power to detect effects with r2 ≥ 0.22. The magnitude of individual effects of SNPs for each gene is illustrated in Table 2. The average r2 value for the three genes deemed significant by Perm-gw was 0.58; average r2 was 0.48 for the ten genes called significant by the Perm-cis analysis, and average r2 was 0.27 for all 63 genes that had a significant cis- signal according to at least one of the methods. It should be noted that estimates of effect size are often biased upwards in QTL studies [27]. Overall, our analysis can only detect large effects, and larger sample sizes are needed for the detection of smaller but not necessarily less important regulatory effects.
Table 2 Genes with Significant cis and trans Associations
Specific Examples of Associations
The most highly significant association signal detected was with SNPs cis- to the SERPINB10 gene, located in a region of Chromosome 18 that harbors five additional members of the serine (or cysteine) proteinase inhibitor, clade B (ovalbumin) gene family (Figure 3A). The additive model identified SNPs with highly significant association with transcript levels of SERPINB10 distributed across two LD blocks; furthermore both probes for this gene exhibited the same significant signal (unpublished data). The majority of significantly associated SNPs are located immediately upstream of the transcription start site, while others are in the coding locus itself and immediately downstream (Figure 3A). Linkage disequilibrium is strong in this region (Figure 3A) and thus the stretch of SNPs exhibiting a similar highly significant association may actually be generated by only one or a few variants that cause the difference in expression. The fact that there are significantly associated cis- SNPs in two regions of high LD surrounding this locus, suggests the possibility of two independent regulatory variants.
Figure 3 Examples of cis- Associations from the Genome-Wide and High-Density SNP Maps
(A) Genomic location of associated SNPs close to the SERPINB10 gene. Custom tracks in the UCSC genome browser (http://genome.ucsc.edu) show the location of the Illumina probe and proximal SNPs in the context of genome annotation. The lower horizontal black line indicates the −log10
p threshold where the corresponding q-value is 0.05 (i.e., any SNPs with values −log10
p that meet or exceed this threshold are significant at the q = 0.05 level), and the upper line is the Bonferroni genome-wide threshold. Additional tracks describe known genes, first-exon and promoter predictions, conserved transcription factor binding sites, Gencode genes, RNA polymerase 2, and Transcription factor 2 binding sites, identified by Affymetrix ChIP/chip experiments, and Sp1 and Sp3 binding sites identified by Stanford's ChIP/chip experiments. Consensus conserved elements are shown in the final track. HapMap LD information below is for the CEU individuals and suggests that there are two conserved haplotype clusters in this region.
Figure 3 Continued
(B) Genomic location of associated SNPs close to the TMEM8 and MRPL28 genes. Note the correlation between the p-values for the two genes. Custom tracks in the UCSC genome browser show the location of the Illumina probe and proximal SNPs in the context of genome annotation. The lower horizontal black line indicates the −log10
p threshold where the corresponding q-value is 0.05 (i.e., any SNPs with values −log10
p that meet or exceed this threshold are significant at the q = 0.05 level), and the upper line is the Bonferroni genome-wide threshold. Additional tracks describe known genes, first-exon and promoter predictions, conserved transcription factor binding sites, Gencode genes, RNA polymerase 2, and Transcription factor 2 binding sites, identified by Affymetrix ChIP/chip experiments, and Sp1 and Sp3 binding sites identified by Stanford's ChIP/chip experiments. Consensus conserved elements are shown in the final track. HapMap LD information below is for the CEU individuals (http://genome.ucsc.edu) [39].
One gene that is consistently significant in genome-wide corrections is TMEM8. The gene is found on Chromosome 16 and exhibits significant p-values with the highest being 1.3 × 10−10 for SNPs very close to the coding sequence. An intriguing observation is that another gene immediately 3′ to TMEM8, MRPL28 is also associated with the same SNPs but with higher but still significant p-values (6.0 × 10−7) (Figure 3B). TMEM8 encodes a transmembrane protein whereas MRPL28 encodes a ribosomal protein and is a gene transposed from the mitochondrial genome, so there is no apparent biological relationship between them. Both genes are in the same orientation so an intriguing scenario is that there are transcripts spanning both genes that are driven from a single promoter, and this is the reason for such high correlation in expression. Such phenomena are now commonly observed in the human and other genomes [28]. However, expressed sequence tag data do not support this idea since none of them show a transcript that spans both genes. Another exciting scenario, from an evolutionary point of view, is that MRPL28 has “landed” at some point in time in the vicinity of TMEM8 and recruited already existing regulatory regions in the region, one of which was a TMEM8 regulatory region. Subsequent segregating variation in this regulator may have then affected the expression of both genes.
SNPs in Hybridization Probes
For each of the 40 genes showing at least one significant association according to the FDR-cis analysis, we considered the possibility of the influence on expression levels of SNPs located at the probe binding sequence. We searched the sequences of all probes for known sequence variants and examined the entry in dbSNP (version 125; http://www.ncbi.nlm.nih.gov/projects/SNP) to assess the likelihood of the variant being a real SNP. We identified a total of 18 variants in sequences of probes for 12 of the 40 genes; three of these genes are in the set of ten loci that exhibited significant cis- signal according to all cis methods. Of these 18 variants, eight had no confidence that they were actual SNPs (i.e., there is no frequency information for any population). Two SNPs had no frequency information, but were identified twice by two different centers. Eight variants are SNPs with genotype frequencies, only three of which have been genotyped in the HapMap CEU population. Of the three SNPs genotyped in the HapMap project, one is monomorphic in CEU and the other two are in high LD with the SNPs exhibiting the strongest significant association in cis for the genes AXIN1 and HRMT1L1, suggesting that the observed cis associations for these two genes may be due to the SNP presence in the probe sequence. Other authors have noted that even with 60-mer probes, a single SNP underlying the probe sequence can lead to spurious cis- associations [29]. Although this is a technical artifact, it illustrates that the experimental and statistical methodologies described here are able to detect signals of differential hybridization with one nucleotide mismatch in the 50-mer probes.
Discussion
Our analysis suggests that there is an abundance of common genetic variation that explains gene expression differences among individuals. From a total of 374 genes with above-background signal interrogated in 60 individuals, we can detect genetic variants with significant effects on transcript level for ten to 40 genes in cis. Trans signals in three genes are only supported by one of the two statistical methods employed for genome-wide analysis (Bonferroni and permutations) and are more likely to be false positives. We therefore observed that the majority of detectable signals are caused by SNPs located cis- to the gene, and the signal is consistent with an additive effect of the causal variant. This observation is consistent with previously published studies that found a significant proportion of the intraspecific differences in transcript level are located cis- to the gene in question [8,10]. The fact that only a small fraction of the genes had significant signal is likely a result of reduced power due to the small sample size. Moreover, testing additional cell types is likely to reveal more regulatory variants. In addition, when we define significance thresholds by focusing on proximal SNPs we detect many more significant associations. This suggests that if we apply differential weighting of SNPs based on proximity to the gene we may be able to enhance our ability to detect significant cis effects [30]. These models are still in an exploratory stage. An intriguing possibility is that some of these significant regions indicate the presence of an additional copy of the gene (cis or trans) due to copy number polymorphisms [CNPs; 31,32], and in fact we have identified the haplotype with the additional copy of the gene. For example, C16orf15 which is significant only for Perm-cis-enrich overlaps with a known copy number variant (http://projects.tcag.ca/variation). Our methodology may thus help identify the map position of excess gene copies. Such cases will be examined when we have copy number polymorphism data for these individuals. Other alternative explanations are that the mapped SNPs tag variants that affect mRNA stability, mRNA trafficking, and other post-transcriptional effects.
In order to assess alternative approaches to dealing with the massive multiple testing problem faced here, we have applied three standard methodologies. The fact that there is substantial overlap of the signals detected from all three suggests that most signals in our analysis are robust and more likely to be true positives, given the distinct theoretical basis for the Bonferroni, permutation, and FDR tests. Such a test of robustness to assumptions in multiple test correction may ultimately be necessary when the number of tests is so large. This will become an even bigger problem when one accounts for potential interactions between SNPs, something we have not addressed in the present study. Given the fact that each phenotype has its own properties of variance and inheritance, it seems unlikely that genome- and experiment-wide thresholds provide the optimal means for assessing significance. Experimental validation of such targets is the next priority [29], and in order for it to be informative one has to consider not only the top few signals but sample from a wide range of significance levels. In fact, it is not yet obvious that what appears to be the most statistically significant effect is the most biologically relevant effect. True validation will require the elucidation of the complete biological effect, rather than a first-pass, low sensitivity, experimental validation procedure.
Perhaps the most promising aspect of studies of gene expression variation mapping is the impact it can have on interpretation of functional genomic information and functional variation. For instance, most of the experimental methodologies that identify regulatory regions in humans rely on experimental procedures that, although they can elucidate the regulatory potential of a region (e.g., binding assays, transfection assays, etc.), they are not able to pinpoint the gene which interacts with the candidate regulatory region. If one couples these experimental procedures with the methodology described in the present study, direct links between regulatory regions and genes can be identified, which will facilitate biological interpretations. In addition, such an approach can greatly facilitate the interpretation of genotype-phenotype (disease) associations when the mapped regions fall into non-coding regions of the genome with no annotated functional DNA.
We have demonstrated that it is feasible to map genetic variants affecting gene expression by genome-wide association testing. The HapMap samples, which have been genotyped for more than 3.5 million SNPs, offer a powerful resource for such gene expression studies as well as for other studies that measure quantitative traits at the cellular level. It has been hypothesized that the genetic basis of complex traits and diseases is largely regulatory [33,34] and such studies may point to potential disease variants. The sets of genes screened in this study are of a strategic nature. The ENCODE regions will soon have accumulated large amounts of functional data, some of which are relevant to gene regulation [20]. We already see patterns of correlation between the location of the associated SNPs and the regulatory active regions (see Figure 3A), which will greatly enhance our ability to interpret the effect of associated variants. Also, Trisomy of HSA21 causes Down syndrome (i.e., gene dosage effect) and the elucidation of expression allelic variants will facilitate the discovery of genes associated with the variable Down syndrome phenotypes such as the congenital heart defect present in 40% of patients with Down syndrome [35]) The 10 Mb region of chr20 is known to be associated with type II diabetes and obesity [36], and the discovery of allelic expression variants may reveal potential causal candidates for this association. Mapping of gene expression phenotypes in cell lines of heavily genotyped samples will provide a baseline resource that will greatly facilitate the fine mapping of disease variants in human populations [37]. We have demonstrated that the use of multiple methods for statistical inference greatly facilitates the interpretation of results and generates stronger candidates for experimental follow-up. Our results have important implications, both for the regulatory landscape of the human genome, as well as for the choice of the type of variation one needs to interrogate for disease association studies.
Materials and Methods
RNA preparation.
Total RNA was extracted from the 60 HapMap parental CEU lymphoblastoid cell lines (Coriell, Camden, New Jersey, United States). Two one-quarter scale Message Amp II reactions (Ambion, Austin, Texas, United States) were performed for each RNA extraction using 200 ng of total RNA. Biotin-16-UTP (Perkin Elmer, Wellesley, California, United States) made up half of the UTP used in the in vitro transcription (IVT) reaction. cRNA yields were quantified using RiboGreen (Invitrogen, Carlsbad, California, United States). 1 μg of the cRNA was then hybridized to an array.
Gene expression quantification.
We designed a custom expression array bead pool with 1,433 unique bead types (two for each of 630 genes, plus controls) each with ~50,000 50-mer probes using Illumina Bead Arrays (Illumina, San Diego, California, United States). Our custom arrays include 321 protein coding ENCODE genes (excluding Chromosome 21), 191 protein coding genes from Chromosome 21, and 118 protein coding gene transcripts from 20q12–13.2. Bead pools were added to bundles of 50,000 fiber optic filaments which were arrayed into an 8×12 Sentrix Array Matrix (SAM) so that 96 arrays could be run in parallel. Each bead type (probe) is present on a single array on average 30 times. Each of the two IVT reactions from the 60 samples was hybridized to three arrays, so that each cell line had six replicate hybridizations. Twelve cell lines had two extra replicates run from one of their IVT reactions, giving them eight replicates, for a total of 384 arrays. cRNA was hybridized to arrays for 18 h at 55 °C before being labelled with Cy3-streptavidin (Amersham Biosciences, Little Chalfont, United Kingdom) and scanned with a Bead Station (Illumina).
Post-experimental raw data processing. Normalization.
With the Illumina bead technology, a single hybridization of RNA from one cell line to an array produces on average approximately 30 intensity values for each of 1,433 bead types. These background-corrected values for a single bead type are subsequently summarized by the Illumina software and output to the user as a set of 1,433 intensity values for each individual hybridization. These raw data were then normalized on a log scale using a quantile normalization method [38]. The expression measurements from the 384 array bundles then have the same distribution.
Averaging replicates.
In our experiment, each cell line was hybridized to six or eight arrays, thus resulting in six or eight reported intensity values (as averages of the values from the 30 beads per probe) for each of the 1,433 bead types. To combine data from our multiple replicate hybridizations, we averaged these normalized intensity values for each bead type to obtain a single value for each of the 1,433 bead types for each individual. These averages (for each probe, across replicates for each individual) of normalized average values are the values used in subsequent analyses.
Association analyses.
The whole genome association analysis employed 753,712 SNP genotypes with minor allele frequency above 5% from the HapMap 5k map (version 16b). Of the 1,405 probes (excluding control probes), we chose the 688 most variable probes (corresponding to 374 unique genes) to use in the association analyses after excluding probes with extreme outliers.
For each of the selected probes and for each SNP, we fitted the following model: the genotype Xi of individual i at the given SNP may be classified as one of three states: Xi = 0, 1, or 2 for homozygous rare, heterozygous and homozygous common alleles, respectively. For this additive model, we fitted a linear regression of the form
Yi = b0 + b1Xi + ɛi where Yi is the normalized log-expression level of the probe for individual i, i = 1,…, 60, and ɛi are independent normally distributed random variables with mean 0 and constant variance. We report the nominal, parametric p-value of the test of no association, i.e., b1 = 0.
Multiple-test correction.
We have employed three approaches for multiple-test correction, namely Bonferroni, assignment of significance based on permutation of phenotypes to the genotypes, and FDR. The first two were applied to both genome-wide analysis and cis analysis (1 Mb) and the FDR was applied only to cis analysis (1 Mb) due to the computational limitations, since the number of genome-wide p-values was too large to be analyzed.
Supporting Information
Figure S1 Chromosomal Location of 63 Genes
Chromosome banding, karyotype cartoon showing the location of the 321 ENCODE genes, the 191 Chromosome 21 genes, and the 118 genes from 20q12–13.2.
(35 KB PPT)
Click here for additional data file.
Figure S2 Correlation between Gene Expression Phenotype Measurements as Quantified by Pairs of Probes Corresponding to the Same Gene
Included are pairs of probes from the set of 688 analyzed probes.
(45 KB PPT)
Click here for additional data file.
Figure S3 Examples of cis- Associations from the Genome-Wide and High-Density SNP Maps; CPNE1 Locus
(127 KB PDF)
Click here for additional data file.
Accession Numbers
The expression data reported in this paper have been deposited in the Gene Expression Omnibus (GEO) (http://www.ncbi.nlm.nih.gov/geo) database (Series Accession Number GSE3612). The accession numbers of the genes mentioned herein include: AXIN1 (NM_181050.1), CPNE1 (NM_152926.1), CSTB (NM_000100.2), HRMT1L1 (NM_001535.1), MRPL28 (NM_006428.2), SERPINB10 (NM_005024.1), and TMEM8 (NM_021259.1).
We would like to thank Tim Cutts for informatics support, Maryline Gagnebin for laboratory support, Mark Gibbs for assistance with probe selection and technical guidance, Mark Dunning for assistance with GEO data formatting, the International HapMap Consortium for making the data available, and the referees for constructive comments. We acknowledge support from the Wellcome Trust to ETD and PD, Swiss National Foundation to SEA, NHGRI ENCODE grant to ETD and SEA, and from NIH grants GM065509 and HG03229 to AGC and HG02790 to ST. ST is a Royal Society-Wolfson Research Merit Award holder.
Competing interests. The authors have declared that no competing interests exist.
Author contributions. BES and MSF performed the experiments and preliminary analysis. BES performed the statistical analysis. AGC, ST, and ETD advised and helped with the statistical analysis. MJM ran the permutation analysis. SD, RL and SEA contributed the cell line RNA and participated in the initial design of the project. BK assisted in the initial steps of setting up the Illumina platform and extraction of data. SH assisted with data format and experimental design issues. ST performed the normalization. PD and ETD led the project. BES and ETD wrote the manuscript and all other authors contributed sections and/or provided comments.
Abbreviations
CEUUtah residents with ancestry from Northern and Western Europe
ChIPchromatin immunoprecipitation
FDRfalse discovery rate
LDlinkage disequilibrium
SNPsingle nucleotide polymorphism
==== Refs
References
Stranger BE Dermitzakis ET 2005 The genetics of regulatory variation in the human genome Hum Genomics 2 126 131 16004727
Storey JD Akey JM Kruglyak L 2005 Multiple locus linkage analysis of genome-wide expression in yeast PLoS Biol 3 e267. DOI: 10.1371/journal.pbio.0030267 16035920
Brem RB Kruglyak L 2005 The landscape of genetic complexity across 5,700 gene expression traits in yeast Proc Natl Acad Sci U S A 102 1572 1577 15659551
Brem RB Yvert G Clinton R Kruglyak L 2002 Genetic dissection of transcriptional regulation in budding yeast Science 296 752 755 11923494
Yvert G Brem RB Whittle J Akey JM Foss E 2003 Trans-acting regulatory variation in Saccharomyces cerevisiae and the role of transcription factors Nat Genet 35 57 64
Bystrykh L Weersing E Dontje B Sutton S Pletcher MT 2005 Uncovering regulatory pathways that affect hematopoietic stem cell function using 'genetical genomics' Nat Genet 37 225 232 15711547
Chesler EJ Lu L Shou S Qu Y Gu J 2005 Complex trait analysis of gene expression uncovers polygenic and pleiotropic networks that modulate nervous system function Nat Genet 37 233 242 15711545
Schadt EE Monks SA Drake TA Lusis AJ Che N 2003 Genetics of gene expression surveyed in maize, mouse, and man Nature 422 297 302 12646919
Hubner N Wallace CA Zimdahl H Petretto E Schulz H 2005 Integrated transcriptional profiling and linkage analysis for identification of genes underlying disease Nat Genet 37 243 253 15711544
Morley M Molony CM Weber TM Devlin JL Ewens KG 2004 Genetic analysis of genome-wide variation in human gene expression Nature 430 743 747 15269782
Monks SA Leonardson A Zhu H Cundiff P Pietrusiak P 2004 Genetic inheritance of gene expression in human cell lines Am J Hum Genet 75 1094 1105 15514893
Cheung VG Spielman RS Ewens KG Weber TM Morley M 2005 Mapping determinants of human gene expression by regional and genome-wide association Nature 437 1365 1369 16251966
Ardlie KG Kruglyak L Seielstad M 2002 Patterns of linkage disequilibrium in the human genome Nat Rev Genet 3 299 309 11967554
Pastinen T Sladek R Gurd S Sammak A Ge B 2004 A survey of genetic and epigenetic variation affecting human gene expression Physiol Genomics 16 184 193 14583597
Yan H Yuan W Velculescu VE Vogelstein B Kinzler KW 2002 Allelic variation in human gene expression Science 297 1143 12183620
Risch N Merikangas K 1996 The future of genetic studies of complex human diseases Science 273 1516 1517 8801636
Wang WY Barratt BJ Clayton DG Todd JA 2005 Genome-wide association studies: Theoretical and practical concerns Nat Rev Genet 6 109 118 15716907
The International HapMap Consortium 2003 The International HapMap Project Nature 426 789 796 14685227
Altshuler D Brooks LD Chakravarti A Collins FS Daly MJ 2005 A haplotype map of the human genome Nature 437 1299 1320 16255080
ENCODE Project Consortium 2004 The ENCODE (ENCyclopedia Of DNA Elements) Project Science 306 636 640 15499007
Kuhn K Baker SC Chudin E Lieu MH Oeser S 2004 A novel, high-performance random array platform for quantitative gene expression profiling Genome Res 14 2347 2356 15520296
Thomas DC Haile RW Duggan D 2005 Recent developments in genome-wide association scans: A workshop summary and review Am J Hum Genet 77 337 345 16080110
Churchill GA Doerge RW 1994 Empirical threshold values for quantitative trait mapping Genetics 138 963 971 7851788
Storey JD Tibshirani R 2003 Statistical significance for genome-wide studies Proc Natl Acad Sci U S A 100 9440 9445 12883005
Lettice LA Heaney SJ Purdie LA Li L de Beer P 2003 A long-range Shh enhancer regulates expression in the developing limb and fin and is associated with preaxial polydactyly Hum Mol Genet 12 1725 1735 12837695
Doerge RW Churchill GA 1996 Permutation tests for multiple loci affecting a quantitative character Genetics 142 285 294 8770605
Beavis WD 1998 QTL analysis: Power, precision, and accuracy Paterson AH editor. Molecular dissection of complex traits New York CRC Press 145 162
Parra G Reymond A Dabbouseh N Dermitzakis ET Antonarakis SE 2005 Tandem chimerism as a means to increase protein complexity in the human genome Genome Research In press.
Doss S Schadt EE Drake TA Lusis AJ 2005 Cis-acting expression quantitative trait loci in mice Genome Res 15 681 691 15837804
Beaumont MA Rannala B 2004 The Bayesian revolution in genetics Nat Rev Genet 5 251 261 15131649
Iafrate AJ Feuk L Rivera MN Listewnik ML Donahoe PK 2004 Detection of large-scale variation in the human genome Nat Genet 36 949 951 15286789
Sebat J Lakshmi B Troge J Alexander J Young J 2004 Large-scale copy number polymorphism in the human genome Science 305 525 528 15273396
Rockman MV Wray GA 2002 Abundant raw material for cis-regulatory evolution in humans Mol Biol Evol 19 1991 2004 12411608
Carroll SB 2005 Endless forms most beautiful: The new science of evo devo and the making of the animal kingdom (New York) W. W. Norton & Company 350 p.
Park SC Mathews RA Zuberbuhler JR Rowe RD Neches WH 1977 Down syndrome with congenital heart malformation Am J Dis Child 131 29 33 138359
Ghosh S Watanabe RM Valle TT Hauser ER Magnuson VL 2000 The Finland-United States investigation of non-insulin-dependent diabetes mellitus genetics (FUSION) study. I. An autosomal genome scan for genes that predispose to type 2 diabetes Am J Hum Genet 67 1174 1185 11032783
Cheung VG Spielman RS 2002 The genetics of variation in gene expression Nat Genet 32 Suppl 522–525
Bolstad BM Irizarry RA Astrand M Speed TP 2003 A comparison of normalization methods for high density oligonucleotide array data based on variance and bias Bioinformatics 19 185 193 12538238
Kent WJ Sugnet CW Furey TS Roskin KM Pringle TH 2002 The Human Genome Browser at UCSC Genome Res 12 996 1006 12045153
|
16362079
|
PMC1315281
|
CC BY
|
2021-01-05 07:59:37
|
no
|
PLoS Genet. 2005 Dec 16; 1(6):e78
|
utf-8
|
PLoS Genet
| 2,005 |
10.1371/journal.pgen.0010078
|
oa_comm
|
==== Front
J CarcinogJournal of Carcinogenesis1477-3163BioMed Central London 1477-3163-4-211628394510.1186/1477-3163-4-21ResearchCancer incidence in the south Asian population of California, 1988–2000 Jain Ratnali V [email protected] Paul K [email protected] Arti [email protected] California Cancer Registry, Public Health Institute, Fresno, California, USA2 University of California, San Diego, Moores UCSD Cancer Center, Cancer Prevention and Control, La Jolla, California, USA3 University of California, San Francisco, Fresno Medical Education Program, Fresno, California, USA4 California Cancer Registry, Public Health Institute, Sacramento, California, USA2005 10 11 2005 4 21 21 11 2 2005 10 11 2005 Copyright © 2005 Jain et al; licensee BioMed Central Ltd.2005Jain 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
Although South Asians (SA) form a large majority of the Asian population of U.S., very little is known about cancer in this immigrant population. SAs comprise people having origins mainly in India, Pakistan, Bangladesh and Sri Lanka. We calculated age-adjusted incidence and time trends of cancer in the SA population of California (state with the largest concentration of SAs) between 1988–2000 and compared these rates to rates in native Asian Indians as well as to those experienced by the Asian/Pacific Islander (API) and White, non-Hispanic population (NHW) population of California.
Methods
Age adjusted incidence rates observed among the SA population of California during the time period 1988–2000 were calculated. To correctly identify the ethnicity of cancer cases, 'Nam Pehchan' (British developed software) was used to identify numerator cases of SA origin from the population-based cancer registry in California (CCR). Denominators were obtained from the U.S. Census Bureau. Incidence rates in SAs were calculated and a time trend analysis was also performed. Comparison data on the API and the NHW population of California were also obtained from CCR and rates from Globocan 2002 were used to determine rates in India.
Results
Between 1988–2000, 5192 cancers were diagnosed in SAs of California.
Compared to rates in native Asian Indians, rates of cancer in SAs in California were higher for all sites except oropharyngeal, oesophageal and cervical cancers. Compared to APIs of California, SA population experienced more cancers of oesophagus, gall bladder, prostate, breast, ovary and uterus, as well as lymphomas, leukemias and multiple myelomas. Compared to NHW population of California, SAs experienced more cancers of the stomach, liver and bile duct, gall bladder, cervix and multiple myelomas. Significantly increasing time trends were observed in colon and breast cancer incidence.
Conclusion
SA population of California experiences unique patterns of cancer incidence most likely associated with acculturation, screening and tobacco habits. There is need for early diagnosis of leading cancers in SA. If necessary steps are not taken to curb the growth of breast, colon and lung cancer, rates in SA will soon approximate those of the NHW population of California.
==== Body
Background
The south Asian (SA) population of United States was 1,893,723 in the year 2000 [1], and between 1990 and 2000 this population grew in size by 106%. Persons with origins in India, Pakistan, Bangladesh, and Sri Lanka are classified as SA and they are now the third largest Asian subgroup in the United States, comprising 16% of all U.S. Asians. Approximately 21% of SAs in the U.S. reside in California, the state with the largest concentration of SAs. From 1990 to 2000, the number of SAs living in California increased from 168,457 to 343,731 (104% increase) [2,3]. 90% of SAs are Asian Indian (people with origins in India). In the year 2000 the SA population of California comprised 1.15% of the total population and this proportion is increasing.
There are no published studies on the incidence of cancer among SAs in United States, except for one study which reported breast and colon cancer incidence in Asian Indians and that analysis was based on a very small sample size [4]. Another study by Divan et al. reports the current available literature on this issue and emphasizes the need to conduct more studies on cancer incidence and mortality [5]. The reason for lack of cancer studies in this population may be multiple; including controversy regarding which communities are included under the title 'South Asian', the relatively recent growth of this community in the US, and the belief that SAs are part of a 'model minority' and therefore have better health status than other minority groups. In previous studies all Asians have been grouped into one category, which may mask important differences in incidence and survival among various subgroups.
Most of the cancer studies in SAs residing outside of south Asia have been done in the UK or Canada [6]. Many cancer studies have been conducted in the SA population of UK, mainly because they form the largest ethnic minority of UK. Much attention has been focused on breast and lung cancer epidemiology [7-11]. Studies focusing on multiple cancer sites are few [11,12] although some attention has been given to childhood cancers, mainly because childhood cancers are increasing with time [13-16].
Initial studies suggested that English SA rates for all sites combined were lower than the non-SA rates but higher than Indian subcontinent rates (especially for lung cancer in males, breast cancer in females, and lymphomas in both sexes). But a sub-site analysis revealed that, English South Asian rates were significantly higher than the non-SA rates for Hodgkins disease in males, and oral, esophageal, thyroid, leukemias in females, and cancers of the pharynx, liver and gall bladder in both sexes [12].
Recent studies in UK indicate, that younger SA, particularly children are at increased risk of cancer than the non-SA population and although generally cancer rates have fallen over the last decade, they are increasing among SAs [11].
Studies on cancer in the SA population of Canada pertain primarily to cancer screening, and no studies on cancer incidence have been reported [17,18].
Studies of cancer incidence in immigrant populations can provide valuable insights into etiology and changes towards the pattern of disease seen in the host country may indicate environmental factors in etiology [19]. Therefore, in this analysis we have calculated age adjusted rates for cancer in the SA population of California and compared these rates to native Asian Indians (people living in India) as well as the Asian/Pacific Islander (Asian/PI) and non-Hispanic White (NHW) populations of California in the same time period. We also conducted a time trend analysis to study the patterns of cancer incidence in this population for the period 1988–2000. Where appropriate, we have also compared these rates to those reported in Great Britain.
Methods
The California Cancer Registry (CCR), a population-based registry, commenced operation in 1988. The methodology of the CCR has been fully described by Morris et al. [20]. The CCR collects information on all cancers except for non-melanoma skin cancers and in situ cancers of the uterine cervix. Information on several demographic variables, diagnostic variables (including stage at diagnosis, tumor size, histology and grade of tumor), and first course of treatment are collected for all cases. Cases are routinely coded with regard to anatomic stage of disease using the general summary stage schema for 1988–1993 [20], and SEER extent of disease for 1994–1997 [21]. Race and ethnicity are categorized into four mutually exclusive groups in the CCR database: White, non-Hispanic, Black, non-Hispanic, Hispanic, and Asian/Pacific Islander. Under the last category there are further breakdowns for several Asian ethnic groups, including the category 'Asian Indian/Pakistani', which includes people of SA origin.
Our analysis included cancer cases diagnosed during the period 1988–2000. Incidence rates were calculated for this population for all major sites and several specific cancer types. Due to small numbers for some of the cancer sites, the rates for individual years were grouped into three-year categories to reduce the instability of rates. In addition, an age-adjusted trend analysis of the rates was completed for the period 1988–2000 to determine the Annual Percentage Change (APC) (using the non-weighted least squares approach) along with p-values for APCs.
Ethnic Classification
The SA group is heterogeneous, not only in national origin, sub-ethnicity (and therefore heritable features), and religion, but also in specific details of pertinent lifestyle including alcohol, tobacco, and various levels of vegetarianism. Secondly, individual hospitals, from where most cancer cases are identified by the CCR, do not have the resources to correctly categorize race/ethnicity. Hence many SA cancer patients may be classified as "Asian, not otherwise specified" by the hospital.
Due to the above situation, a British developed software program called 'Nam Pehchan' [22] (literally means name identification in Hindi) was used in this study in order to address the issue of misclassification of race/ethnicity. This software is a computer program for the identification of names, which originate in the Indian subcontinent and Sri Lanka, which collectively we call here "South Asia". It provides a reasonably accurate way of identifying people belonging to "South Asian" and "Other" ethnic groups. It also identifies the religious and linguistic origins of the names where possible. Both surnames and forenames can be matched against the program's stored lists. Given the possibility that different elements of a name may meet with varying recognition from the lookup table, the final result is not simply "South Asian" or "not South Asian", but rather a numeric code indicating the outcome of the search and match process. Knowing the limitations of this program [23], we used this software program, as well as birthplace and a visual case-by-case review to correctly identify approximately 5,200 cancer cases of SA origin, from the 106, 653 Asian/Pacific Islander cancer database at CCR, 1988–2000. We identified 30% more SA cases as compared to CCR (CCR identified approximately 4000 SA cases in the same time period).
Calculation of incidence rates
Numerators, comprised of all newly diagnosed cancer cases, were derived by applying the Nam Pehchan software to all cancer cases classified as Asian/Pacific Islander by the CCR, 1988–2000. The numerators were coupled with age, gender and yearly specific denominator data for the SA population in California (population counts) obtained from the U.S. Census Bureau. Detailed population counts and demographic characteristics for SA subgroups for both the 1990 and 2000 decennial census are available from the US Census Bureau [2,24]. Electronic population data by age and sex for all SA subgroups were identified and obtained. Hard copy population data for the California 1990 SA subgroups were also identified and key-entered and are available at the cancer registry. Using these census data sets, interpolation between the two decennial censuses was completed and extrapolation back to the years 1988 and 1989 was completed to create the best estimates of the SA subgroups at risk on an age and sex specific basis. The interpolation and extrapolation was done assuming a linear growth in the SA population subgroups. Finally, the subgroup estimates were combined on an age and sex specific basis for each individual year from 1988–2000 to form one SA population group for each individual year. Using these data age-specific and age-adjusted cancer incidence rates were calculated for the time period 1988–2000. We used the 2000 U.S (5-year groups) population as the standard population.
For purpose of comparison between cancer rates in native Asian Indians (living in India) and SAs in California, we calculated Age Standardized Rates (ASRs), using the world standard for the California SAs and compared them to ASRs in India, obtained from the Globocan 2002 [25]. Globocan is a publication of the International Association of Research for Cancer (IARC), and rates for India are for the time period 1993–1997, and cover eight regional registries in India. We used rates from India as our comparison parameter, as 90% of SAs in the U.S. are of Asian Indian origin. In addition, we calculated incidence rate ratios (IRRs) by taking a ratio of California SA ASRs and Indian ASRs, calculated Confidence Intervals (CIs) and determined the significance [26].
Grouped analysis
Rates for the period of 1988–2000 were divided into 4 time periods by grouping the years of diagnosis into four categories, namely 1988–1991, 1992–1994, 1995–1997 and 1998–2000. Incidence rates were calculated for each of these time periods. We also compared these rates to the Asian/PIs as well as the NHW population of California for the same time periods.
Time Trend analysis
We performed a time trend analysis for each of the cancer sites separately for males and for females, using the 'age-adjusted trend analysis' feature of SEER-STAT [27]. For this purpose, we used the annual data versus the categorized grouped data. We calculated the Annual Percentage Change (APC) (identifies the percent change by computing the slope of the best-fitting regression line around the data points-rates for each individual years in this case) and p-values for APCs.
Results
In total, 5192 cases of cancer were diagnosed in SA population of California between 1988–2000, including 2411 males, and 2781 females. The median age at diagnosis of cancer was 63 years in males and 54 years in females. A comparison of overall age-adjusted invasive cancer incidence rates for the three ethnic groups revealed that the SA average annual incidence rate was 307.5/100,000, compared to 325.2/100,000 for Asian/PI and 489.1/100,000 for NHW (Figure 1). In the recent years the overall invasive cancer rates for California SAs have been higher than those of the Asian/PIs of the state.
Figure 1 Comparison of invasive cancers, all sites combined, in the SA1, NHW2 and API3 population of California, 1988–2000. * Rates are per 100,000 and age-adjusted to the U.S. 2000 standard population (18 age groups). 1 south Asian. 2 White, non-Hispanic. 3 Asian/Pacific Islanders.
Table 1 summarizes the cancer counts by major cancer sites and Figures 2 and 3 show the top five leading cancers and their trends in the SA males and females respectively. Leading cancers in SA males include prostate, colorectal, urinary system, lung and bronchus, and lymphomas. The leading cancer in SA females is breast cancer followed by colorectal, uterine, ovarian and cervical cancer. In this section we have categorized cancers into two groups namely; common cancers (cancers common to males and females) and gender specific cancers (reproductive organ cancers).
Table 1 Cancer counts in the south Asian population of California, by cancer sites, 1988–2000.
Male and female Male Female
All Sites 5,192 2,411 2,781
Oral Cavity and Pharynx 160 95 65
Esophagus 52 26 26
Stomach 128 71 57
Colon and Rectum 471 285 186
Liver and Intrahepatic Bile Duct 104 70 34
Gallbladder 48 12 36
Pancreas 83 47 36
Lung and Bronchus 296 188 108
Skin excluding Basal and Squamous 61 38 23
Breast 981 6 975
Cervix Uteri 270 0 270
Corpus and Uterus 145 0 145
Ovary 155 0 155
Prostate 661 661 0
Testis 34 34 0
Urinary System 244 189 55
Brain and Other Nervous System 141 76 65
Thyroid 148 41 107
Hodgkin Lymphoma 56 32 24
Non-Hodgkin Lymphoma 247 143 104
Myeloma 86 49 37
Leukemia 225 138 87
Figure 2 Top five leading cancers with trends in California south Asian males, 1988–2000. * Rates are per 100,000 and age-adjusted to the U.S. 2000 standard population (18 age groups).
Figure 3 Top five leading cancers with trends in California south Asian females, 1988–2000. * Rates are per 100,000 and age-adjusted to the U.S. 2000 standard population (18 age groups).
Comparison of cancer incidence between California SAs and native Asian Indians
Age standardized rates for California SAs and those for India as well as Incidence Rate Ratios (with statistical significance) are presented in Table 2. IRR of more than one indicates that California SAs are at higher risk of developing that particular cancer than the native Asian Indians. Overall, California SA males and females are at double risk for developing cancer than native Asian Indians.
Table 2 IRRs and site specific cancer ASRs in India and South Asians in California, 1993–1997.
SITE SOUTH ASIAN MALES SOUTH ASIAN FEMALES
India ASR1 California ASR2 IRR3 India ASR California ASR IRR
All Sites 99 199.6 2.0 104.4 195.7 1.9
Oral Cavity 12.8 5.1 0.4 7.5 4.6 0.6
Esophagus 7.6 2.6 0.3 1.9 2.4 1.3
Stomach 5.7 6 1.1 2.8 5.2 1.9
Colon and Rectum 4.7 22.8 4.9 3.2 13.4 4.2
Liver & Intrahepatic Bile Duct 2.3 5.3 2.3 1.1 3.1 2.8
Pancreas 1.4 4.5 3.2 0.8 4.5 5.6
Lung and Bronchus 9 13.9 1.5 2 9.9 5.0
Prostate 4.6 69.9 15.2 ~ ~ ~
Breast ~ ~ ~ 19.1 66.6 3.5
Cervix Uteri ~ ~ ~ 30.7 8.4 0.3
Corpus Uteri ~ ~ ~ 1.7 10 5.9
Ovary ~ ~ ~ 4.9 10.4 2.1
Urinary Bladder 3.2 9.4 2.9 ~ ~ ~
Kidney and Renal Pelvis 1.2 4.9 4.1 0.5 2.4 4.8
Brain & other CNS 2.6 5.4 2.1 1.6 3.6 2.3
Thyroid 1 1.5 1.5 1.9 7.6 4.0
Non-Hodgkin Lymphoma 3.2 11.2 3.5 1.7 10.3 6.1
Myeloma 1 4.6 4.6 0.6 3.1 5.2
Leukemia 3.1 9.9 3.2 1.9 6.8 3.6
1Rates for India obtained from Globocan 2002, rates are an average of 8 regional cancer registries (1993–1997).
2 Age standardised rates (world population), rates are per 100,000 population.
3 Incidence Rate Ratios, ratio of rates in SAs in California vs Indian cancer rates.
IRRs in bold indicate significantly elevated IRRs in California SAs (p < 0.05).
~ indicates cases less than 10 for California SAs, hence rates are not significant enough to report.
Common cancers
California SAs were at lower risk of oropharyngeal and esophageal cancers than the native Asian Indian population, which occur very commonly in India. The California SA population was at higher risk for gastrointestinal cancers (namely colorectal, hepatic, and pancreatic cancers). They were also at higher risk for hematopoietic and lymphoreticular and endocrine malignancies. The SA population of California also experienced a higher risk for other organ systems such as, urinary system and brain & CNS cancers.
Gender specific cancers
SA men experienced 15 fold risk of prostate cancer than the native Asian Indian population. California SA females experienced higher risk of all reproductive organ cancers except cervical cancer.
Comparison of incidence rates between SAs and Asian/PIs of the state of California
Incidence rates and time rends between 1988–2000, for California SAs as well as the Asian/PI and NHW population are presented in Table 3 and Table 4.
Table 3 Comparison of cancer rates1 in the SA2, API3 & NHW4 population of California, 1988–2000.
Males Females
Cancer site time period SA API NHW SA API NHW
oral cavity and pharynx 1988–1991 14.6 13.2 20.6 11 7.1 9
1992–1994 8.3 12.9 19.5 8.2 6.8 8.2
1995–1997 11.7 13.2 18.9 6.4 6 8.1
1998–2000 7.7 11.8 17.7 8 6 7.2
esophagus 1988–1991 7.9 6.3 6.4 2 1.4 2.3
1992–1994 1.9 4.8 6.8 4.1 1.2 2.1
1995–1997 5.7 4.9 7.2 3.9 1.1 2.3
1998–2000 4.1 3.4 7.5 6 1.4 2.3
stomach 1988–1991 16.6 24.9 12.9 4.9 15.9 5.4
1992–1994 8.6 24.9 11.4 7.6 15.5 4.6
1995–1997 11.8 24.2 11 9.8 13.8 4.4
1998–2000 10.8 20.5 9.7 10 12.2 4.4
liver and intrahepatic bile duct 1988–1991 10.4 21.6 4.2 5 7.2 1.9
1992–1994 11 22.6 5 4.3 9.2 2.1
1995–1997 9.8 24.4 5.6 5.3 9.1 2.3
1998–2000 10.4 24.5 6.3 6.4 8.5 2.5
gall bladder 1988–1991 1 1.6 0.8 9.4 3.3 1.2
1992–1994 1.9 1.6 0.7 9.1 2.4 1.1
1995–1997 2.4 1.5 0.6 3.2 2.2 1.1
1998–2000 1.2 1.2 0.6 3.6 1.8 1.1
pancreas 1988–1991 6.3 12.4 13.3 6.6 8.6 10.2
1992–1994 4.4 11 12.8 7.3 6.6 9.9
1995–1997 7.4 9.6 12.5 6.9 7.9 9.4
1998–2000 10.6 8.9 11.8 3.2 7.5 9.5
lung and bronchus 1988–1991 33.5 65.3 101.9 19 28.3 59.3
1992–1994 26.8 64.7 92.6 21.7 26.9 59.5
1995–1997 18.5 60.8 86.5 18 28.4 60
1998–2000 35.2 59 79 27 27.3 57.2
colon and rectum 1988–1991 42.6 57 78.8 24.6 41.9 54.6
1992–1994 36.8 58.5 71.4 21.5 39.6 49.3
1995–1997 46.2 58.3 67.7 24.6 39.1 47.9
1998–2000 35.6 52.4 64.1 34.5 40.1 46.7
brain and other CNS (central nervous system) 1988–1991 8.2 4 9.4 7.4 3.6 6.5
1992–1994 10.5 4.7 9.1 5.7 3.7 6.4
1995–1997 3.7 3.8 8.9 3.9 3.3 6.1
1998–2000 7.3 4 8.8 8.3 2.6 6.1
urinary bladder 1988–1991 22.3 15 43.5 5.2 4.4 10.7
1992–1994 14.8 17 42.3 1.6 4.3 9.9
1995–1997 18 14.8 41.2 3 4.4 10.1
1998–2000 18.6 15.1 40.9 3.2 3.7 9.9
kidney and renal pelvis 1988–1991 8.8 6.6 14.6 2.8 3.8 6.8
1992–1994 6.2 7 14.2 3.3 3.4 7
1995–1997 7.9 8.3 14.8 3.2 3.9 7.3
1998–2000 12.5 8.1 14.8 3.8 3.7 7.2
endocrine (thyroid) 1988–1991 4 4.2 4.2 7.6 9.7 7.7
1992–1994 1.3 4.7 3.9 8.7 11.2 8
1995–1997 2.2 4.6 4.1 9.4 11.5 8.8
1998–2000 3.2 4.3 4.5 8.5 12.4 10.2
non-Hodgkins lymphoma 1988–1991 11 15.8 24.4 10.5 10.6 14.6
1992–1994 13.8 16.7 25.7 16.2 10.6 14.7
1995–1997 15.2 15.9 25.7 14.7 10.7 15.7
1998–2000 21.1 15.7 23.3 11 11.4 15.9
Hodgkins lymphoma 1988–1991 2 1.5 3.6 1.5 0.6 2.8
1992–1994 2.4 1.5 3.4 1.9 0.8 2.7
1995–1997 1.7 1.2 3.4 1.5 0.8 2.8
1998–2000 2.2 1.1 3.3 1.5 1 2.6
leukemias 1988–1991 10.4 10.9 18 9.6 6.6 10.2
1992–1994 13.4 10.4 17.5 11.4 7 9.7
1995–1997 9.6 9.6 16.9 9 6.3 9.8
1998–2000 16.1 9.2 15.4 8.4 5.8 9
multiple myelomas 1988–1991 10.5 4.8 6.3 5.1 3.3 4.2
1992–1994 5.3 4.5 6.4 7.9 2.4 3.9
1995–1997 9.8 5.2 6.4 2.9 3.2 4.1
1998–2000 5.6 3.7 5.7 5.5 3 3.6
skin, excluding basal & squamous cell cancer 1988–1991 8.2 3.9 44.4 2.5 2 20.5
1992–1994 2.7 4.2 45.8 1.1 1.8 21.8
1995–1997 4.4 3.9 50 1.9 1.6 28.3
1998–2000 4 2.8 53.1 2.3 2.5 31.8
1 Rates are per 100,000 and age-adjusted to the U.S. 2000 standard population (5 year age-groups).
2 South Asian.
3 Asian/Pacific Islander.
4 white, non-Hispanic.
Table 4 Comparison of rates1 of reproductive organ cancers in the SA2, API3 & NHW4 population of California.
Cancer site time period SA API NHW
MALES
prostate 1988–1991 112.5 78.2 165
1992–1994 127.8 109.4 197
1995–1997 111.8 90 153.2
1998–2000 110.7 87.9 152.2
testicular 1988–1991 1.5 1.5 6.5
1992–1994 2.2 1.9 6.5
1995–1997 0.8 1.6 6.4
1998–2000 1.9 1.9 7.1
FEMALES
breast (in situ) 1988–1991 8.1 10.2 19.9
1992–1994 14.1 12.3 21.7
1995–1997 13.2 17.1 25.3
1998–2000 14.6 21.6 29.4
breast (invasive) 1988–1991 75.2 76 144
1992–1994 75.5 79 142.3
1995–1997 97.6 87 146.5
1998–2000 98.1 91.9 150.8
ovary 1988–1991 14.3 12.9 19.5
1992–1994 15.1 13.3 19.3
1995–1997 19.4 13.4 18.2
1998–2000 15.1 12.2 18.2
cervix 1988–1991 11.7 14.7 9.3
1992–1994 12.1 15.8 8.6
1995–1997 9.5 13.1 8.4
1998–2000 10 9.8 7.9
uterus and corpus 1988–1991 17.8 13.9 28.1
1992–1994 19 14.8 27.1
1995–1997 11 14.8 27.1
1998–2000 22.2 15.9 25.3
1 Rates are per 100,000 and age-adjusted to the U.S. 2000 standard population (5 year age-groups).
2 south Asian.
3 Asian/Pacific Islander.
4 white, non-Hispanic.
Common cancers
In general, the SA population of California experienced more brain & CNS cancers, hematopoietic and lymphoreticular cancers than the Asian/PI population of the state. SA females also experienced higher oropharyngeal, esophageal and gall bladder cancer than the Asian/PI women of California. As regards to other cancer sites, the SA population of California was at equal or lower risk than the Asian/PIs of the state.
Gender-specific cancers
SA males experienced more prostate cancer than the Asian/PI males and SA females experienced more reproductive organ cancers than the Asian/PI women, except for cervical cancer.
Comparison of SA rates with NHWs of the state
Common cancers
The SA population of California experienced more Gastro Intestinal cancers (mainly hepatic, gall bladder and stomach cancers) and myelomas than the NHW population of the state. SA females experienced more oropharyngeal and esophageal cancers than the NHW women. SA males experienced recent increase in leukemia incidence as compared to the NHW males.
Gender specific cancers
As far as the reproductive cancers are concerned, the SA population was at lower risk of these cancers than the NHW population of the state, except for cervical cancer.
Trends over time
Overall, SA males experienced a decreasing trend of all cancers combined, over the time period of 1988–2000 (APC = -1, p = 0.2), while SA females experienced an increasing trend (APC = 0.9, p = 0.4). Table 5, shows the APCs and their p-values for each individual cancer in the SA population.
Table 5 Table showing annual percentage change along with significance in cancer sites in the SA population of California, 1988–2000.
SITE MALE FEMALE
APC* P-VALUE APC P-VALUE
All Sites -1.0 0.2 0.9 0.4
common cancers
Oral Cavity and Pharynx -3.0 0.5 -4.0 0.1
Esophagus -1.3 0.9 7.2 0.3
Stomach -1.9 0.5 5.6 0.3
Liver and Intrahepatic Bile Duct 2.1 0.7 1.2 0.86
Gallbladder ~ ~ -9.9 0.1
Pancreas 3.5 0.5 -1.2 0.8
Colon and Rectum -1.3 0.3 3.5 0.3
Colon excluding Rectum -3.2 0.1 11.2* <0.05
Lung and Bronchus -2.9 0.3 0.8 0.8
Skin excluding Basal & Squamous -9.3 0.1 ~ ~
Urinary Bladder 0 1 ~ ~
Kidney and Renal Pelvis 8.1 0.2 ~ ~
Brain and other CNS -1.4 0.7
NHL 6 0.1 0.8 0.9
multiple myelomas -2.8 0.6 0.6 0.9
leukemias 2.6 0.5 -0.1 1
Thyroid ~ ~ -1.1 0.8
Reproductive organ cancers
Breast (In situ) ~ ~ 8.8* <0.05
Breast (Inavasive) ~ ~ 2.3 0.2
Cervix Uteri ~ ~ -2.1 0.3
Corpus and Uterus ~ ~ 1 0.8
Ovary ~ ~ -0.2 0.9
Prostate -0.5 0.6 ~ ~
* Annual Percentage change, numbers in red indicate significant p-values.
~ Annaul percentage change could not be calculated.
Common cancers
The SA population of California experienced a significantly decreasing trend of oropharyngeal cancers. On the other hand they experienced an increasing trend of hepatic and renal cancers. In addition, SA males experienced an increasing trend of hematopoietic & lymphoreticular cancers (NHL, multiple myelomas, leukemias) and brain & other CNS cancers. SA females experienced an increasing trend of gastrointestinal cancers (esophageal colon, hepatic, and stomach), lung and thyroid cancers.
Gender-specific cancers
As far as the reproductive organs were concerned, SA females experienced an increasing trend of breast and uterine cancers. All other sites experienced either a decreasing or steady trend over time.
Discussion
The present study reveals several unique cancer patterns among SAs in California. Firstly, the median age at diagnosis of cancer in this population is 58 years compared to 68 years for all other races [28]. Secondly, the most common cancers in the Indian subcontinent are not the most common cancers in SAs of California. The most common cancers among men in India are oral cavity and pharynx, lung, esophagus, laryngeal and stomach cancers [25]. In India, cervical cancer is most common in women, followed by breast, oral cavity, esophagus and ovarian cancer [25]. In India about half the cases among men and one-fifth cases among women are in cancer sites affected by tobacco use (tobacco smoking as well as tobacco chewing) [29], which was not seen in SAs of California.
Common cancers (cancers common to both males and females)
Oropharyngeal cancers
Our findings indicate that California SAs are at lower risk of oral and esophageal cancers than the native Asian Indians. This directly reflects the general tendency of the SA immigrants to avoid use of tobacco products (especially chewing 'paan' (tobacco rolled up in betel nut leaves) and smoking 'bidi' (cigarette made out of tobacco leaves, with no filters) in a foreign country. Besides, majority of SA immigrants in California tend to be educated and do not have such habits even in South Asia.
Esophagus cancer
Esophageal cancer is increasing in SA females and is higher than both NHW and Asian/PI females. Such findings of increasing trend are not seen in the SA males. This finding also seems contradictory to the general decreasing trend of oropharyngeal cancers, as esophageal and oropharyngeal cancers share similar etiologies. The etiology of esophageal cancer is mainly associated with consumption of tobacco (smoking or smokeless) and alcohol. In addition Barrett's esophagus, diet and nutrition, reflux disease also play an important role in etiology of esophageal cancer [30,31]. There are no published studies about smoking/tobacco/alcohol use prevalence in the SA population in the U.S. Because of lack of such data we cannot correlate our findings with the smoking prevalence. The rise of esophageal cancer in California SA females as well as histological subtype evaluation of this cancer is needed.
Stomach cancer
IRRs suggest that California SA females are at a higher risk for stomach cancer than native Asian Indian females, but this is not true for males. The time trend analysis suggests that male stomach cancer is decreasing, but female stomach cancer is on the rise. Infections with Helicobacter pylori and genetic predisposition of host have been suggested to be the most important causes of stomach cancer in general population [32,33].
Cancers of the liver and intrahepatic bile duct
These cancers are of common occurrence in Asians. HBV (hepatitis B virus) infection, with and without aflatoxin exposure, and alcoholic liver cirrhosis are responsible for most cases of hepatocellular cancer in developing countries [34]. There is widespread contamination of foods with aflatoxin and moderately high prevalence of HBV and hepatitis C (HCV) virus-related chronic liver disease in India [35]. IRRs suggest that California SA population is at higher risk (more than two-fold) of hepatic cancers than native Asian Indians. Our findings are similar to the studies done in the past in UK on migrants of Indian ethnicity as well as British ethnicity, to the UK [16,36].
Gall bladder cancer
The major causative factors for gall bladder cancer include gallstones and genetic susceptibility, and liver flukes in Asian countries have also been suggested to be causative [37]. In one study done in India, the prevalence of gallstones in adult population was 6.12% (3.07% in males, 9.6% in females) [38]. All these above stated factors could explain our finding of much higher rates in the SA population than Asian/PI or NHW population. Similar findings have been reported by studies in SA immigrants to the UK [12,36,39]. Nevertheless, it is encouraging that there is a significantly decreasing trend of this cancer in California SAs.
Colon and rectal cancer
Both SA males and females of California experienced more than four-fold risk of developing this cancer compared to the native Asian Indian population. Studies in the general population estimate that 13% of this cancer can be attributed to being physically inactive, 12% to eating a Western style diet, and 8% to having a first degree relative with colorectal cancer [40]. The diet of Asian Indians in the United States has changed from one featuring low-fat, high-fiber foods to one characterized by higher-fat animal protein, low fiber, and high levels of saturated fat. There is an increased tendency among Asian Indians in America to consume fast foods and convenience foods [41]. The significantly rising trend of colon cancers seen in SA females, which is otherwise a low-risk population, may be related to migration and subsequent acculturation and adoption of Western diet and lifestyle.
Lung and bronchus cancers
As compared to the native Asian Indian rates, the SAs of California are at higher risk for this cancer. The five-fold risk in California SA females as compared to the native Asian Indian females and an increasing trend is noteworthy. A decreasing trend of lung cancer in SA males is not in concordance with a recent study done in the UK SA population, which reports recent increase in incidence of lung cancer in both SA men as well as women [7].
Non-Hodgkin's Lymphomas (NHL)
IRRs suggest that the California SAs are at a much higher risk (3–6 fold higher risk) of developing NHL than native Asian Indians. In addition, an increasing trend of NHL has been observed in the SA population of California. While the incidence of NHL has doubled in the U.S., etiology of lymphomas remains elusive. Epidemiological studies suggest the role of hereditary factors, immunosuppression, viruses (HIV, EBV, HTLV, H.pylori, HHV8, HCV), chemical and agricultural exposures and other factors in the etiology of NHL [42]. Recent studies have also associated menstrual and reproductive factors (higher parity and early menarche offer a protective effect for NHL) with risk of NHL [43,44]. Lack of immune stimulation/challenge ('hygiene hypothesis') [45] and acculturation could explain the higher risk seen in this population.
Leukemias
Three-fold higher risk of developing leukemias in California SAs as compared to the native Asian Indians, and a rising trend of this cancer over time shows similarity with results from UK SA studies [14,16,19]. Types of leukemias and their causes vary widely and are age dependant. Further investigation, especially age specific and leukemia subtype analysis is needed into this finding.
Multiple myelomas
IRRs suggest that California SA population experience a much higher risk (four-five fold) of developing myelomas than the native Asian Indians, as well as higher rates than the Asian/PI or NHW population of California. Risk factors for multiple myelomas include, monoclonal gammopathy of unknown significance, chronic immune stimulation (as in infections with tuberculosis, malaria, hepatitis, etc.), autoimmune disorders, and occupational exposures [46]. Every year, approximately two million persons in India develop tuberculosis, and incidence of malaria is 2–3 million cases per year [47,48]. Exposure to these chronic diseases before migration could explain the high rates of myelomas seen in California SAs.
Findings of elevated risk of haematopoietic and lymphoreticular malignancies (lymphomas, leukemias and myelomas) in SAs after migration needs further investigation. Similar results have been reported in SA immigrants of UK [12,16,36].
Thyroid cancer
IRRs indicate that California SA females are four times more likely to get thyroid cancer than Indian females; this is not true in males. The incidence of congenital hypothyroidism and prevalence of goiter in India is much higher than the worldwide average [49]. A large fraction of the Indian population suffers from iodine deficiency disorders [50]. The major etiological factors for thyroid cancers have been iodine deficiency and ionizing radiation [51-53]. We cannot explain the higher IRR observed in California SA females.
Brain and other nervous system cancers
California SAs experienced higher IRRs of these malignancies as compared to native Asian Indians. SA males experience higher rates of these malignancies than the Asian/PIs and SA females recently experienced higher rates than Asian/PIs as well as NHWs. This finding is not in concordance with the other studies done in the UK SA population [14,16]. These cancers are infrequent in India and frequent amongst the U.S. Whites, making the SA population a low-risk population [54,55]. In spite of being a low-risk population, higher IRRs and rates of these cancers observed in SAs need further investigation.
Gender-specific cancers
Prostate cancer
Prostate cancer is the most common cancer in SA males and has increased from 1988–2000. California SA males experienced fifteen fold-increased risk of this cancer as compared to Indian males. Also, rates are higher in California SA males than in Asian/PIs of California. Epidemiological studies suggest that endogenous risk factors like family history, androgens, race, aging, oxidative stress and exogenous factors including diet and environmental agents have been associated with this cancer [56]. Other studies suggest that screening for this cancer has dramatically increased the number of men with local disease at diagnosis [57]. The fifteen-fold risk of prostate cancer in this population as compared to the native Asian Indians could be explained by early detection (measurement of serum PSA), rather than true differences in underlying risk. The other factors explaining this difference could be lead-time, case identification, detection and reporting biases.
Breast cancer
Breast cancer is the number one cancer in the California SA females and they are 3.5 times more likely to develop this cancer as compared to native Asian Indian females. Our time-trend analysis suggests that, although in situ breat cancer diagnosis has significantly increased, invasive breast cancer diagnosis has increased alarmingly more in SA than in Asian/PIs and NHWs. In the general population major risk factors include, late maternal age at first parity (>30 years of age), having one child vs. 4, use of oral contraceptives (OCs), use of hormone replacement therapy (HRT), obesity and alcohol [58-60]. Adoption of above-mentioned lifestyle practices by SA women and inadequate screening could be related to the increase in breast cancer in this population.
Cervical Cancer
Although HPV has been proposed as the first identified necessary cause of cervical cancer [61,62], we attribute the decreasing trend and very low IRRs of cervical cancer in California SA women to screening success. California SA women are getting screened at very early stages and hence treated completely as compared to the Indian women (cervical cancer ranks number one in India).
Ovarian and uterine cancers
Risk factors for epithelial ovarian cancer include older age, being White, positive family history, nulliparity, infertility, and obesity (high saturated fat and carbohydrate intake), postmenopausal HRT and use of cosmetic talc. Conversely, preventive factors include OC use, vegetable consumption, gravidity, lactation, tubal ligation, and hysterectomy. Genetic influence also plays a role, women with mutations in the BRCA1 or BRCA2 genes having an elevated risk [63-65]. Rates of this cancer in the SA women are higher than the ASIAN/PIs and almost approximating those of NHWs. Almost two-fold elevated risk of ovarian cancer in California SA women compared to native Asian Indian women can be explained by all the adoption of above mentioned western life-style factors.
Similarly uterine/endometrial cancer is a disease of the developed world. Epidemiological studies have shown that majority of the incidence can be attributed to excess body weight (in turn due to 'unopposed estrogens'), lack of physical activity, exogenous hormones and chronic hyperinsulinemia along with genetic predisposition [66,67]. California SA women face a five-fold risk of this cancer as compared to the native Asian Indians and they show much higher rates than the Asian/PIs and their rates seem to be fast approaching the NHWs of the state. Clearly, acculturation can explain these findings.
Limitations
Certain limitations in the methods employed in this study deserve comment. The assumption of a linear growth of population may not be completely tenable, and various factors such as birth/death rates and immigration/migration related issues could impact patterns of population growth. However, communication with staff at the Los Angeles County Cancer Surveillance program indicated that interpolation performed well when compared to more complex methods of estimation based on year/race/ethnicity/sex and county specific population estimates obtained from the California Department of Finance (state agency in California charged with maintaining intercensal population figures) [personal communication, Dr. Lihua Liu, USC/CSP, December, 2003].
While performing incidence studies on sub-ethnic populations, the issue of small number of cases is inevitable. This could create instability of rates, especially in analyses pertaining to trends over time. To overcome this, forming groups and performing a grouped analysis in those groups was completed.
Conclusion
Our findings are in general agreement with studies completed in the UK and suggest a strong role for acculturation, screening and lifestyle factors in explaining the patterns of cancer in SA in California. Minor disagreements with findings in UK studies are to be expected, as there are minor underlying differences in methodology. For example, some studies have used absolute numbers for comparison or a proportionate approach for comparison. But most of the studies have reported incidence rates based on data available from the cancer registries and census bureaus/corresponding organizations in UK (with whom we have compared our data).
More studies are needed to evaluate gender differences in this population, especially the rising trend of gastrointestinal cancers seen in SA females vs. males, needs more investigation. Our study also reveals the need for additional screening measures and early diagnosis in this population. Our overall impression is that, if measures are not taken to improve screening, and curb smoking in this population and if the current conditions prevail, the rates of colon, lung, and breast cancer in the SA population will approximate those of California NHWs.
We have presented a general picture of cancer in the SA population in this paper. It is beyond the scope of this paper to discuss subtypes of each cancer. Hence we conclude that more studies are needed on this issue and subtype analysis of cancer sites needs to be conducted.
List of Abbreviations used
SA: south Asian
NHW: non-Hispanic White
API: Asian/Pacific Islanders
APC: Annual percentage change
CCR: California Cancer Registry
Authors' contributions
RVJ conceptualized and designed the study, as well as carried out the data analysis and prepared the manuscript. PKM was responsible for the study design, acquisition of funding and data, as well as interpretation of the data. APP helped prepare the manuscript and gave technical advice.
==== Refs
U.S. Census Bureau. The Asian Population by detailed group: 2000
1990 Census of Population: General Population Characteristics, California
U.S. Census Bureau. Public Use Presentation Library:The Asian Population: 2000 (September 2002)
Blesch KS Davis F Kamath SK A comparison of breast and colon cancer incidence rates among native Asian Indians, US immigrant Asian Indians, and whites J Am Diet Assoc 1999 99 1275 1277 10524396 10.1016/S0002-8223(99)00313-2
Divan HA Cancer incidence and mortality in Asian Indians: a review of literature from the United States, South Asia, and beyond Asian Am Pac Isl J Health 2003 10 73 85 15509148
Islam N Zojwalla N Cancer A Brown Paper: The Health of South Asians in the United States SAPHA, October 2002
Smith LK Peake MD Botha JL Recent changes in lung cancer incidence for south Asians: a population based register study Bmj 2003 326 81 82 12521972 10.1136/bmj.326.7380.81
Velikova G Booth L Johnston C Forman D Selby P Breast cancer outcomes in South Asian population of West Yorkshire Br J Cancer 2004 90 1926 1932 15138473 10.1038/sj.bjc.6601795
dos Santos Silva I Mangtani P De Stavola BL Bell J Quinn M Mayer D Survival from breast cancer among South Asian and non-South Asian women resident in South East England Br J Cancer 2003 89 508 512 12888822 10.1038/sj.bjc.6601097
Farooq S Coleman MP Breast cancer survival in South Asian women in England and Wales J Epidemiol Community Health 2005 59 402 406 15831690 10.1136/jech.2004.030965
Smith LK Botha JL Benghiat A Steward WP Latest trends in cancer incidence among UK South Asians in Leicester Br J Cancer 2003 89 70 73 12838302 10.1038/sj.bjc.6600973
Winter H Cheng KK Cummins C Maric R Silcocks P Varghese C Cancer incidence in the south Asian population of England (1990-92) Br J Cancer 1999 79 645 654 10027344 10.1038/sj.bjc.6690102
Flavell KJ Biddulph JP Powell JE Parkes SE Redfern D Weinreb M Nelson P Mann JR Young LS Murray PG South Asian ethnicity and material deprivation increase the risk of Epstein-Barr virus infection in childhood Hodgkin's disease Br J Cancer 2001 85 350 356 11487264 10.1054/bjoc.2001.1872
McKinney PA Feltbower RG Parslow RC Lewis IJ Glaser AW Kinsey SE Patterns of childhood cancer by ethnic group in Bradford, UK 1974-1997 Eur J Cancer 2003 39 92 97 12504664 10.1016/S0959-8049(02)00488-4
McKinney PA Feltbower RG Parslow RC Lewis IJ Picton S Kinsey SE Bailey CC Survival from childhood cancer in Yorkshire, U.K.: effect of ethnicity and socio-economic status Eur J Cancer 1999 35 1816 1823 10673997 10.1016/S0959-8049(99)00173-2
Cummins C Winter H Maric R Cheng KK Silcocks P Varghese C Batlle G Childhood cancer in the south Asian population of England (1990-1992) Br J Cancer 2001 84 1215 1218 11336473 10.1054/bjoc.2001.1775
Ahmad F Cameron JI Stewart DE A tailored intervention to promote breast cancer screening among South Asian immigrant women Soc Sci Med 2005 60 575 586 15550305 10.1016/j.socscimed.2004.05.018
Ahmad F Stewart DE Predictors of clinical breast examination among South Asian immigrant women J Immigr Health 2004 6 119 126 15269515 10.1023/B:JOIH.0000030227.41379.13
Powell JE Parkes SE Cameron AH Mann JR Is the risk of cancer increased in Asians living in the UK? Arch Dis Child 1994 71 398 403 7826109
Morris C Cohen R Perkins C Allen M Kwong S Schlag R Wright W Cancer in California, 1988-1996. 1999 Sacramento, CA, California Health Department Of Health Services, Cancer Surveillance Section
Surveillance, Epidemiology, and End Results Program (1992). SEER Extent of Disease -1988 Coding Instructions, Second Edition. Bethesda, MD: National Cancer Institute NIH Pub No 92-2313
Nam Pehchan 2002 Version 2.1 Bradford, City of Bradford Metropolitan District Council and Bradford Health Authority
Cummins C Winter H Cheng KK Maric R Silcocks P Varghese C An assessment of the Nam Pehchan computer program for the identification of names of south Asian ethnic origin J Public Health Med 1999 21 401 406 11469361 10.1093/pubmed/21.4.401
United States Census Bureau, Census 2000
Ferlay J Bray F Pisani P Parkin DM GLOBOCAN 2002: Cancer Incidence, Mortality and Prevalence Worldwide.
Breslow NE Day NE Statistical methods in cancer research. Volume II--The design and analysis of cohort studies IARC Sci Publ 1987 1 406
Surveillance Research Program, National Cancer Institute SEER*Stat software Version 5.3.1
Kwong S Perkins CI Morris CR Cohen R Allen M Schlag R Wright WE Cancer in California: 1988-1998. 2000 Sacramento, California Health Department of Health Services, Cancer Surveillance Section
Gajalakshmi V Shanta V Swaminathan R Cancer Registration in India. Asian Pacific J Cancer Prev, 2, IACR Supplement, 13-20
Blot WJ McLaughlin JK Devesa SS Fraumeni JF Jr Schottenfeld D, Fraumeni JF Jr Cancers of the oral cavity and pharynx Cancer Epidemiology and Prevention 1995 Second 666 680
Enzinger PC Mayer RJ Esophageal cancer N Engl J Med 2003 349 2241 2252 14657432 10.1056/NEJMra035010
Correa P The biological model of gastric carcinogenesis IARC Sci Publ 2004 301 310 15055303
Nardone G Morgner A Helicobacter pylori and gastric malignancies Helicobacter 2003 8 Suppl 1 44 52 14617217 10.1046/j.1523-5378.2003.00160.x
El-Serag HB Hepatocellular carcinoma: an epidemiologic view J Clin Gastroenterol 2002 35 S72 8 12394209 10.1097/00004836-200211002-00002
Dhir V Mohandas KM Epidemiology of digestive tract cancers in India. III. Liver Indian J Gastroenterol 1998 17 100 103 9695391
Swerdlow AJ Marmot MG Grulich AE Head J Cancer mortality in Indian and British ethnic immigrants from the Indian subcontinent to England and Wales Br J Cancer 1995 72 1312 1319 7577489
Fraumeni JF JrDevesa SS McLaughlin JK Stanford JL Schottenfeld D, Fraumeni JF Jr Biliary tract cancer Cancer Epidemiology and Prevention 1995 Second 794 805
Khuroo MS Mahajan R Zargar SA Javid G Sapru S Prevalence of biliary tract disease in India: a sonographic study in adult population in Kashmir Gut 1989 30 201 205 2649414
Barker RM Baker MR Incidence of cancer in Bradford Asians J Epidemiol Community Health 1990 44 125 129 2370499
Slattery ML Diet, lifestyle, and colon cancer Semin Gastrointest Dis 2000 11 142 146 10950460
Kulkarni KD Food, culture, and diabetes in the United States Clinical Diabetes 2004 22 190 192
Fisher SG Fisher RI The epidemiology of non-Hodgkin's lymphoma Oncogene 2004 23 6524 6534 15322522 10.1038/sj.onc.1207843
Nelson RA Levine AM Bernstein L Reproductive factors and risk of intermediate- or high-grade B-Cell non-Hodgkin's lymphoma in women J Clin Oncol 2001 19 1381 1387 11230482
Zhang Y Holford TR Leaderer B Boyle P Zahm SH Zhang B Zou K Morton LM Owens PH Flynn S Tallini G Zheng T Menstrual and reproductive factors and risk of non-Hodgkin's lymphoma among Connecticut women Am J Epidemiol 2004 160 766 773 15466499 10.1093/aje/kwh278
Vijh AK Inverse trend between estimated worldwide frequency of major cancers and inferred infectious burdens of populations: possible role of adaptive immune system Med Hypotheses 2004 62 880 888 15142641 10.1016/j.mehy.2003.12.026
Herrington LJ Weiss NS Olshan AF Schottenfeld D, Fraumeni JF Jr Multiple myeloma Cancer Epidemiology and Prevention 1995 Second 946 970
Lal S Dhillon GP Aggarwal CS Epidemiology and control of malaria Indian J Pediatr 1999 66 547 554 10798110
Progress toward tuberculosis control--India, 2001 MMWR Morb Mortal Wkly Rep 2002 51 229 232 11925019
Desai MP Disorders of thyroid gland in India Indian J Pediatr 1997 64 11 20 10771808
Sood A Pandav CS Anand K Sankar R Karmarkar MG Relevance and importance of universal salt iodization in India Natl Med J India 1997 10 290 293 9481104
Przybylik-Mazurek E Szybinski Z [Biological risk factors for differentiated thyroid cancer] Wiad Lek 2001 54 Suppl 1 157 162 12182020
Szybinski Z Huszno B Rachtan J Przybylik-Mazurek E Buziak-Bereza M Trofimiuk M [Epidemiology of thyroid cancer in Poland] Wiad Lek 2001 54 Suppl 1 106 116 12182013
Lawal O Agbakwuru A Olayinka OS Adelusola K Thyroid malignancy in endemic nodular goitres: prevalence, pattern and treatment Eur J Surg Oncol 2001 27 157 161 11289751 10.1053/ejso.2000.1085
Preston-Martin S Descriptive epidemiology of primary tumors of the brain, cranial nerves and cranial meninges in Los Angeles County Neuroepidemiology 1989 8 283 295 2586698
Preston-Martin S Mack W.J Schottenfeld D, Fraumeni JF Jr Neoplasms of the nervous system Cancer Epidemiology and Prevention 1995 Second 1231 1281
Bostwick DG Burke HB Djakiew D Euling S Ho SM Landolph J Morrison H Sonawane B Shifflett T Waters DJ Timms B Human prostate cancer risk factors Cancer 2004 101 2371 2490 15495199 10.1002/cncr.20408
Brawley OW Knopf K Merrill R The epidemiology of prostate cancer part I: descriptive epidemiology Semin Urol Oncol 1998 16 187 192 9858324
Vatten L [Female sex hormones increase the risk of breast cancer] Tidsskr Nor Laegeforen 1998 118 2969 2974 9748837
Wenten M Gilliland FD Baumgartner K Samet JM Associations of weight, weight change, and body mass with breast cancer risk in Hispanic and non-Hispanic white women Ann Epidemiol 2002 12 435 434 12160603 10.1016/S1047-2797(01)00293-9
Singletary KW Gapstur SM Alcohol and breast cancer: review of epidemiologic and experimental evidence and potential mechanisms JAMA 2001 286 2143 2151 11694156 10.1001/jama.286.17.2143
Liu T Wang X Waterbor JW Weiss HL Soong SJ Relationships between socioeconomic status and race-specific cervical cancer incidence in the United States, 1973-1992 J Health Care Poor Underserved 1998 9 420 432 10073217
Bosch FX de Sanjose S Human papillomavirus in cervical cancer Curr Oncol Rep 2002 4 175 183 11822990
Sagae S Mori M [Ovarian cancer] Gan To Kagaku Ryoho 2001 28 179 183 11242643
Sagae S Mori M Moore MA Risk Factors for Ovarian Cancers: Do Subtypes Require Separate Treatment in Epidemiological Studies? Asian Pac J Cancer Prev 2002 3 5 16 12718602
Goodman MT Howe HL Descriptive epidemiology of ovarian cancer in the United States, 1992-1997 Cancer 2003 97 2615 2630 12733127 10.1002/cncr.11339
Akhmedkhanov A Zeleniuch-Jacquotte A Toniolo P Role of exogenous and endogenous hormones in endometrial cancer: review of the evidence and research perspectives Ann N Y Acad Sci 2001 943 296 315 11594550
Kaaks R Lukanova A Kurzer MS Obesity, endogenous hormones, and endometrial cancer risk: a synthetic review Cancer Epidemiol Biomarkers Prev 2002 11 1531 1543 12496040
|
16283945
|
PMC1315316
|
CC BY
|
2021-01-04 16:39:19
|
no
|
J Carcinog. 2005 Nov 10; 4:21
|
utf-8
|
J Carcinog
| 2,005 |
10.1186/1477-3163-4-21
|
oa_comm
|
==== Front
J Inflamm (Lond)Journal of Inflammation (London, England)1476-9255BioMed Central London 1476-9255-2-161631646710.1186/1476-9255-2-16ResearchCharacterization of Toll-like receptors in primary lung epithelial cells: strong impact of the TLR3 ligand poly(I:C) on the regulation of Toll-like receptors, adaptor proteins and inflammatory response Ritter Mirko [email protected] Detlev [email protected] Andreas [email protected] Peter [email protected] Department of Pulmonary Research, Boehringer-Ingelheim Pharma GmbH & Co. KG, Birkendorfer Straβe, 88937 Biberach a.d. Riss, Germany2005 29 11 2005 2 16 16 4 8 2005 29 11 2005 Copyright © 2005 Ritter et al; licensee BioMed Central Ltd.2005Ritter 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
Bacterial and viral exacerbations play a crucial role in a variety of lung diseases including COPD or asthma. Since the lung epithelium is a major source of various inflammatory mediators that affect the immune response, we analyzed the inflammatory reaction of primary lung epithelial cells to different microbial molecules that are recognized by Toll-like receptors (TLR).
Methods
The effects of TLR ligands on primary small airway epithelial cells were analyzed in detail with respect to cytokine, chemokine and matrix metalloproteinase secretion. In addition, the regulation of the expression of TLRs and their adaptor proteins in small airway epithelial cells was investigated.
Results
Our data demonstrate that poly(I:C), a synthetic analog of viral dsRNA, mediated the strongest proinflammatory effects among the tested ligands, including an increased secretion of IL-6, IL-8, TNF-α, GM-CSF, GRO-α, TARC, MCP-1, MIP-3α, RANTES, IFN-β, IP-10 and ITAC as well as an increased release of MMP-1, MMP-8, MMP-9, MMP-10 and MMP-13. Furthermore, our data show that poly(I:C) as well as type-1 and type-2 cytokines have a pronounced effect on the expression of TLRs and molecules involved in TLR signaling in small airway epithelial cells. Poly(I:C) induced an elevated expression of TLR1, TLR2 and TLR3 and increased the gene expression of the general TLR adaptor MyD88 and IRAK-2. Simultaneously, poly(I:C) decreased the expression of TLR5, TLR6 and TOLLIP.
Conclusion
Poly(I:C), an analog of viral dsRNA and a TLR3 ligand, triggers a strong inflammatory response in small airway epithelial cells that is likely to contribute to viral exacerbations of pulmonary diseases like asthma or COPD. The pronounced effects of poly(I:C) on the expression of Toll-like receptors and molecules involved in TLR signaling is assumed to influence the immune response of the lung epithelium to viral and bacterial infections. Likewise, the regulation of TLR expression by type-1 and type-2 cytokines is important considering the impact of exogenous and endogenous TLR ligands on Th1 or Th2 driven pulmonary inflammations like COPD or asthma, respectively.
==== Body
Background
In addition to its barrier function, the airway epithelium plays an important role for the immune response in the lung [1]. Thus, the lung epithelium is a major source of cytokine, chemokines and other inflammatory mediators that affect the adaptive and innate immune response and therefore modulate inflammatory diseases like COPD or asthma [2]. The innate immune functions of the lung epithelium are critical for the local host defense providing protection of the airways and lung parenchyma from microbial colonization and infections. Besides the induction of acute or chronic pulmonary inflammation, increased bacterial load and viral infections of the lung lead to severe exacerbations of diseases like COPD [3,4] or asthma [5] most probably due to a biased release of inflammatory mediators.
The family of Toll-like receptors (TLR) plays a key role in pathogen recognition and induction and regulation of the innate and adaptive immune response. Airway epithelial cells express TLRs and activation of TLRs on epithelial cells has been shown to induce the production of several cytokines, chemokines and antimicrobial peptides [6-8]. The importance of TLRs for the host defense in the lung has been demonstrated by the increased susceptibility of TLR knockout mice towards viral or bacterial infections. For example, TLR2 deficient mice have been shown to be highly susceptible to infection by Staphylococcus aureus, Borrelia burgdorferi and Streptococcus pneumonia [9,10]. In addition to their prominent role in host response and innate immunity, TLRs play an important role for the adaptive immune system and the regulation of a Th1/Th2 balance [11]. This is thought to have a strong impact for Th2 biased allergic disease like asthma. Whereas the TLR1/TLR2 specific agonist Pam3CSK4 and high levels of the TLR4 ligand E. coli LPS have beneficial effects in asthma animal models most probably due to re-equilibrium of the cytokine pattern and the induction of a Th1 response [12-14], low dose of LPS and the TLR2 ligand peptidoglycan bias the immune response toward a Th2 phenotype and lead to aggravation of experimental asthma [12,15]. In Th1 associated inflammatory disease like COPD, activation of TLRs either by exogenous or endogenous ligands is likely to result in disease exacerbations due to a biased proinflammatory response.
In the current study, we aimed to get a better understanding of the role of TLRs in airway epithelial cells and the consequences for pulmonary inflammatory diseases. Therefore, we performed a detailed analysis of cytokine and chemokine secretion by primary small-airway epithelial cells (SAEC) induced by the activation with different TLR ligands. We also analyzed the matrix metalloproteinase (MMP) release of the stimulated SAEC. Furthermore, the regulation of the expression of TLRs and their adaptor proteins in primary airway epithelial cells (SAEC) stimulated with TLR ligands or Th1 and Th2 cytokines was investigated. We could demonstrate that among the different TLR ligands evaluated poly(I:C), an analog of viral dsRNA and a ligand for TLR3, is the most potent proinflammatory stimulus for lung epithelial cells regarding the secretion of cytokines, chemokines and MMPs. Poly(I:C) induced an increased expression of its receptor TLR3 and has pronounced effects on the expression of other TLRs and proteins involved in TLR signaling in SAEC. In addition, we demonstrate that the expression of TLRs and their signaling proteins in SAEC is strongly regulated by type-1 and type-2 cytokines. These findings are thought to have a major effect on the impact of bacterial and viral infections in type-1 or type-2 biased pulmonary inflammations like COPD or asthma, respectively. In summary, our results provide further insight in the role of TLRs in the immune response of the lung epithelium to viral and bacterial infections and their contribution to virus and bacteria induced exacerbations of pulmonary diseases.
Methods
Antibodies
Polyclonal antibodies to TRIF and TIRAP and mAb to TOLLIP were obtained from Alexis (Lausen, Switzerland). Polyclonal antibodies to IRAK-2 and IRAK-3 were purchased from Chemicon (Temecula, CA). Polyclonal antibodies to IRAK-1 and IRAK-4 were from Active Motif (Rixensart, Belgium) or Upstate (Waltham, MA), respectively. The polyclonal anti-MyD88 antibody was obtained from Santa-Cruz Biotechnology (Santa-Cruz, CA). The anti-TLR1 (GD2.F4) mAb, anti-TLR2 mAb (TL2.3), anti-TLR3 mAb (TLR3.7) used for immunofluorescence was purchased from Alexis. The anti-TLR5 mAb and anti-TLR6 mAb (86B1153) were obtained from Imgenex (San Diego, CA) and Kamiya (Seattle, WA), respectively. The neutralizing goat polyclonal anti-IFN-β antibody was from R&D Systems (Wiesbaden, Germany).
Cell Culture
Primary SAEC (normal human small airway epithelial cells) as well as all the basal media and growth supplements were obtained from Clonetics (San Diego, CA). Cells were cultivated according to the instructions of the manufacturer on plastic dishes or flasks (BD Bioscience, Heidelberg, Germany). Passage number was kept to less than four passages from original stocks. SAEC cells were maintained in small airway epithelial cell basal medium (SAGM) supplemented with 52 μg/ml bovine pituitary extract, 0.5 ng/ml human recombinant EGF, 0.5 μg/ml hydrocortisone, 0.5 μg/ml epinephrine, 10 μg/ml transferrin, 5 μg/ml insulin, 0.1 ng/ml retinoic acid (RA), 6.5 ng/ml triiodothyronine, 50 μg/ml Gentamicin/Amphotericin-B (GA-1000) and 50 μg/ml fatty acid-free bovine serum albumin (BSA). 24 h before stimulation of the cells medium was replaced by basal medium only supplemented with GA-1000, RA and BSA. SAECs of three different donors were used for the study.
Stimulation of small-airway epithelial cells (SAEC)
All cytokines were obtained from R&D Systems (Wiesbaden, Germany). The following cytokine concentrations were used for stimulation of SAECs: 50 ng/ml TNF-α, 10 ng/ml IL-1β, 10 ng/ml IL-4, 20 ng/ml IL-13 or 20 ng/ml IFN-γ.
To evaluate the effects of TLR activation, SAEC were stimulated with 5 μg/ml E. coli 0111:B4 LPS (ultrapure), 10 μg/ml S. aureus peptidoglycan (PGN), 10 μg/ml S. cerevisiae zymosan, 200 ng/ml MALP-2 (macrophage activating lipopeptide-2), 200 ng/ml Pam3CSK4 (palmitoyl-3-cysteine-serine-lysine-4) (all obtained from Invivogen, (San Diego, CA), 10 μg/ml poly(I:C) (SIGMA, Munich, Germany) or 20 ng/ml S. muenchen flagellin (Calbiochem, Darmstadt, Germany).
24 h before stimulation of the cells, medium was replaced by basal medium only supplemented with GA-1000, retinoic acid and BSA. Subsequently, cells were incubated for 6 h or 24 h with the indicated stimuli in basal SAGM medium supplemented with GA-1000, retinoic acid and BSA.
For blocking experiments, cells were pre-incubated with a functional blocking anti-TLR3 mAb (clone TLR3.7; 20 μg/ml) [16], an goat anti-IFN-β antibody (1 – 5 μg/ml), an isotype control IgG1 (20 μg/ml) or a goat IgG (1 – 5 μg/ml) for 1 h and subsequently stimulated with 5 μg/ml poly(I:C) for 6 h. IP-10 and IFN-β secretion were measured using an IP-10 or IFN-β ELISA (R&D Systems), respectively.
RNA preparation
RNA extraction from cells was carried out according to the manufacturer's instructions using the RNeasy Mini Kit (Qiagen, Hilden, Germany). Purity and integrity of the extracted RNA was assessed on the Agilent 2100 bioanalyzer with the RNA 6000 Nano LabChip reagent set (Agilent Technologies, Palo Alto, CA).
Real-time quantitative RT-PCR
Primers and TaqMan probes for human TLR1 – TLR5 and TRAM (table 1) were designed using the PrimerExpress Software 2.0 (Applied Biosystems, Darmstadt, Germany). The probes used for detection in real-time PCR were labeled with 6-carboxyfluorescein (FAM) at their 5'-terminal and were quenched with 6-carboxytetramethylrhodamine (TAMRA) on their 3'-terminal. Primers and TaqMan probes for human β-actin, TLR6 – TLR10, IRAK-1, IRAK-2, IRAK-3, IRAK-4, TRIF, TIRAP, TOLLIP, I-TAC, IP-10 and MyD88 were obtained from Applied Biosystems (Assay-On-Demand). GAPDH mRNA levels were measured using a JOE labeled human GAPDH TaqMan PDAR endogenous control reagent kit (Applied Biosystems).
Table 1 Primers and probes used for quantitative real-time RT-PCR
Oligonucleotide Sequence
TLR1 (FP) 5'-CCCATTCCGCAGTACTCCATT-3'
TLR1 (RP) 5'-TTTCCTTGGGCCATTCCA-3'
TLR1 (TP) 5'-CAGTTATCACAAGCTCAAAAGTCTCATGGCCA-3'
TLR2 (FP) 5'-TGTGAAGAGTGAGTGGTGCAAGT-3'
TLR2 (RP) 5'-ATGGCAGCATCATTGTTCTCAT-3'
TLR2 (TP) 5'-TGAACTGGACTTCTCCCATTTCCGTCTTTT-3'
TLR3 (FP) 5'-CCTGGTTTGTTAATTGGATTAACGA-3'
TLR3 (RP) 5'-GAGGTGGAGTGTTGCAAAGGTAGT-3'
TLR3 (TP) 5'-CCCATACCAACATCCCTGAGCTGTCAA-3'
TLR4 (FP) 5'-AGCTCTGCCTTCACTACAGAGACTT-3'
TLR4 (RP) 5'-GCTTTTATGGAAACCTTCATGGA-3'
TLR4 (TP) 5'-CCCGGTGTGGCCATTGCTGC-3'
TLR5 (FP) 5'-GCACTTTTATCAATTGGCTTAATCAC-3'
TLR5 (RP) 5'-AACGAGTCAGGGTACACACAATATATG-3'
TLR5 (TP) 5'-CAATGTCACTATAGCTGGGCCTCCTGCAG-3'
TRAM (FP) 5'-CAGTGCTCTTACCCAGATGGA-3'
TRAM (RP) 5'-TCTGATAATCGATGACAGACTTCA-3'
TRAM (TP) 5'-CTGCCTGTGTTTCAATTCACGAAGCT-3'
FP: forward primer; RP: reverse primer; TP: FAM and TAMRA-labeled TaqMan probe
TaqMan PCR assays were performed on an ABI Prism 9600 Sequence Detection System (Applied Biosystems) as a one-step RT-PCR using the EZ-RT-PCR Reagent Kit (Applied Biosystems) and 40 ng of RNA. Final Mn2+ concentrations were optimized for each assay and varied between 2.5 mM and 4 mM. Assays were performed in 384-well optical plates and run in duplicates or triplicates. To quantify the results obtained by real-time PCR, we used a calibration curve. Serial dilutions of human lung or spleen RNA were used as standards and were run in parallel to the samples. The Sequence Detector Software SDS 2.0 (Applied Biosystems) was used for data analysis. The results were normalized to endogenous controls (GAPDH or β-actin).
Immunofluorescence
Immunocytochemistry for TLRs was performed on human primary small airway epithelial cells grown on collagen-I coated chamber slides (BD Bioscience) in completely supplemented SAGM. 24 h before stimulation of the cells medium was replaced by basal medium only supplemented with GA-1000, retinoic acid and BSA. Subsequently, cells were stimulated for 24 h with the indicated compounds. For staining cells were fixed and permeabilized by incubation with BD Cytofix/Cytoperm solution (BD Bioscience). The anti-TLR1 (GD2.F4) mAb, anti-TLR2 mAb (TL2.3), anti-TLR3 mAb (TLR3.7) and anti TLR5 (19D759.2) mAbs were used at a concentration of 2 – 5 μg/ml. Antibody dilutions and washing was performed in BD Perm/Wash solution (BD Bioscience). To rule out non-specific staining, a matching isotype negative control was used instead of the TLR specific antibody. Binding of the primary antibody was detected using an Alexa488-labeled anti-mouse antibody (Molecular Probes, Eugene, OR). Nuclei were stained with propidium iodide.
Measurement of cytokine, chemokine and MMP secretion
To analyze the secretion of chemokines and cytokines by SAEC, cells were stimulated for 24 h with different ligands for TLRs as described above. Cell supernatants were cleared by centrifugation, supplemented with a protease inhibitor mix (Complete™, Roche) and stored at -80°C. Cytokines and chemokines were detected using a cytokine array (RayBio® Human Cytokine Antibody Array III, RayBiotech, Norcross, GA) following the manufacturer's instructions. A complete list of cytokine antibodies present on the array is available under .
For quantification of chemokines, MMPs multiplex assays were performed in duplicates using three different dilutions of the cell supernatants. Assays were performed according to the manufacturer's instructions (SearchLight™ Array, Sample Testing Service, Pierce, Woburn, MA).
Western blotting
Western blotting was performed according to standard procedures. Briefly, proteins were separated by SDS-PAGE under reducing conditions and blotted on PVDF membranes. Membranes were blocked in PBS/5% milk powder/0.1% Tween-20. Primary and secondary antibodies were diluted in PBS/5% milk powder/0.1% Tween-20. Antibody binding was detected using a HRP labeled secondary antibody and SuperSignal® West Pico chemiluminescence substrate (Pierce).
Results
TLR triggered secretion of cytokines and chemokines by SAEC
Since the lung epithelium is a major source of chemokine and cytokine secretion in various viral and bacterial pulmonary infections [1,17], we analyzed which types of cytokines and chemokines are released from the small airway epithelial cells (SAEC) in response to activation by different TLR ligands.
Using a real-time RT-PCR approach we could demonstrate that human SAEC constitutively express mRNA of TLRs 1–6, whereas expression of TLRs 7–10 was not detected (data not shown). Therefore, SAECs were stimulated for 24 h with different ligands of TLR1/TLR2 or TLR2/TLR6 heterodimers (PGN, zymosan, Pam3CSK4, MALP-2), TLR3 (poly(I:C)), TLR4 (E. coli LPS) or TLR5 (flagellin). After stimulation, cytokine and chemokine levels in the cell culture supernatants were analyzed using a cytokine array (Fig. 1) and multiplex ELISA systems (Fig. 2A).
Figure 1 Cytokine and chemokine secretion by stimulated primary small airway epithelial cells (SAEC). SAEC were stimulated for 24 h with different TLR ligands including poly(I:C), flagellin, zymosan, peptidoglycan (PGN) and macrophage activating lipopetide-2 (MALP-2) and cell supernatants were analyzed using a cytokine antibody array. Cytokines and chemokines were detected by chemiluminescence. Results were compared to untreated controls. The figure shows representative results of three independent experiments. (C) internal positive control, (1) GRO, (2) IL-8, (3) GM-CSF, (4) GRO-α, (5) IL-6, (6) MCP-1, (7) RANTES, (8) TARC, (9) TNF-α.
Figure 2 Chemokine secretion by stimulated primary small airway epithelial cells (SAEC). (A) SAEC were stimulated for 24 h with different TLR ligands including poly(I:C), flagellin, zymosan and macrophage activating lipopetide-2 (MALP-2) and cell supernatants were analyzed using multiplex ELISAs. Results are shown as fold changes relative to untreated controls. The figure shows representative results of three independent experiments. Stimulation of the cells with LPS or PGN had no significant effect on the secretion of the indicated chemokines (data not shown). (B) Inhibition of poly(I:C) induced IP-10 secretion by a monoclonal, functional blocking anti-TLR3 antibody. SAEC were stimulated in the presence of an anti-TLR3 antibody or an IgG1 isotype control with 5 μg/ml poly(I:C) for 6 h and IP-10 secretion was analyzed in triplicates using an IP-10 ELISA. Results were compared to untreated controls and poly(I:C) stimulated cells in the absence of an antibody. (C) IFN-β secretion of SAEC after stimulation with poly(I:C). SAEC were stimulated with 5 μg/ml poly(I:C) for 6 h or 24 h and IFN-β secretion was analyzed in triplicates using an IFN-β ELISA. Results were compared to untreated controls. (D) Inhibition of poly(I:C) induced IP-10 and I-TAC expression by a goat polyclonal functional blocking anti-IFN-β antibody. SAEC were stimulated in the presence of a goat anti-IFN-β antibody (1 or 5 μg/ml; gray bars) or a goat IgG control (1 or 5 μg/ml; white bars) with 5 μg/ml poly(I:C) for 6 h. IP-10 or I-TAC expression was analyzed by real-time RT-PCR. Expression data were normalized using β-actin as endogenous control and are shown as fold changes relative to untreated controls. Results were compared to poly(I:C) stimulated cells in the absence of an antibody (black bars).
As shown in Fig. 1, stimulation of SAEC with the TLR3 ligand poly(I:C) resulted in the most pronounced inflammatory response regarding the secretion of cytokines and chemokines including an increased release of IL-6, IL-8, TNF-α, GM-CSF, MCP-1, RANTES, TARC and GRO-α. In contrast to poly(I:C), ligands for TLR2 (MALP-2, PGN, zymosan) or TLR5 (flagellin) induced only an increased secretion of IL-8 compared to the untreated control cells. In all experiments SAEC showed no response to stimulation with LPS regarding the secretion of cytokines or chemokines (data not shown).
To screen an additional panel of chemokines multiplex ELISAs were performed (Fig. 2A). These results supported our initial findings that activation of TLR3 by poly(I:C) leads to a strong proinflammatory response that is characterized by an increased secretion of MIP-3α, MIP-3β and GRO-α and by a very strong induction of the IFN-inducible chemokines IP-10 (CXCL10) and ITAC (CXCL11) (Fig. 2A). In accordance with the induction of IFN-inducible chemokines, we detected an elevated secretion of IFN-β after 6 h of poly(I:C) stimulation (Fig. 2C). The strong induction of IP-10 secretion could be inhibited by a functional blocking anti-TLR3 antibody demonstrating the involvement of cell-surface expressed TLR3 in the response of SAEC to poly(I:C) (Fig. 2B). Likewise, the increased IP-10 and I-TAC expression in poly(I:C) stimulated cells were significantly inhibited by a neutralizing anti-IFN-β antibody indicating an IFN-β dependent mechanism of IP-10 and I-TAC induction by poly(I:C) (Fig. 2D).
In comparison to poly(I:C), stimulation of SAEC with the TLR5 ligand flagellin resulted in a less pronounced secretion of the IFN-inducible chemokine IP-10, but induced the secretion of similar levels of MIP3α, MIP3β and GRO-α (Fig. 2A). MALP-2, a specific ligand for TLR2/TLR6 heterodimers, increased the secretion of IP-10, MIP-3α and GRO-α. The TLR2 ligand zymosan induced an elevated secretion of GRO-α (Fig. 2A). Other TLR2 ligands including Pam3CSK4 and PGN had no significant impact on the secretion of the analyzed chemokines (data not shown). Again SAEC showed no response to stimulation with the TLR4 ligand LPS (data not shown).
TLR triggered secretion of matrix metalloproteinases by SAEC
Since increased levels of matrix metalloproteinases (MMPs) are found in chronically inflamed tissues in diseases like COPD and are thought to contribute to the pathophysiology of the diseases, we analyzed the secretion of MMPs by SAEC and the effect of TLR activation on this process. SAECs were stimulated for 24 h with different ligands for TLRs and MMP concentrations in the cell culture supernatants were analyzed using a multiplex ELISA (Fig. 3).
Figure 3 Matrix metalloproteinase (MMP) and TIMP secretion by stimulated primary small airway epithelial cells (SAEC). SAEC were stimulated for 24 h with different TLR ligands including poly(I:C), flagellin, macrophage activating lipopetide-2 (MALP-2) and zymosan and cell supernatants were analyzed using MMP multiplex ELISAs. Results are shown as fold changes relative to untreated controls. The figure shows representative results of three independent experiments.
These results demonstrate that the TLR3 ligand poly(I:C) induced an markedly increased secretion of type-I collagenases MMP-1, MMP-8, MMP-13 as well as an increased release of the type-IV collagenase MMP-9 (Gelatinase B) and the stromelysin MMP-10. Secretion of MMP inhibitors TIMP-1 and TIMP-2 was not elevated. TLR5 activation by flagellin slightly increased the release MMP-1, MMP-9, MMP-10 and induced a strongly elevated secretion of MMP-13 by SAEC. Activation of SAECs with TLR2 ligands had only a minor effect on MMP or TIMP secretion. The TLR2 ligand zymosan induced the release of MMP-1 and the MMP inhibitor TIMP-2. MALP-2, a ligand for TLR2/TLR6 heterodimers, increased the secretion of MMP-13, whereas the TLR1/TLR2 specific ligand Pam3CSK4 and the TLR4 ligand LPS had no significant effects on the secretion of the analyzed MMPs or TIMPs (data not shown). These data demonstrate that among the tested TLR ligands, poly(I:C) induced the highest MMP secretion by activated lung epithelial cells. This is in accordance with the highly increased cytokine and chemokine secretion induced by poly(I:C) and emphasizes the strong proinflammatory impact of the TLR3 ligand poly(I:C) for SAEC.
Regulation of TLR expression in SAEC by different ligands for TLRs
To analyze whether TLR activation has an effect on the expression of TLRs in SAECs, cells were stimulated for 6 h or 24 h with ligands for TLR2 (PGN, zymosan, Pam3CSK4, MALP-2), TLR3 (poly(I:C)), TLR4 (E. coli LPS) or TLR5 (flagellin) and TLR mRNA and protein expression was measured by quantitative real time RT-PCR and immunofluorescence. As shown in Fig. 4, the TLR3 ligand poly(I:C) had the most potent effects on the mRNA expression of TLRs in SAEC. We detected an increased mRNA expression of TLR3 after poly(I:C) stimulation that was accompanied with an increased TLR3 protein expression (Fig. 5).
Figure 4 Toll-like Receptor (TLR) mRNA expression by stimulated primary small-airway-epithelial cells (SAEC). SAEC were stimulated for 6 h (black bars) or 24 h (gray bars) with different TLR ligands including LPS, flagellin, poly(I:C), macrophage activating lipopetide-2 (MALP-2) and zymosan. TLR1 – TLR6 mRNA expression was analyzed by quantitative RT-PCR. Results were normalized using β-actin as endogenous control and are shown as fold changes relative to untreated controls. TLR6 protein expression in unstimulated (U) and poly(I:C) stimulated (I:C) SAEC was analyzed by Western blotting and is shown as an insert in the diagram.
Figure 5 Toll-like Receptor (TLR) expression in unstimulated and stimulated primary small airway epithelial cells (SAEC). SAEC were stimulated for 24 h with poly(I:C) (middle) or with TNF-α in combination with IFN-γ (right). TLR1, TLR2, TLR3 and TLR5 protein expression was analyzed by immunofluorescence of fixed and permeabilized cells (green). Nuclei were stained with propidium iodide (red). Results were compared to unstimulated cells (left).
TLR1 mRNA expression was down-regulated after 6 h of poly(I:C) stimulation followed by an up-regulation after 24 h of poly(I:C) stimulation (Fig. 4) that was associated with an elevated TLR1 protein expression (Fig. 5). We also observed a very strong up-regulation of TLR2 mRNA and protein expression in poly(I:C) stimulated SAECs (Fig. 4 and 5). In contrast, mRNA and protein expression of TLR6 was strongly decreased by poly(I:C) (Fig. 4). Likewise, we observed a strong down-regulation of TLR5 mRNA (Fig. 4) and protein (Fig. 5) expression by poly(I:C). In contrast to poly(I:C), other TLR ligands had only minor effects on the regulation of TLR expression in SAECs. We detected no induction of TLR7 – TLR10 gene expression by stimulation of SAECs with poly(I:C) or other TLR ligands (data not shown).
Regulation of the expression of TLR signaling molecules by different ligands for TLRs
Signaling by TLRs is initiated by the recruitment of a specific set of adaptor proteins [18]. Since the use of different adaptor proteins provides a mechanism to modulate and specify the response of individual TLRs, we analyzed the regulation of the expression of the known TLR adaptor proteins in small-airway epithelial cells. To examine the impact of different TLR ligands on the expression of the adaptor proteins, cells were stimulated for 6 h or 24 h with different ligands for TLR2 (PGN, zymosan, Pam3CSK4, MALP-2), TLR3 (poly(I:C)), TLR4 (E. coli LPS) or TLR5 (flagellin) and TLR adaptor expression was measured by quantitative real time RT-PCR (Fig. 6) and Western blotting (Fig. 7). As shown in Fig. 6, the TLR3 ligand poly(I:C) had the most potent effects on the mRNA expression of TLR adaptor proteins. Poly(I:C) induced an increased mRNA expression of the common adaptor protein MyD88 and the TLR3 and TLR4 specific adaptor protein TRIF after 24 h of poly(I:C) stimulation. In contrast, the gene expression of TOLLIP, a negative regulator of TLR signaling, was decreased by poly(I:C) (Fig. 6). In accordance with the reduced mRNA expression, TOLLIP protein levels were found to be decreased by poly(I:C) stimulation (Fig. 7A). Surprisingly, Western blot analysis demonstrated that despite the elevated TRIF mRNA expression, TRIF protein levels are strongly decreased after 3 h, 6 h and 24 h of poly(I:C) stimulation indicating that TRIF is cleaved or degraded following TLR3 signaling (Fig 7A and 7B).
Figure 6 mRNA expression of genes involved in TLR signaling in stimulated primary small airway epithelial cells (SAEC). SAEC were stimulated for 6 h (black bars) or 24 h (gray bars) with different TLR ligands including LPS, flagellin, poly(I:C), macrophage activating lipopetide-2 (MALP-2) and zymosan. MyD88, TRIF, TOLLIP and IRAK-2 mRNA expression was analyzed by quantitative RT-PCR. Results were normalized using β-actin as endogenous control and are shown as fold changes relative to untreated controls. No changes of the gene expression of TRAM, TIRAP, IRAK-1, IRAK-3 and IRAK-4 were observed in the stimulated cells (data not shown).
Figure 7 Expression of proteins involved in TLR signaling in stimulated primary small-airway-epithelial cells (SAEC). (A) SAEC were untreated (control) or stimulated for 24 h with the indicated TLR ligands. MyD88, TOLLIP, TRIF, TIRAP, IRAK-1 and IRAK-2 protein expression was analyzed by Western blotting. A vimentin loading control is shown below. (B) SAEC were stimulated for 1 h, 3 h or 6 h with poly(I:C) (I:C) and TRIF protein levels were analyzed by Western blotting (upper panel). Results were compared to untreated controls (U) and to IRAK-1 protein expression (lower panel).
Interleukin-1 receptor-associated kinases (IRAKs) are part of the TLR signaling cascade and are involved in the activation of NFκB and the MAP kinase pathway following TLR activation [19]. Due to their most up-stream location in the TLR signaling cascade, we aimed to analyze the regulation of these key signaling proteins in SAECs by TLR ligands. IRAK expression was measured by quantitative real time RT-PCR and Western blotting. As shown in Fig. 6 and Fig. 7A, the TLR3 ligand poly(I:C) had a pronounced effect on the mRNA and protein expression of IRAKs. We found a strongly increased mRNA (Fig. 6) and protein (Fig. 7A) expression IRAK-2 after 24 h of poly(I:C) stimulation, whereas the protein expression of IRAK-1 was strongly decreased by 24 h of poly(I:C) most probably due to an increased IRAK-1 degradation following enhanced TLR signaling (Fig. 7A) since we found no evidence for a transcriptional regulation of IRAK-1 by poly(I:C) (data not shown). IRAK-1 degradation was detectable after 24 h of poly(I:C) stimulation and succeeded TRIF cleavage or degradation (Fig. 7B). This is in agreement with the finding that IRAK-1 activation is down-stream of TRIF in the TLR3 signaling pathway. Poly(I:C) and other TLR ligands had no effect on the mRNA or protein expression of IRAK-3 or IRAK-4 (data not shown).
Regulation of TLR expression in SAEC by Th1 and Th2 cytokines
Since TLR ligands are thought to induce a pronounced type-1 immune response and since the cytokine milieu might affect the TLR expression in SAEC, we investigated the regulation of TLR expression by different type-1 or type-2 cytokines. Therefore, SAEC were stimulated for 6 h or 24 h with cytokines or combinations of cytokines that are associated with a type-1 or type-2 inflammatory response and the expression of TLRs was analyzed by real-time RT-PCR (Fig. 8) and immunofluorescence (Fig. 5). Type-1 cytokines including IL-1β, TNF-α and IFN-γ induced a strong up-regulation of TLR2 mRNA (Fig. 8) and protein expression that was most pronounced by the co-stimulation with TNF-α and IFN-γ (Fig. 5). Likewise, TNF-α in combination with IFN-γ induced an elevated mRNA (Fig. 8A) and protein expression of TLR1 (Fig. 5) and strongly reduced the expression of TLR6 indicating that this cytokine combination favors a signaling through TLR1/TLR2 heterodimers.
Figure 8 mRNA expression of Toll-like Receptors (TLR) in stimulated primary small-airway-epithelial cells (SAEC). (A) SAEC were stimulated for 6 h (black bars) or 24 h (gray bars) with different type-1 cytokines. TLR1 – TLR6 mRNA expression was analyzed by quantitative RT-PCR. Results were normalized using GAPDH as endogenous control and are shown as fold changes relative to untreated controls. (B) SAEC were stimulated for 6 h (black bars) or 24 h (gray bars) with different type-2 cytokines. TLR1, TLR2 and TLR4 mRNA expression was analyzed by quantitative RT-PCR. Results were normalized using GAPDH as endogenous control and are shown as fold changes relative to untreated controls. No changes of the gene expression of TLR3, TLR5 and TLR6 were observed by stimulation of SAEC with type-2 cytokines (data not shown).
Surprisingly, in contrast to the very strong induction of TLR2 mRNA by IFN-γ, both TLR1 and TLR6, which are known to form functional heterodimers with TLR2, were found to be down-regulated by IFN-γ stimulation in the absence of TNF-α (Fig. 8A). Likewise, we observed a strong down-regulation of TLR5 mRNA (Fig. 8A) and protein expression (Fig. 5) after 24 h of TNF-α and IFN-γ stimulation. In contrast to the poly(I:C) mediated up-regulation of TLR3 mRNA and protein expression, we detected no effect of type-1 cytokines on the expression level of TLR3 (Fig. 5 and 8A).
Type-2 cytokines like IL-4 or IL-13 in combination with TNF-α induced in a synergistic manner a strong up-regulation of TLR2 mRNA expression (Fig. 8B). Similar, there is an increase in TLR1 mRNA expression after stimulation with IL-4 or IL-13 in combination with TNF-α, whereas there is no effect on the TLR6 mRNA level (data not shown) indicating that these conditions favor a signaling through TLR1/TLR2 heterodimers. We found no effect on type-2 cytokines on the mRNA expression of TLR3 or TLR5 (data not shown).
In addition to the cytokine milieu, kinetic seems to play an important role in the regulation of TLR4 mRNA expression. Whereas type-1 cytokines induce a rapid up-regulation of TLR4 mRNA levels after 6 h of stimulation that return to a basal level after 24 h of stimulation (Fig. 8A), type-2 cytokines induce a slow increase of TLR4 mRNA levels that is detectable after 24 h of stimulation (Fig. 8B).
Regulation of TLR signaling proteins in SAEC by Th1 and Th2 cytokines
To analyze the regulation of TLR signaling molecules by type-1 or type-2 cytokines, SAEC were stimulated for 6 h or 24 h with cytokines or combinations of cytokines that are associated with a type-1 or type-2 inflammatory response and the expression of TLR signaling molecules was analyzed by real-time RT-PCR (Fig. 9) and Western blotting (Fig. 10). The mRNA expression of MyD88 was found to be rapidly up-regulated after 6 h of stimulation with IFN-γ alone or in combination with IL-1β or TNF-α. Similarly, the mRNA expression of TRIF was induced by the type-1 cytokine IFN-γ alone or in combination with IL-1β or TNF-α (Fig. 9). In contrast to type-1 cytokines, stimulation of the cells with Th2 associated cytokines had no effect on MyD88 or TRIF mRNA levels (data not shown). The mRNA expression data of MyD88 and TRIF correlated very well with the MyD88 and TRIF protein levels as determined by Western blotting (Fig. 10). We found no effect of Th1 or Th2 cytokines on the expression level of the adaptor proteins TRAM, TIRAP and TOLLIP in SAEC.
Figure 9 mRNA expression of genes involved in TLR signaling in stimulated primary small-airway-epithelial cells (SAEC). SAEC were stimulated for 6 h (black bars) or 24 h (gray bars) with different cytokines. MyD88, TRIF, IRAK-2 and IRAK-3 mRNA expression was analyzed by quantitative RT-PCR. Results were normalized using GAPDH as endogenous control and are shown as fold changes relative to untreated controls. No changes of the gene expression of TRAM, TIRAP, TOLLIP, IRAK-1 and IRAK-4 were observed in the cytokine stimulated cells (data not shown).
Figure 10 Expression of proteins involved in TLR signaling in stimulated primary small-airway-epithelial cells (SAEC). SAEC were stimulated for 24 h with the indicated cytokines. MyD88, TRIF, TOLLIP, TIRAP, IRAK-2 and IRAK-3 protein expression was analyzed by Western blotting. A β-actin loading control is shown below. No changes of the protein expression of IRAK-1 and IRAK-4 were observed in the stimulated cells (data not shown).
Likewise, the regulation of IRAKs by TLR ligands by type-1 or type-2 cytokines was analyzed by real-time RT-PCR and Western blotting. As shown in Fig. 9, there is a strong regulation of IRAK-2 and IRAK-3 on the level of mRNA expression. IRAK-2 mRNA expression is highly increased following stimulation of SAEC with type-1 cytokines like IL-1β, TNF-α or IFN-γ. Moreover, there is a synergistic up-regulation of IRAK-2 expression induced by stimulation with IFN-γ together with IL-1β or TNF-α. In contrast to Th1 cytokines, type-2 cytokines like IL-4 or IL-13 had no effect on the IRAK-2 expression and displayed no synergism with TNF-α regarding the up-regulation of IRAK-2 mRNA (data not shown). The mRNA expression data of IRAK-2 closely correlated with the IRAK-2 protein levels as determined by Western blotting (Fig. 10).
Considering the regulation of IRAK-3 in SAECs, we found a slightly increased mRNA expression after stimulation of the cells for 6 h with IL-1β or TNF-α, that was further increased by the simultaneously stimulation with IFN-γ. After 24 h of stimulation with IFN-γ in the presence of IL-1β or TNF-α, IRAK-3 mRNA expression decreased to the basal level. In contrast, stimulation of the cells with IL1β or TNF-α alone resulted in a sustained increased IRAK3 mRNA expression that was also detectable after 24 h of stimulation. Type-2 cytokines like IL-4 or IL-13 only slightly increased IRAK-3 mRNA levels and had no major effect on the TNF-α induced up-regulation of IRAK-3 mRNA (data not shown). The mRNA expression data of IRAK-3 correlated with the IRAK-3 protein levels as determined by Western blotting (Fig. 10). We found no evidence for a regulation of IRAK-1 and IRAK-4 by type-1 or type-2 cytokines in SAEC (data not shown).
Discussion
Bacterial and viral exacerbations play a crucial role in a variety of lung diseases including COPD or asthma most probably due to a biased release of pro-inflammatory mediators. Since the lung epithelium is a major source of inflammatory molecules [1,2,17], we aimed to analyze which cytokines and chemokines are released from the lung epithelium in response to activation by different microbial molecules that are recognized by Toll-like receptors. To characterize the effects of TLR ligands under well controlled in vitro conditions, we have chosen primary small airway epithelial cells (SAEC) as a model to study the inflammatory response of the lung epithelium.
Among the TLR ligands evaluated in this study, poly(I:C), a synthetic analog of viral dsRNA and a well characterized ligand for TLR3 [20], mediated the most potent proinflammatory effects in SAEC. Cytokines and chemokines induced by poly(I:C) included IL-6, IL-8, GM-CSF, TNF-α, MIP-3α, GRO-α, IFN-β and IFN-inducible genes like IP-10, ITAC and RANTES, which is in accordance with the activation of IRF-3 following TLR3 stimulation [21]. Likewise, Guillot et al. reported an elevated secretion of IL-6, IL-8 IFN-β and RANTES in poly(I:C) stimulated BEAS-2B cells (human bronchial epithelial cell line) [8]. In accordance with an elevated IFN-β secretion detectable after 6 h of poly(I:C) stimulation, the expression of the IFN-inducible IP-10 and ITAC could be significantly blocked by an IFN-β neutralizing antibody demonstrating the induction of an IFN-response in SAEC by poly(I:C).
Immunofluorescence staining of permeabilized and non-permeabilized SAEC demonstrated a low cell-surface expression of TLR3 and revealed that most of TLR3 protein is found at an intracellular compartment (data not shown). Nevertheless, the strong induction of IP-10 secretion by poly(I:C) could be markedly inhibited by a functional blocking anti-TLR3 antibody demonstrating a key role of cell-surface expressed TLR3 for the response of SAEC to poly(I:C).
The poly(I:C) induced secretion of this set of chemokines and cytokines is part of a pronounced Th1 response leading to the recruitment and activation of neutrophils, macrophages and Th1 cells. Therefore, this strong proinflammatory type-1 immune response mediated by poly(I:C) is very likely to contribute to viral exacerbations found in type-1 pulmonary diseases like COPD. Surprisingly, poly(I:C) also induced the secretion of TARC, which is a well known chemoattractant of Th2 cells [22]. In addition, IL-8, RANTES and GM-CSF have been shown to be involved in the recruitment and survival of eosinophils [23] and contribute to a Th2 immune response. These findings imply a role for TLR3 stimulation not only for Th1 but also for Th2 mediated pulmonary immune responses like allergic asthma.
Flagellin, a ligand for TLR5, is a compound of bacterial flagellae and has been shown to be strong mediator for pulmonary inflammations [24]. However, flagellin stimulation of SAEC induced only a subset of chemokines or cytokines found to be elevated by poly(I:C) stimulation including IL-6, IL-8, MIP-3α and GRO-α. Interestingly, flagellin also induced IP-10 secretion in SAEC pointing to the induction of IFN-dependent genes following TLR5 activation. In contrast to poly(I:C) and flagellin, ligands for TLR2 were less efficient in stimulating cytokine or chemokine release by SAEC. Among the TLR2 ligands tested, MALP-2, a specific ligand for TLR2/TLR6 heterodimers was the strongest inflammatory stimulus. The precise molecular reasons for the low response of SAEC to TLR2 activation and the observed differences between various TLR2 ligands need further experiments, but might be in part due to the observed low level of TLR2 expression in unstimulated SAEC.
Although we detected a low TLR4 mRNA expression in SAEC, we observed no inflammatory response of the cells to LPS stimulation. This is in accordance with recent findings published by Monick et al. [25] showing that primary airway epithelial cells are unresponsive to endotoxin exposure under normal conditions. Since the airway epithelium – like the intestine epithelium – is in constant contact with multiple pathogen-related antigens like LPS, the hyporesponsiveness of these epithelial cells to LPS stimulation [25,26] is assumed to provide a mechanism to dampen the inflammatory response to the constant LPS exposure and to prevent the chronic inflammation of the tissue.
In addition to an elevated secretion of cytokines and chemokines, poly(I:C) also triggered a markedly increased secretion of matrix metalloproteinase (MMP) by SAECs. Increased levels of MMPs are found in chronically inflamed tissues in diseases like COPD and are thought to contribute to the pathophysiology of the disease by matrix degradation in tissue remodeling and emphysema [27]. Our results demonstrate that activation of TLR3 by poly(I:C) triggered an markedly increased secretion of the type-I collagenases MMP-1, MMP-8, MMP-13 as well as the release of the type-IV collagenase MMP-9 and the stromelysin MMP-10. Likewise, flagellin increased the release MMP-1, MMP-9, MMP-10 and MMP-13. Elevated protein levels of MMP-1, MMP-8 and MMP-9 were found in BALF or lung parenchyma of patients with emphysema [28,29]. Although alveolar macrophages and neutrophils are considered to be the major source of MMPs in the respiratory tract, our data demonstrate that SAEC secrete a variety of MMPs in in response to TLR stimulation and therefore are an additional source for increased proteolytic activity in infected airways that might contribute to lung emphysema.
The results discussed above emphasize the strong inflammatory properties of the TLR3 ligand poly(I:C) regarding the activation of lung epithelial cells that are likely to contribute to virus-induced exacerbations of pulmonary diseases like COPD, asthma or lung fibrosis. To analyze the molecular mechanism that is responsible for the strong inflammatory effects of poly(I:C), we evaluated the expression pattern of TLRs and proteins involved in TLR signaling following poly(I:C) stimulation. These data demonstrated that poly(I:C) induced an elevated expression of TLR3 mRNA and protein expression. In addition, poly(I:C) stimulation up-regulated the mRNA expression of the TLR3 adaptor protein TRIF [30,31], whereas the expression of TOLLIP, a negative regulator of TLR signaling [32], was found to be down-regulated by poly(I:C). This transcriptional regulatory mechanism is assumed to promote TLR3 signaling and to contribute to the strong inflammatory effects induced by poly(I:C). An interesting feature of TLR3 signaling in SAEC is the decrease of TRIF protein levels following poly(I:C) stimulation. The exact mechanism that is responsible for the decreased TRIF protein expression is not known, but might be related to protein cleavage or degradation. Protein degradation of TLR signaling proteins has been demonstrated for IRAK-1 [33], IRAK-4 [34] and MyD88 [35]. Interestingly, cleavage of TRIF by the viral protease NS3/4A was recently described by Li et al [36]. A similar proteolytic mechanism might be involved in the regulation of TRIF levels in poly(I:C) stimulated SAEC and might provide a mechanism to negatively regulate the cellular response to poly(I:C). However, currently we can not exclude that a poly(I:C) induced modification of TRIF might prevent the antibody detection of modified TRIF in our Western blot experiments. We found that TRIF degradation or modification was detectable after 3 h of poly(I:C) stimulation of SAEC and proceeded the degradation of IRAK-1, which is in agreement that IRAK-1 activation is down-stream of TRIF in TLR3 signaling.
In addition to the increased expression of TLR3, poly(I:C) has also a strong impact on the expression of other TLRs in SAEC. We could demonstrate that poly(I:C) strongly increased the expression of TLR2, whereas the expression of TLR5 and TLR6 was down-regulated by poly(I:C) stimulation. TLR1 expression was found to be strongly down-regulated after 6 h of poly(I:C) stimulation followed by an elevated mRNA and protein expression of TLR1 after 24 h of stimulation. The increased expression of TLR1 and TLR2 and the simultaneous down-regulation of TLR6, indicates that TLR3 activation favors a signaling through TLR1/TLR2 heterodimers rather than TLR2/TLR6 dimers. Furthermore, we could demonstrate that poly(I:C) stimulation results in an increased expression of IRAK-2. IRAK-2 has been shown to interact with TIRAP [37], which is involved in TLR2 and TLR4 signaling [38,39]. Although IRAK-2 posses no kinase activity, overexpression of IRAK-2 has been shown to promote NFκ-B activation [40]. Therefore, these poly(I:C) induced effects are likely to affect TLR2 and TLR4 mediated immune response in infected airways that have to be analyzed more carefully. Likewise, the down-regulation of TLR5 by poly(I:C) and the consequence for pulmonary infections with flagellated bacteria needs further investigations.
Our data demonstrate a pronounced regulation of the expression of TLRs and TLR signaling proteins in SAEC by type-1 and type-2 cytokines, which is important considering the impact of exogenous (pathogen associated) or endogenous TLR ligands on Th1 or Th2 driven pulmonary inflammations like COPD or asthma, respectively. Our data demonstrate that IFN-γ stimulation of SAEC results in the induction of an elevated IRAK-2, MyD88 and TRIF mRNA and protein expression and leads to the down-regulation of TLR1, TLR5 and TLR6 expression. There is a strong synergism of IFN-γ and TNF-α that results in a highly elevated expression of TLR2 and an increased expression of TLR4. Likewise, TNF-α alone or in combination with IFN-γ up-regulates TLR1 expression in SAEC. These data indicate that a type-1 cytokine milieu (i.e. TNF-α together with IFN-γ) promotes TLR2 and TLR4 signaling due to an up-regulation of the TLR1/TLR2 and TLR4 expression as well as due to an enhanced expression of MyD88, TRIF and IRAK-2. On the other side, type-1 cytokines are likely to inhibit TLR5 and TLR2/TLR6 signaling due to the strong down-regulation of TLR5 and TLR6 expression in SAEC. The observed up-regulation of IRAK-3, a negative regulator of TLR signaling [41], by type-1 cytokines might represent a self-limiting mechanism helping to control the inflammatory response of SAEC.
We also analyzed the influence of a Th2 cytokine milieu on the mRNA expression of TLR and their signaling or proteins. These data demonstrate that type-2 cytokines like IL-4 and IL-13 induce an elevated gene expression of TLR4 in SAEC and, together with TNF-α, up-regulate the gene expression of TLR1 and TLR2 in a synergistic manner. Therefore, this cytokine milieu is likely to promote signaling through TLR1/TLR2 heterodimers. The elevated expression of TLR4 and TLR1/TLR2 induced by type-2 cytokines is assumed to modulate the impact of bacterial infections on Th2 associated pulmonary diseases like asthma. In this regard, it is interesting to note that bacterial TLR2 and TLR4 ligands like lipopeptides or LPS have been shown to modulate the immune response in animal models of allergic asthma. Whereas the TLR1/TLR2 specific ligand Pam3CSK4 and high concentrations of the TLR4 ligand E. coli LPS have beneficial effects in asthma animal models [12-14], low-dose LPS and the TLR2 ligand peptidoglycan bias the immune response toward a Th2 phenotype and lead to aggravation of experimental allergic asthma [12,15].
Conclusion
In summary, our data demonstrate that among the evaluated TLR ligands, poly(I:C), a synthetic analog of double-stranded viral RNA mediates the strongest proinflammatory effects in primary small airway epithelial cells (SAEC) with respect to cytokine and chemokine secretion and MMP release of the cells. These inflammatory features of poly(I:C) are thought to contribute to viral exacerbations of pulmonary inflammations including COPD and asthma. Furthermore, poly(I:C) modulates the gene expression of other TLRs in SAEC, what is likely to have a strong influence on the response of the lung epithelium to stimulation with additional TLR ligands during viral and bacterial infections. In addition, our data demonstrate a pronounced regulation of the expression of TLRs and TLR signaling proteins by a type-1 or type-2 cytokine milieu. The regulation of TLR expression in small airway epithelial cells by type-1 and type-2 cytokines is important considering the impact of exogenous (pathogen associated) or endogenous TLR ligands on Th1 or Th2 driven pulmonary inflammations like COPD or asthma, respectively.
List of abbreviations
COPD: chronic obstructive pulmonary disease; dsRNA: double-stranded RNA; IRAK: Interleukin-1 receptor associated kinase; MALP-2: macrophage activating lipopeptide-2; MMP: matrix metalloproteinase; Pam3CSK4: synthetic tripalmitoylated lipopeptide Pam3CysSer(Lys)4; PGN: peptidoglycan; poly(I:C): polyinosine-polycytidylic acid; SAEC: small airway epithelial cell.
Competing interests
The work was supported by Boehringer-Ingelheim Pharma GmbH & Co. KG, Germany.
Authors' contributions
MR conceived of the experiments, carried out the experimental work of the study and drafted the manuscript. PS conceived of the experiments, participated in the design and direction of the study and revised the manuscript. AW and DM made substantial contribution to the data interpretation and helped to revise the manuscript. All authors read and approved the final manuscript.
==== Refs
Bals R Hiemstra PS Innate immunity in the lung: how epithelial cells fight against respiratory pathogens Eur Respir J 2004 23 327 33 14979512 10.1183/09031936.03.00098803
Holgate ST Lackie PM Davies DE Roche WR Walls AF The bronchial epithelium as a key regulator of airway inflammation and remodelling in asthma Clin Exp Allergy 1999 29 90 5 10421830 10.1046/j.1365-2222.1999.00016.x
Greenberg SB Allen M Wilson J Atmar RL Respiratory viral infections in adults with and without chronic obstructive pulmonary disease Am J Respir Crit Care Med 2000 162 167 73 10903237
Zalacain R Sobradillo V Amilibia J Barron J Achotegui V Pijoan JI Llorente JL Predisposing factors to bacterial colonization in chronic obstructive pulmonary disease Eur Respir J 1999 13 343 8 10065679 10.1034/j.1399-3003.1999.13b21.x
Cohen L Castro M The role of viral respiratory infections in the pathogenesis and exacerbation of asthma Semin Respir Infect 2003 18 3 8 12652448 10.1053/srin.2003.50001
Hertz CJ Wu Q Porter EM Zhang YJ Weismuller KH Godowski PJ Ganz T Randell SH Modlin RL Activation of Toll-like receptor 2 on human tracheobronchial epithelial cells induces the antimicrobial peptide human beta defensin-2 J Immunol 2003 171 6820 6 14662888
Sha Q Truong-Tran AQ Plitt JR Beck LA Schleimer RP Activation of airway epithelial cells by toll-like receptor agonists Am J Respir Cell Mol Biol 2004 31 358 64 15191912 10.1165/rcmb.2003-0388OC
Guillot L Le Goffic R Bloch S Escriou N Akira S Chignard M Si-Tahar M Involvement of toll-like receptor 3 in the immune response of lung epithelial cells to double-stranded RNA and influenza A virus J Biol Chem 2005 280 5571 80 15579900 10.1074/jbc.M410592200
Takeuchi O Hoshino K Akira S Cutting edge: TLR2-deficient and MyD88-deficient mice are highly susceptible to Staphylococcus aureus infection J Immunol 2000 165 5392 6 11067888
Wetzler LM The role of Toll-like receptor 2 in microbial disease and immunity Vaccine 2003 21 S55 S60 12763684 10.1016/S0264-410X(03)00201-9
Dabbagh K Lewis DB Toll-like receptors and T-helper-1/T-helper-2 responses Curr Opin Infect Dis 2003 16 199 204 12821808
Eisenbarth SC Piggott DA Huleatt JW Visintin I Herrick CA Bottomly K Lipopolysaccharide-enhanced, toll-like receptor 4-dependent T helper cell type 2 responses to inhaled antigen J Exp Med 2002 196 1645 51 12486107 10.1084/jem.20021340
Patel M Xu D Kewin P Choo-Kang B McSharry C Thomson NC Liew FY TLR2 Agonist Ameliorates Established Allergic Airway Inflammation by Promoting Th1 Response and Not via Regulatory T Cells J Immunol 2005 174 7558 63 15944255
Rodriguez D Keller AC Faquim-Mauro EL de Macedo MS Cunha FQ Lefort J Vargaftig BB Russo M Bacterial lipopolysaccharide signaling through Toll-like receptor 4 suppresses asthma-like responses via nitric oxide synthase 2 activity J Immunol 2003 171 1001 8 12847273
Chisholm D Libet L Hayashi T Horner AA Airway peptidoglycan and immunostimulatory DNA exposures have divergent effects on the development of airway allergen hypersensitivities J Allergy Clin Immunol 2004 113 448 54 15007346 10.1016/j.jaci.2003.12.011
Matsumoto M Kikkawa S Kohase M Miyake K Seya T Establishment of a monoclonal antibody against human Toll-like receptor 3 that blocks double-stranded RNA-mediated signaling Biochem Biophys Res Commun 2002 293 1364 9 12054664 10.1016/S0006-291X(02)00380-7
Diamond G Legarda D Ryan LK The innate immune response of the respiratory epithelium Immunol Rev 2000 173 27 38 10719665 10.1034/j.1600-065X.2000.917304.x
Yamamoto M Takeda K Akira S TIR domain-containing adaptors define the specificity of TLR signaling Mol Immunol 2004 40 861 8 14698224 10.1016/j.molimm.2003.10.006
Janssens S Beyaert R Functional diversity and regulation of different interleukin-1 receptor-associated kinase (IRAK) family members Mol Cell 2003 11 293 302 12620219 10.1016/S1097-2765(03)00053-4
Alexopoulou L Holt AC Medzhitov R Flavell RA Recognition of double-stranded RNA and activation of NF-kappaB by Toll-like receptor 3 Nature 2001 413 732 8 11607032 10.1038/35099560
Doyle S Vaidya S O'Connell R Dadgostar H Dempsey P Wu T Rao G Sun R Haberland M Modlin R Cheng G RF3 mediates a TLR3/TLR4-specific antiviral gene program Immunity 2002 17 251 63 12354379 10.1016/S1074-7613(02)00390-4
Sandoval-Lopez G Teran LM TARC: novel mediator of allergic inflammation Clin Exp Allergy 2001 31 1809 12 11737030 10.1046/j.1365-2222.2001.01268.x
Lampinen M Carlson M Hakansson LD Venge P Cytokine-regulated accumulation of eosinophils in inflammatory disease Allergy 2004 59 793 805 15230810 10.1111/j.1398-9995.2004.00469.x
Liaudet L Szabo C Evgenov OV Murthy KG Pacher P Virag L Mabley JG Marton A Soriano FG Kirov MY Bjertnaes LJ Salzman AL Flagellin from gram-negative bacteria is a potent mediator of acute pulmonary inflammation in sepsis Shock 2003 19 131 7 12578121 10.1097/00024382-200302000-00008
Monick MM Yarovinsky TO Powers LS Butler NS Carter AB Gudmundsson G Hunninghake GW Respiratory syncytial virus up-regulates TLR4 and sensitizes airway epithelial cells to endotoxin J Biol Chem 2003 278 53035 44 14565959 10.1074/jbc.M308093200
Abreu MT Vora P Faure E Thomas LS Arnold ET Arditi M Decreased expression of Toll-like receptor-4 and MD-2 correlates with intestinal epithelial cell protection against dysregulated proinflammatory gene expression in response to bacterial lipopolysaccharide J Immunol 2001 167 1609 16 11466383
Parks WC Shapiro SD Matrix metalloproteinases in lung biology Respir Res 2001 2 10 9 11686860 10.1186/rr33
Betsuyaku T Nishimura M Takeyabu K Tanino M Venge P Xu S Kawakami Y Neutrophil granule proteins in bronchoalveolar lavage fluid from subjects with subclinical emphysema Am J Respir Crit Care Med 1999 159 1985 91 10351949
Finlay GA Russell KJ McMahon KJ D'Arcy EM Masterson JB FitzGerald MX O'Connor CM Elevated levels of matrix metalloproteinases in bronchoalveolar lavage fluid of emphysematous patients Thorax 1997 52 502 6 9227714
Oshiumi H Matsumoto M Funami K Akazawa T Seya T TICAM-1, an adaptor molecule that participates in Toll-like receptor 3-mediated interferon-beta induction Nat Immunol 2003 4 161 7 12539043 10.1038/ni886
Yamamoto M Sato S Hemmi H Hoshino K Kaisho T Sanjo H Takeuchi O Sugiyama M Okabe M Takeda K Akira S Role of adaptor TRIF in the MyD88-independent toll-like receptor signaling pathway Science 2003 301 640 3 12855817 10.1126/science.1087262
Zhang G Ghosh S Negative regulation of toll-like receptor-mediated signaling by Tollip J Biol Chem 2002 277 7059 65 11751856 10.1074/jbc.M109537200
Li L Cousart S Hu J McCall CE Characterization of interleukin-1 receptor-associated kinase in normal and endotoxin-tolerant cells J Biol Chem 2000 275 23340 5 10811644 10.1074/jbc.M001950200
Hatao F Muroi M Hiki N Ogawa T Mimura Y Kaminishi M Tanamoto K Prolonged Toll-like receptor stimulation leads to down-regulation of IRAK-4 protein J Leukoc Biol 2004 76 904 8 15258191 10.1189/jlb.0504277
Naiki Y Michelsen KS Zhang W Chen S Doherty TM Arditi M Transforming growth factor-beta differentially inhibits MyD88-dependent, but not TRAM- and TRIF-dependent, lipopolysaccharide-induced TLR4 signaling J Biol Chem 2005 280 5491 5 15623538 10.1074/jbc.C400503200
Li K Foy E Ferreon JC Nakamura M Ferreon AC Ikeda M Ray SC Gale M JrLemon SM Immune evasion by hepatitis C virus NS3/4A protease-mediated cleavage of the Toll-like receptor 3 adaptor protein TRIF Proc Natl Acad Sci U S A 2005 102 2992 7 15710891 10.1073/pnas.0408824102
Fitzgerald KA Palsson-McDermott EM Bowie AG Jefferies CA Mansell AS Brady G Brint E Dunne A Gray P Harte MT McMurray D Smith DE Sims JE Bird TA O'Neill LA Mal (MyD88-adapter-like) is required for Toll-like receptor-4 signal transduction Nature 2001 413 78 83 11544529 10.1038/35092578
Horng T Barton GM Flavell RA Medzhitov R The adaptor molecule TIRAP provides signalling specificity for Toll-like receptors Nature 2002 420 329 33 12447442 10.1038/nature01180
Yamamoto M Sato S Hemmi H Sanjo H Uematsu S Kaisho T Hoshino K Takeuchi O Kobayashi M Fujita T Takeda K Akira S Essential role for TIRAP in activation of the signalling cascade shared by TLR2 and TLR4 Nature 2002 420 324 9 12447441 10.1038/nature01182
Muzio M Ni J Feng P Dixit VM IRAK (Pelle) family member IRAK-2 and MyD88 as proximal mediators of IL-1 signaling Science 1997 278 1612 5 9374458 10.1126/science.278.5343.1612
Kobayashi K Hernandez LD Galan JE Janeway CA JrMedzhitov R Flavell RA IRAK-M is a negative regulator of Toll-like receptor signaling Cell 2002 110 191 202 12150927 10.1016/S0092-8674(02)00827-9
|
16316467
|
PMC1315317
|
CC BY
|
2021-01-04 16:36:23
|
no
|
J Inflamm (Lond). 2005 Nov 29; 2:16
|
utf-8
|
J Inflamm (Lond)
| 2,005 |
10.1186/1476-9255-2-16
|
oa_comm
|
==== Front
BMC PediatrBMC Pediatrics1471-2431BioMed Central London 1471-2431-5-451632976010.1186/1471-2431-5-45Research ArticleThe impact of national and international guidelines on newborn care in the nurseries of Piedmont and Aosta Valley, Italy Guala Andrea [email protected] Roberta [email protected] Mauro [email protected] Claudio [email protected] Claudio [email protected] Guido [email protected] Gianni [email protected] Neonatal Piedmont Group [email protected] S.O.C. di Pediatria, Ospedale SS Pietro e Paolo, ASL 11, Borgosesia, Italy2 Clinica Pediatrica, Dipartimento di Scienze Mediche, Università del Piemonte Orientale, Novara, Italy3 Cattedra di Neonatologia, Dipartimento Materno-Infantile, Università di Torino, Torino, Italy2005 5 12 2005 5 45 45 19 12 2004 5 12 2005 Copyright © 2005 Guala et al; licensee BioMed Central Ltd.2005Guala 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
Care procedures for preventing neonatal diseases are carried out according to nurseries' traditions and may be not consistent with the evidence based medicine issues.
Methods
A multi-centric survey was conducted in 2 Regions located in NW Italy (Piedmont and Aosta Valley) in order to collect information on some healthy newborn care procedures. During 2001, a questionnaire was sent to the chief pediatrician in charge to the all 33 nurseries of the region asking the methods used during 2000 as prevention of ophthalmia neonatorum, early and late hemorrhagic disease of newborn, umbilical cord care and recommendations of vitamin D administration. Thereafter, during 2004 the same questionnaire was sent to the 34 chief pediatrician of nurseries to evaluate if the procedures were changed during 2003 according to guidelines. The nurseries care for 32,516 newborns in 2000 and 37,414 in 2003.
Results
Aminoglycoside eyes drops as prevention of ophthalmia neonatorum were the first choice in both periods (23 out 33 nurseries in 2000 and 24 out 34 in 2003 p > 0.05; the corresponding figures for newborns were18,984 out 32,516 newborns vs. 28,180 out of 37,414 p < 0.05). The umbilical cord care was carried out with alcohol in 12/33 centers (13,248 newborns) and dry gauze in 3/33 centers (2,130 newborns) in 2000, the corresponding figures in 2003 were 6/34 centers (p > 0.05), (6,380 newborns, p < 0.05) and 12/34 centers (p < 0.05), (18,123 newborns, p < 0.05). The percentage of newborns receiving of i.m. vitamin K. at birth increased during the study period (15,923/32,104 in 2000 vs. 19,684/37,414 in 2003, p < 0.01), but not the number of nurseries (16 in 2000 and 17 in 2003 p > 0.05). The numbers of parents of newborns who receive the recommendations of oral vitamin K during the first months life decreased from 2000 (25,516/30,606) to 2003 (29,808/37,414, p < 0.01) as well as for Vitamin D recommendation (14,582/30,616 in 2000 vs. 11,051/37,414 in 2003, p < 0.01). Oral vitamin K during the first months of life was recommended by 25 nurseries in 2000 and 27 in 2003 (p > 0.05), the corresponding figures for Vitamin D were 15 and 14 (p > 0.05).
Conclusion
In the present study a large variability of procedures among the nurseries was observed. During the study periods, guidelines and evidence based medicine issues have only partially modified the neonatal care procedures In Piedmont and Aosta Valley nurseries. These observations suggest to implement local forum/consensus conference to standardized procedures as much as possible.
==== Body
Background
Immediately after delivery healthy newborn underwent care procedures to prevent some diseases. Usually these interventions are done according to nurseries' common place and traditions which sometimes are not consistent with established guidelines and/or with evidence based medicine issues.
The key aims of this descriptive multi-centric survey are: 1) to compare the procedures applied to during 2000 and during 2003 in Piedmont and Aosta Valley nurseries for the prevention of ophthalmia neonatorum, of early and late hemorrhagic disease of newborn (HDN) through the administration of vitamin K, of rickets through the recommendation of vitamin D and for the care of the umbilical cord and 2) to evaluate the impact of Italian guidelines [14], when available, and/or international ones [1,2,5,18] on the newborns care.
Methods
In 2001 and 2004 an identical questionnaire (see Additional file 1) was sent to the chief pediatrician and ward nurse of all nurseries located in 2 Regions of North West Italy (Piedmont and Aosta Valley) in order to collect information on procedures used to prevent the ophthalmia neonatorum, the early and late HDN, the rickets and methods employed for the umbilical cord care and, in addition, the number of newborns born in each hospital in 2000 and in 2003. The forms, composed of 10 closed items and 9 open ones, were filled by the pediatrician in charge.
In 2000, there were 33 nurseries operating (1 was temporary closed), whereas in 2003 there were 34 active nurseries in Piedmont and Aosta Valley Regions. The number of newborns born in each hospital ranged from 400/year to 4000/year. For these reason the "weight" of each nurseries included in the present study is very different. In addition, in Italy, the policy is that there is just one way of doing things in each hospital, mandated by the chief pediatrician. For these reason we used as unit of analysis both the number of the nurseries and the number of newborns.
The differences among the distributions of nurseries who used different approach and of newborns who received different procedures were statistically analyzed using chi-square test.
Results
During 2000, in Piedmont and Aosta Valley 32,516 newborns were born in the 33 nurseries. For 2003, the corresponding figures were 37,414 newborns born in the 34 nurseries. About 25% of newborns were cared in 2 nurseries operating in Turin, the largest city of both Regions.
All but one colleagues answered questions on the prevention of ophthalmia neonatorum in 2000, while all questionnaires were filled in 2003. The majority of the nurseries used amino glycoside as prophylaxis (23 in 2000 and 24 in 2003 p > 0.05; caring on 18,984 and 28,180 newborns respectively, p < 0.05). Over 5400 newborns (16%) received chloramphenicol eye-drops in 2000 and in the second period over 6700 (18%) (p < 0.05, table 1).
Table 1 Ophthalmia neonatorum prophylaxis: drugs used in the nurseries of Piedmont and Aosta Valley, during 2000 and 2003
Year of survey 2000 2003
Drug N. nurseries N. newborns % N. nurseries N. newborns %
Tobramycin 12 15,168 47.2 17 20,879 55.8
Cloramphenicol 7 5,457 17.0 8 6,720 18.0
Gentamicin 6 3,816 11.8 7 7,301 19.5
Clortetraciclyne 3 3,614 11.1 - - -
Other 4 4,129 12.9 2 2,514 6.7
Total 32 32,104 100.0 34 37,414 100.0
Data not available 1 412 1.3
Table 2 shown the distribution of procedures used for the umbilical cord care. In 2000, the umbilical cord care was done with alcohol in 12 nurseries (who cared 13,248 newborns, 41%), in 4 nurseries (6,265 newborns, 19%) with micronized silver spray, in 4 nurseries (1,681 of newborns, 5%) with eosine and only in 3 nurseries (2,130 newborns, 7%) with dry gauze. In 2003, 12 nurseries (18,123 of newborns, 48 %) used for cord stump care dry gauze (comparing the number of newborns 2000 vs 2003, p < 0.05), 7 nurseries (5,613 newborns,15 %) arnica/echinacea and only 6 nurseries (6380 newborns,17 %) alcohol. The number of centers who used dry gauze or arnica/echinacea increased from 2000 to 2003 (p < 0.05), whereas the number of centers who used alcohol decreased (p < 0.05).
Table 2 Umbilical cord care used in the nurseries of Piedmont and Aosta Valley, during 2000 and 2003
Year of survey 2000 2003
Methods N. nurseries N. newborns % N. nurseries N. newborns %
Alcohol 12 13,248 40.6 6 6,380 17.0
Silver micronized 4 6,265 19.2 2 1,364 3.6
Eosine 2% 4 1,681 5.2 2 990 2.6
Arnica/echinacea 3 3,221 9.9 7 5,613 15.0
Clorexidine 3 2,313 7.1 1 1,197 3.3
Dry gauze 3 2,130 6.5 12 18,123 48.5
Peroxid water 2 1,357 4.2 3 2,559 6.8
Salicylic sugar 1 1,216 3.7 - - -
Triple dye 1 1,165 3.6 1 1,188 3.2
Total 33 32,516 100.0 34 37,414 100.0
All colleagues but one replied on early HDN prophylaxis used in 2000, while all replied in 2003 (table 3). In 2000, vitamin K was administered per os to 16,261 newborns (50%) cared in 17 nurseries. Among these 17 centers, 13 used intramuscolar (i.m.) vitamin K in pre-term newborns, in low birth weight babies and when there was evidence of fetal distress. In 2003, 17 nurseries used vitamin K per os (17,730 newborns, 47 %).
Table 3 Early hemorrhagic disease of newborn prophylaxis according to vitamin K schedule in the nurseries of Piedmont and Aosta Valley during 2000 and 2003
Year of survey 2000 2003
Administration N. nurseries N. newborns % N. nurseries N. newborns %
Oral 17 16,261 50.6 17 17,730 47.4
Intramuscular 15 15,923 49.4 17 19,684 52.6
Total 32 32,104 100.0 34 37,414 100.0
Data not available 1 412 1.3
Data on late HDN are shown in table 4. In 2000, 25 centers prescribed oral vitamin K for 3–6 months. Among the 15 nurseries, where vitamin K prophylaxis was done by i.m. after the birth, only 6 did not prescribe additional oral vitamin K dose. In 2003, 27 nurseries administered oral vitamin K. Among the 17 centers that used i.m. vitamin K after the birth, 7 did not suggest further oral doses. The number of newborn who received the recommendation of oral vitamin K increase from 25,516 during 2000 to 29,808 thereafter (p < 0.05), whereas the number of nurseries was not statistically different in the two periods.
Table 4 Late hemorrhagic disease of newborn: prophylaxis with oral vitamin K suggested by the nurseries of Piedmont and Aosta Valley, during 2000 and 2003
Year of survey 2000 2003
Oral vitamin K suggested N. nurseries N. newborns % N. nurseries N. newborns %
Yes 25 25,516 83.4 27 29,808 79.7
No 6 5,170 16.6 7 7,606 20,3
Total 31 30,606 100.0 34 37,414 100.0
Data not available 2 1,910 5.9
In 2000, at discharge the health care providers of 15 centers recommended vitamin D to all newborns (14,582 newborns, 48 %) (regardless pigmentation of skin and period of birth) for different period of time ranging between 3 months to 1 year (table 5). Among the 16 centers that did not recommend routinely the use of vitamin D, 10 suggested vitamin D at the discharge of pre-term or small for gestational age newborns, 2 suggested only to babies born in winter, 1 suggested only to dark-skinned newborns and 3 never. In 2003, 14 nurseries (11,051 newborns, 30 %) prescribed vitamin D to all newborns (comparing the number of newborns 2000 vs. 2003, p < 0.05), whereas the number of nurseries was not statistically different in the two periods. Among centers in 2003, that did not routinely suggest vitamin D, 11 prescribed it only to pre-term newborns, 4 only to dark-skinned babies and 5 never.
Table 5 Rickets prophylaxis: vitamin D suggested at discharge in the nurseries of Piedmont and Aosta Valley during 2000 and 2003
Year of survey 2000 2003
Oral vitamin D suggested N. nurseries N. newborns % N. nurseries N. newborns %
yes 15 14,582 47.6 14 11,051 29.5
no 16 16,104 52.4 20 26,363 70.5
Total 31 30,606 100.0 34 37,414 100.0
Data not available 2 1,910 5.9
Discussion
Clinical guidelines provide specific recommendations for practice. Various organization developed systematic programs to produce consensus statements and standards for good medical care. Under the movement of evidence based medicine, systematic literature review and linkage of recommendations to supporting evidence became essential. Nevertheless, despite the availability of the same body of evidence (MEDLINE and The Cochrane Library) recommendations differ and the translation in good practice is not linear [4,12]. Our study reports an heterogeneity of procedures in the nurseries of 2 Regions located in NW of Italy (Piedmont and Aosta Valley) to prevent potentially severe diseases in healthy newborns. In every nursery the prophylactic interventions are mainly related to long lasting habit/commonplace and previously developed experiences. During the study period, many guidelines or EBM reviews on the newborns care have been released both from Italian Health Care Society both from international Society. In addition a variety of published papers, meta-analysis reports, recommendations developed by ad-hoc committees are available, for each of the four prophylactic interventions described in the present study [1,2,5,6,14,15,17,18].
According to recommendations of the Canadian Society of Pediatrics [5] and to the guidelines issued by the WHO [15], the choice of antibiotic for preventing ophthalmia neonatorum should be related to local epidemiology of the most frequent bacteria.
Guidelines have been developed by the WHO [17] and by Cochrane Library [18] for the umbilical cord care and an Italian evidence based study was recently published [11]. Delay in the cord stump detachment might cause parents' anxiety [3] and different procedures to cord care have been compared [8,13].
A lot of studies and reviews are available on the prevention of early HND: the i.m. via is effective as the oral administration without severe side-effects [7]. However the low compliance of the oral administration did can not completely exclude the late HND [9].
Recommendations of the American Society of Pediatrics are available on vitamin D administration to prevent rickets [2]. The prevention might begin during pregnancy [6]. In Piedmont, due to the immigration from Africa in recent years, rickets is diagnosed almost exclusively in dark skinned children (personal data) and the prescription of vitamin D during early after birth is not mandatory in non dark skinned infants.
During 2000, in the large majority of nurseries the used procedures differ from the guidelines developed by the Italian scientific Society. Many scientific literature review and recommendations from several pediatric academies have been published after 2000 and this might justify the variability of conduct observed during the first period of the our survey. In the 2003 (second period of the present study) the results showed a more strict adherence to the available guidelines. The 2 largest nurseries, which care about one quarter of newborns born each year in Piedmont, changed the procedures according to evidence of available reviews. While the medium size (1000–2000 newborns/year) or small size (less than 1000 newborns/year) nurseries show a more slow trend to adapt their care standard according to guidelines. In any case, the health provider conduct showed a large variability even if Italian guidelines, implemented by the scientific societies, were published [14].
Besides, medical prescription or diagnostic/therapeutic pathway are not exclusively tied to scientific evidences. Drug prescriptions are not only a medical act coming from diagnosis evidence, but are often strongly influenced by psychological, cultural and socio-economic elements, as recently confirmed in Italy by APE (Attitudini Prescrittive in Pediatria) study [10,16].
Conclusion
It is unclear why clinician fail to follow the guidelines. Although the quality of evidence-based guidelines in pediatrics is high [19], it is always noticed the gap between the scientific evidence and the update and the application of that evidence in practice [20]. Reliance on guidelines may prevent the development of serious co-morbidities. The large variability of preventive intervention observed in our survey might be an opportunity for implementing a forum among the pediatricians and the neonatologists to developing or reinforce evidence-based guidelines. Health care organization should increase their educational efforts. Actually, during 2005 and 2006 a number of meetings are planned and supported by the regional piedmont section of the Italian Society of Pediatrics, by the Italian Society of Neonatology and by the Associazione Culturale Pediatri to achieve this goal.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
AG had the primary responsibility for protocol development and writing the manuscript. RG contributed to elaboration of data. MZ, CM, CF, GB contributed to the writing the manuscript. GP supervised the design and the execution of the study, and contributed to the writing the manuscript.
The Neonatal Piedmont Group: Besenzon L. (Saluzzo, Savigliano), Bonenti G. (Rivoli), Campra D. (Borgosesia), Castella E. (Tortona), Castelli G. (Ceva, Mondovì), Cattaneo G. (Moncalieri), Cattrini C. (Domodossola, Verbania), Cocuzza S. (Alba, Bra), Coscia A. (Torino), De Franco S. (Novara), De Marinis S. (Cuneo), Favetta S. (Chivasso), Frigerio M. (Torino), Galligani L. (Biella), Giaretto G. (Courgnè, Ivrea), Gomirato G. (Torino), Grazia G. (Pinerolo), Guala G. (Torino), Lancione F. (Chieri), Machado E. (Aosta), Maganuco S. (Ciriè, Venaria Reale), Papili F. (Casale Monferrato), Pesce F. (Alessandria), Provera S. (Vercelli), Rigardo S. (Acqui Terme, Novi Ligure), Roberi P. (Carmagnola), Rossi E. (Torino), Vitali M. (Borgomanero), Zannino L. (Asti).
Pre-publication history
The pre-publication history for this paper can be accessed here:
Supplementary Material
Additional File 1
the additional file nidi questionario.doc report the questionnaire used for the present study.
Click here for file
==== Refs
American Academy of Pediatrics Controversies concerning Vitamin K and the newborn Pediatrics 2003 112 191 192 12837888 10.1542/peds.112.1.191
American Academy of Pediatrics Prevention of rickets and vitamin D deficiency: new guidelines for vitamin D intake Pediatrics 2003 111 908 210 12671133 10.1542/peds.111.4.908
Anhalt H Marino RV Rosenfeld W Retained umbilical stump: clinical approaches and separation anxiety Am J Dis Child 1992 46 1413 1414 1456242
Burggers JS van Everdingen JJE Beyond the evidence in clinical guidelines Lancet 2004 364 392 393 15288718 10.1016/S0140-6736(04)16778-0
Canadian Paediatric Society Recommendations for the prevention of neonatal ophthalmia. Paediatr Child Health 2002 7 480 483
Comité de Nutrition La supplementation en vitamine D durant la grossesse: une nécessité Arch Pédiatr 1995 2 373 376 10.1016/0929-693X(96)81161-4
Fear NT Roman E Ansell P Simpson J Day N Eden OB Vitamin K and childhood cancer: a report from the United Kingdom Childhood Cancer Study Brit J Cancer 2003 89 1228 1231 14520451 10.1038/sj.bjc.6601278
Guala A Pastore G Garipoli V Agosti M Vitali M Bona G The time of umbilical cord separation in healthy full-term newborns: a controlled clinical trial of different cord care practices Eur J Pediatr 2003 162 350 351 12692719
Hansen KN Minousis M Ebbesen F Weekly oral vitamin K prophylaxis in Denmark Acta Paediatr 2003 92 802 805 12892158 10.1080/08035250310002858
Linee Guida Appropriatezza clinica ed organizzativa degli interventi di tonsillectomia e/o adenoidectomia Medico e Bambino 2003 7 441 457
Lo Iacono G Trizzino A Buzzetti R Umbilical cord care at birth: commonplace, traditions and EBM in paediatrics Ital J Pediatr 2002 28 271 274
Parker KL Wyatt DT Blethen SL Baptista J Price L Screening girls with Turner syndrome: the national cooperative growth study experience J Pediatr 2004 143 133 135 12915841
Pezzati M Biagioli EC Martelli E Gambi B Biagiotti R Rubaltelli FF Umbilical cord care: the effect of eight different cord-care regimens on cord separation time and other outcomes Biol Neonate 2002 81 38 44 11803175 10.1159/000047182
Rapisardi G Pierattelli M Tamburlini G Raccomandazioni per l'assistenza alla madre in puerperio e al neonato Ital J Pediatr 2000 26 232 243
Schaller UC Klauss V Is Credè's prophylaxis for ophthalmia neonatorum still valid? Bull WHO 2001 79 262 264 11285676
Studio APE, 2003
Zupan J Garner P Topical umbilical cord care after birth The Cochrane Library 2001 4 1 23
Boluyt N Lincke CR Offringa M Quality of evidence-based pediatric guidelines Pediatrics 2005 115 1378 1391 15867050 10.1542/peds.2004-0575
Simclair JC Evidence-based therapy in neonatology : distilling the evidence and applying it in practice Acta Paediatr 2004 93 1146 1152 15384873 10.1080/08035250410015619
|
16329760
|
PMC1315318
|
CC BY
|
2021-01-04 16:31:06
|
no
|
BMC Pediatr. 2005 Dec 5; 5:45
|
utf-8
|
BMC Pediatr
| 2,005 |
10.1186/1471-2431-5-45
|
oa_comm
|
==== Front
Global HealthGlobalization and Health1744-8603BioMed Central London 1744-8603-1-161631367010.1186/1744-8603-1-16DebateExploring ethical considerations for the use of biological and physiological markers in population-based surveys in less developed countries Pappas Gregory [email protected] Adnan A [email protected] Chairman, Department of Community Health Science, Aga Khan University, 3500 Stadium Road, Karachi, Pakistan2 Assistant Professor, Dept. of International Health and Berman Bioethics Institute, Johns Hopkins University Bloomberg School of Public Health, 615 N. Wolfe Street, Baltimore, MD 21205, USA2005 28 11 2005 1 16 16 6 4 2005 28 11 2005 Copyright © 2005 Pappas and Hyder; licensee BioMed Central Ltd.2005Pappas and Hyder; 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 health information needs of developing countries increasingly include population-based estimates determined by biological and physiological measures. Collection of data on these biomarkers requires careful reassessment of ethical standards and procedures related to issues of safety, informed consent, reporting, and referral policies. This paper reviews the survey practices of health examination surveys that have been conducted in developed nations and discusses their application to similar types of surveys proposed for developing countries.
Discussion
The paper contends that a unitary set of ethical principles should be followed for surveys around the world that precludes the danger of creating double standards (and implicitly lowers standards for work done in developing countries). Global ethical standards must, however, be interpreted in the context of the unique historical and cultural context of the country in which the work is being done. Factors that influence ethical considerations, such as the relationship between investigators in developed and developing countries are also discussed.
Summary
The paper provides a set of conclusions reached through this discussion and recommendations for the ethical use of biomarkers in populations-based surveys in developing countries.
population based health surveysethical standardsless developed countries
==== Body
Introduction
The national health information needs of developing countries are increasingly relying on the collection of biological and physiological measures [1-5]. The use of these biomarkers in population-based surveys has led to a call for a review of the ethical standards under which surveys are conducted in less-developed nations [6]. Controversies over clinical research conducted in developing countries have intensified the scrutiny of all research being conducted in these settings [7]. This debate has reaffirmed that the global human rights and medical ethical principles – including fidelity, truthfulness, confidentiality, autonomy, and beneficence, – must be carefully reviewed in the name of research, surveillance, monitoring, and evaluation [8-10]. Ethical issues in population-based surveys include informed safety, informed consent, confidentiality, and reporting findings of testing. Globalization of ethical principles must be interpreted within the particulars of national context in which the scientific study takes place.
Development of rapid, low cost, diagnostic kits and portable, robust technology has made possible a new generation of population-based health surveys [11]. While standards and safeguards for health interview surveys in developing countries have been in place for a number of years, the use of these new technologies adds a layer of new ethical issues to health surveys. The ethical issues raised by the collection of biomarkers in large-scale examination surveys, and standards for implementation have been developed over the past four decades in national population-based surveys done in developed countries [12]. The purpose of this paper is to review the current ethical practices and procedures used to guide survey work in the developed world, and discuss the implication of such practices for work in developing countries. The manuscript refers to written documents on these issues and draws upon health examination survey work of the authors in the United States and in less developed countries [3,13].
The article appears in three sections. The first section lays out the ethical principles that bind which are raised by health examination surveys, and reviews current practices. The use of these so called biomarkers has a long history in the National Health and Nutrition Examination Survey (NHANES) of the Centers for Disease Control and other major epidemiological studies (Framingham, Alameda county) [14,15]. These population-based health surveys have dealt with the following types of ethical issues: safety of the survey to subjects and workers in survey, obtaining appropriate informed consent, confidentiality of information collected in the survey, the reporting of findings of the health examination to the participant, the use of stored samples for research, and the provision of health care as part of a survey. While ethical principles may be global, implementation of those principles must be careful considered within local contexts in which the health examination survey takes place. In the second section these ethical issues will be addressed as they relate to conducting health examination surveys in less developed countries. This section emphasizes situations in which global standards have to be interpreted in a national or local context. While the spirit of the standard may be toward uniformity of procedures, implementation of surveys must consider many local conditions. Implementation of global standards in resource poor settings has created feedback and frequently challenges interpretation of those standards. The final section is a set of recommendations to be considered by national survey planners and donor agencies. While guidelines for health examination surveys have been developed over a long period in the US and Europe, those standards are under continued review as science and practice evolve. The globalization of standards creates a challenge for those standards and for feedback onto standards as practiced in the resource rich settings. Scientific advances and technological innovations will continue to require review of standards and procedures of ethical conduct in health examination surveys.
I. Ethical Issues in Health Examination Surveys
This section reviews current practices being used in population-based health surveys in the US and other countries. Ethical practices for conducting health examinations surveys have developed over the past four decades in the United States and Europe related to safety, informed consent, confidentiality, the reporting of findings, and long-term sample storage in developed countries.
Safety issues in surveys consider consequences of collection of biological specimens and clinical testing for survey participants and for data collectors. Biomarkers that are minimally invasive are preferred (e.g., blood pressure measurement, venous blood draw or urine collection). Potentially dangerous tests (high intensity x-rays) have usually been avoided. The use of minimally invasive, as opposed to potentially dangerous procedures has important implications to the risk/benefit ratio of the survey. Safety standards for specimen collectors typically follow clinical standards, and universal precautions for laboratory work have been adopted for field use.
Informed consent in health examination surveys has been most frequently obtained in writing in developed nations. Concerns about obtaining informed consent in special populations have lead to accommodations or modified procedures. In special populations – those with low levels of literacy, different cultural traditions, a context of inequities, or where there is lack of health services, – informed consent requires special consideration. An empirical literature has developed around the informed consent process that raises concern about what participants understand or remember about the consent process [16]. Judgment about the appropriate level of effort to achieve informed consent hinges on the risks and benefits of the survey. For example, clearly worded, plain language, consent has been deemed acceptable in surveys that convey minimal to low risk and additional efforts to ensure complete comprehension of the purpose of the survey are not routinely made.
Maintaining the confidentiality of information from survey participants is an important responsibility including specific components, such as privacy during interviews (particularly around sensitive issues) and prevention of inadvertent disclosure of identity, in analysis and data release products. Concerns about confidentiality are increased with biomarkers because they have the potential for disclosing sensitive health information (e.g., disease status, diabetes, HIV) leading to the misuse of this information to profile individuals for social and health (insurance) purposes. Legal standards to ensure confidentiality have been adopted by governmental statistical agencies [17].
The report of findings of individual laboratory or physiological testing back to survey participants has historically been done in three categories: standard reporting, routine referral, and urgent referral. For example, normal levels of blood pressure are reported to a subject in a standard way. Elevated levels are reported and recommendations or follow-up are made (referrals). Very high levels require urgent referral; recommendations for action are made for urgent action.
When investigators know, or can know, the identity of the tested individual, they are required to report back to the survey participant unless the consent process specified otherwise. A person can consent that a result will not be reported. Consent for not reporting has been considered appropriate when the result of the test has little or poorly understood significance to an individual's health. An example of such a test was the assay for homocystene that was done as part of the NHANES III. The laboratory results for homocystene (that has been under study as a risk factor for cardiovascular diseases) were not reported to participants or their physicians because its physiological or clinical significance was not known. Disease markers that have specific and potentially important health consequences must always be reported to the individual if the investigators can identify the individual and connect the results of the test to that individual. An alternative that has been used to address this issue is to "anonomize" the samples by stripping the identifiers from the sample in a way that makes it impossible to identify the survey participant. In either case, the consent statement should make clear whether the results will or will not be reported back to participants.
Implementation of reporting also depends on a variety of issues in a survey setting. Laboratory analyses that are not performed in the field can not therefore be reported to survey subjects at the time of the survey. Reporting at a time after the initial field work can be a major challenge for large scale surveys. It may also be desirable to share the test results of an individual with a physician designated by the survey participant. Reporting of results in a form that is understandable to survey participants needs to be given high priority when implemented. Reports of survey findings to respondents may also include recommendations of appropriate actions to be taken based on the findings.
Long-term storage of biological specimens collected in population-based surveys has also been common and raises other ethical concerns. Intention to store samples should be included in the consent process. Frequently, biological samples are stored without specific plans for analysis and because testing may take place in the remote future (years, even decades in the future) it cannot be anticipated what tests it will be possible to conduct. It may not be feasible to guarantee that the investigators could reliably contact the person from whom the sample was obtained.
Long term storage also has created the possibility for genetic testing of samples collected in health examination surveys. Standards are still involving the United States and Europe [2]. The U.S. National Center for Health Statistics/CDC/DHHS has approved of six protocols for genetic testing of specimens collected as part of national populations based household surveys (Health and Nutrition Examination Survey, NHANES, III) [18]. Consent in that survey included broad permission for doing unplanned future testing. Genetic tests had not been planned in the original survey and was not specifically included in the consent. With the extensive involvement of the IRB genetic testing has begun under protocols that are considered exploratory. White cells were used to create over 8000 cell lines for future genetic testing. The samples were "anonomized" so that no investigator, including those inside the government can link the sample to the identity of a sample person. The on-going involvement of the NCHS IRB will hopefully lead to creation of guidelines that may be useful to other research settings.
II. Ethical Issues in Examination Surveys in Developing Countries
Universal standards may be the preferable approach to guiding ethical conduct in surveys. Application of universal standards for ethical behavior in survey work in diverse settings, however, requires careful consideration. Implementation of universal standards must consider cultural settings, the health care delivery system of the country, national legal frameworks, and the context of relationships of investigators. This section will discuss ethical issues raised by large-scale surveys in less developed countries that take biological and physiological measures.
Safety
International safety guidelines for the collection of biological and physiological measurements in surveys are the norm for surveys conducted in less developed countries. However, as with clinical practice, achieving these safety guidelines is frequently difficult. Financial constraints in less developed countries often lead to clinical practices that do not follow universal precautions. Lack of appropriate equipment and lack of training may leads to clinical practices that put patients and health care workers at risk. The clinical realities and experience of those who are recruited to work in a health survey has consequences for the implementation of safety guidelines. Safety guidelines may be difficult to enforce in surveys with data collectors drawn from clinical settings in which universal precautions are not observed. Conflicts between international advisors and country implementation staff may surface around expenditures on equipment that is considered unnecessary (latex gloves, disposal facilities). Clear contractual agreements and supervision concerning safety precautions are critical to the maintenance of safety standards.
Informed Consent
Many aspects of consent in examination surveys are similar to standards used in interview surveys in less developed countries. Standards for informed consent must be adapted to cultural circumstances of the country in which the surveys are conducted [19]. Verbal documented consent is routinely used in settings of low literacy. In addition, cultural hierarchies may demand that approval is obtained from the authority figures in the local context (e.g., village elders, head of family). Translation problems must be addressed; often many of the words needed to explain the survey or a procedure do not translate into local languages. The language of consent forms needs to address the specific settings [20].
While coercion or inappropriate inducement is considered a violation of consent, financial incentives in developed country settings are a common practice [21]. Financial incentives are seen as a benefit to off-set lost wages or the opportunity cost of time spent and effort required for participation. The poverty in many countries in which surveys are conducted, means that even the smallest financial incentive may raise concern about undue inducement to participate. Levels of financial incentives for participation that is optimal without being coercive is an empirical issue and must be established for each setting.
Much of the contemporary controversy concerning ethics of research done in less developed countries comes out of clinically oriented research or drug efficacy trials. A distinction between experimental medical research and population-based health surveys helps clarify this discussion of ethical guidelines. The risk inherent in clinical research imposes a responsibility on the investigator for the well being of human subjects. In contrast, surveys collect data that describe health conditions of a population, or monitor and evaluate population-based programs, and usually are of minimal to low risk to human subjects. The risk of population-based surveys is typically related to the burden of questions, time, and sample of body fluid or tissue. The difference between clinical studies and survey research must be made clear to institutional review boards. Including survey expertise on the membership of an institutional review board is a good way to inform these boards of the particular issues raised by survey research.
While these ethical principles are well established, specific procedures (e.g., wordings, verbal versus written, assurance of comprehension) are very much tied to the specific context of the research (complexity of the issues that require consent, risks and benefits). Globally agreed upon principles do not simplify the process of establishing informed consent, and a sometimes lengthy consultation is needed. However, complexities of informed consent are common to resource rich and resource poor settings.
Confidentiality
The practical considerations related to confidentiality in highly developed countries and the less developed countries may differ due to legal requirements, cultural issues, and logistical constraints of the field research. The primary concern in the US and other developed countries is that survey data is not used to disclose the identity of an individual. Linkage of data systems and profiling (by health care providers or insurers) presents concerns about confidentiality in some settings. These sorts of concerns are usually of low priority in less developed countries because such risks are usually minimal or non-existent. In many less developed countries people may not have specific addresses or standardized naming systems. These conditions make it difficult to locate persons after the initial contact of a survey. In a practical sense this reality creates confidentiality and represents a safe guard. In addition, stripping data of unique identifiers and data control procedures are practices that help maintain confidentiality and should be adopted universally [22].
Perhaps more critical in developing countries is privacy in the field setting. Inadvertent disclosure of medical information to family or neighbors during the data collection process in the fields setting may be the major privacy issues raised by field surveys in less developed countries. Surveys conducted in less developed countries, particularly in rural settings and in household surveys, make privacy in data collection a problem. Lack of privacy in household survey settings has consequences on data quality and potential breaches of confidentiality. Communal ways of living, large families, joint households and other elements mean that people congregate quickly and individual interviews are challenging. Training of field staff, careful supervision, and quality control in field work related to privacy standards, are essential if privacy is to be maintained in field settings.
Reporting Results
The reporting of findings of individual biological and physical measurements to survey participants is determined, in part, by the nature of the parameter and the clinical meaning (significance) of the parameter which have universal implication. The social meaning of the disease or condition, the social context in which the report is made and the nature of the health care delivery system, which differ in countries, determine the appropriate manner in which the reporting is done in less developed countries. Health examination surveys in less developed countries have followed U.S. formats for reporting of test results by reporting three categories: standard reporting, routine referral, and urgent referral.
The social meanings of a report of findings in a particular country context may influence the way that a report of a finding is provided. For example leprosy or epilepsy in stigmatized some cultures require additional consideration in reporting. Privacy in reporting would need to be heightened in these contexts.
The policy for reporting the findings of biological and physiological measures to participants of population-based surveys must also consider where the sample testing actually takes place and how the reporting take place. Measures taken in the field for which results are immediately available (blood pressure, rapid testing of blood) can be reported as a routine part of field work. Specimens collected in the field and tested in central laboratory require other reporting procedures. Telephoning or mailing reports of findings is the preferable policy but this may not be feasible in many less developed countries. Long distances, difficulty with travel, and difficulties with the re-identification of sample persons make reporting of centralized testing very difficult.
An option exists to resolve the issues of difficulties raised by having to return to households with test results done after the initial visit. Obtaining consent not to report the particular finding has been acceptable solution. In conditions for which treatments are not available, reporting may not be useful to the sample person (e.g., HSV2) and may not be necessary. Nonetheless consent for not reporting has been recommended. Ethical review committees must weigh the risks and benefits of these types of trade-offs in surveys. The risks and benefits of participating in the survey must be considered with a clear definition of instances under which not reporting findings would be considered appropriate.
The nature of the health care delivery system (that is, the level of its development) is also critical to the report of findings. Reporting a finding and recommending an action that is not a realistic possibility for the patient is a usual challenge that policy for report of findings must face. Referral to the existing standard of care in the country is the usual policy that has been proposed. More on the issue of the responsibilities of survey researchers to provide services is discussed in the next section
Service and the Provision of Care
The finding of abnormal results of a biological or physiological measure in a population-based survey may necessitate referral. In less developed countries the policy for referral must consider the types of services that are available to the survey population. A referral to a poorly functioning health care delivery system is problematic. The notion that the prevailing standard of care in the country is a sufficient referral has been challenged by some [23]. These sorts of challenges have been faced in the context of clinical research in which investigators from countries with higher levels of services have offered lower levels of care. In clinical research, investigators may have a relationship with the study populations over a period of years and subject the population to circumstances (drugs) that are not risk free. The referral to prevailing standards of care in the country may not be considered adequate. However, since population-based surveys make only transient contact with the participants (that is a clinical relationship is not established or intended) and because risk is generally low, relying on the prevailing standard of health care in the areas may be appropriate.
Does the absence of adequate medical services imply that tests in surveys should not be performed? The question might be reframed, "To what extent is a report of a disease or condition considered to be a benefit when treatment is not immediately available?" Risks of knowledge, in and of itself, may be considered minimal. It can be stated that the primary purpose of data collected in a national survey is not to provide benefit at the individual level, but to collect information to benefit the whole population in the long term. Thus, a discussion of the benefits from a survey needs to balance the immediate benefit to participants and to the nation.
Ethical considerations about beneficence are often mixed with humanitarian concerns to provide health care services to those in need. Historically in the United States the consideration of risks and benefits of most surveys has not lead to recommends that require that health care be provided to survey participants. The health care needs of participants and their communities fuels considerations to use survey work as a vehicle to provide care. The doctors or nurses on the survey team may be the only health care professional that is actually available in a resource poor setting. However, building a transient health care provision system onto a survey can not take the place of national health care system development. In addition, ethical considerations about coercion or inducements must be considered. Survey participation should not be perceived by the participants as a criterion for receiving services.
Special considerations for HIV testing
HIV testing in national population-based surveys in less developed countries has been successfully completed in a number of less developed countries [4,5,24]. The inclusion of HIV as a biomarker requires special consideration because of the stigma attached to the disease and the life threatening nature of the disease. Disclosure of HIV status is a life and death issue in some countries making the standards for confidentiality in such surveys very high. Confidentiality must be maintained in the field (protection of disclosure to family and neighbors in communities) and in data sets (protection of survey participants from use of the data set to identify HIV positive individuals). The burden of maintaining confidentiality among field staff members in the context of rapid testing in the field is a real concern. In a number of research protocols, it would be important to debate the obligation to report the finding if the researchers can know the HIV status of a person in a study.
De-linked protocols ensure the confidentiality of the data, but disappoint advocates of voluntary counseling and testing (VCT) who look at reporting results to survey participants as a service. Indeed, HIV test results (received voluntarily) can be considered as a public health good and as a benefit to the individual. Reconciliation of needs for confidentiality and service has been found in the provision of VCT services in conjunction with surveys that collect de-linked data. This necessitates repeat testing of those who want to know their HIV status.
III. Discussion and recommendations
This discussion of the ethics related to the incorporation of biomarkers into population-based surveys is a starting point for agencies conducting health surveys in less developed countries and those providing technical assistance and funding. Institutional responses to ethical conduct of science in less developed countries have been emphasized here and this institutional capacity should be considered integral to the capacity building in health that is supported by many international donor agencies. Developing countries must have their own sustainable institutional frameworks to discuss, develop, and implement ethical protections which are morally sound and nationally relevant [25]. International harmonization of ethical principles and procedures as they relate to population-based surveys using biomarkers need to be established. A set of standards for ethical practice may be preferable to the development of different standards for developed and less developed settings. Current practices in use may be appropriate but need to be explicitly discussed and debated for adoption in less developed settings. While implementation of surveys will differ in various countries, the principles underlying their conduct should not differ. Ethical issues raised by population-based surveys are distinct from those raised by clinical research. Defining the benefit of survey work to participants needs to be made explicit but the mission of a survey should not be confused with a clinical mission. At the same time, the knowledge or specific information a sample person learns about their health status should not be considered an inducement for survey participation.
Development of capacity for ethical review of survey research in less developed countries should be a part of technical assistance from high income countries. While ethical principles may be global, procedures must be adapted to local conditions, and local IRBs must be able to ensure that ethical principles are translated into implementation plans. Survey activities would be well served by the development of more sophisticated Institutional Review Boards in host countries that undertake survey work. Development agencies and their contractors may wish to invest in this development, and capacity building in the health sector should formally include work in this area. The composition of IRBs should also include those who understand the technical aspects of survey work, as the particulars of research are essential in the appropriate evaluation of ethical procedures and standards.
Population-based surveys frequently have political consequences that must be anticipated by donors and host countries. While these considerations are important, they should not become confused with the ethical standards that frame survey design. The institutional integrity of IRBs in donor and in host-recipient countries is essential. The power differential between donor and host institutions must be recognized when institutional rules are established. A practical recommendation is for the cost of the ethical review to be borne by the research partner from the high income country as part of the technical assistance offered in a particular research project. Studies funded by rich countries should fund ethical review of surveys in poor countries in such a way that level of functioning and autonomy of local IRBs (Institutional Review Boards) can be ensured.
International donor agencies should facilitate collection of research experience and dissemination of such experience. An exploration of such practices can be done to establish contemporary practices. The National Center for Health Statistics of the Centers for Disease Control and Prevention (CDC) has conducted the National Health and Nutrition Examination Survey (NHANES) in the United States for over 30 years. The written policies and procedures of the NHANES represent the standard for the use of health assessments in population-based surveys in the United States.
Because many biological specimens can be used for genetic testing, specimens being collected in less developed countries will draw great interest. One practical recommendation to address this inevitability is to follow the approach that is being used in the United States that has allowed provisional use of samples for genetic testing in an exploratory setting under on-going scrutiny of an IRB. This proposed exploratory study of ethical standards should be well supervised and well staffed by IRBs in both the donor country and in the host country. A well conducted study will maximize the learning concerning ethical standards for genetic testing and safe guard the rights of the sample persons. IRBs should require intimate involvement in such exploratory work with frequent reporting back to evaluate the practical issues raised by genetic testing. Who owns genetic materials once it is collected is a unique issue that must be addressed. Experience with genetic testing in less developed countries may have implication for a discussion that is increasingly global in nature.
Acknowledgements
The authors would like to thank Wilbur Hadden and Geraldine McQuillan of the National Center for Health Statistics/ CDC/ Department of Health and Human Services for assistance on this paper. The two peer reviewers are also to be thanked for their excellent comments.
==== Refs
Boerma J Holt E Black R Measurement of Biomarkers in Surveys in Developing Countries: Opportunities and Problems Popul Dev Rev 2001 27 303 14 10.1111/j.1728-4457.2001.00303.x
Finch C Vaupel J Kinsella K (Eds) Cells and Surveys: Should Biological Measures Be Included in Social Science Research? Committee on Population 2001 National Research Council National Academy Press: Washington DC
Uzbekistan National Health Survey 2004 2004 ORC Macro: Calverton, MD
Dominican Republic Demographic and Health Survey 2002 2002 ORC Marco: Calverton, MD
Mali Demographic and Health Survey 2001 2001 ORC Macro: Calverton, MD
Biological and Clinical Data Collection in Population Surveys in Less Developed Countries Summary of a meeting held by MEASURE Evaluation National Academy of Sciences: January 24–25, 2000; Washington DC
London L Ethical Oversight of Public Health Research: Can Rules and IRBs Make a Difference in Developing Countries? AJPH 2002 92 1079 84
CIOMS International Ethical Guidelines for Biomedical Research Involving Human Subjects [draft] 2002 Geneva, Switzerland: Council of International Organizations of Medical Sciences (CIOMS)
Butler D Call for North/South code of research ethics Nature 2000 406 337 10935609 10.1038/35019216
Tangwa GB Moral Agency, Moral Worth and the Question of Double Standards in Medical Research in Developing Countries Developing World Bioeth 2001 1 156 162 12872761 10.1111/1471-8847.00022
Fisher G Pappas G Limb M Prospects Problems and Prerequisites for National Health Examination Surveys in Less Developed Countries Soc Sci Med 1996 42 1639 1650 8783426 10.1016/0277-9536(95)00319-3
Wagener DK Ethical considerations in the design and execution of the National and Hispanic Health and Nutrition Examination Survey (HANES) Environ Health Perspect 1995 103 75 80 7635116
Pakistan Medical Research Council Health Profile of the People of Pakistan Islamabad 1997
McQuillan G Speers MA Burt V Applying the Federal Regulations for Protecting Human Subjects in Survey Research: The NHANES Experience [unpublished manuscript] NBAC archive 2000
Schulte Hunter Rothman Ethical and Social Issues in the Use of Biomarkers in Epidemiologic Research IARC Sci Publ 1997 142 313 318 9354930
Brody BA Making informed consent meaningful IRB 2001 23 1 5 12737171 10.1385/IR:23:1:01
Zion D Gillam L Loff B The Declaration of Helsinki, CIOMS and the ethics of research on vulnerable populations Nat Med 2000 6 615 617 10835665 10.1038/76174
64 FR 29321-29325 National Health and Nutrition Examination Survey Program DNA Specimens; guidelines Federal Register Centers for Disease Control and Prevention June 1, 1999
Castillo FA Limiting Factors Impacting on Voluntary First Person Informed Consent in the Philippines Developing World Bioeth 2002 2 21 27 10.1111/1471-8847.00032
Yoder Stanley P Konaté MK Obtaining Informed Consent for HIV Testing: The DHS Experience in Mali 2002 ORC Macro: Calverton, MD
Findlay JS Schaible WL A Study of the Effect of Remuneration Upon Response in the Health and Nutrition Examination Survey Vital and Health Statistical Series 2 1975 Washington DC: GPO 76 1341 [DHEN (Series Editor): No. 67]
Oonagh C Empty ethics: the problem with informed consent Sociol Health Illn 2003 25 768 92 16523548 10.1046/j.1467-9566.2003.00369.x
Angell M The ethics of clinical research in the Third World N Engl J Med 1997 337 847 849 9295243 10.1056/NEJM199709183371209
Smith CB Battlin MP Jacobson JA Are there characteristics of Infectious Diseases that raise special ethical issues? Developing World Bioeth 2004 4 1 16 15086371 10.1111/j.1471-8731.2004.00064.x
Macpherson CC Searching for Consensus Bioethics 2004 18 283 292 15341040 10.1111/j.1467-8519.2004.00395.x
|
16313670
|
PMC1315319
|
CC BY
|
2021-01-04 16:39:02
|
no
|
Global Health. 2005 Nov 28; 1:16
|
utf-8
|
Global Health
| 2,005 |
10.1186/1744-8603-1-16
|
oa_comm
|
==== Front
Health Res Policy SystHealth Research Policy and Systems1478-4505BioMed Central London 1478-4505-3-71630067210.1186/1478-4505-3-7ResearchA systematic review of existing national priorities for child health research in sub-Saharan Africa Swingler George H [email protected] James H [email protected] William M [email protected] Félix [email protected] Martin M [email protected] School of Child and Adolescent Health, University of Cape Town, South Africa2 Directorate of Primary Health Care, University of Cape Town, South Africa3 Department of Paediatrics and Child Health, University of Nairobi, Kenya4 Department of Paediatrics, Faculty of Medicine and Biomedical Sciences, University of Yaoundé, Cameroon5 Department of Paediatrics, University of Calabar, Nigeria2005 21 11 2005 3 7 7 22 6 2005 21 11 2005 Copyright © 2005 Swingler et al; licensee BioMed Central Ltd.2005Swingler 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
We systematically reviewed existing national child health research priorities in Sub-Saharan Africa, and the processes used to determine them.
Methods
Collaborators from a purposive sample of 20 WHO-AFRO Region countries, assisted by key informants from a range of governmental, non-governmental, research and funding organisations and universities, identified and located potentially eligible prioritisation documents. Included documents were those published between 1990 and 2002 from national or nationally accredited institutions describing national health research priorities for child health, alone or as part of a broader report in which children were a clearly identifiable group. Laboratory, clinical, public health and policy research were included. Two reviewers independently assessed eligibility for inclusion and extracted data.
Results
Eight of 33 potentially eligible reports were included. Five reports focused on limited areas of child health. The remaining three included child-specific categories in reports of general research priorities, with two such child-specific categories limited to reproductive health. In a secondary analysis of Essential National Health Research reports that included children, though not necessarily as an identifiable group, the reporting of priorities varied markedly in format and numbers of priorities listed, despite a standard recommended approach. Comparison and synthesis of reported priorities was not possible.
Conclusion
Few systematically developed national research priorities for child health exist in sub-Saharan Africa. Children's interests may be distorted in prioritisation processes that combine all age groups. Future development of priorities requires a common reporting framework and specific consideration of childhood priorities.
==== Body
Background
Africa experiences a huge burden of childhood disease in a context of limited resources for health care and research. Sixty five percent of the burden of disease in sub-Saharan Africa in 1990 was attributable to childhood conditions [1]. In 2001 28 of the 30 countries with the highest under-5 mortality rates were in Africa, and the under-5 mortality rate for sub-Saharan Africa was almost 25 times the average rate for industrialised countries [2]. Improvements in child health in Africa have been attributed to the findings of research – such as in vitamin A deficiency, malaria and mother to child transmission of HIV – and further research has been described as fundamental to further improvements [3]. Research is also important in guiding cost-effective policymaking. Because of severely limited resources, prioritisation of research is essential. A recent report of the status of health research in Africa highlighted the need for attention to research priority setting [4]. It cited colonial government interests as having determined priorities in the region prior to independence, and suggested that in the post-independence period priority setting has been haphazard, and determined by institutions or individuals rather than based on country or regional needs.
This study systematically reviews existing national child health research priorities in Sub-Saharan Africa, and the processes used to determine them.
Methods
Identification of reports on national child health research priorities
We took a purposive sample of 20 of the 45 countries in sub-Saharan Africa (corresponding to the WHO-AFRO Region, excluding Algeria). This sample was selected to provide an overall sample that, at face value, represented sub-Saharan Africa geographically, linguistically and with respect to the most important determinants of child health. It included all 15 WHO-AFRO countries in sub-Saharan Africa that had previously participated in the Africa Regional Consultative Process with respect to the status of health research in Africa (Benin, Burkina Faso, Burundi, Cameroon, Cote d'Ivoire, Ethiopia, Guinee Conakry, Kenya, Mali, Mauritius, Nigeria, Senegal, South Africa, Tanzania and Zambia) [4]. For this study, five more countries (Angola, Chad, Democratic Republic of Congo, Ghana and Zimbabwe) were added to improve on representiveness. The total population in the countries included in this survey represented 80% of the total population of sub-Saharan Africa [5], and 76% of the MEDLINE-indexed research output on child health from the region over the study period.
For the primary search, collaborators in each country identified personnel in national Ministries of Health, universities, research institutes, non-governmental organisations and funding organisations who had participated in national health research forums, or who were considered to be knowledgeable about national health research priorities by virtue of their professional positions. These 'key informants' were then surveyed by means of a pre-piloted mailed self-administered questionnaire. They were first contacted by telephone, or in person when judged appropriate. Country collaborators collected questionnaires, clarified responses in the questionnaire and followed up non-respondents (generally first by mail, fax or email, with non-respondents thereafter followed up by telephone and direct contact). They also obtained and forwarded to the authors copies of all available reports that potentially met the study inclusion criteria.
MEDLINE was also searched in order to locate additional indexed reports from the sampled countries. The MeSH terms "Health Services Research", "Health Priorities", "Health Policy", "Nutrition Policy", "Policy Making", text words "policy", "research", and "priorit*", and MeSH country terms, were used.
A secondary search was performed for Essential National Health Research (ENHR) reports on priority setting processes from all 45 countries in the sub-region, whether sampled for the primary analysis or not. This broad search was restricted to ENHR reports because the central promotion and monitoring of the process was expected to enable reliable identification of reports from central sources. The standard process recommended for setting priorities and compiling reports was also expected to enable a comparison of sampled and non-sampled countries. The search was augmented by information obtained from the website of the Council on Health Research for Development (COHRED) [6], and from members of the African Health Research Forum, COHRED, ENHR focal points, and the authors' personal networks.
Inclusion criteria for reports on national child health research priorities
Documents eligible for inclusion were all reports or other formal documents, dated from 1990 to 2002, from a national or nationally accredited institution, describing national research priorities for child health, alone or as part of a broader report in which children were an identifiable group. There was no language restriction.
Childhood covered any age from birth to 18 years. The term "research" referred to basic (laboratory), clinical, public health or policy research. "Child health" included both health or nutritional conditions (e.g. diarrhoea, Vitamin A deficiency) or determinants of child health (e.g. breastfeeding, tobacco smoke), provided that studies of the determinants included health-related outcomes or associations. Reports could prioritise both between different health conditions, within a single health condition, or between risk factors for childhood disease or malnutrition.
Critical assessment of reports on national child health research priorities
The quality of reports was assessed using criteria modified from those proposed by the Global Forum for Health Research for consideration when setting health research priorities [7]. viz. consideration of i) burden of disease; ii) determinants of disease; iii) the burden of determinants of disease; iv) the present level of knowledge, and v) the cost-effectiveness of interventions. In addition, the interest groups participating in the setting of priorities were recorded for use in the determination of the breadth of participation.
Potentially eligible reports identified by country collaborators were independently assessed for inclusion by two reviewers, with disagreements resolved through consensus.
Data extraction and statistical analysis
Pre-specified data were extracted independently by two reviewers onto a pre-designed data extraction form. Disagreements were resolved by consensus. French language documents were examined by French speaking reviewers. Data extracted included date of report, the presence of the inclusion criteria, the health conditions and determinants considered, interest groups and institutions participating, funders and the presence of the quality criteria specified above. Listed priorities were extracted verbatim for later analysis, translated where necessary from French into English by a bilingual author (FT).
Quantitative synthesis was not attempted.
Ethical approval
Ethical approval was obtained from the Research Ethics Committee of the University of Cape Town (ref 251/2002).
Results
Two hundred and fifteen key informants responded to the questionnaire survey. Thirty three potentially eligible reports were identified from 14 of the 20 sampled countries (Figure 1). The 14 countries from which reports were identified generated 99.4% of the MEDLINE-indexed research output on child health from the sample for the period under review. Eight studies met the study inclusion criteria [8-15] (Table 1). Of the 25 excluded reports, 12 would have been eligible if priorities had been reported with children as an identifiable group [16-28]. Characteristics of included studies plus those that would have been eligible if children had been an identifiable group are shown in Table 2.
Figure 1 Identification of studies.
Table 1 Reports identified
Included TOTAL
Yes No*
Burkina Faso 3 (3) 3
Cameroon 1 (1) 1
Cote d' Ivoire 1 (0) 1
Ethiopia 1 (0) 1
Ghana 1 (1) 1
Guinea 1 (1) 1
Kenya 2 (1) 2
Mali 1 (1) 1
Nigeria 2 1 (0) 3
Senegal 1 1
South Africa 4 10 (2) 14
Tanzania 2 (1) 2
Zambia 1 1
Zimbabwe 1 (1) 1
TOTAL 8 25 (12) 33
* The numbers in brackets represent reports that would have been eligible if child priorities had been separately identifiable
Table 2 Included reports
Name of Report Country Date Scope
Child health priorities separately identifiable
Priorities for Health Research in Nigeria.[8] Nigeria 2001 General
Handbook for Health Systems/Operations Research at Local Government Level.[9] Nigeria 1993 Health systems
Interim findings on the National PMTCT Pilot Sites.[10] South Africa 2002 HIV/AIDS
Workshop on an integrated policy for school health.[11] South Africa 1997 School health
Chronic Disease of Childhood. Workshop proceedings.[12] South Africa 1999 Chronic diseases
Saving Babies 2001. Second perinatal care survey of South Africa.[13] South Africa 2001 Perinatal
Programme National de Recherche en Santé (National Programme of Health Research).[14] Senegal 2001 General
Zambia National Health Research Agenda.[15] Zambia 1999 General
Child health priorities not distinguished from adult priorities
Les programmes d'intervention nutritionnelle au Burkina Faso (Nutritional intervention programmes in Burkina Faso).[16] Burkina Faso 2001 Nutrition
Actes du Premier Symposium sur la Recherche Nationale en Santé au Burkina Faso (Proceedings of the first symposium on national essential health research in Burkina Faso).[17] Burkina Faso 1997 General
Plan National d'Action pour la Nutrition [version revisée]. (National Action Plan for Nutrition [revised version]).[18] Burkina Faso 2001 Nutrition
Recommendations of the National Symposium on Medical Research in Cameroon.[19] Cameroon 2002 General
Medium Term Health Strategy: Towards Vision 2020.[20] Ghana 1995 General
Atelier de réactualisation des priorités nationales en matière de recherche en santé en République de Guinée (National workshop on definition of health research priorities).[21] Guinea 2000 General
Conceptual Framework for Essential National Health Research in Kenya.[22,23] Kenya 1994 General
First National Symposium on Health Research Priority Setting in Mali.[24] Mali 2001 General
Foresight Health Report.[25] South Africa 1999 General
Proceedings of the First Essential National Health Research Congress on Priority setting.[26] South Africa 1996 General
Tanzania Essential National Health Research Priority Setting Workshop. Final Report.[27] Tanzania 1999 General
The Essential National Health Research.[28] Zimbabwe 1995 General
The extent to which included reports satisfied the pre-specified quality criteria is shown in Table 3, together with reports that would have been eligible if priorities had been reported with children as an identifiable group. Six (30%) of 20 reports satisfied three or more of the five quality criteria, with the number of criteria met ranging from zero to five. Table 4 lists the interest groups participating in the prioritisation processes of the reports
Table 3 Quality criteria met by included reports
Included reports (%), n = 8 Reports that would have been eligible without the requirement of child-specific priorities (%), n = 20
Consideration of: Yes No Unclear Yes No Unclear
burden of disease 6 (75) 2 (25) 0 (0) 14 (70) 4 (20) 2 (10)
determinants of disease 5 (62.5) 2 (25) 1 (12.5) 10 (50) 8 (40) 2 (10)
burden of determinants of disease 2 (25) 5 (62.5) 1 (12.5) 2 (10) 16 (80) 2 (10)
existing knowledge 4 (50) 4 (50) 0 (0) 10 (50) 10 (50) 0 (0)
cost-effectiveness of interventions 2 (25) 5 (62.5) 1 (12.5) 5 (25) 14 (70) 1 (5)
Table 4 Interest groups participating in the prioritisation processes of included reports
Included reports
(%), n = 8 Reports that would have been eligible without requirement of child-specific priorities
(%), n = 20
Yes No Unclear Yes No Unclear
Researchers 6 (75) 1 (12.5) 1 (12.5) 16 (80) 2 (10) 2 (10)
Institutions of learning 5 (62.5) 2 (25) 1 (12.5) 15 (75) 3 (15) 2 (10)
Health managers 6 (75) 1 (12.5) 1 (12.5) 16 (80) 2 (10) 2 (10)
Policy makers 6 (75) 1 (12.5) 1 (12.5) 17 (85) 2 (10) 1 (5)
Non-governmental organisations 3 (37.5) 4 (50) 1 (12.5) 12 (60) 6 (30) 2 (10)
Consumers 2 (25) 5 (62.5) 1 (12.5) 9 (45) 9 (45) 2 (10)
Four of the eight studies with children as an identifiable group addressed limited areas of child health only i.e. mother to child transmission of HIV, school health, chronic diseases of childhood, and perinatal care [10-13]. All four were from South Africa. The remaining four reports, from Nigeria (two), Senegal and Zambia, covered research priorities for both children and adults and included specific categories of child health [8,9,14,15]. One of these four reports included a section dealing with general child health [15], one dealt with maternal and child health in a report limited to health systems research priorities [9], one with "Adolescent health and sexuality" [8], and one with "Pregnancy and delivery, pregnancy and oral cavity diseases, mother and child care, adolescent health" limited entirely to the pre-natal period [14]. The wide variation in the areas of interest of the eight reports in the primary analysis precluded any attempt at synthesis of priorities.
Eighteen reports from 12 of 20 sampled countries reported ENHR processes (Figure 1). Of the other eight countries in the sample, there was no record of ENHR processes having been initiated in three countries, and confirmation of no report of priorities in a fourth [6]. ENHR documents were thus identified from 12 (75%) of the 16 countries from which they were potentially available by the primary search. The secondary search for ENHR reports from all 45 countries in the sub-region identified four potentially eligible reports from two countries not included in the primary sample (Malawi and Uganda), two of which were found to be eligible. In total therefore, 14 reports from 14 countries met the secondary inclusion criteria; two from non-sampled countries (Malawi and Uganda) [8,14,15,17,19-24,26-30]. All of the reports dealt with general priorities. Four reports covering research priorities for both children and adults included specific categories on child health (from Nigeria, Senegal, Uganda and Zambia) [8,14,15,30]. Only two reports listed research priorities for child health as a whole. Except for possible overlap in the area of nutritional interventions, there were no specific priorities common to both [15,30]. The two other reports were those described above that dealt with reproductive health [8,14].
Despite a standard ENHR process for prioritisation, there was wide variation in the frameworks for the categorisation of research priorities, and in the number of categories and sub-categories in each report. Some reports listed only specific health conditions, while others categorised by other, but varying, frameworks such as health systems, public health and socio-cultural issues (Table 5). The listed priorities included one to five tiers of categorisation (median two). The number of first-tier categories of priorities varied from three to 26 (median five). Of the 10 reports with second-tier categories, the total number of priorities listed in the second-tier categories varied from 10 to 90 (median 28). Three of the 14 reports offered a scoring or ranking of research priorities. There were too few ENHR reports from non-sampled countries identified in the secondary search to compare priorities of sampled with non-sampled countries.
Table 5 Types of health-related categories listed in first- and second-tier headings of reports
First tier headings (n = 14) Second tier headings (n = 10)
Specific health conditions 11 (78.6%) 10 (100%)
Determinants of health conditions 0 (0%) 5 (50%)
Nutritional issues 6 (42.9%) 9 (90%)
Public health issues 12 (85.7%) 10 (100%)
Research issues 3 (21.4%) 3 (30%)
Discussion
This study aimed to provide as valid and replicable an overview of existing child health research priorities as possible, and to describe the processes whereby the priorities were set. To do so, a pre-specified systematic approach was followed, adapted from the process now widely accepted in healthcare practice [31,32]. This approach to health research priorities does not adequately address nuances of meaning and local context, and thus represents an incomplete picture. Although insufficient in itself, we suggest that this analytical approach is an essential component of the assessment of research prioritisation processes, particularly because of the shortcomings of existing processes identified by this study.
Prioritisation reports were identified from the 14 countries that generated 99.4% of the MEDLINE-indexed child health research output of the entire sample of 20 countries. If it is assumed that the countries producing the bulk of published research also generate the bulk of prioritisation processes, the study appears to have identified prioritisation reports from the countries in which they might be expected to exist. ENHR documents were obtained from 12 of 16 countries from which they were potentially available, representing a minimum 75% success rate in identifying ENHR documents. The sample itself was limited to 20 of the 45 countries in the WHO-AFRO region (excluding Algeria), representing 80% of the total population under study and 76% of the MEDLINE-indexed research articles on child health from the region. The data collected therefore appear to be reasonably representative of the population under study.
The quality of reports was variable, meeting between zero and five of the five quality criteria modified from the approach suggested by the Global Forum for Health Research for the development of priorities [7]. However, these criteria were not proposed primarily as quality assessment criteria and have not to our knowledge been validated as such. The representation on prioritisation teams was generally broad.
The most striking finding of this review is the dearth of systematically developed national research priorities in child health and child nutrition in sub-Saharan Africa. Only eight documents that offered child-specific priorities were identified from the sampled countries. In the bulk of identified prioritisation processes children were included but research priorities were not considered, or reported, in a manner that enabled a separate assessment of childhood priorities. This is concerning, given the differences in the health issues confronting children and adults, and the 65% of burden of disease in sub-Saharan Africa in 1990 attributable to conditions occurring in children [1].
Of the general reports listing research priorities for all ages, only three included child-specific categories. Two of these covered reproductive health only; one antenatal care only (i.e. pregnant children). It is unlikely that the reproductive health of children is the only priority for child health research. The prominence of reproductive health as a priority for children is probably a distortion due to overlap with adult research priorities, and suggests that children's interests are not adequately represented in processes that combine all ages. Only one (ENHR) report from the sample presented priorities for overall child health, with one additional ENHR report identified from a non-sampled country. No specific listed research priorities were common to both countries (Zambia and Uganda), possibly reflecting the variation in reporting even when a relatively standard process is used.
The secondary analysis of ENHR reports, regardless of whether children were dealt with separately, identified other obstacles to the comparison and synthesis of childhood research priorities. Even for ENHR processes, which use a standard recommended approach, the reporting of priorities varied markedly in format, with different methods of categorisation and numbers of listed priorities. The large number of research priorities listed themselves required prioritisation, but only three of 14 reports provided a ranking or weighting. These factors made comparison and synthesis of reported research priorities very difficult. A common conceptual framework for the reporting of priorities would greatly facilitate a meaningful overview of research priorities. If such a framework makes specific provision for child health and child nutrition, it could improve specific consideration and reporting of such priorities.
The shortcomings in current prioritisation processes and the difficulties encountered in synthesising research priorities raise the question of whether it is appropriate to attempt to develop sub-regional and regional research priorities by synthesising existing national priorities. This approach has the advantage of utilising existing work that is in any case necessary at national level, and of building on locally developed priorities. However, for this to be a viable approach, considerable structural changes in current processes are necessary. Challenges for the coherent development of research priorities appear to include the development of national health research system assessments that use a common conceptual framework and include specific consideration of research priorities for children.
Alternatives to this approach include regional or sub-regional application of centrally developed global research priorities, or a more qualitative and consultative synthesis of national priorities. The former involves a top-down approach that may compromise local applicability, while both processes are vulnerable to the problems of incomplete representation.
Conclusion
Few systematically developed national research priorities for child health exist in sub-Saharan Africa. Children's interests may be distorted in prioritisation processes that combine all age groups. Future development of priorities requires a common reporting framework and specific consideration of childhood priorities.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
All authors participated in the conception and design of the study, and the acquisition, analysis and interpretation of data. GS wrote the first draft of the manuscript, and all authors participated in its critical revision for important intellectual content. All authors have seen and approved the final version. GS and JI obtained funding and co-ordinated the study. GS is the guarantor.
Funding
The study was commissioned by the Child Health and Nutrition Research Initiative (CHNRI) of the Global Forum for Health Research and supported by a contribution from the World Bank through its grant facility to the Global Forum. CHNRI commented on the study design, but played no role in the collection, analysis or interpretation of data, the writing of the report or the decision to submit the paper.
Acknowledgements
We acknowledge the advice and support of Prof. Marian Jacobs, Prof. Kim Mulholland and Ms Griet Onsea in planning this study.
==== Refs
Murray CJL Lopez AD The Global Burden of Disease, a comprehensive assessment of mortality and disability from diseases, injuries, and risk factors in 1990 and projected to 2020 1996 Harvard: Harvard Univ Press
United Nations Children's Fund (UNICEF) The State of the World's Children 2003 New York 2002
Dabis F Orne-Gliemann J Perez F Leroy V Newell ML Coutsoudis A Working Group on Women and Child Health. Improving child health: the role of research BMJ 2002 324 1444 7 12065272 10.1136/bmj.324.7351.1444
Mugambi M Onsea G Regional Consultative Process Africa Report prepared for International Conference on Health Research for Development, 2000 Oct 10–13, Bangkok
World Health Organization World Health Report: 2002: Reducing risks, promoting healthy life Geneva 2002
Council on Health Research for Development (COHRED) ENHR activities and reports
Global Forum For Health Research The 10/90 Report On Health Research 2000 Geneva 2000
Federal Ministry of Health Priorities for Health Research in Nigeria 2001 Federal Republic of Nigeria
National Primary Health Care Development Agency of Nigeria Handbook for health systems/operations research at local government level Nigeria 1993
Health Systems Trust and Department of Health Interim findings on the National PMTCT Pilot Sites. Department of Health, South Africa, 2002
Child Health Policy Institute, University of Cape Town Workshop on an integrated policy for school health Cape Town 1997
Child Health Policy Institute, University of Cape Town Chronic Disease of Childhood Workshop proceedings Cape Town 1999
MRC Unit for Maternal and Infant Health Strategies, PPIP Users, and the National Department of Health, South Africa Saving Babies 2001 Second perinatal care survey of South Africa South Africa 2001
Ministère de la Sante et de la Prévention Programme National de Recherche en Santé (PNRS) Senegal [National Programme of Health Research] 2001 [In French]
Central Board of Health Zambia National Health Research Agenda Zambia 1999
Les programmes d'intervention nutritionnelle au Burkina Faso [Nutritional intervention programmes in Burkina Faso] 2001 [In French]
Ministère de la Sante & Ministère des Enseignements Secondaire, Supérieur et de la Recherche Scientifique Actes du Premier Symposium sur la Recherche Nationale en Santé au Burkina Faso [Proceedings of the first symposium on national essential health research in Burkina Faso] 1997 [In French]
Plan National d'Action pour la Nutrition (version revisée) Burkina Faso [National Action Plan for Nutrition (revised version)] 2001 [In French]
Recommendations of the National Symposium on Medical Research in Cameroon Cameroon 2002
Republic of Ghana Medium Term Health Strategy: Towards Vision 2020 Accra 1995
Atelier de réactualisation des priorités nationales en matière de recherche en santé en République de Guinée, 26–29 Avril 2000, Conakry [National workshop on definition of health research priorities] [In French]
Abdullah MS Conceptual framework for essential national health research in Kenya Proceedings of the National Health Research Plan Kenya, 25–27 May 1994, Mombasa, Kenya 1994
Ministry of Health First National Congress on Quality Improvement in Health Care, Medical Research and Traditional Medicine Nov 19–23, 2001, Nairobi, Kenya
Ministry of Health First National Symposium on Health Research Priority Setting in Mali 2–4 Aug 2001, Bamako, Mali
Department Arts, Culture, Science and Technology Foresight Health Report South Africa 1999
Directorate Research Coordination and Management, Department of National Health Proceedings of the First Essential National Health Research Congress on Priority setting 14–15 Nov 1996, Pretoria, South Africa
National Institute for Medical Research Tanzania Essential National Health Research Priority Setting Workshop Final Report 15–21 Feb 1999, Arusha, Tanzania
Medical Research Council of Zimbabwe The Essential National Health Research Priority Setting Process 22–25 Aug 1995, Nyanga, Zimbabwe
Essential National Health Research Priority Setting Workshop 18–19 Jun 2001, Mangochi, Malawi
Uganda Ministry of Health and Uganda National Council for Science and Technology A report of a workshop on the National Health Research Plan for the period 1997–2001 1–2 Sep 1997, Kampala, Uganda
Cook DJ Mulrow CD Haynes RB Systematic reviews: synthesis of best evidence for clinical decisions Ann Intern Med 1997 126 376 380 9054282
Petticrew M Systematic reviews from astronomy to zoology: myths and misconceptions BMJ 2001 322 98 101 11154628 10.1136/bmj.322.7278.98
|
16300672
|
PMC1315320
|
CC BY
|
2021-01-04 16:37:16
|
no
|
Health Res Policy Syst. 2005 Nov 21; 3:7
|
utf-8
|
Health Res Policy Syst
| 2,005 |
10.1186/1478-4505-3-7
|
oa_comm
|
==== Front
Lipids Health DisLipids in Health and Disease1476-511XBioMed Central London 1476-511X-4-291630574110.1186/1476-511X-4-29ResearchIncorporation of branched-chain fatty acid into cellular lipids and caspase-independent apoptosis in human breast cancer cell line, SKBR-3 Wongtangtintharn Sawitree [email protected] Hirosuke [email protected] Hironori [email protected] Masashi [email protected] Takayoshi [email protected] Teruyoshi [email protected] United Graduate School of Agricultural Sciences, Kagoshima University, 1-21-24, Korimoto, Kagoshima 890-0065, Japan2 Center of Molecular Bioscience, University of the Ryukyus, Nishihara, Okinawa 903-0213, Japan3 Department of Clinical Laboratory Medicine, University of the Ryukyus Hospital, School of Medicine, Uehara 207, Nishihara, Okinawa 903-0125, Japan4 Department of Applied Biological Sciences, Saga University, Saga 840-8502, Japan2005 23 11 2005 4 29 29 27 10 2005 23 11 2005 Copyright © 2005 Wongtangtintharn et al; licensee BioMed Central Ltd.2005Wongtangtintharn 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
13-Methyltetradecanoic acid (13-MTD), an iso-C15 branched- chain saturated fatty acid, has been shown to induce apoptotic cell death of numerous human cancer cells. However, the mechanism for the induction of apoptosis has not been fully understood. This study described the incorporation of 13-MTD into cellular lipid of SKBR-3 breast cancer cells and apoptosis related event to gain more insight into the mechanism action of this fatty acid.
Results
Treatment of SKBR-3 cells with 13-MTD lowered the cell viability and induced apoptosis. Proportion of 13-MTD in the glycerolipids increased to saturation level within 6 hours. Triacylglycerol contained 13-MTD in higher concentration than phospholipid with positional preference to sn-2. 13-MTD caused no changes in the caspase activity and its gene expression. Furthermore, addition of caspase-inhibitor to culture medium did not prevent the cells from the cytotoxicity of 13-MTD. No-increase in the cellular calcium level was also noted with 13-MTD treatment. However, 13-MTD disrupted the mitochondrial integrity in 4 hours, and increased the nuclear translocation of apoptosis inducing factor.
Conclusion
These results showed that 13-MTD disrupted the mitochondrial integrity, and induced apoptosis via caspase-independent death pathway.
branched-chain fatty acidincorporationglycerolipidapoptosiscaspase-independent
==== Body
Background
The branched-chain fatty acid (BCFA) or alcohols has been reported to suppress the proliferation or development of tumor cells [3]. Recently, 13-methyltetradecanoic acid (13-MTD), an iso-C15 saturated fatty acid, was purified as an anti-cancer agent from a soy-fermentation product [4]. 13-MTD inhibited cell proliferation and induced apoptotic cell death in many cancer cells such as prostate carcinoma (DU-145), leukemia (K563) and mammary adenocarcinoma (MCF-7) [4].
Induction of apoptosis was characterized by many morphological features including membrane blebbing, cytoplasmic and nuclear condensation, DNA fragmentation and formation of apoptotic bodies [5,6]. Our previous study showed that 13-MTD synthetically lowered the fatty acids biosynthesis and reduced the precursor of free fatty acid for phospholipid synthesis in human breast cancer cell (SKBR-3) [3]. However, the mechanism for the induction of apoptosis by 13-MTD has not been fully understood.
Epidemiological studies have indicated that fatty acid composition of the diet correlated with the risk of mammary cancer [7-9]. The level of stearic acid in tumor membrane phosphatidylcholine has been considered to be an independent tumor marker of breast cancer prognosis [10]. Furthermore, the involvement of lipogenic enzyme in oncogenesis has suggested that the membrane lipids could be a factor for the modulation of tumor growth [11,12]. It is therefore possible that the incorporation of 13-MTD into membrane or storage lipid may play a role for the induction of apoptosis in the cancer cells. Thus the aim of this study was to describe the incorporation of 13-MTD into cellular glycerolipid and its positional distribution in the lipid molecules. Furthermore, this study shed the light on the effect of 13-MTD on apoptosis related events to gain more insight into the mechanism actions of this fatty acid.
Results
The effect of 13-MTD on the relative viability of SKBR-3 cells are shown in figure 1. Consistent with the previous results, 13-MTD lowered the cell viability compared with the straight chain fatty acid counterpart, but to a lesser extent than cis-9, trans-11 conjugated linoleic acid (CLA), a well known anticancer fatty acid. Quantitation of apoptotic cells found that 13-MTD as well as CLA increased the number of apoptotic cells even after treatment of 4 hours (Fig. 2).
Figure 1 Effect of 13-MTD on the relative viability of SKBR-3 cells. Cells were treated with the fatty acid samples for 24 h, and the relative viabilities were determined by MTS assay kit according to the manufacture's instruction. Data are mean ± SE of 8 analyses. Data not sharing the same letter within each concentration are significantly different at p < 0.05 by Tukey-Kramer test.
Figure 2 Effect of 13-MTD on the induction of apoptotic cell death in SKBR-3 cells. Cells were pre-cultured for 24 h and treated with the fatty acid samples for another 4 h, and apoptotic cells were measured by Anexin V staining. Number of Anexin V stained cells in approximately 100 cells was counted and expressed as percentages of apoptotic cells. Data are mean ± SE of 4 analyses. Data not sharing the same letter shows statistically significant difference at p < 0.05 by Tukey-Kramer test.
Figure 3 illustrates the incorporation of 13-MTD, CLA and myristic acid into cellular glycerolipid, TG (A) and PL (B). Proportions of 13-MTD and CLA in both PL and TG increased with the incubation time. 13-MTD was incorporated preferentially into TG rather than PL. The proportion of 13-MTD in both PL and TG increased to the saturation level in 6 h. As was the case of 13-MTD, preference for TG was noted with the incorporation of CLA into cellular lipid. However, the proportion of CLA in PL was much higher than that of 13-MTD, showing that CLA accumulated in membrane lipid to a greater extent than 13-MTD.
Figure 3 Time course of 13-MTD incorporation into triacylglycerol (A) and phospholipids (B) of SKBR-3 cells. SKBR-3 cells grown to 80% confluency were incubated with the fatty acid samples for 3 to 24 h, and their proportions in the total fatty acids were measured by gas-chromatograph. Data are mean ± SE of triplicate analyses. Data not sharing the same letter shows statistically significant difference at p < 0.05 by Tukey-Kramer test.
Fatty acids taken up by the cells undergo esterification to form membrane PL or storage TG, depending probably on their chemical structure. The metabolic fate of fatty acid thus showed diversity with their chemical structure, and resulted in uneven positional distribution in TG or PL molecules [17]. TG contained almost equal proportions of 13-MTD at sn-2 position and at sn-1,3 positions (Fig 4A). However, this was not the case for CLA: sum of sn-1 and sn-3 position was higher than the proportion at sn-2 position.
Figure 4 Positional distribution of 13-MTD in triacylglycerol (A) and phospholipids (B) of SKBR-3 cells. Cells were grown to 80% confluency, and incubated with the fatty acid samples for 6 hours. Phospholipid and triacylglycerol fractions were hydrolyzed respectively with lipase and phospholipase A2, and positional distribution of sample fatty acid was determined as described in the method section. Data are mean ± SE of triplicate analyses. Data not sharing the same letter shows statistically significant difference at p < 0.05 by Tukey-Kramer test. Asterisk shows the statistical significance between position 1,3 and position 2 by Student's t-test.
Preferential esterification of 13-MTD to sn-2 position was noted for PC, while this trend was more clearly shown for the case of CLA (Fig. 4B). Incorporation of 13-MTD into PC was higher than that into PE. CLA distributed evenly between PC and PE.
Caspases are involved in signal transduction leading to apoptotic cell death [18,19]. Caspase 3 acts at the final step of caspase cascade to cleave proteins involved in cytoskeletal and nuclear structure, resulting in the cell shrinking and DNA fragmentation of late-stage apoptosis. 13-MTD has been demonstrated to be an anticancer agent to induce apoptosis in a variety of cancer cells. However, 13-MTD had no effect on the activity of caspase 3 in SKBR-3 cells. (Fig. 5A) In contrast to 13-MTD, CLA significantly elevated the caspase 3 activity after 2 and 4 h of incubation. Despite lack of caspase 3 activation, 13-MTD reduced the cell viability to the same extent as CLA did (Fig. 5B).
Figure 5 Effect of 13-MTD on caspase-3 activity (A) and relative viability (B) of SKBR-3 cells. Cells were incubated with the fatty acid samples (0.25 mM) for 2 or 4 h. Caspase activity in the cell lysate and the cell viability were determined as described in the method section. Data are mean ± SE of triplicate analyses. Data not sharing the same letter shows statistically significant difference at p < 0.05 by Tukey-Kramer test.
To confirm the non-involvement of caspase pathway, cytotoxicity and apoptosis induction were studied in the presence of caspase 3 inhibitor (Z-DEVD-FMK) and negative control for caspase-3 inhibitor (Z-FA-FMK). The negative control (Z-FA-FMK) only inhibits cysteine protease, and has no inhibitory effect on apoptosis mediated by caspase. Presence of caspase-3 inhibitor did not alleviate the cytotoxicity of 13-MTD (Fig. 6A), and also did not prevent the cells from induction of apoptosis (Fig. 6B). However, presence of the caspase inhibitor, but not the negative control, almost abolished the cytotoxicity of CLA (Fig. 6A).
Figure 6 Effect of caspase-3 inhibitor on cytotoxicity (A) and induction of apoptosis (B) by13-MTD. Cells pre-cultured for 24 h were incubated with PBS (no inhibitor), caspase-3 inhibitor (20 μM Z-DEVD-FMK) and negative control inhibitor (20 μM Z-FA-FMK) for 60 min and then further incubated with 0.25 mM of fatty acids samples for 4 h. On completion of the incubation, cell viability was determined by MTS assay (A), and induction of apoptosis was assayed as described in legend to Fig. 2 (B). Data are mean ± SE of triplicates analyses. Data not sharing the same letter within the treatment shows the statistically significant difference at p < 0.05 by Tukey-Kramer test.
Effect of 13-MTD on the expression of several regulatory genes involved in the apoptosis induction was studied (data not shown). Addition of fatty acid samples evenly decreased the expression of p53 gene with no difference between 13-MTD and myristic acid treatment. CLA increased mRNA level of caspase 3 as well as its activity. 13-MTD had no effect on the expression of caspase 8, apoptosis inducing factor (AIF), Bax, Bad and Bcl-XL (Fig. 7).
Figure 7 Effect of 13-MTD on the expression of several regulatory genes involved in the apoptosis induction in SKBR-3 cells. Cells pre-cultured for 24 hours were treated with the fatty acids for 4 h, and gene expression were measure by real time PCR. The relative change in gene expression was quantified essentially based on the 2 -ΔΔCT methods. Data are mean ± 95% confidence intervals of triplicates analyses. Data not sharing the same letter shows the statistically significant difference at p < 0.05 by Tukey-Kramer test.
It has been demonstrated that hydrogen peroxide mediates the induction of apoptosis in response to several external stimuli [20,21], and the supplementation of the medium with catalase prevent the cells from apoptosis induction [22,23]. Addition of catalase to the culture medium did not protect or rescue the cells from cytotoxicity of 13-MTD (Fig. 8). In agreement with the previous result, presence of catalase in the culture medium protected the cells from the cytotoxicity of CLA (Fig. 8).
Figure 8 Cytotoxicity of 13-MTD to SKBR-3 cells in the presence or absence of catalase. Cells were pre-cultured for 24 h. Catalase (100 units/ml) was added at indicated time of incubation with 0.25 mM of fatty acid samples, and the relative cell viability was measured after 24 h by MTS assay. Data are mean ± SE of 8 analyses. Asterisk shows the statistically significant difference from PBS or C14:0 at p < 0.05 by Tukey-Kramer test.
Several lines of studies have demonstrated that short chain or medium chain fatty acid increased the cellular Ca2+ mobilization via G-protein coupled orphan receptors [24,25]. Furthermore, excess loading of mitochondria with Ca2+ results in abnormal mitochondrial metabolism, which triggers the programmed cell death [26]. No mobilization of Ca2+ occurred when the cells were loaded with 13-MTD (Fig. 9). Oleic acid included as positive control stimulated the cellular Ca2+ mobilization as well as the case for CLA (Fig. 9).
Figure 9 Effect of 13-MTD on cellular calcium mobilization in SKBR-3 cells. SKBR-3 cells were seeded into black 96-well plate at 105 cells/well, and incubated for 1 day. Cells were loaded with 4 μM Fluo 3-AM for 30 min, and subsequently stimulated with 0.125, 0.25 and 0.5 mM fatty acids in 0.8% Tween 80 in PBS. The increases in fluorescence were measured 25 times with 0.15 s intervals. Data are mean ± SE of 38 replicate analyses. Data not sharing the same letter are significantly different at p < 0.05 by Tukey-Kramer test.
As to induction of apoptosis, mitochondria play a pivotal role in the signal transduction pathways [27]. To examine mitochondrial membrane integrity, MitoCapture reagent was used to stain the apoptotic cells after treatment of the cells with 13-MTD. In healthy cells, MitoCapture accumulates and aggregates in the mitochondria, and gives off a bright red fluorescence, while in apoptotic cells, MitoCapture can not enter mitochondria due to altered mitochondrial transmembrane potential, and therefore remain predominantly in the cytosol in its monomer form fluorescing green. After 4 hours of treatments with 13-MTD, the MitoCapture dye fluoresced predominantly green was indicating the disruption of mitochondrial integrity (Fig. 10). Myristic acid slightly increased the staining of green fluorescence compared with control vehicle buffer, but to much lower extent than 13-MTD did. CLA treatment for 4 hours liberated the cells from the bottom of the culture dishes into medium, and resulted in staining green fluorescence (Fig. 10).
Figure 10 Effect of 13-MTD on mitochondrial membrane integrity of SKBR-3 cells. Cells cultured for 24 h on coverlips were incubated with 0.25 mM fatty acid sample for 4 h, and loaded with MitoCapture (Biovision, USA) reagent for 20 min at 37°C under 5% CO2 atmosphere. Cells were observed with a fluorescence microscope using band-pass filter. MitoCapture aggregates and fluoresces red in the mitochondria of healthy cells. However, in apoptotic cells, MitoCapture cannot accumulate in the mitochondria, it remains as monomers in the cytoplasm, and fluoresces green. Counting of mitochondria disrupted cells in total number of 40 cells were made for 4 to 6 separate microscopic fields, and the percentages are shown (B). Data not sharing the same letter are significantly different at p < 0.05 by Tukey-Kramer test.
The most notorious apoptotic factors released from permeabilized mitochondria are cytochrome C and AIF. However, the results shown in the foregoing studies may rule out the involvement of cytochrome C in the death pathway of 13-MTD induced apoptosis. AIF was therefore the most plausible candidate of death signal transducer for 13-MTD induced apoptosis. Figure 11 shows the immunodetection of AIF punctuated apoptotic nucleus after treatment of the cells with 13-MTD. Treatment of the cells with vehicle buffer (control) showed few nuclear staining of AIF (Fig. 11). However, the percentage of AIF punctuated apoptotic nuclei was significantly increased by 13-MTD treatment (Fig. 11). This percentage was increased to some extent by the treatment of myristic acid. However the difference between control and myristic acid was statistically insignificant.
Figure 11 Effect of 13-MTD on nuclear translocation of AIF in SKBR-3 cells. Cells were grown for 24 h on coverslips and treated with 0.25 mM fatty acid samples for 6 h. After the treatment, cells were stained with Hoechst 33258 and AIF as described in the method section (A). Counting of AIF punctuated nuclei in total number of 40 cells were made for 4 to 6 separate microscopic fields, and the percentages of AIF translocated cells are shown (B). Data not sharing the same letter are significantly different at p < 0.05 by Tukey-Kramer test. Arrow shows the AIF translocated nucleus.
Discussion
This study described the incorporation of 13-MTD into cellular lipid, and the apoptosis related events after treatment of the cells with this fatty acid. 13-MTD induced apoptosis after 4 hours of treatment (Fig. 2), and reduced the cell viability (Fig. 1). The cytotoxicity of 13-MTD may be challenged by a claim that its toxic effect is due to non-specific detergent properties of fatty acids [28]. However, myristic acid, the straight chain fatty acid counterpart, had no significant cytotoxicity against SKBR-3 cells. Furthermore, the detergent property of 13-MTD may have limited significance in the current experiment because the sample fatty acids were dissolved in a detergent buffer containing 0.8% Tween 80. The vehicle buffer (PBS containing 0.8%Ttween 80) displayed negligible effect on the cell viability (data not shown). These observations negate the possibility that the cytotoxicity of 13-MTD is due to non-specific detergent properties of 13-MTD.
Structural membrane glycerolipids of animal cells contain almost exclusively 14 to 22 carbon fatty acids of straight chain with up to three methylene-interrupted double bonds. These fatty acids were referred to as usual fatty acids. Contrary to these fatty acids, 13-MTD is a fatty acid with branched carbon chain, and can be classified as unusual fatty acid.
Fatty acid entered the cancer cells undergo esterification to form TG or PL, and the production of TG and PL shares a common pathway referred to as the glycerol-3-phosphate pathway. In plant cells, unusual fatty acids are sequestered into storage lipid TG, and have no structural functions [29]. In current study, the concentration of 13-MTD in TG was higher than that found in PL. This observation may favor the above view that the fatty acids of which chemical structure deviate from the common fatty acids are edited out from membrane PL, and channeled into TG (Fig. 3). The targeting of 13-MTD to TG may be primarily explained by the substrate specificity of the enzyme involved in the biosynthesis of PL and TG: thioesterase, acyltransferase and cytidine diphosphocholine diacylglycerol choline phosphotransferase (CDP-CPT). Of the enzymes, CDP-CPT catalyzes the reversible conversion of DG to PC. Therefore, this enzyme could potentially prevent DG species containing 13-MTD from moving into phospholipid pool. Thus it become of interest to clarify the biological machinery to channel 13-MTD into storage lipid pool.
CLA known as anticancer fatty acid was also preferentially incorporated into TG as well as the case for 13-MTD. However, PL contained much higher concentration of CLA compared with 13-MTD, suggesting that CLA rather than MTD was more compatible with the membrane structure. Conversely, PL containing 13-MTD may not be acceptable to membrane structure, and for this reason the large proportion of 13-MTD may be edited out to accumulate as TG. In this context, it is noteworthy that the membrane fluidity is a factor to regulate the activities of membrane-bound enzymes [30,31]. Thus it is plausible that 13-MTD incorporated into PL affects the mitochondrial membrane fluidity because of its branching chain structure, and hence modulates the enzyme activities leading to abnormal mitochondrial metabolism.
Positional distribution of 13-MTD also showed its characteristics in the preference toward positon-2. In the case of TG, sum proportion of 13-MTD at position 1 and 3 was almost equal to that at position 2, showing that the individual proportion at position 1 or 3 was lower than that at position 2 (Fig. 4). Same trend in the positional distribution was also seen with the case of structural lipid PC. This characteristic also may be explained by the enzyme specificity involved in the biosynthesis of these lipids.
This study reinforced the cytotoxicity of 13-MTD. 13-MTD, as well as the known anticarcinogenic fatty acid CLA, induced apoptotic cell death of breast cancer cells. Several lines of studies demonstrated that CLA induced apoptosis via caspase-dependent mechanism [29,32-34]. However, the mechanism action of 13-MTD differed from that of CLA. Incubation of the cells with 13-MTD induced no changes in the caspase activity and its gene expression (Figs. 5 and 7). Furthermore, inclusion of caspase-3 inhibitor in the culture medium did not prevent the cells from the cytotoxicity of 13-MTD (Fig. 6). These findings suggested that 13-MTD induced apoptosis via caspase-independent signal transduction.
The mitochondria are playing central role in both caspase-dependent and caspase-independent death pathway [26,35]. Mitochondria respond to multiple death stimuli which induce mitochondrial membrane permeabilization, and cause the release of apoptotic molecules such as cytochrome C or apoptosis inducing factor (AIF). Incubation of the breast cancer cells for 4 h with 13-MTD caused disruption of mitochondrial integrity (Fig. 10). This observation suggested that 13-MTD permeabilized the mitochondrial membrane, and proapoptotic molecules were released from mitochondria to trigger the chromatin condensation. These proapoptotic molecules can also be released from mitochondria by cellular perturbations that cause a sudden increase in intracellular calcium level [25]. However, 13-MTD induced no mobilization of calcium hence no increase in the intracellular calcium concentration (Fig. 9). Thus 13-MTD appeared to induce apoptosis without changing cellular calcium level.
The most notorious apoptotic factors released from permeabilized mitochondria are cytochrome C and AIF. However, the results shown in this study (Figs. 5-6) may rule out the involvement of cytochrome C in the death pathway of 13-MTD induced apoptosis. AIF was therefore the most plausible candidate of death signal transducer for 13-MTD induced apoptosis. Immunocytochemical analysis supported this view. The percentage of AIF punctuated apoptotic nuclei was increased by 13-MTD treatment (Fig. 11), suggesting that nuclear AIF translocation could be the main death signaling system. More recently, it has been demonstrated that permeabilized mitochondria can also release other proapoptotic factors such as Smac/Diablo [36,37] or HtrA2/Omi [38-40]. Involvement of these factors in the action mechanisms of 13-MTD should be explored in the coming study.
AIF is usually present in the intermembrane space of mitochondria, and translocated from mitochondria to nucleus to initiate apoptosis [41]. In most cases of apoptotic cell death, the release of AIF appeared to coincide with the loss of mitochondrial membrane integrity [42-44]. Permeabilization of mitochondrial membrane triggers the death pathway by releasing the proapoptotic molecules. The liberation mode and rate of these molecules appeared to depend on the death stimulus [44,45]. In the case of 13-MTD treatment, translocation of AIF may precede the release of cytochrome C, and play a crucial role in the progress of cell death. AIF translocation after 13-MTD treatment may be accompanied by the release of cytochrome C. However, this event usually occurs during the end stage of cell death, and has been considered to be limited significance for the completion of apoptosis [46].
13-MTD differed from CLA in that this unusual fatty acid preferentially induced translocation of AIF rather than release of cytochrome C. This feature may bear relevance with its chemical structure which appeared to be less tolerable to the membrane structure than CLA. Phospholipids isolated from 13-MTD treated cells contained 13-MTD to some extent, suggesting that similar changes in the fatty acid composition took place in the phospholipids of mitochondrial membrane. Incorporation of phospholipids containing 13-MTD into membrane structure may impose a change in the membrane environment as is the case of death stimulus. The physicochemical properties of CLA are similar to those of unsaturated fatty acid, and may be more acceptable as membrane component. Thus the change in the membrane lipid profile elicited by incorporation of 13-MTD alternatively may affect the permeability of mitochondrial membrane, and result in the preceded release of AIF. The physicochemical properties of 13-MTD thereby may be a critical determinant to induce AIF translocation.
Majority of proapoptotic stimuli, including anticancer drug require a caspase dependent death pathway. It has been shown that AIF determines the chemoresistance of non-small-cell lung cancer cells [47]. In this case, activation of caspase cascade was insufficient to kill cancer cells, and an AIF-mediated death pathway played an important role for the induction of apoptosis [47]. Although it may be difficult to achieve mM concentration of 13-MTD in humans, present findings warrant further investigation to develop additional chemotherapeutic agent that is effective to kill the cancer cells resistant to the induction of apoptosis by conventional anticancer drug.
Conclusion
In conclusion, we report that the 13-MTD was incorporated into glycerolipids of SKBR-3 cells with the preference for triglycerides rather than phospholipids, suggesting that incorporation of phospholipids containing 13-MTD into membrane structure influenced the membrane environment as are the case of death stimulus. Treatment with 13-MTD induced no change in the activity and expression of caspase 3. Furthermore, 13-MTD induced no mobilization of cellular calcium. However, 13-MTD altered the mitochondrial transmembrane potential and induced AIF translocation from mitochondrial to nuclear after 4 hrs of treatment. These results supported the view that incorporation of 13-MTD into cellular lipids triggered apoptosis via caspase-independent pathway.
Materials and methods
Chemicals
Enzyme phospholipase A2 was purchased from Sigma Chemicals Co.(Tokyo, Japan) and pancreatic lipase from Funakoshi(Tokyo, Japan). The primers were obtained from Hokkaido System Science Co., Ltd (Hokkaido, Japan). The RNA isolation kit was purchased from Wako (Osaka, Japan) and cDNA synthesis kit from Invitrogen(Tokyo, Japan). Other chemicals were all guaranteed grade, and obtained from domestic suppliers.
Cell and tissue culture
Human breast cancer cell line SKBR-3 was purchased from Dainippon pharmaceuticals Co. (Osaka, Japan). Cells were maintained and sub cultured according to the supplier's recommendation. Culture medium for SKBR-3 was McCOY's 5A containing 10% FBS. Cells were cultured in a humidified atmosphere of 5% CO2 at 37°C
Cell cytotoxicity assay
Test substances of free fatty acids were neutralized and dissolved in PBS containing 0.8% tween 80. Cell cytotoxicity titration curve was constructed with serial dilution of the test substances in a 96-well microplate. Cells seeded at density of 700 cells/well were incubated with serially diluted test substances, and the viable cell numbers were determined by MTS assay according to the manufacture's instruction (CellTiter® AQueous Non-Radioactive Cell Proliferation Assay, Promega Co., Madison, USA). Cell cytotoxicity was thus expressed as the relative viability against control cells treated only with the vehicle solutions.
Quantitation of apoptotic cells
SKBR-3 cells (1 × 104 cells/dish) were pre-incubated in the medium for 2 hours, and cultured further with fatty acid samples for 4 hours. At the end of the treatment, cells were washed with PBS and resuspended in 100 μl of binding buffer containing Anexin V (MBL, Nagoya, Japan) and incubated for 30 min. Cells were washed with binding buffer, and apoptotic cells stained with Anexin V were counted with a fluorescence microscope. Data were expressed as percentages of apoptotic cells in the total cells.
Lipid extraction and fractionation
SKBR-3 cells were seeded in 90 × 17 mm petridish at density of 1 × 104 /dish and grown to 80% confluence. Cells were subsequently incubated with 0.25 mM fatty acids dissolved in PBS containing 0.8% Tween 80 for 3, 6 and 24 h, respectively. After incubation, surface of the cultures were washed twice with PBS. Cells were scraped, centrifuged for 10 min at 700 × g and extracted with 1 ml of chloroform/methanol (2:1, by vol.) at 40°C for 60 min. The extract was split into chloroform and methanol water layer by addition of 0.2 ml water. Lipids partitioned into chloroform layer were concentrated and spotted onto HPTLC plates (Merck, Darmstadt, Germany) for a separation of triacylglycerol, (TG) and phospholipid (PL) fractions as described previously [3]. In the case of separation of phospholipids, the HPTLC plates were developed stepwise with chloroform/acetone/methanol (9:0.5:0.5, by vol.), chloroform/ethyl acetate/methanol/2-propanol/triethylamine/water (6.4: 1.5: 1.7: 0.05: 0.3: 0.2, by vol.). Phosphatidylethanolamine (PE) and phosphatidylcholine (PC) were identified by comparison of Rf (retardation factor) value with authentic standards. TG and PL factions were extracted and methanolysed with methanolic hydrochloric acid as described elsewhere [14]. The fatty acid methyl esters produced were analyzed by a gas chromatograph (Model GC-2010; Shimadzu, Kyoto, Japan) with a capillary column (DB-FFAP column 0.25 mm ID × 30 m; J&W Scientific, CA, USA). Injector and detector temperatures were 210°C. The oven temperature was programmed as follows: the initial temperature was 50°C for 1 min, raised to 100°C at 10°C/min, thereafter to final temperature of 240°C at 5°C/min, and kept for 26 min. The chromatograms were analyzed by GC Solution Software (Shimadzu).
Positional distribution of fatty acid
The method of enzymatic hydrolysis was used to analyze the positional distribution of fatty acid in PL and TG. PE and PC fractions were hydrolyzed with phospholipase A2 from bee venom (Sigma Chemicals Co.) About 1 mg of PE or PC in 1 ml diethyl ether was treated with 0.5 ml (850 U) phospholipase A2 dissolved in 1 M Tris buffer (pH 7.5) containing 4 mM calcium chloride. The enzyme reaction was monitored by examining the reaction products by HPTLC. Fractions of PE or PC were completely hydrolyzed to give rise to lyso PE or lyso PC, respectively. The released free fatty acid and lyso PE or lyso PC were analyzed for fatty acid compositions of sn-2 and sn-1, respectively.
In the case of TG, pancreatic lipase (Funakoshi, Japan) was used for the enzymatic hydrolysis. About 50 μg of TG was mixed stepwise with 0.5 ml of 1 M Tris buffer (pH 8.0), 50 μl of 2.2%calcium chloride and 0.125 ml of 0.05% cholic acid sodium salt. The mixture was incubated at 37°C for 1 min. Pancreatic lipase (40 μl, 1 U/μl) was added to this solution and shaken vigorously at 37°C for 1 h. The reaction was stopped by the addition of 0.5 ml of ethanol and 6 M HCl, and extracted with 4 ml of diethyl ether. The extract was washed with 1.5 ml of distilled water and dried over sodium sulphate. The reaction products, free fatty acid (FFA) and monoacylglycerol (MG), were separated by HPTLC with hexane/diethyl ether/acetic acid (5:5:0.1, by vol.), and analyzed for fatty acid composition of n-1,3 and n-2, respectively.
Measurement of caspase activity
Cells were incubated with fatty acid sample for 2 or 4 hours and collected to measure caspase activity. Cells were lysed in buffer containing 10 mM Tris-HCl (pH 7.5), 10 mM Na2H2PO4/Na2HPO4, 130 mM NaCl, 1% Triton X-100 and 10 mM sodium pyrophosphate. The cell lysates (45 μl) were mixed with 5 μl of fluorogenic caspases-3 substrate (500 μM) DEVD-R110 (Roche Diagnostics, Mannheim, Germany) in 96-well micro titer plate according to the manufacturer's instruction. The plate was incubated at 37°C for 1 h, and then fluorescence was monitored with excitation and emission wavelengths at 485 and 535 nm, respectively.
Real time PCR
Total RNA was extracted from 80% confluent SKBR-3 cells by RNA isolation kit, (Isogen, Wako, Osaka, Japan). A first strand cDNA synthesis was conducted by cDNA synthesis kit (Invitrogen, Japan). The sequences of the forward and reverse primers used were; caspases 3 (5'-GCCTGAGCAGAGACATGACTCA-3 and 5'-TCATCCACACATACCAGTGCGA-3), caspases 8 (5'-GATATATCCCGGATGAGGCTGAC-3 and 5'-TGACTGGATGTACCAGGTTCCC-3), P53 (5'-GGGATGTTTGGGAGATGTAAGAAATG-3 and 5'-GTGGGCCCCTACCTACCTAGAATG-3), AIF (5'-AGTAGTTTGCCCACAGTTGGTGTT-3 and 5'-TCACTCTCTGATCGGATACCAGTTC-3), Bad (5'-CAGTGACCTTCGCTCCACATC-3 and 5'-AAGGAGACAGCACGGATCCTC-3), Bax (5'-TCTGACGGCAACTTCAACTGG-3 and 5'-AGCCCATGATGGTTCTGATCA-3), Bcl-XL (5'-CCTCAGCTGCCTCACTTCCTA 3 and 5'-CCATAGCTGTTCCTGATAGCTCC-3), and GAPD (5'-TCTGGTAAAGTGGATATTGTTGCC-3 and 5'-CCTTCTTGATGTCATATTTGGC-3).
Each 20 μl PCR reaction contained 0.5 U DNA polymerase (TaKaRa, Kyoto, Japan), 20 mM Tris-HCl (pH 8.4), 100 mM KCl, 0.1 mM EDTA, 1 mM DTT, 0.5% Tween 20, 0.5% Nonident P-40, 50% glycerol), 2.5 mM dNTP, 3 μg cDNA, 10 pmol primers and 0.2 μl of 1 × working solution of SYBR Green II (Takara, Shiga, Japan). The real time quantitative PCR was conducted in the thermocycler (iCycler, BioRad, Hercules, CA). The temperature program for real time analysis consisted of 30 cycles of 90 °C for 30 s, 58 °C for 30 s and final extension step at 72 °C for 30 s. Threshold cycle (CT) numbers were calculated during PCR amplification by using the iCycler data analysis software. The house-keeping GAPD genes were internal control to normalize the PCR for the amount of template cDNA. Relative change in gene expression was quantified essentially based on the 2-ΔΔCT methods as described previously [15]. The 2-ΔΔCT denotes the relative change of target gene expressed in sample-treated cells compared with that in reference cells [15]. Because the normalized concentration of cDNA template was used in this experiment, 2-ΔCT represents the relative change of gene expression between treated and untreated cells. The equation that describes the relative change in threshold cycle is therefore
ΔCT = CT (sample)-CT (reference)
where CT is the threshold cycle reflecting the cycle number at which the fluorescence generated within a reaction process crosses the threshold level. In this study, sample cDNAs were serially diluted, and the mean CT and its 95% confidence interval were estimated by linear regression.
Measurement of cellular calcium mobilization
SKBR-3 cells were seeded into black 96-well plate at 105 cells/well, and incubated for 1 day. The cells were washed twice with Hank's balanced salt solution (HBSS), and loaded with 4 μM Fluo 3-AM for 30 min in HBSS. The plates were washed with HBSS, transferred to microplate reader (Wallac 1420 Multilabel Counter, Parkin Elmer Co., Turka, Finland), and the basal fluorescence levels were recorded. Cells were loaded with 0.25 mM fatty acids samples, mixed for 30 s, and the increases in fluorescence were measured 25 times with 0.15 s intervals. Excitation and emission wavelength were 485 and 535 nm, respectively.
Immunocytochemistry
For immunocytochemical analysis, cells grown on coverslips were fixed with 4% paraformaldehyde in PBS for 20 min, and washed twice with PBS. The cells were further permeabilized with 0.1% Triton X-100 in PBS for 20 min. After two washes with PBS, cells were incubated with diluted (1/200) rabbit anti-AIF (Chemicon International Co., CA, USA) in PBS containing 0.1% Tween 20(PBST) for 2 hours at room temperature. The cells were washed three times with PBST, and incubated with FITC-labeled goat anti-rabbit IgG (Biosource International Co., CA, USA) in PBST for 2 hours. After three washes with PBST, the cells were loaded with 1 μg/ml of Hoechst 33258 to stain DNA. Coverslips were then placed on a slide glass with cell-side down, and were covered with nail lacquer to prevent them from drying. The fluorescent images were visualized by conventional (Model BX41, Olympus Co., Tokyo, Japan) or confocal fluorescence microscopy (Radiance 2100, Bio-Rad Laboratories, Inc. Tokyo, Japan).
Statistical analyses
Data are represented as mean ± SE except otherwise stated. The statistical significance was evaluated by Tukey-Kramer tests [16] or Student's t-test. The criterion for statistical significance was p <0.05.
List of abbreviations
13-MTD, 13-methyltetradecanoic acid; BCFA, branched-chain fatty acid; C14:0, myristic acid; C18:0, stearic acid; HPTLC, high-performance thin layer chromatography; PL, phospholipid; TG, triglycerides; MG, monoacylglycerol; DG, diglyceride; PE, phosphatidylethanolamine; PC, phosphatidylcholine; FFA, free fatty acid; PBS, phosphate-buffered saline; FBS, fetal bovine serum, FA, fatty acid; PCR, polymerase chain reaction; IgG, immunoglobulin G; HBSS, Hanks' balanced salt solution; cDNA, complementary DNA; Rf, retention factor; DMSO, dimethyl sulfoxide; DTT, dithiothreitol; RNA, ribonucleic acid; SE, standard error; SEM, standard error of the mean; TAG, triacylglycerol.
Authors' contributions
WS and OH participated in study design, and drafted the manuscript. IH carried out the real time PCR. IM participated in immunochemical studies. TT and TY helped to draft the manuscript. All authors read and approved the final manuscript.
==== Refs
Cinatl J JrCinatl J Driever PH Kotchetkov R Pouckova P Kornhuber B Schwabe D Sodium valproate inhibits in vivo growth of human neuroblastoma cells Anticancer Drugs 1997 8 958 963 9436639
Yang P Collin P Madden T Chan D Sweeney-Gotsch B McConkey D Newman RA Inhibition of proliferation of PC3 cells by the branched-chain fatty acid, 12-methyltetradecanoic acid, is associated with inhibition of 5-lipoxygenase Prostate 2003 55 281 291 12712407 10.1002/pros.10243
Wongtangtintharn S Oku H Iwasaki H Toda T Effect of branched-chain fatty acids on fatty acid biosynthesis of human breast cancer cells J Nutr Sci Vitaminol(Tokyo) 2004 50 137 143 15242018
Yang Z Liu S Chen X Chen H Huang M Zheng J Induction of apoptotic cell death and in vivo growth inhibition of human cancer cells by a saturated branched-chain fatty acid, 13-methyltetradecanoic acid Cancer Res 2000 60 505 509 10676625
Ziegler U Groscurth P Morphological features of cell death News Physiol Sci 2004 19 124 128 15143207 10.1152/nips.01519.2004
Rogalinska M Alterations in cell nuclei during apoptosis Cell Mol Biol Lett 2002 7 995 1018 12511968
Chan PC Ferguson KA Dao TL Effects of different dietary fats on mammary carcinogenesis Cancer Res 1983 43 1079 1083 6825080
Carroll KK Hopkins GJ Dietary polyunsaturated fat versus saturated fat in relation to mammary carcinogenesis Lipids 1979 14 155 158 106196
Tinsley IJ Schmitz JA Pierce DA Influence of dietary fatty acids on the incidence of mammary tumors in the C3H mouse Cancer Res 1981 41 1460 1465 7214328
Bougnoux P Chajes V Lanson M Hacene K Body G Couet C Le Floch O Prognostic significance of tumor phosphatidylcholine stearic acid level in breast carcinoma Breast Cancer Res Treat 1992 20 185 194 1571571 10.1007/BF01834624
Chajes V Lanson M Fetissof F Lhuillery C Bougnoux P Membrane fatty acids of breast carcinoma: contribution of host fatty acids and tumor properties Int J Cancer 1995 63 169 175 7591199
Fonck K Scherphof GL Konings AW Control of fatty acid incorporation in membrane phospholipids. X-ray-induced changes in fatty acid uptake by tumor cells Biochim Biophys Acta 1982 692 406 414 6816280
Poon R Richards JM Clark WR The relationship between plasma membrane lipid composition and physical-chemical properties. II. Effect of phospholipid fatty acid modulation on plasma membrane physical properties and enzymatic activities Biochim Biophys Acta 1981 649 58 66 6272860
Christie WW Christie WW Structural analysis of lipids by means of enzymatic hydrolysis. In: Lipid Analysis:isolation, separation, identification, and structural analysis of lipids. 1982 2nd Pergamon press Germany
Livak KJ Schmittgen TD Analysis of relative gene expression data using real-time quantitative PCR and the 2(-ΔΔCT) Method Methods 2001 25 402 408 11846609 10.1006/meth.2001.1262
Ludbrook J On making multiple comparisons in clinical and experimental pharmacology and physiology Clin Exp Pharmacol Physiol 1991 18 379 392 1914240
Ramirez M Amate L Gil A Absorption and distribution of dietary fatty acids from different sources Early Hum Dev 2001 65 S95 S101 11755040 10.1016/S0378-3782(01)00211-0
Philchenkov A Caspases: potential targets for regulating cell death J Cell Mol Med 2004 8 432 444 15601572
Fulda S Debatin KM Apoptosis signaling in tumor therapy Ann N Y Acad Sci 2004 1028 150 126 15650241 10.1196/annals.1322.016
Kim DK Cho ES Um HD Caspase-dependent and -independent events in apoptosis induced by hydrogen peroxide Exp Cell Res 2000 257 82 88 10854056 10.1006/excr.2000.4868
Hancock JT Desikan R Neill SJ Role of reactive oxygen species in cell signalling pathways Biochem Soc Trans 2001 29 345 350 11356180 10.1042/BST0290345
Arimura T Kojima-Yuasa A Suzuki M Kennedy DO Matsui-Yuasa I Caspase-independent apoptosis induced by evening primrose extract in Ehrlich ascites tumor cells Cancer Lett 2003 201 9 16 14580681 10.1016/S0304-3835(03)00440-3
Yonezawa T Katoh K Obara Y Existence of GPR40 functioning in a human breast cancer cell line, MCF-7 Biochem Biophys Res Commun 2004 314 805 809 14741707 10.1016/j.bbrc.2003.12.175
Briscoe CP Tadayyon M Andrews JL Benson WG Chambers JK Eilert MM Ellis C Elshourbagy NA Goetz AS Minnick DT Murdock PR Sauls HR JrShabon U Spinage LD Strum JC Szekeres PG Tan KB Way JM Ignar DM Wilson S Muir AI The orphan G protein-coupled receptor GPR40 is activated by medium and long chain fatty acids J Biol Chem 2003 278 11303 11311 12496284 10.1074/jbc.M211495200
Berridge MJ Bootman MD Lipp P Calcium- a life and dDeath signal Nature 1998 395 645 648 9790183 10.1038/27094
Bras M Queenan B Susin SA Programmed cell death via mitochondria: different modes of dying Biochemistry 2005 70 231 239 15807663
Oku H Wongtangtintharn S Iwasaki H Toda T Conjugated linoleic acid (CLA) inhibits fatty acid synthetase activity in vitro Biosci Biotechnol Biochem 2003 67 1584 6158 12913306 10.1271/bbb.67.1584
Millar AA Smith MA Kunst L All fatty acids are not equal: discrimination in plant membrane lipids Trends Plant Sci 2000 5 95 101 10707074 10.1016/S1360-1385(00)01566-1
Majumder B Wahle KW Moir S Schofield A Choe SN Farquharson A Grant I Heys SD Conjugated linoleic acids (CLAs) regulate the expression of key apoptotic genes in human breast cancer cells FASEB J 2002 16 1447 1449 12205043
Davis PJ Poznansky MJ Modulation of 3-hydroxy-3-methylglutaryl-CoA reductase by changes in microsomal cholesterol content or phospholipid composition Proc Natl Acad Sci USA 1987 84 118 121 3467343
Ardail D Lerme F Louisot P Further evidence for both functional and structural microcompartmentation within the membranes of two associated organelles, mitochondrion and endoplasmic reticulum Biochem Biophys Res Commun 1990 31 878 885 2176492 10.1016/S0006-291X(05)80868-X
Miller A Stanton C Devery R Cis 9, trans 11- and trans 10, cis 12-conjugated linoleic acid isomers induce apoptosis in cultured SW480 cells Anticancer Res 2002 22 3879 3887 12553008
Bergamo P Luongo D Rossi M Conjugated linoleic acid-mediated apoptosis in Jurkat T cells involves the production of reactive oxygen species Cell Physiol Biochem 2004 14 57 64 14976407 10.1159/000076927
Miglietta A Bozzo F Bocca C Gabriel L Trombetta A Belotti S Canuto RA Conjugated linoleic acid induces apoptosis in MDA-MB-231 breast cancer cells through ERK/MAPK signalling and mitochondrial pathway Cancer Lett 2005 7 1 9
Kim R Recent advances in understanding the cell death pathways activated by anticancer therapy Cancer 2005 103 1551 1560 15742333 10.1002/cncr.20947
Du C Fang M Li Y Li L Wang X Smac, a mitochondrial protein that promotes cytochrome c-dependent caspase activation by eliminating IAP inhibition Cell 2000 102 33 42 10929711 10.1016/S0092-8674(00)00008-8
Taylor AB Smith BS Kitada S Kojima K Miyaura H Otwinowski Z Ito A Deisenhofer J Crystal structures of mitochondrial processing peptidase reveal the mode for specific cleavage of import signal sequences Structure 2001 9 615 625 11470436 10.1016/S0969-2126(01)00621-9
Suzuki Y Imai Y Nakayama H Takahashi K Takio K Takahashi R A serine protease, HtrA2, is released from the mitochondria and interacts with XIAP, inducing cell death Mol Cell 2001 8 613 621 11583623 10.1016/S1097-2765(01)00341-0
Hegde R Srinivasula SM Zhang Z Wassell R Mukattash R Cilenti L DuBois G Lazebnik Y Zervos AS Fernandes-Alnemri T Alnemri ES Identification of Omi/HtrA2 as a mitochondrial apoptotic serine protease that disrupts inhibitor of apoptosis protein-caspase interaction J Biol Chem 2002 277 432 438 11606597 10.1074/jbc.M109721200
Verhagen AM Ekert PG Pakusch M Silke J Connolly LM Reid GE Moritz RL Simpson RJ Vaux DL Identification of DIABLO, a mammalian protein that promotes apoptosis by binding to and antagonizing IAP proteins Cell 2000 102 43 53 10929712 10.1016/S0092-8674(00)00009-X
Otera H Ohsakaya S Nagaura Z Ishihara N Mihara K Export of mitochondrial AIF in response to proapoptotic stimuli depends on processing at the intermembrane space EMBO J 2005 24 1375 86 15775970 10.1038/sj.emboj.7600614
Susin SA Lorenzo HK Zamzami N Marzo I Snow BE Brothers GM Mangion J Jacotot E Costantini P Loeffler M Larochette N Goodlett DR Aebersold R Siderovski DP Penninger JM Kroemer G Molecular characterization of mitochondrial apoptosis-inducing factor Nature 1999 397 441 446 9989411 10.1038/17135
Ferri KF Jacotot E Geuskens M Kroemer G Apoptosis and karyogamy in syncytia induced by the HIV-1-envelope glycoprotein complex Cell Death Differ 2000 7 1137 1139 11288697 10.1038/sj.cdd.4400748
Cregan SP Fortin A MacLaurin JG Callaghan SM Cecconi F Yu SW Dawson TM Dawson VL Park DS Kroemer G Slack RS Apoptosis-inducing factor is involved in the regulation of caspase-independent neuronal cell death J Cell Biol 2002 158 507 517 12147675 10.1083/jcb.200202130
Arnoult D Parone P Martinou JC Antonsson B Estaquier J Ameisen JC Mitochondrial release of apoptosis-inducing factor occurs downstream of cytochrome c release in response to several proapoptotic stimuli J Cell Biol 2002 159 923 929 12486111 10.1083/jcb.200207071
Yu SW Wang H Poitras MF Coombs C Bowers WJ Federoff HJ Poirier GG Dawson TM Dawson VL Mediation of poly (ADP-ribose) polymerase-1-dependent cell death by apoptosis-inducing factor Science 2002 297 259 263 12114629 10.1126/science.1072221
Gallego MA Joseph B Hemstrom TH Tamiji S Mortier L Kroemer G Formstecher P Zhivotovsky B Marchetti P Apoptosis-inducing factor determines the chemoresistance of non-small-cell lung carcinomas Oncogene 2004 23 6282 6291 15286713 10.1038/sj.onc.1207835
|
16305741
|
PMC1315321
|
CC BY
|
2021-01-04 16:39:18
|
no
|
Lipids Health Dis. 2005 Nov 23; 4:29
|
utf-8
|
Lipids Health Dis
| 2,005 |
10.1186/1476-511X-4-29
|
oa_comm
|
==== Front
BMC Cell BiolBMC Cell Biology1471-2121BioMed Central London 1471-2121-6-421632116510.1186/1471-2121-6-42Research ArticleImmunochemical, biomolecular and biochemical characterization of bovine epithelial intestinal primocultures Rusu Dorina [email protected] Suzanne [email protected] Olivier [email protected] Jacques [email protected] Guy [email protected] Department of Biochemistry and General Physiology, University of Liege, Institute of Chemistry B6C, B-4000 Liege (Sart-Tilman), Belgium2 Department of Biology, University of Namur (FUNDP), Rue de Bruxelles, 61, B-5000 Namur, Belgium3 Department of Infectious and Parasitic Diseases/Bacteriology, Faculty of Veterinary Medicine, University of Liege, Boulevard de Colonster B43, B-4000 Liege (Sart-Tilman), Belgium2005 1 12 2005 6 42 42 24 8 2005 1 12 2005 Copyright © 2005 Rusu et al; licensee BioMed Central Ltd.2005Rusu 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
Cultures of enterocytes and colonocytes represent valuable tools to study growth and differentiation of epithelial cells. In vitro models may be used to evaluate passage or toxicity of drugs, interactions of enteropathogenes bacteria strains with intestinal epithelium and other physiologic or pathologic phenomenon involving the digestive tract.
Results
Cultures of bovine colonocytes and jejunocytes were obtained from organoid-enriched preparations, using a combination of enzymatic and mechanical disruption of the intestine epithelium, followed by an isopicnic centrifugation discarding most single cells.
Confluent cell monolayers arising from plated organoids exhibited epithelium typical features, such as the pavement-like structure, the presence of apical microvilli and tight junctions. Accordingly, cells expressed several markers of enterocyte brush border (i.e. maltase, alkaline phosphatase and fatty acid binding protein) as well as an epithelial cytoskeleton component (cytokeratin 18). However, enterocyte primocultures were also positive for the vimentin immunostaining (mesenchyme marker). Vimentin expression studies showed that this gene is constitutively expressed in bovine enterocytes. Comparison of the vimentin expression profile with the pattern of brush border enzymes activities, suggested that the decrease of cell differentiation level observed during the enterocyte isolation procedure and early passages of the primoculture could result from a post-transcriptional de-repression of vimentin synthesis. The low differentiation level of bovine enterocytes in vitro could partly be counteracted adding butyrate (1–2 mM) or using a glucose-deprived culture medium.
Conclusion
The present study describes several complementary approaches to characterize bovine primary cultures of intestinal cells. Cultured cells kept their morphologic and functional characteristics during several generations.
==== Body
Background
Intestinal epithelium is organized as a single layer which covers the luminal side of this part of the digestive tract. Cells that form this inner cover present specialisations according to their role in the digestive function but they have also common functions and are continuously renewed by the stem cell proliferation. In vivo, these stem cells, known as progenitors of all cell types, are mostly located in the lower third of the epithelium crypts, which are epithelial invaginations into the lamina propria. Cells differentiation is associated with their migration from the depth of the crypts to the top of the villi and is followed by the cells death and desquamation into the intestinal lumen (for review, see: [1,2]).
Enterocyte cultures represent valuable tools to assess the passage and/or toxicity of drugs, as well as the molecular mechanisms operating in pathologies caused by infectious agents known to affect the intestinal epithelium integrity (i.e. microvilli effacing microbial strains). However, the successful establishment of an intestinal cell culture is hampered by the high rate of cell death occurring when isolating them from the epithelium (for review: see [3]) and the difficulty to select proliferating cells to ensure several cell generations in vitro.
Although several investigators have already developed culture methods of intestinal cells for animal species such as the mouse [4-8], the rabbit [9-11] or the pig [12,13], until now only two studies reported the production of primary cultures from intestinal bovine cells. More precisely, Dibb-Fuller and coworkers [14] established a procedure for obtaining such cultures from ileum and colon. Föllmann and collaborators [15] also used the bovine colon as a source of intestinal cell cultures.
Various strategies were reported in order to isolate enterocytes from the colon or the small bowel. Among the oldest methods, the mechanical dissociation [16] provided rapidly a viable cell preparation but was frequently associated with a fibroblast contamination. The chelating methods also generated isolated epithelial cells retaining their morphologic characteristics. However, while first attempts seemed to affect cell surface receptors [17-19], further applications of chelating agents for a short time were suitable for the production of uncontaminated cultures of human colonocytes [20]. Matrisperse, a non enzymatic solution initially designed to isolate epithelial cells grown on Engelbreth-Holm-Swarm (EHS) biomatrix, also allowed the dissociation of the integral villus epithelial lining from human intestinal biopsies that produced a confluent monolayer in vitro [21]. The enzymatic digestion provided a cell preparation made up mainly of organoids (crypt-like cell aggregates) that were shown to successfully reconstitute either a monolayer of epithelial cells in vitro or crypt-villus structures in vivo following grafting in various tissues [15,20,22-27]. In spite of these numerous efforts, little is known about the cell differentiation evolution over culture generations.
The present study describes several complementary approaches to characterize bovine primary cultures of intestinal epithelial cells called jejunocytes and colonocytes as they were respectively isolated from adult jejunum and spiralled colon. Cultures were initiated using an organoid-enriched suspension obtained by a multi-step method. Results of vimentin expression in both types of enterocyte primary cultures are discussed in terms of differentiation status of cells.
Results
Morphological features of jejunocyte and colonocyte cultures
The method developed to isolate cell material from the bovine jejunum and spiraled colon led to the production of suspension enriched in undissociated cell aggregates (or organoids; Figure 1A) adhering to collagen coated culture flasks. They give rise to circular proliferating foci (Figure 1B). Within 5 days the initial foci enlargement had led to the fusion of cell plaques in one confluent layer (Figure 1C). Confluent monolayers arising from the initial seeding of cultures presented a heterogeneous aspect (Figure 1C) due to the presence of residual multicellular organoids dispersed in the newly formed monolayer. With the first passage jejunocyte and colonocyte cultures acquired a homogeneous pavement-like aspect typical of epithelial sheets. We generally carried out 7 to 9 passages (with split ratio 1:2 or 1:3) over 4 to 5 weeks in both cell series (jejunocytes and colonocytes) without major changes of cell morphology. During early passages (1–3), the confluence recovery was achieved within 2 to 3 days and then a progressive slowdown in the cell proliferation rate was observed until cessation. Confluent monolayers of both cell types exhibited domes, appearing as round blurred arrays corresponding to fluid entrapment between the flask wall and the basolateral side of the monolayer, resulting from ionic regulations by functional epithelial cells (Figure 1D). This observation is in agreement with dome formation in cultures of the human intestinal cell lineage Caco-2 [28].
Figure 1 Morphology of cell culture. Aspect, under phase contrast microscopy, of primoculture evolution of bovine intestinal enterocytes, starting from (A) adherent organoid giving larger (B) proliferating foci that joined together into (C) confluent monolayers presenting typical (D) close-up of a confluent monolayer showing a dome. Bar: 25 μm.
Ultrastructural analyses of cultured cells revealed apical tight junctions (Figure 2A) between cells forming the monolayer of both cell types. Cultured jejunocytes and colonocytes also exhibited a few apical microvilli (Figure 2A, B).
Figure 2 Ultrastructure of jejunocytes and colonocytes. Electron micrographs of monolayer cultures of bovine jejunocytes (A) and colonocytes (B). Note the presence of microvilli (MV) and tight junctions (TJ). Magnification: 6,000 ×.
To ascertain the identity and functionality of cultured monolayers, cells components from various culture passages were submitted to immunolabelling (immunocytochemistry and Western blot). These data were compared to results from gene expression studies (RT-PCR) and measurements of specific activities of brush border associated enzymes.
Cell characterization using immunodetection
Antibodies directed against cytokeratins (intermediary type II filaments specifically expressed in epithelia) and vimentin (intermediary type III filaments) were used to distinguish cells of epithelial origin from contaminating fibroblasts. Double immunolabeling of cryosections of the bowel wall showed the presence of the cytokeratin positive cells all over the lining epithelium and vimentin positive cells in the submucosa (Figure 3A). Cytokeratins showing the characteristic disposition of intermediate filaments were also recognized in obtained monolayers. Surprisingly, double staining vimentin-cytokeratin (Figure 3B) clearly showed that cultured cells expressed both epithelial and mesenchymal markers. However, the double staining cytokeratin-α-actin (Figure 3C) distinguished separate cells populations. The few α-actin positive cells, probably contaminant myofibroblasts (and/or smooth muscle cells) were released from the gut wall during the dissociation procedure and not completely eliminated by the sorbitol centrifugation or the selective attachment. Indirect immunofluorescence revealed also the membrane distribution of the epithelial specific antigen (ESA, a cell surface glycoprotein) in cultured enterocytes (Figure 4A) and intracellular distribution of the cytokeratin 18 (Figure 4B). E-Cadherin representing one of the proteins located at the adherence junctions was also detected, but a mis-localization was observed. The cells did not present the characteristic staining pattern to the cell periphery, but a granular distribution into the cytoplasm (data not shown).
Figure 3 Immunostaining of cryosections and primocultures. Double-immunostainings: (A) pan-cytokeratin (FITC conjugate) and vimentin (revealed by a Cy3 conjugate) in small intestine cryosections; and (B) in enterocyte primocultures; (C) pan-cytokeratin (FITC conjugate) and alpha actin (revealed by a TRITC conjugate) in of bovine enterocyte primocultures.
Figure 4 Immunostaining of bovine enterocytes. Indirect immunofluorescence of (A) the epithelium specific antigen and (B) of cytokeratin 18 in bovine enterocytes.
Western blot analyses also confirmed the presence of epithelial characters in cultured cells. Pan-anti-cytokeratin and anti-cytokeratin 18 antibodies recognised the same major 45 KDa protein (Figure 5A, B) in homogenates from: (1) freshly scrapped intestinal epithelium, (2) organoid suspensions used to seed cultures, (3) intestinal cell cultures of both types, and in the (4) positive control Caco-2 cells. No cytokeratin staining was noticeable in 3T3 fibroblasts. The size uniformity of stained products in all tested epithelial cells was also confirmed using the pan-cytokeratin antibody (major signal at 45 KDa, Figure 5C, D). In addition, blots were also positive for another epithelial cell marker, the E-cadherin (around 35 KDa; Figure 5G, H). Compared to the expected size (120 KDa), the anti-E cadherin target likely presented the size of a protein domain (35 KDa, Figure 5H). The antibody used was directed against the intracellular domain of this protein.
Figure 5 Western blot of cell markers in jejunum, colon, jejunocyte and colonocyte proteins. Immunodetection of cytokeratin-18 (A, B), others acidic and basic cytokeratins by an anti-pan cytokeratin antibody (C, D), α-actin 1A4 (E, F) and E-cadherin (G, H) and vimentin (I, J) in cellular extracts separated by SDS-PAGE (10%) and transferred on PVDF membranes. A, C, E, I were jejunum and jejunocyte samples. B, D, F, J were colon and colonocyte samples. Lanes: 1, freshly scrapped epithelia; 2, organoids suspension; 3, primary culture of bovine enterocytes; 4, cultured enterocytes after the first passage; 5, cultured enterocytes after the second passage; 6, broad range protein standards molecular weight ranging from 5.7 KDa to 198 Kda; 7, cultured 3T3-fibroblast used as positive control for vimentin and α-actin staining and 8, cultured Caco-2 cells used as positive control for cytokeratin and E-cadherin staining.
In agreement with immunocytochemistry results, vimentin and α-actin presented distinct distribution in sample homogenates. Indeed, the anti-vimentin targeted the same proteins in all bovine intestinal cell cultures and in the control 3T3 fibroblasts (Figure 5I, J), while α-actin was scarcely detected in more advanced passages (Figure 5E, F). Taken together, western blot and immunocytochemistry results strongly suggested that vimentin is expressed in primary cultures of epithelial intestinal cells from bovine jejunum and colon. However, vimentin was not detected by western-blot in Caco-2 cell proteins (Figure 5I, J). Although immunocytochemistry results for α-actin indicated a very low contamination of bovine epithelial cell cultures by mesenchymal cells, western blot results indicated that this phenomenon is to low to be detected in most culture stages.
Gene expression analyses of enterocyte markers
RT-PCR analyses allowed us to investigate the expression of bovine enterocyte markers namely: villin (actin-caping protein in microvilli), zonula occludens (ZO1, associated with the cytoplasmic surface of tight junctions), fatty acid binding protein (FABP), small intestine peptidase (IP) and E-Cadherin. Expressions studies were also used to confirm the presence of the vimentin transcript in bovine cells. Similarly to western blot analyses, gene expressions studies were carried out on total RNA extracted from bovine samples of: (1) freshly scrapped epithelia, (2) organoid suspensions used to seed cultures and (3) intestinal cell cultures of both types. As can be seen in figure 6 (A–G), cDNA amplification products presented the expected size (see methods) using specific primers for coding regions of the epithelial markers, such as the villin gene (Figure 6C), the ZO1 gene (Figure 6D) and the E-cadherin (Figure 6G), that were uniformly expressed in cells from primary cultures of bovine enterocytes. Besides these structural components expected in any epithelium, we also investigated the expression of functional markers of digestive epithelia. Among them, the gene coding for the fatty acid binding protein (FABP, Figure 6B), presented a constant expression in both jejunocytes and colonocytes in vitro. The second tissue specific marker studied was the gene coding for the intestinal peptidase (IP, Figure 6E). In this case, the expression was restricted to jejunocyte samples, as logically expected on the basis of the functional specialization of the jejunum (nutrient processing and absorption). The IP expression could then be considered as suitable jejunocyte marker. In accordance with immunostaining results, samples from jejunocyte and colonocyte cultures expressed the vimentin gene (Figure 6F). Interestingly, while western blots analyses failed to reveal vimentin either in freshly removed epithelia or in organoids suspensions used to seed cultures, the transcript of vimentin gene was clearly detected in those samples, indicating that the vimentin is constitutively expressed in jejunum, as well as in colon from the bovine intestine. However, western blot results suggested that vimentin synthesis seemed to be restricted in cultured cells with a progressive increase over the culture passage number (Figure 5I, J).
Figure 6 Gene expession analysis in bovine intestinal epithelium, organoids and cultured cells. RT-PCR analysis of several gene expression in bovine intestinal epithelia, isolated organoids and cultured cells. The fragment size (bp) of amplified cDNA fragment was determined after migration on agarose (1%) gels and subsequent staining with ethidium bromide. Primers directed against (A) β-actin, amplicon size 485 bp, used as a quality control of cDNA obtained by reverse transcription. Amplifications of epithelial and enterocyte specific markers were obtained using primers directed against (B) FABP, 568 bp, (C) villin, 384 bp, (D) ZO1, 272 bp, (E) small intestinal peptidase, 276 bp, (F) vimentin, 548 bp and (G) E-cadherin, 308 bp. Lanes: 1, freshly scrapped colon epithelium; 2, organoid suspension from colon; 3, primary culture of colonocytes; 4, cultured colonocytes after the first passage; 5, freshly scrapped jejunum epithelium; 6, organoid suspension from jejunum; 7, primary culture of jejunocytes; 8, cultured jejunocytes after the first passage; 9, molecular weight standard in bp.
Specific activities of two brush border-associated enzymes
The chosen enzymes were a disaccharidase (maltase) and the alkaline phosphatase. As for gene expression studies, enzyme activities were investigated in bovine samples of: (1) freshly scrapped epithelia, (2) organoid suspensions used to seed cultures and (3) intestinal cell cultures of both types. Assays performed on fresh epithelial tissue confirmed that jejunum and colon differed in respect to enzyme specific activities (SA), the jejunum presenting the highest level. Maltase SA (i.e. differentiation marker) clearly decreased in primary cultures of both cells types (Figure 7A).
Figure 7 Enzymatic characterization of fresh mucosa, seeded cells, primary culture and cultured cells. Specific activities of (A) maltase and (B) alkaline phosphatase from cell homogenates of fresh mucosa (FM), seeded cells (CS), primary cultured cells (P0), cultured cells after the first passage (P1) and cultured cells after the second passage (P2). Homogenates were made at confluence of the monolayer. Black boxes represent jejunum cells and open boxes colon cells. Results are expressed as mean values ± SD for three culture samples. Inside the same organ group, boxes sharing the same supercript letter are significantly different (p < 0.05, N = 3).
Furthermore, the cell isolation procedure selected disaccharidase depleted material, since organoid suspensions presented a 50% reduction in regard to the fresh epithelium preparation. It is interesting to note that, compared to the organoids suspension, jejunocytes from the first culture passage did not present a drastic reduction of the maltase SA. However, with subsequent culture passages the decreasing of disaccharidase SA was progressing toward a stable low level. Similar results were obtained from measurements of the intestinal alkaline phosphatase SA (IAP, Figure 7B), a ubiquitous enzyme that is a marker of brush border in intestinal epithelial cultures. Indeed, as expected from functional differences between jejunum and colon in vivo, epithelia homogenates, organoids suspensions, and cultures at first passages from bovine jejunum presented higher SA for this enzyme. Moreover, similarly to what was noted for maltase SA, IAP presented a strong decrease from fresh tissue to subcultured cells, so that jejunocyte and colonocyte SA values joined together to the same low level as the culture passage number increased.
Regarding a given passage number, the maltase SA decreased over the culture duration (Figure 8A). These results reflected a loss of cell differentiation in vitro compared to the in vivo level. Adjustments of cell culture medium were made to improve the differentiation status of jejunocytes and colonocytes in vitro. To this end the glucose was substituted by inosin in the culture medium; this condition was described to stimulate the acquisition of the enterocyte differentiated phenotype in vitro [29]. As seen in figure 8B, the use of a inosin-containing glucose-free culture medium during 7 days, led to an increase of maltase activity. However this effect diminished as the passage number increased, suggesting that glucose substitution by inosin should be complemented by other culture medium modifications. Addition of sodium butyrate (1–2 mM), a substance thought to promote enterocyte differentiation through a stimulation of the CDX2 homeobox gene expression [30], proved to be efficient to stimulate maltase SA (Figure 8C).
Figure 8 Enzymatic characterization of confluent cultures. Specific activity of maltase (A) over the age of a confluent culture of jejunocytes at the first passage, (B) over the increasing passage number of confluent cultures of jejunocytes (open boxes) compared to the same culture during 7-days in a inosin-containing glucose-deprived culture medium (black boxes) or (C) maintained 7 days in a medium added of sodium butyrate (1–2 mM). Results are expressed as mean values ± SD for three culture samples. ND =not detected, ***p < 0.001.
Discussion
Morphological data presented in the present study indicated that primary cultures of bovine enterocytes isolated from colon and jejunum presented characteristics of epithelial cells, such as a typical pavement-like aspect, the formation of domes and apical tight junctions and microvilli in confluent cultures. These bovine intestinal cells were shown to express in vitro epithelial cell markers such as brush border enzymes (maltase and alkaline phosphatase) and the epithelium typical cytoskeleton proteins, the cytokeratins. The enterocyte isolation procedure developed in the present study proved efficient to avoid a noticeable contamination from α-actin-positive cells (presumably myofibroblasts) in the two first steps of each type of culture (initial primary culture and first passage). The first passage also corresponded to cells presenting still a substantial differentiation level (in terms of brush border enzyme activities). To this respect, the first passage of each type of cultures appeared to be suitable to establish immortalized cell lines, a task which is now underway in our laboratory.
The functional differentiation state of the colon cells in culture could be estimated by the activities of drug metabolizing enzymes [31]. Indeed, bovine colon epithelial cell culture were characterized, as freshly isolated cells, by cytochrome P450 1A1-associated 7-ethoxyresorufin O-deethylase activity as well as by prostaglandin H-synthase-mediated production of prostaglandin E2. Activities of phase II enzymes (i.e. N-acetyltransferase 1) were also observed in colon epithelial cell cultures.
By contrast to most reports about mammalian enterocyte primocultures, data accumulated in the present study using (1) cell immunocytochemistry, (2) western blot and (3) gene expression analyses, showed that intestinal cell cultures from bovine jejunum and colon co-expressed epithelial markers and vimentin, an embryonic cytoskeleton filament that is expressed only in mesenchymal cells after birth. A second fibroblast marker, the α-actin was scarcely detected, indicating that the vimentin strong expression of culture samples did not reflect a culture contamination by mesenchymal cells. In addition, the vimentin distribution pattern did not exactly fit with the gene expression of this protein. Indeed, the immunodetection was negative in homogenates of fresh epithelia from jejunum and colon, as well as in organoid suspension used to seed each culture type, while the protein was essentially detected in samples of culture after the second passage. By contrast, the vimentin gene transcript appeared in all bovine samples analysed, including the undissociated epithelium. It seemed then that a post-transcriptional basal inhibition of vimentin synthesis has been suppressed in vitro. In agreement with this hypothesis, a previous research dedicated to the development of an intestinal lineage from the porcine intestine [12] led to the production of vimentin-positive cultures. In view of this result, authors postulated that cultured cells had undergone a "mesenchymal transformation" in vitro. Similarly, a re-expression of vimentin has been reported in epithelial cells under pathological conditions in vivo, as well as in primary culture [32]. Vimentin re-expression in vitro could be instrumental in the maintenance of cell structure and/or functions of specific proteins such as the ones associated to membrane lipid rafts [33].
Besides the possible mesenchymal transformation of epithelial cells in vitro, accumulating data are now in favour of a natural incidence of vimentin in the undissociated intestine epithelium. For instance, specialized M-cells found in the epithelium covering the intestine Peyer's patches were identified, among other criteria, by vimentin immunostaining [34-39]. These "Microfold-cells" are almost devoid of microvilli at their apical side as they are specialized in the transport of particulate antigens from the gut lumen to the underlying lymphoid tissue, where specific immune responses could take place. Although the structure and functions of these cells seem to be broadly admitted it is still unclear whether epithelial cells leaving the epithelium crypts are predetermined as M-cells or whether their particular phenotype developed from differentiated enterocytes at the lymphoid tissue vicinity. In favour of a possible local induction of M-cells differentiation, in vitro experiments using Caco-2 cells showed that the enterocyte typical phenotype could be converted in a M-like phenotype adding B lymphocytes to the basolateral side of the epithelium [40-42]. Regardless the aforementioned debate concerning the M-cells differentiation, enterocyte-lymphocyte co-culture experiments clearly demonstrated that a non-bacterial environmental factor could cause the brush border effacement of a vimentin-negative enterocyte (Caco-2 cells) leading to the phenotype of an intestinal cell type with a vimentin synthesis potential (M-like cell). Supporting the role of lymphocytes as phenotype conversion inducers, vimentin-positive cells were observed scattered throughout the villus epithelium of the rabbit small intestine, with the protein location extending from the perinuclear region to the cell membrane touching intraepithelial lymphocytes [43]. Additional indications that vimentin could be a marker of "differentiation variants" of enterocytes, came from a recent study that pointed out, in the ordinary epithelium villi of the rabbit ileum, a vimentin-positive enterocyte type sharing M-cell morphological features (brush border poor cells), but most probably representing a distinct cell type. Indeed, these so-called "cup-cells", differed from M-cells at two levels: they bound distinctive lectins and they did not take up microbeads instilled in the ileal lumen [44].
Conclusion
Altogether, M-cells and Cup-cells features in vivo, as well as the experimental effacement of the enterocyte brush-border in vitro suggest that a vimentin re-expression could be a marker of cellular dedifferentiation. In accordance with this hypothesis, primary cultures of bovine enterocytes obtained in the present study exhibited a low differentiation level. A poor differentiated level of cultured intestinal cells has already been described by several authors using intestinal biopsies from the human [23,45,46], the mouse [25,46] or the rat [47,48]. This differentiation deficiency could stem from the preferential selection of less differentiated cells (i.e. the ones belonging to the proliferating areas of epithelial crypts) by the isolation procedure. A low differentiation level in vitro may also be due to the suboptimal conditions of culture. Accordingly, obtaining a primary culture of fully differentiated intestinal cells from human foetal gut, Perreault and Beaulieu [21] failed to detect vimentin neither by immunocytochemistry, nor by western-blot. Further studies would then be designed to improve the differentiation level of bovine enterocytes in vitro and then allow us to verify if this is associated with an inhibition of vimentin synthesis. Results presented in the present study, suggest that the combination of at least two culture medium modifications would be the first step of this study: the glucose substitution by inosin and the addition of butyrate in the culture medium.
Methods
Cell isolation
Intestinal fragments of proximal region of jejunum and of spiralled colon from adult animals were obtained from a local slaughterhouse. Tissue removals were made in agreement with relevant local animal welfare laws, guidelines and policies. After several washes in warm (37°C) divalent ion free PBS (PBS: 13.7 mM NaCl, 0.27 mM KCl, 0.43 mM Na2HPO4, 0.14 mM KH2PO4, pH 7.4), fragments were washed in warm PBS supplemented with 1% Antibiotic-Antimycotic solution (Gibco/BRL), 2.7 mg/ml D-Glucose (Sigma), and 4 mM L-Glutamine (Gibco/BRL). Animal organs were then immersed in the same warm supplemented PBS and transported within 30 minutes to the laboratory. All following steps were performed under a laminar flux hood (Heraeus Instruments). After several washes, fragments of intestine were filled with PBS containing 1 mM 1.4-dithiothreitol (ICN Biomedicals Inc.) and, after closing their extremities, incubated for 5 minutes in a shaking bath at 37°C, to get the epithelium surface rid of mucous before performing the cell isolation procedure. Then this incubation medium was replaced by the digesting solution consisting in supplemented PBS added of collagenase (Sigma; 300 U/ml) and dispase (Gibco/BRL; 0.1 mg/ml in PBS), for 15 minutes (for jejunum) or 20 minutes (for colon) at 37°C in a shaking bath. Using again this medium, a second digestion step lasting 45 minutes (for jejunum) or 60 minutes (for colon) was carried out in the same conditions. The lumen content (bearing mostly dead cells, as demonstrated using a viability test described below) was discarded after enzymatic incubations. Then, each intestinal segment was longitudinally wide opened and the pre-digested epithelium was scraped from the digestive mucosa using a sterile scalpel blade. The resulting material was incubated in PBS containing 1 mg/ml dispase for 10 minutes whilst active pipetting movements were done to help the dissociation of epithelium fragments. Cells and organoids (i.e. cell aggregates) were pelleted by centrifugation at 140 × g for 3 minutes.
As the large amount of single cells present in the pellet is likely to include contaminant lymphocytes (initially located in basolateral spaces between epithelial cells) and fibroblasts (released from the conjunctive subepithelial layer), the following two methods were used to impoverish the preparation in those cells: the pellet was suspended in 30 ml of Dulbecco's modified Eagle's (D-MEM) medium containing 2% of sorbitol (Sigma) and centrifuged at 50 × g for 3 minutes (isopicnic centrifugation). Under these conditions, a large part of the single cells remained on the top of the sorbitol density cushion whereas organoids accumulated at the bottom of the tube. The supernatant was then discarded and the pellet was washed in sorbitol containing D-MEM and centrifuged at 50 × g for 3 minutes. The pellet wash was repeated for about 5 times, to obtain a clear supernatant. The final pellet content, not completely devoid of single cells, was then allowed to settle down on a cell culture surface, under conditions favouring the adhesion of possible residual fibroblasts (selective attachment): the medium used to suspend the pellet was D-MEM supplemented with a high amount of FBS (10%; Gibco/BRL) and the suspension was incubated for 1 hour at 37°C in a 175 cm2 culture flask (Greiner Bio-One) devoid of any coating permitting the attachment of epithelial cells. Unattached material was then removed from the flask, centrifuged at 140 × g for 3 minutes and finally the pellet suspended in the medium used to culture epithelial cells in collagen coated flasks (see below).
Cell culture and subculture
The culture medium used was high glucose D-MEM (Gibco/BRL) supplemented with 100 nM hydrocortisone (Sigma), 20 nM triiodothyronine (Sigma), 1 ng/ml Epidermal Growth Factor (Sigma), 1 μg/ml insulin (Actrapid, Novo Nordisk A/S, Denmark), 10 μg/ml Acid linoleic/Albumin (Sigma), 1% Glutamax (Gibco/BRL) and 1% Non Essential Amino-Acids (NEAA, Gibco/BRL) [49], with 1% Antibiotic-Antimycotic solution and 2% FBS (Hyclone Perbio Sciences). The organoids were seeded at a high density, in collagen I (Roche Diagnostics; 17 μg/cm2) coated culture flasks (to permit epithelial cell adhesion), and the first medium change was done after 20 hours. The cultures were maintained at 37°C in a humidified incubator (Heraeus Instruments) in a 5% CO2 atmosphere. Medium was changed every 2 or 3 days. Confluent cells were subcultured at a split ratio of 1:2. To this end, cells were detached by incubation for 3 to 5 minutes, at 37°C, with 40 μl/cm2 trypsin/EDTA (Gibco/BRL). Cells were harvested in a 10 fold greater volume of supplemented culture medium and enzyme was washed away by centrifugation at 140 × g for 3 minutes. The pellet was suspended in the supplemented culture medium and the cell suspension redistributed in collagen coated culture flasks. Samples from both cell culture types (colonocytes or jejunocytes) were frozen and stored in liquid nitrogen, at 1.5 106 cells per vial, in 1 ml of D-MEM containing 10 % DMSO (Merck) and 10 % FBS (Hyclone Perbio Sciences). We checked that frozen cells were suitable to reconstitute a culture.
Cell viability
Viability of freshly isolated material was tested with a mixture of 10 mg/ml ethidium bromide (Sigma) and 5 mg/ml acridin orange (Sigma) in PBS. The mixture was added 1/1000 to the cell suspension and immediately observed using a fluorescence microscope (Nikon Eclipse TE 200). Under these conditions dead cells nuclei were stained in red while nuclei of living cells corresponded to green spots.
Immunocytochemistry
Several antibodies were used to characterize the primary cell cultures, namely: monoclonal anti-pan-cytokeratin (mixture of clones C-11, PCK-26, CY-90, KS1A3, M20, A53 and B/A2, Sigma), monoclonal anti-E-cadherin (Becton Dickinson Biosciences), monoclonal anti-α smooth muscle actin (clone 1A4, Sigma), monoclonal anti-vimentin (clone V9, Sigma), monoclonal anti-epithelium specific antigen (ESA, Sigma) and monoclonal anti-cytokeratin peptide 18 (clone KS-BA2, Sigma). Prior to immunostaining, cells were fixed with 5% paraformaldehyde into PBS for 5 minutes at room temperature. Cells were then permeabilized by incubation for 15 minutes with 0.2% Triton X-100 (Roche) dissolved in PBS containing 5% goat serum (Gibco/BRL) to block non-specific binding. After 3 washes in PBS, for 5 minutes each, cells were incubated for 60 minutes at room temperature with the primary antibody in the blocking solution. Dilutions used were those recommended by the manufacturer. Cells were washed three times, for 5 minute each, in PBS and then incubated with the secondary antibody for 30 minutes at room temperature in darkness. The secondary antibody was anti-mouse IgG conjugated to FITC (Sigma) or TRITC to detect the antigen-antibody complex. After another three washes in PBS, cells were observed with a fluorescent microscope (Nikon Eclipse TE 200). In a few cases, direct immunofluorescence was performed using an anti-pan-cytokeratin (clone C11) FITC conjugate (Sigma) and anti-vimentin (clone V9) Cy3 conjugate (Sigma). For epithelia specific staining, Caco-2 cells (human carcinoma colonic cells, received from Prof. YJ Schneider, Catholic University of Louvain, Belgium) were used as positive controls. For all other immunodetections, 3T3 fibroblasts (a generous gift from Prof. E. Heinen, Histology Laboratory, University of Liege, Belgium) and bovine cell samples submitted to the staining procedure, but omitting the incubation with primary antibody, were used as negative controls.
Western Blot
Samples of fresh mucosa and cell suspension used for starting the cultures were prepared by washing the cells twice into PBS and suspending the pellet in ultrapure water (1 ml). Samples of confluent cultured cells were prepared by discarding the culture medium, rinsing the cell monolayer twice with PBS and then cell scraping in ultrapure water (300 μl ultrapure water for a 25 cm2 culture flask). Samples were sonicated (30 sec on ice, Sonic Power Company Cell Disrupter) to accomplish cell lysis. Twenty μg of proteins from each sample were separated by electrophoresis on 10% SDS-PAGE [50] and transferred to Bio Trade PVDF Transfer Membrane (0.45 μm, Pall Corporation, Life Sciences). After 1 hour incubation at 37°C in the blocking solution (PBS containing 0.2% of Tween 20 and 5% of milk powder; Nestle), membranes were incubated for 1 hour at 37°C (or overnight at 4°C) with primary antibody (see immunocytochemistry section) diluted in the blocking solution according the manufacturer's instruction. Three washes in the blocking solution were followed by the secondary antibody incubation. The secondary antibody was a peroxydase conjugate (Antimouse IgG peroxydase conjugate; Sigma) diluted at 1:1000 in the blocking solution. The blots were developed using the peroxydase substrate 3,3-diaminobenzidine (DAB; Sigma) containing 0.03% H2O2 (Merck). Cultured Caco-2 cells and 3T3 fibroblasts, submitted to the same treatment, were respectively used as positive and negative controls for epithelial immunostaining.
Gene expression analysis
Gene expression analyses of cell samples were realized using the combined mRNA reverse transcription-polymerase chain reaction (RT-PCR). Total RNA was extracted using the guadinium thiocyanate protocol [51]. Prior to reverse transcription RNA was treated with DNAse (DNAse RQ1; Promega) in the appropriate DNAse buffer (RQ1 Buffer; Promega) for 30 minutes at 37°C to break up the possible contaminant DNA. Reverse transcription of mRNA was performed from 2 μg of total RNA, in presence of RNAse inhibitor (RNAguard 40 U/μl; Promega) using oligo-dT primers (Oligo dT-15 Primer, Promega), deoxynucleotides (dNTP 10 mM, Promega), Mo-MuLV reverse transcriptase (200 U/μl, Promega) and the reverse transcriptase buffer (Promega) in a 150 μl final volume. Reverse transcription was performed at 42°C for 1 h. Primers used for the specific amplification of cDNA fragments corresponding to coding segments of the following proteins were: villin forward 5'-ACC-TTC-ACA-GGC-TGG-TTC-CT-3' and reverse 5'-GGT-TTT-GTT-GCT-TCC-AT-3' (amplification product size: 384 bp); intestinal peptidase (IP): forward 5'-GCG-ATT-ATG-CCC-CTT-TCA-TT-3' and reverse 5'-CAG-CCT-GCA-GGA-AGC-T-3' (amplification product size: 276 bp); Fatty Acids Binding Protein (FABP) forward: 5'-TTC-AGC-AGC-TGG-TAG-GAA-3' and reverse 5'-TAA-CCA-AAG-AGA-TGA-CCC-TA-3' (amplification product size: 276 bp); Zonula Occludens 1 (ZO1) forward: 5'-GCG CTG AAA GAA GCA ATT CA-3' and reverse: 5'-AAA CAT GGT TCT GCC TCA TC-3' (amplification product size: 272 bp); vimentin: forward 5'-CCG-GAG-CTA-CGT-GAC-CAC-AT-3' and reverse 5'-CTC-GGC-TTC-CTC-TCT-CTG-AA-3' (amplification product size: 540 bp); E-cadherin: forward 5'-CGC-ACA-ACA-AAA-TGT-TCA-CC-3' and reverse 5'-CAT-TGG-TGA-CTG-GGT-CTG-TG-3' (amplification product size: 308 bp). RNA extraction and RT-PCR success were checked through the co-detection of the constantly expressed gene β-actin, using the primer pair forward: 5'-AGA-AAA-TCT-GGC-ACC-ACA-CC-3' and reverse: 5'-GTC-AGG-CAG-CTC-GTA-GCT-CT-3' (amplification product size 485 bp). PCR comprised 35 cycles, each cycle consisting in the succession of a denaturing step at 95°C for 1 minute, a primer annealing step (at 50 to 55°C, depending on the primer pair) for 1 minute and a primer elongation step 72°C for 1 minute. They were preceded by a first denaturing step for 3 minutes at 95°C and followed by a final elongation step at 72°C for 5 minutes. The PCR reaction contained 1 to 2 μl of cDNA, 1 μmol/L of each primer (Eurogentec, Belgium), 1.5 mmol/L of MgCl2 (Promega), 200 μmol/L of deoxynucleotide triphosphates (dNTP 10 mM, Promega) and 0.5 units of Taq polymerase (Promega) in the appropriate buffer (Promega). Control reactions omitting the cDNA were included and yielded negative response.
Amplification products were visualized after electrophoresis on 1% agarose (Sigma) gels in Tris Borate EDTA buffer pH 8 (TBE; 89 mM Tris base, Sigma; 89 mM boric acid, Merck; 2 mM Na EDTA, Merck) stained with ethidium bromide (1‰ w/v, Sigma). Samples were dissolved in electrophoresis loading buffer (Promega). Simultaneous migration of DNA molecular weight standards (1 Kb ladder, Promega) allowed the size determination (in bp) of amplification products.
Enzymatic activities
Samples of fresh mucosa, cell suspension used to start the cultures and epithelial cell cultures were prepared as described for western blot analysis. Maltase activity (brush border disaccharidase) was measured according to the spectrophotometric method developed by Dahlqvist [52]. Alkaline phosphatase was estimated using 4-nitrophenyl phosphate as substrate (1.5 mM; Acros Organics) in ethyldiethanolamine buffer (150 mM; Sigma) and dinitrophenolphosphate (Sigma) as standard. After a 20 minute incubation time, the reaction was stopped by a 1 M NaOH solution. The absorbance of the enzyme reaction product was measured at 415 nm. Enzymatic activities were reported to the protein content measured by the method of Bradford [53] and expressed as specific enzymatic activities (SA, μmol/min.g).
Electron microscopy
For transmission electron microscopy, cells were grown in collagen I coated culture wells. Cultured cells were washed twice with PBS and then fixed in 2.5% glutaraldehyde solution at 4°C for 1 hour. Two other PBS washes were followed by a post-fixation in 2% osmium tetroxyde for 1.5 h. Then samples were dehydrated in serial ethanol dilutions, scrapped from their plastic support and embedded in epon-epoxy resin. Ultrathin sections were counterstained with uranyl acetate were fixed on grids and observed under a transmission electron microscope (JEOL JEM-100CX) at 80 kV.
Statistics
Results are reported as mean ± standard deviation (SD). With regard to heteroscedasticity, statistical analysis was performed using Newman-Student-Keuls or Kruskal-Wallis ANOVA. A p-value below 0.05 was considered as statistically significant.
Authors' contributions
DR carried out all the experimental analysis, participated to the design of the study and to the manuscript drafting. SL designed the study, initial culture protocols and molecular biology methods and supervised the first author in the paper writing with a major contribution to the discussion section. OP performed the statistical analysis, realized the artwork and helped to draft the manuscript. JM contributed to the design of the study and helped the access to animal tissues used in this study. GD participated to the design of the study and helped to draft the manuscript. All authors read and approved the final manuscript.
Acknowledgements
This work was supported by a Belgian Ministry for the Middle Class and Agriculture grant (S-6046). Part of results was presented as posters in the following meetings: Euroregional Life Sciences Conference (Maastricht, The Netherlands, November 2002) and "Bioforum" (University of Liege, Belgium, May 2003).
==== Refs
Booth C Potten CS Gut instincts: thoughts on intestinal epithelial stem cells J Clin Invest 2000 105 1493 1499 10841502
Clatworthy JP Subramanian V Stem cells and the regulation of proliferation, differentiation and patterning in the intestinal epithelium: emerging insights from gene expression patterns, transgenic and gene ablation studies Mech Dev 2001 101 3 9 11231054 10.1016/S0925-4773(00)00557-8
Evans GS Flint N Potten CS Primary cultures for studies of cell regulation and physiology in intestinal epithelium Annu Rev Physiol 1994 56 399 417 8010745 10.1146/annurev.ph.56.030194.002151
Autrup H Stoner GD Jackson F Harris CC Shamsuddin AK Barrett LA Trump BF Explant culture of rat colon: a model system for studying metabolism of chemical carcinogens In Vitro 1978 14 868 877 102584
Kedinger M Simon-Assmann PM Lacroix B Marxer A Hauri HP Haffen K Fetal gut mesenchyme induces differentiation of cultured intestinal endodermal and crypt cells Dev Biol 1986 113 474 483 2868951 10.1016/0012-1606(86)90183-1
Kondo Y Rose I Young GP Whitehead RH Growth and differentiation of fetal rat small intestinal epithelium in tissue culture. Relationship to fetal age Exp Cell Res 1984 153 121 134 6734734 10.1016/0014-4827(84)90454-3
Pretlow TP Stinson AJ Pretlow TG Glover GL Cytologic appearance of cells dissociated from rat colon and their separation by isokinetic and isopyknic sedimentation in gradients of Ficol J Natl Cancer Inst 1978 61 1431 1438 281552
Quaroni A Development of fetal rat intestine in organ and monolayer culture J Cell Biol 1985 100 1611 1622 3988803 10.1083/jcb.100.5.1611
Benya RV Schmidt LN Sahi J Layden TJ Rao MC Isolation, characterization, and attachment of rabbit distal colon epithelial cells Gastroenterology 1991 101 692 702 1650317
Hata Y Ota S Nagata T Uehara Y Terano A Sugimoto T Primary colonic epithelial cell culture of the rabbit producing prostaglandins Prostaglandins 1993 45 129 141 8430223 10.1016/0090-6980(93)90028-6
Reddy PM Sahi J Desai G Vidyasagar D Rao MC Altered growth and attachment of rabbit crypt colonocytes isolated from different developmental stages Pediatr Res 1996 39 287 294 8825802
Kaeffer B Bottreau E Velge P Pardon P Epithelioid and fibroblastic cell lines derived from the ileum of an adult histocompatible miniature boar (d/d haplotype) and immortalized by SV40 plasmid Eur J Cell Biol 1993 62 152 162 8269973
Velge P Kaeffer B Bottreau E Van LN The loss of contact inhibition and anchorage-dependent growth are key steps in the acquisition of Listeria monocytogenes susceptibility phenotype by non-phagocytic cells Biol Cell 1995 85 55 66 8882519 10.1016/0248-4900(96)89127-1
Dibb-Fuller MP Best A Stagg DA Cooley WA Woodward MJ An in-vitro model for studying the interaction of Escherichia coli O157:H7 and other enteropathogens with bovine primary cell cultures J Med Microbiol 2001 50 759 769 11549177
Follmann W Weber S Birkner S Primary cell cultures of bovine colon epithelium: isolation and cell culture of colonocytes Toxicol In Vitro 2000 14 435 445 10963960 10.1016/S0887-2333(00)00033-3
Moyer MP Culture of human gastrointestinal epithelial cells Proc Soc Exp Biol Med 1983 174 12 15 6634700
Hahn U Extracellular matrix proteins in small-intestinal cell cultures Scand J Gastroenterol Suppl 1988 151 70 78 3227319
Hahn U Stallmach A Hahn EG Riecken EO Basement membrane components are potent promoters of rat intestinal epithelial cell differentiation in vitro Gastroenterology 1990 98 322 335 2295387
Pang G Buret A O'Loughlin E Smith A Batey R Clancy R Immunologic, functional, and morphological characterization of three new human small intestinal epithelial cell lines Gastroenterology 1996 111 8 18 8698229 10.1053/gast.1996.v111.pm8698229
Pedersen G Saermark T Giese B Hansen A Drag B Brynskov J A simple method to establish short-term cultures of normal human colonic epithelial cells from endoscopic biopsy specimens. Comparison of isolation methods, assessment of viability and metabolic activity Scand J Gastroenterol 2000 35 772 780 10972184 10.1080/003655200750023471
Perreault N Beaulieu JF Primary cultures of fully differentiated and pure human intestinal epithelial cells Exp Cell Res 1998 245 34 42 9828099 10.1006/excr.1998.4221
Booth C O'Shea JA Potten CS Maintenance of functional stem cells in isolated and cultured adult intestinal epithelium Exp Cell Res 1999 249 359 366 10366435 10.1006/excr.1999.4483
Brandsch C Friedl P Lange K Richter T Mothes T Primary culture and transfection of epithelial cells of human small intestine Scand J Gastroenterol 1998 33 833 838 9754731 10.1080/00365529850171495
Hansen T Borlak J Bader A Cytochrome P450 enzyme activity and protein expression in primary porcine enterocyte and hepatocyte cultures Xenobiotica 2000 30 27 46 10659949 10.1080/004982500237802
Macartney KK Baumgart DC Carding SR Brubaker JO Offit PA Primary murine small intestinal epithelial cells, maintained in long-term culture, are susceptible to rotavirus infection J Virol 2000 74 5597 5603 10823867 10.1128/JVI.74.12.5597-5603.2000
Seidelin JB Horn T Nielsen OH Simple and efficient method for isolation and cultivation of endoscopically obtained human colonocytes Am J Physiol Gastrointest Liver Physiol 2003 285 G1122 G1128 14613919
Slorach EM Campbell FC Dorin JR A mouse model of intestinal stem cell function and regeneration J Cell Sci 1999 112 Pt 18 3029 3038 10462519
Schreider C Peignon G Thenet S Chambaz J Pincon-Raymond M Integrin-mediated functional polarization of Caco-2 cells through E-cadherin--actin complexes J Cell Sci 2002 115 543 552 11861761
Panja A Goldberg S Eckmann L Krishen P Mayer L The regulation and functional consequence of proinflammatory cytokine binding on human intestinal epithelial cells J Immunol 1998 161 3675 3684 9759892
Domon-Dell C Wang Q Kim S Kedinger M Evers BM Freund JN Stimulation of the intestinal Cdx2 homeobox gene by butyrate in colon cancer cells Gut 2002 50 525 529 11889074 10.1136/gut.50.4.525
Birkner S Weber S Dohle A Schmahl G Bolt HM Follmann W Activities of drug metabolizing enzymes in bovine colon epithelial cell cultures Arch Toxicol 2003 77 621 629 14508638 10.1007/s00204-003-0490-7
Terzi F Maunoury R Colucci-Guyon E Babinet C Federici P Briand P Friedlander G Normal tubular regeneration and differentiation of the post-ischemic kidney in mice lacking vimentin Am J Pathol 1997 150 1361 1371 9094992
Runembert I Queffeulou G Federici P Vrtovsnik F Colucci-Guyon E Babinet C Briand P Trugnan G Friedlander G Terzi F Vimentin affects localization and activity of sodium-glucose cotransporter SGLT1 in membrane rafts J Cell Sci 2002 115 713 724 11865027
Gebert A Hach G Bartels H Co-localization of vimentin and cytokeratins in M-cells of rabbit gut-associated lymphoid tissue (GALT) Cell Tissue Res 1992 269 331 340 1384978 10.1007/BF00319625
Jepson MA Simmons NL Hirst GL Hirst BH Identification of M cells and their distribution in rabbit intestinal Peyer's patches and appendix Cell Tissue Res 1993 273 127 136 8364955 10.1007/BF00304619
Jepson MA Clark MA Simmons NL Hirst BH Epithelial M cells in the rabbit caecal lymphoid patch display distinctive surface characteristics Histochemistry 1993 100 441 447 8163387 10.1007/BF00267824
Lelouard H Reggio H Mangeat P Neutra M Montcourrier P Mucin-related epitopes distinguish M cells and enterocytes in rabbit appendix and Peyer's patches Infect Immun 1999 67 357 367 9864237
Onori P Franchitto A Sferra R Vetuschi A Gaudio E Peyer's patches epithelium in the rat: a morphological, immunohistochemical, and morphometrical study Dig Dis Sci 2001 46 1095 1104 11341655 10.1023/A:1010778532240
Takeuchi T Gonda T Cellular kinetics of villous epithelial cells and m cells in rabbit small intestine J Vet Med Sci 2004 66 689 693 15240944 10.1292/jvms.66.689
Gullberg E Leonard M Karlsson J Hopkins AM Brayden D Baird AW Artursson P Expression of specific markers and particle transport in a new human intestinal M-cell model Biochem Biophys Res Commun 2000 279 808 813 11162433 10.1006/bbrc.2000.4038
Kerneis S Bogdanova A Kraehenbuhl JP Pringault E Conversion by Peyer's patch lymphocytes of human enterocytes into M cells that transport bacteria Science 1997 277 949 952 9252325 10.1126/science.277.5328.949
Lo D Tynan W Dickerson J Scharf M Cooper J Byrne D Brayden D Higgins L Evans C O'Mahony DJ Cell culture modeling of specialized tissue: identification of genes expressed specifically by follicle-associated epithelium of Peyer's patch by expression profiling of Caco-2/Raji co-cultures Int Immunol 2004 16 91 99 14688064 10.1093/intimm/dxh011
Iwatsuki H Ogawa C Suda M Vimentin-positive cells in the villus epithelium of the rabbit small intestine Histochem Cell Biol 2002 117 363 370 11976909 10.1007/s00418-002-0389-z
Ramirez C Gebert A Vimentin-positive cells in the epithelium of rabbit ileal villi represent cup cells but not M-cells J Histochem Cytochem 2003 51 1533 1544 14566025
Deveney CW Rand-Luby L Rutten MJ Luttropp CA Fowler WM Land J Meichsner CL Farahmand M Sheppard BC Crass RA Deveney KE Establishment of human colonic epithelial cells in long-term culture J Surg Res 1996 64 161 169 8812628 10.1006/jsre.1996.0323
Whitehead RH Demmler K Rockman SP Watson NK Clonogenic growth of epithelial cells from normal colonic mucosa from both mice and humans Gastroenterology 1999 117 858 865 10500068 10.1016/S0016-5085(99)70344-6
Fritsch C Simon-Assmann P Kedinger M Evans GS Cytokines modulate fibroblast phenotype and epithelial-stroma interactions in rat intestine Gastroenterology 1997 112 826 838 9041244 10.1053/gast.1997.v112.pm9041244
Tavelin S Milovic V Ocklind G Olsson S Artursson P A conditionally immortalized epithelial cell line for studies of intestinal drug transport J Pharmacol Exp Ther 1999 290 1212 1221 10454497
Schneider YJ Optimisation of hybridoma cell growth and monoclonal antibody secretion in a chemically defined, serum- and protein-free culture medium J Immunol Methods 1989 116 65 77 2644356 10.1016/0022-1759(89)90314-1
Laemmli UK Cleavage of structural proteins during the assembly of the head of bacteriophage T4 Nature 1970 227 680 685 5432063 10.1038/227680a0
Chomczynski P Sacchi N Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction Anal Biochem 1987 162 156 159 2440339 10.1016/0003-2697(87)90021-2
Dahlqvist A Method for assay of intestinal disaccharidases Anal Biochem 1964 7 18 25 14106916 10.1016/0003-2697(64)90115-0
Bradford MM A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding Anal Biochem 1976 72 248 254 942051 10.1016/0003-2697(76)90527-3
|
16321165
|
PMC1315322
|
CC BY
|
2021-01-04 16:31:29
|
no
|
BMC Cell Biol. 2005 Dec 1; 6:42
|
utf-8
|
BMC Cell Biol
| 2,005 |
10.1186/1471-2121-6-42
|
oa_comm
|
==== Front
Nutr Metab (Lond)Nutrition & Metabolism1743-7075BioMed Central London 1743-7075-2-321630067410.1186/1743-7075-2-32ResearchInsulin induces a positive relationship between the rates of ATP and glycogen changes in isolated rat liver in presence of glucose; a 31P and 13C NMR study Baillet-Blanco Laurence [email protected] Marie-Christine [email protected] Henri [email protected] Vincent [email protected] Jean-Louis [email protected] Service de Diabétologie-Nutrition, Hôpital du Haut-Lévêque, Avenue de Magellan, F-33604 Pessac, France2 Centre de Résonance Magnétique des Systèmes Biologiques, UMR 5536 CNRS-Université Bordeaux 2, 146 rue Léo Saignat, F-33076 Bordeaux Cedex, France2005 21 11 2005 2 32 32 23 8 2005 21 11 2005 Copyright © 2005 Baillet-Blanco et al; licensee BioMed Central Ltd.2005Baillet-Blanco 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
There is an emerging theory suggesting that insulin, which is known to be the predominant postprandial anabolic hormone, is also a major regulator of mitochondrial oxidative phosphorylation in human skeletal muscle. However, little is known about its effects in the liver. Since there is a theoretical relationship between glycogen metabolism and energy status, a simultaneous and continuous investigation of hepatic ATP and glycogen content was performed in intact and isolated perfused liver by 31P and 13C nuclear magnetic resonance (NMR) The hepatic rates of ATP and glycogen changes were evaluated with different concentrations of insulin and glucose during continuous and short-term supply.
Results
Liver from rats fed ad libitum were perfused with Krebs-Henseleit Buffer (KHB)(controls) or KHB containing 6 mM glucose, 30 mM glucose, insulin alone, insulin + 6 mM glucose, insulin + 30 mM glucose. In the control, glycogenolysis occurred at a rate of -0.53 ± 0.021 %·min-1 and ATP content decreased at a rate of -0.28 ± 0.029 %·min-1. In the absence of insulin, there was a close proportional relationship between the glycogen flux and the glucose concentration, whereas ATP rates never varied. With insulin + glucose, both glycogen and ATP rates were strongly related to the glucose concentration; the magnitude of net glycogen flux was linearly correlated to the magnitude of net ATP flux: fluxglycogen = 72.543(fluxATP) + 172.08, R2 = 0.98.
Conclusion
Only the co-infusion of 30 mM glucose and insulin led to (i) a net glycogen synthesis, (ii) the maintenance of the hepatic ATP content, and a strong positive correlation between their net fluxes. This has never previously been reported. The specific effect of insulin on ATP change is likely related to a rapid stimulation of the hepatic mitochondrial oxidative phosphorylation. We propose that variations in the correlation between rates of ATP and glycogen changes could be a probe for insulin resistance due to the action of substrates, drugs or pathologic situations. Consequently, any work evaluating insulin resistance on isolated organs or in vivo should determine both ATP and glycogen fluxes.
insulinATPglycogenoxidative phosphorylationliver
==== Body
Background
In the metformin treatment of insulin resistance-related complications, the mitochondrial effects of the drug are probably crucial in explaining its unique efficacy [1]. Mitochondrial dysfunctions have been reported in the muscle in type 2 diabetes [2] and in age-related insulin resistance [3], suggesting a link between insulin action and oxidative capacity in humans [4]. Thus, in healthy humans, it has been demonstrated that high physiological insulin sustained stimulated muscle protein synthesis and mitochondrial ATP production rate for 8 hr [5]. However, a rapid stimulatory action of insulin on ATP production was not shown.
Owing to its strong capacity for glucose production and utilization, the liver is a key regulator of glucose homeostasis. One of its major functions is to store glucose as glycogen after meals (glycogen synthesis) and to release glucose from this glycogen (glycogenolysis) at the post-absorptive state, which accounts for most endogenous glucose production. Disturbance of this function is thought to play a major role in the hyperglycemia of type 2 diabetes and in other insulin-resistant states. Despite much work on the issue, the effect of insulin on hepatic glycogenosynthesis remains controversial: insulin is known to activate glycogen synthase in vitro [6], but hepatic glycogenesis in vivo seems to need an increase in both insulin and plasma glucose levels.
Glucose is the main energy substrate and its hepatic metabolism can lead to ATP production during glycogenolysis (by cytosolic glycolysis and mitochondrial oxidative phosphorylation) or direct ATP consumption during glycogen synthesis (or indirectly from gluconeogenesis). Despite this strong link, no study to our knowledge has simultaneously addressed the effects of insulin and glucose on the rates of changes of hepatic glycogen and ATP contents. This knowledge gap is particularly regrettable as some studies have reported abnormal hepatic ATP contents in insulin-resistant states as obesity [7] and nonalcoholic steatohepatitis [8].
Nuclear magnetic resonance spectroscopy (NMR) is the only technique that allows non-invasive repetitive and simultaneous quantitation of both glycogen and ATP contents in the isolated liver. Monitoring in real-time makes it possible to calculate the rates of change in metabolites. We chose to work on isolated whole liver, because it is not subject to hormonal or nutritional parameters (in contrast to in vivo models). Moreover, the isolated whole liver is a closer model to physiological conditions than isolated hepatocytes. The initial presence of hepatic glycogen is the prerequisite for the kinetic study of its rate of change. As the glycogen level is very low in the fasting state, this study was performed on isolated livers from fed rats.
The purpose of this work was to explore in the liver the relationship between the glycogen pathway and energy metabolism and their dependence on insulin and/or glucose supply.
Results
(1) Rates of change of liver glycogen content (Table 1)
Table 1 Rates of hepatic glycogen changes and ATP changes (nmol·min-1·g-1 of liver ww).
Experimental conditions
Control (KHB)
Group A,
n = 7 G 6 mM
Group B,
n = 8 G 30 mM
Group C,
n = 8 Insulin
Group D,
n = 5 Insulin + G 6 mM
Group E,
n = 5 Insulin + G 30 mM
Group F,
n = 5
GLYCOGEN -387 ± 73 (a) -64 ± 109*
p = 0.02 vs control
(b) -474 ± 102
NS vs control (a) +52 ± 95**
p = 0.003 vs control
(b) -277 ± 80
NS vs control -431 ± 29
NS vs control -29 ± 73
p < 0.003 vs insulin NS vs G6(a) +139 ± 47
p < 0.00001 vs insulin
p = 0.015 vs G30(a)
ATP -7.28 ± 0.76 -10.14 ± 1.61
NS vs control -5.98 ± 1.22
NS vs control -8.06 ± 1.22
NS vs control -3.38 ± 0.17
p = 0.03 vs insulin
p = 0.01 vs G6 -0.10 ± 0.42
p = 0.003 vs insulin
p = 0.006 vs G30
100% glycogen = 73 ± 8.5 μmol glycosyl units·g-1 of liver ww; 100% ATP = 2.60 ± 0.65 μmol·g-1 of liver ww. The rates of glycogen changes were biphasic in group B and C; the first phase (a) lasted *20 min or **60 min. Mean ± SEM.
Since glycogen has been shown to be nearly 100% NMR-visible [9], changes in the peak area of the C1 resonance of glycogen can be used to monitor and calculate the rate of change in glycogen content in real time. In the absence of glucose in the isotonic perfusion medium (KHB, control group A), a linear decrease in glycogen content occurred without any lag time and its rate of disappearance was -0.53 ± 0.021 %·min-1 (100% being considered as the initial content = 73 ± 8.5 μmol glycosyl units·g-1 liver wet weight) (fig. 1).
Figure 1 Natural abundance 13C NMR spectra of two typical experiments at the beginning (A1 and B1) and after 115 min of perfusion (A2 and B2). A1 and A2 were 13C NMR spectra from a control rat perfused with Krebs-Henseleit buffer alone and B1 and B2 were 13C NMR spectra from rats perfused with both 30 mM glucose and insulin. Major resonances are assigned to (a,g) fatty acid, (b) C-1 glycogen, (c) glucose and glycogen (C-3β, C-5β glucose, glycogen; C-2 glucose, C-3α glucose; C2, C-5α glucose, C-5 glycogen; C-4αβ glucose, glycogen), (d) C-6 glucose, glycogen, (e) choline, (f) ethanolamine. Note the C-1 glycogen (b) decrease in the KHB group (A1 and A2) and the relative stability of the glycogen content of the liver during perfusion of both 30 mM glucose and insulin (B1 and B2).
In the presence of glucose in the KHB medium (Group B and C), the glycogen content tended to remain near the initial level for 20 min (6 mM glucose) and for 50 min with 30 mM glucose. During this step, a slight decrease in glycogen content with a rate of -0.087 ± 0.29 %·min-1 (p = 0.022 vs. control) occurred in the 6 mM glucose group and an increase in glycogen content was observed (+0.071 ± 0.13%·min-1, n = 4, p = 0.002 vs. KHB perfusion) in the 30 mM glucose group (fig 2). This positive balance between glycogenolysis and glycogenosynthesis was about 52 nmoles glucosyl units·min-1·g-1. After this relatively stable initial period, the apparent rate of glycogenolysis was similar in all the groups (-0.65 ± 0.14 and -0.38 ± 0.11%·min-1, in 6 mM and 30 mM glucose groups respectively; NS vs. KHB perfusion).
Figure 2 Rates of liver glycogen and ATP changes in absence of insulin in the perfusate. The results are expressed in %·min-1 (mean ± SEM; 100% initial glycogen content = 73 ± 8.5 μmol glycosyl units·g-1 liver wet weight; 100% initial ATP content = 2.60 ± 0.65 μmol/g liver wet weight). Only the rate of glycogen content increased with the perfusate glucose concentration.
With insulin alone in the KHB at different concentrations (Group D) (60 mU/L, 120 mU/L or 600 mU/L), the decrease in glycogen content was similar to that in the KHB group. The results are expressed with different insulin doses in the same group D for purposes of comparison with the other liver groups (-0.59 ± 0.04%·min-1, n = 9) (fig. 3)
Figure 3 Rates of liver glycogen and ATP changes in presence of insulin in the perfusate. The results are expressed in %·min-1 (mean ± SEM; 100% initial glycogen content = 73 ± 8.5 μmol glycosyl units·g-1 liver wet weight; 100% initial ATP content = 2.60 ± 0.65 μmol/g liver wet weight). The rates of both ATP and glycogen contents increased with the perfusate glucose concentration.
In the presence of both 6 mM glucose and insulin (Group E) the initial liver glycogen content was maintained for a period ranging from 45 to 90 min (slight decrease in glycogen content of -0.04 ± 0.38%·min-1, p = 0.01 vs. insulin group (group D) and NS vs. 6 mM glucose alone (group B). In the presence of both 30 mM glucose and insulin in the medium (group F), we observed throughout perfusion an increase in the rate of glycogen change in the isolated liver (+0.19 ± 0.065%·min-1, p = 0.0001 vs. KHB perfusion; p < 0.0001 vs. insulin alone; NS vs. 30 mM glucose alone) (fig. 3 and 4), leading to a glycogen content 2.6-fold higher than in glucose 30 mM. This positive balance was about 139 nmoles glycosyl units·min-1·g-1.
Figure 4 Time course of the change in hepatic glycogen content throughout the entire protocol perfusion in presence of insulin. The results are presented for three typical individual experiments: insulin alone (group D), insulin + 6 mM glucose (group E) and insulin + 30 mM glucose (group F). The results are expressed as the percentage of initial glycogen content, 100% being the value at the onset of the experiment. Note that the decrease in glycogen content observed with insulin alone was similar to that in the KHB control group.
(2) Rates of Change of liver ATP content (Table 1)
In the absence of insulin, the liver ATP content in the control group (0 glucose) decreased at a rate of -0.28 ± 0.029 %·min-1, near -7.3 nmol. min-1·g-1, 100% ATP corresponding to 2.60 ± 0.65 μmol/g liver wet weight (fig. 2 and 5). Similar rates were observed with 6 mM glucose (group B; -0.39 ± 0.062) or with 30 mM glucose (group C; -0.23 ± 0.047 %·min-1) (fig 2).
Figure 5 31P NMR spectra of 2 typical experiments at the beginning (A1 and B1) and after 115 min of perfusion (A2 and B2). A1 and A2 were 31P NMR spectra from a control rat perfused with Krebs-Henseleit buffer alone and B1 and B2 were 31P NMR spectra from rat perfused with both 30 mM glucose and insulin. Major resonances are assigned to (a) phosphomonoesters, (b) phosphocholine, (c) intracellular inorganic phosphate, (d) glycerol-3-phosphorylcholine, (f) nucleoside-5'-triphosphates (γNTP) and diphosphates (βNDP), (g) α-NTP and α-NDP, (h,i) nicotinamide adenine dinucleotide and uridine-5'-diphosphoglucose, (j) βNTP. The external reference is not shown (18.40 ppm). Note the βATP (j) decrease in the KHB group (A1 and A2) and the relative stability of the ATP content of the liver during the perfusion of both 30 mM glucose and insulin (B1 and B2).
In the presence of insulin alone in the medium (group D), there was also no significant difference compared to the control group. At all insulin concentrations, no difference was observed in ATP evolution so the results are pooled in group D (-0.31 ± 0.047%·min-1, n = 9).
In the presence of both insulin and glucose (group E and F), two different responses were observed: (i) with 6 mM glucose (group E) a decrease in the ATP content (-0.13 ± 0.006 %·min-1) similar to that in the control group (ii) with 30 mM glucose (group F) the rate of ATP decrease was almost zero; -0.004 ± 0.016 %·min-1) (fig 3), so 100% of the initial content was maintained during the first 90 min (fig. 6). This latter rate was significantly lower than with 30 mM glucose alone (p = 0.031) or insulin alone (p < 0.001).
Figure 6 Time course of the change in hepatic ATP content throughout the entire protocol perfusion in presence of insulin. The results are presented for three typical individual experiments: insulin alone (group D), insulin + 6 mM glucose (group E) and insulin + 30 mM glucose (group F). The results are expressed as the percentage of initial ATP content, 100% being the value at the onset of the experiment. Note that in the control KHB group, the 6 mM glucose group and the 30 mM glucose group, the ATP content decreased at a similar rate to the insulin group.
(3) Relationship between rates of hepatic glycogen and ATP changes
Simultaneous analysis of ATP and glycogen contents under the same experimental conditions showed differences in the evolution of the net metabolic flux that was closely dependent on the presence or absence of insulin. In the absence of insulin, there was a close proportional relationship between the glycogen flux and the concentration of glucose supply, whereas the ATP rates were independent of the glucose concentration (fig 2). Conversely, in the presence of both glucose and insulin, glycogen and ATP rates were both strongly related to the perfusate glucose concentration (fig 3). The magnitude of net glycogen flux was thus correlated with the net ATP flux : fluxglycogen = 72.543(fluxATP) + 172.08, R2 = 0.98 (fig 7), and the final glycogen content was significantly higher than in the presence of glucose alone.
Figure 7 Correlation between the rates of ATP and glycogen changes in presence of insulin according to the perfusate glucose concentration. The rates are expressed in nmol·min-1·g-1. The equation for the linear correlation is fluxglycogen = 72.543(fluxATP) + 172.08, R2 = 0.98. To obtain a positive net flux of glycogen (net synthesis), the net ATP consumption must not exceed 2.4 nmol·min-1·g-1.
Discussion
Recent evidence supports the hypothesis that insulin, the key postprandial hormone involved in fuel metabolism, is also a major regulator of mitochondrial oxidative phosphorylation in human skeletal muscle [4,5]. However, little is known about its effects in the liver.
Given the theoretical relationship between glycogen metabolism and energy status, hepatic ATP and glycogen content were investigated in rat liver. Since methods based on extraction of glycogen and/or ATP from liver cannot monitor their evolution in real time in the same intact liver, NMR spectroscopy was used to study the appearance or disappearance of metabolites in the same organ. This work is the first to investigate in real time the rapid effects of glucose and/or insulin on the rates of changes in glycogen and ATP content in the isolated liver of fed rats.
Insulin alone had no effect on the natural decrease in hepatic glycogen and ATP contents during the 2-hr perfusion. The increase in the concentration from 0 to 30 mM of glucose alone in the perfusate maintained liver glycogen in a dose-dependent manner. Moreover, glucose had a glycogen-sparing effect, in agreement with other data [10,11], but had no effect on basal ATP consumption, particularly at 30 mM. Indeed it is known that the glucose concentration must be around 20–30 mM to induce both a decrease in the activity of phosphorylase a below 10% (of total activity) and a subsequent activation of the glycogen synthase [6,12].
Only co-perfusion with insulin and 30 mM glucose led to a glycogen synthesis 2.6-fold (260%) higher than with 30 mM glucose alone, and this positive balance was in agreement with previous results obtained in fasted rats in conditions of neoglucogenesis (addition of alanine and/or lactate) [9] or with fructose [11]. This isolated liver model clearly demonstrates that insulin acts directly on the cells, thereby invalidating the hypothesis of an insensitivity of the liver to insulin developed elsewhere [13]. Since glycogen synthesis implies some UTP (and ATP) use, one might expect an increase in ATP consumption resulting in a reduction of liver ATP content. In fact, the presence of both 30 mM glucose and insulin led to (i) the maintenance of the hepatic ATP content, demonstrating that ATP consumption equaled ATP synthesis, and (ii) a net glycogen synthesis reflecting a positive balance between glycogenolysis and glycogen synthesis. More significant is the evidence of a strong linear correlation between the net fluxes of these two metabolites. We did not differentiate between the reduction of glycogenolysis and the enhancement of glycogenosis, or between the separate rates of ATP production and utilization that led to modifying hepatic glycogen and ATP contents. However, our global kinetic study of the balance between the synthesis and lysis pathways clearly underlines for the first time that it is only in the presence of insulin that the net flux of ATP correlates with the net flux of hepatic glycogen synthesis from exogenous glucose in the isolated liver.
To date, the role of insulin in the correlation of the net hepatic flux of ATP and glycogen has not been described, although previous reports point to a link between carbohydrate metabolism and ATP pathways. Insulin and glucose perfusions during resuscitation of rats from hemorrhagic shock increase the hepatic ATP content [14]. Both liver glycogen and ATP contents are decreased by fasting and exercise [15]. Cortez-Pinto et al [8] reported that recovery from hepatic ATP depletion was reduced with a body mass increase in healthy humans and was severely impaired in subjects with nonalcoholic steatohepatitis, a condition associated with hepatic insulin resistance. This result is corroborated by an inverse correlation observed between hepatic ATP content and Body Mass Index [7].
In 48 hr-cultured hepatocytes in the presence of insulin, a previous study showed an increase in the ATP concentration from 12 hr with a concomitant close correlation between the increases in exogenous glucose transport rate and glycogen content [16]. However, the authors underlined that no clear effect of insulin on glucose transport was found in the isolated 2 hr-perfused liver system. Moreover, the main role of insulin in maintaining the metabolic competence of cultured hepatocytes could be linked to the absolute ATP concentration rather than to the energy charge [17]. In rat hepatocytes and muscle, insulin, which otherwise has no significant effect on respiration, acts specifically on the mitochondrial Krebs cycle, within 30 sec, to stimulate by 30% the oxidation of only carbons 2 and 3 of pyruvate (or acetate) to CO2 [18]. Other authors [19] showed in the isolated hepatic mitochondria of rats treated with insulin for 9 weeks that the hormone improved the function of oxidative phosphorylation by increasing (i) ATP synthase activity, (ii) the ADP/O ratio and (iii) the respiratory control ratio. It may therefore be hypothesized that the rapid action of insulin (in the presence of exogenous glucose) on liver glycogen flux is exerted through the increase in ATP turnover, via (i) an enhancement in glycolytic ATP production and/or (ii) an increase in ATP flux originating from oxidative phosphorylation related to the activation of Krebs cycle and/or (iii) an increase in the oxidative phosphorylation yield.
Conclusion
Beside extending numerous NMR studies using non-physiological conditions (such as clamps to maintain insulin and glucose levels; review in [20]) or expensive 13C-labeling [21], this kinetic study performed by natural abundance NMR in isolated liver describes a simple reliable method to analyze the link between ATP and glycogen.
The main finding was a close metabolic link between hepatic ATP and glycogen induced by insulin in the presence of glucose. Such a major role for energy metabolism is in agreement with the emergent hypothesis suggesting a defect in mitochondrial oxidative phosphorylation in insulin-resistance [22]. We propose that variations in the correlation between rates of ATP and glycogen changes could be a probe for insulin resistance action of substrates, drugs or pathologic situations. Using NMR surface coil techniques, it should be possible to address this question in a similar fashion in vivo in the rat and in clinical investigations in human. Any work evaluating insulin resistance on isolated organs or in vivo should determine both ATP and glycogen fluxes.
Methods
Animal perfusion conditions
Male Wistar rats weighing between 80 and 120 g were used. Rats were obtained from the existing colony in the animal unit of Bordeaux 2 University. They were maintained on standard rat chow and were housed in an environmentally controlled room (temperature, humidity and airflow conditions) with a 12-hour light/dark cycle. The standard non-purified diet contained by weight (g/100 g): 63 starch (corn, sorghum, wheat, oats, barley), 22 protein, 3.5 fat, 6 fiber, 1 vitamin mixture and 4 salt mixture (Ralston, Purina, St. Louis, MO). The rats were fed ad libitum to facilitate glycogen synthesis and visualization of glycogen liver store at the beginning of the experiment. The rats were anesthetized with intraperitoneal barbiturate injection (pentobarbital sodium: 50 mg/kg of body weight). The rat liver was commonly used to investigate energetic metabolism [23] and glycogen turn-over [9]. The liver (4–6 g) from rats was perfused in situ through the portal vein (anterograde perfusion technique) as previously described [23]. The bile duct was cannulated in order to avoid mixing of bile with the surrounding medium. The perfusion was performed at 37°C with 95% O2–5% CO2 gassed Krebs-Henseleit buffer (KHB) as a flow rate (5 ml/min·g wet weight (ww)). The KHB composition was (mmol/L): 120 NaCl, 4.70 KCl, 1.20 MgSO4, 25 NaHCO3, 1.20 K2HPO4-KH2PO4, 1.30 CaCl2, 0.3 mM Na-pyruvate and 2.10 mM Na-lactate (pH = 7.35 at 37°C). The temperature of the perfusion circuit was regulated with a thermostatic bath. The liver (perfused in a non-recirculating mode) was then excised from the rat abdomen and transferred to a 20-mm diameter NMR cell. The perfusate temperature and pH were monitored before entering and after leaving the liver by continuous flow pH electrodes and temperature probes. High-grade chemicals were purchased from Sigma Chemicals (Saint Louis MO) except where otherwise specified.
This study complied with NIH guidelines (national research 1985).
Experimental procedures
Different groups of liver were perfused for 150 min under various medium concentrations of glucose and insulin (Actrapid, Novo Nordisk A/S 2880 Bagsvaerd Danemark). Insulin and glucose concentrations in this work reflect the portal concentration in in vivo conditions [24-26]. 60 mU/L of insulin is known to be the physiological level in the portal vein in the fasting state but this concentration probably increases in the fed state [24]. To study the effect of insulin on the liver, we chose to use (i) the physiological concentration, and (ii) 2- and 10-fold this concentration to mimic the fed portal state. Plasma glucose concentration in rats is around 6 mmol/L but the level in the portal vein during the fed state is higher (around 30 mmol/L) [25,26].
- Group A: control group (KHB alone) (n = 7)
- Group B: D-glucose 6 mM (n = 8)
- Group C: D-glucose 30 mM (n = 8)
- Group D: Insulin 60 mU/L (n = 2)
Insulin 120 mU/L (n = 3)
Insulin 600 mU/L (n = 4)
- Group E: D-glucose 6 mM + insulin 60 mU/L (n = 5)
- Group F: D-glucose 30 mM + insulin 60 mU/L (n = 5)
In the absence of a dose effect of insulin in group D, we performed the last experiments (insulin + glucose i.e. E and F groups) with 60 mU/L insulin concentration. Moreover, in some experiments performed with glucose 30 mM, no difference was observed when insulin concentration was 120 mU/L instead of 60 mU/L (data not shown).
Perfusion lines were saturated by KHB containing insulin at least 30 min before the beginning of perfusion to avoid absorption phenomena during the experiment as reported by others [25].
NMR methodology
The spectra were obtained using a 31P/13C double tuned 20 mm probe operating at 9.4T. Liver ATP content was followed by 31P NMR and carbohydrate content in natural abundance was assessed by 13C NMR. It was therefore possible to calculate the rate of changes of glycogen and ATP content. 31P and 13C NMR spectra were recorded at 161.9 and 100.6 MHz respectively on a DPX400 spectrometer. The magnetic field was adjusted on the water proton signal. 31P NMR spectra were obtained every 2 minutes (240 free induction decays (FID)) without proton decoupling. Radiofrequency pulses (70° flip angle) and 10,000 Hz spectral width were used for data acquisition. 13C NMR spectra were proton-decoupled using a gated bilevel mode. 13C NMR spectra were obtained every 8 min (500FID) resulting from a 66° radiofrequency pulse repeated every second (25,000 Hz spectral width).
Lorentzian line broadening of 15 Hz was applied before Fourier transformation for both 31P and 13C NMR spectra. Chemical shift of phosphorylated metabolites was expressed relative to the position of resonance in the frequency scale of an internal reference set (glycerophosphoryl-choline) at 0.47 ppm.13C chemical shifts were expressed from an external silicone reference (1.45 ppm). During the initial perfusion period, any liver showing an increase in the intensity of inorganic phosphate resonance occurring with a concomitant decrease in NTP signals probably reflecting some partial lobe ischemia was discarded.
Analysis
ATP and glycogen levels were estimated from peak areas and expressed as a percentage of the initial value. Relative changes in ATP levels, reflecting the dynamic changes in ATP liver content, were estimated from changes in the area of the spectral peaks of the β-phosphate of nucleoside triphosphates. βATP represents at least 80% of the NTP (around -18 ppm) [27]. The glycogen signal was characterized by a narrow signal at 100.5 ppm (C-1 resonance of glycosyl subunit in glycogen). The difference between the observed change in C1 resonance glycogen represents an estimate for the amount of glycogen that was broken down or synthesized during the time interval. 13C NMR quantification of glycogen content at the beginning of KHB perfusion was performed from a calibration curve established with oyster glycogen (0 to 185 mM glycosyl units).
Rate was expressed as %·min-1 or nmol. min-1·g-1 and can be calculated from the change in the metabolite signal area, as the linewidth remained constant throughout the experiment:
Rate = 0 : Synthesis = lysis
Rate>0 : Synthesis>lysis
Rate<0 : Synthesis<lysis
Data in the text, table and figures are given as mean ± SEM. Statistical analysis was performed with the software package Excel. A t test was used after a one-way ANOVA to identify significant differences between different perfusate compositions. Probability values of P < 0.05 were considered to be significant.
Abbreviations
ADP, adenosyl di phosphate; ATP, adenosyl tri phosphate; FID, free induction decay; FFA, free fatty acids; G6P, glucose 6 phosphate; KHB, Krebs-Henseleit buffer; NEFA, non-esterified fatty acids; NMR, nuclear magnetic resonance; NTP, nucleoside tri phosphate; UDPG, uridine diphosphoglucose.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
Laurence Baillet-Blanco: contribution to conception and design; acquisition of data; analysis of data and contribution to interpretation of data.
Marie-Christine Beauvieux: contribution to design, acquisition and analysis of data; involved in drafting the manuscript.
Vincent Rigalleau: critical revision.
Henri Gin: contribution to study design; involved in drafting the manuscript.
Jean-Louis Gallis: contribution to study design; analysis of data and large contribution to interpretation of data; involved in drafting the manuscript.
Acknowledgements
The authors are grateful to Ray Cook (Département des Langues Vivantes, Université Bordeaux 2) for revising the manuscript.
==== Refs
Leverve XM Guigas B Detaille D Batandier C Koceir EA Chauvin C Fontaine E Wiernsperger NF Mitochondrial metabolism and type-2 diabetes: a specific target of metformin Diabetes Metab 2003 29 6S88 94 14502105
Kelley DE He J Menshikova EV Ritov VB Dysfunction of mitochondria in human skeletal muscle in type 2 diabetes Diabetes 2002 51 2944 2950 12351431
Petersen KF Befroy D Dufour S Dziura J Ariyan C Rothman DL DiPietro L Cline GW Shulman GI Mitochondrial dysfunction in the elderly: possible role in insulin resistance Science 2003 300 1140 1142 12750520 10.1126/science.1082889
Boirie Y Insulin regulation of mitochondrial proteins and oxidative phosphorylation in human muscle TRENDS Endocrinol Metab 2003 14 393 394 14580754 10.1016/j.tem.2003.09.002
Stump CS Short KR Bigelow ML Schimke JM Nair KS Effect of insulin on human skeletal muscle mitochondrial ATP production, protein synthesis, and mRNA transcripts Proc Natl Acad Sci USA 2003 100 7996 8001 12808136 10.1073/pnas.1332551100
Hers HG The control of glycogen metabolism in the liver Annu Rev Biochem 1976 45 167 190 183599 10.1146/annurev.bi.45.070176.001123
Nair S Chacko VP Arnold C Diehl AM Hepatic ATP reserve and efficiency of replenishing: comparison between obese and nonobese normal individuals Am J Gastroenterol 2003 98 466 471 12591070
Cortez-Pinto H Chatham J Chacko VP Arnold C Rashid A Diehl AM Alterations in liver ATP homeostasis in human nonalcoholic steatohepatitis: a pilot study JAMA 1999 282 1659 1664 10553793 10.1001/jama.282.17.1659
Shulman GI Rothman DL Chung Y Rossetti L Petit WA JrBarrett EJ Shulman RG 13C NMR studies of glycogen turnover in the perfused rat liver J Biol Chem 1988 263 5027 5029 3128534
Parniak M Kalant N Incorporation of glucose into glycogen in primary cultures of rat hepatocytes an J Biochem Cell Biol 1985 63 C333 340
Youn JH Youn MS Bergman RN Synergism of glucose and fructose in net glycogen synthesis in perfused rat livers J Biol Chem 1986 261 15960 15969 3023336
Stalmans W The role of the liver in the homeostasis of blood glucose Curr Top Cellul Reg 1976 11 51 97
Bergman RN Non-esterified fatty acids and the liver: why is insulin secreted into the portal vein? Diabetologia 2000 43 946 952 10952470 10.1007/s001250051474
Chang CG Van Way CW Dhar A Helling T JrHahn Y The use of insulin and glucose during resuscitation from hemorrhagic shock increases hepatic ATP J Surg Res 2000 92 171 176 10896818 10.1006/jsre.2000.5857
Ghanbari-Niaki A Bergeron R Latour MG Lavoie J Effects of physical exercise on liver ATP levels in fasted and phosphate-injected rats Arch Physiol Biochem 1999 107 393 402 10916167 10.1076/1381-3455(199912)107:05;1-5;FT393
Jeejeebhoy KN Ho J Mehra R Bruce-Robertson A Hepatotrophic effects of insulin on glucose, glycogen and adenine nucleotides in hepatocytes isolated from fed adult rats Can J Biochem 1980 58 1004 1011 7006753
Gallo G Mazzei M Voci A Fugassa E Effects of insulin and dexamethasone on adenine nucleotide levels in cultured hepatocytes from adult rat Cell Biochem Funct 1988 6 101 105 3288371 10.1002/cbf.290060204
Bessman SP Mohan C Insulin as a probe of mitochondrial metabolism in situ Mol Cell Biochem 1997 174 91 96 9309671 10.1023/A:1006834408181
Huang Q Shao L Jiang H Miao ZC Shi QD Liu SS Effect of insulin on oxygen free radicals and oxidative phosphorylation in liver mitochondria of diabetic rats Acta Pharmacol Sin 2001 22 455 458 11743896
Shulman GI Cellular mechanisms of insulin resistance in humans Am J Cardiol 1999 84 3J 10J 10418851 10.1016/S0002-9149(99)00350-1
Shulman GI Rothman DL Smith D Johnson CM Blair JB Shulman RG DeFronzo RA Mechanism of liver glycogen repletion in vivo by nuclear magnetic resonance spectroscopy J Clin Invest 1985 76 1229 36 4044833
Petersen KF Dufour S Befroy D Garcia R Shulman GI Impaired mitochondrial activity in the insulin-resistant offspring of patients with type 2 diabetes N Engl J Med 2004 350 664 671 14960743 10.1056/NEJMoa031314
Delmas-Beauvieux MC Gallis JL Rousse N Clerc M Canioni P Phosphorus-31 nuclear magnetic resonance of isolated rat liver during hypothermic ischemia and subsequent normothermic perfusion J Hepatol 1992 15 192 201 1506639 10.1016/0168-8278(92)90035-N
Cardin S Emshwiller M Jackson PA Snead WL Hastings J Edgerton DS Cherrington AD Portal glucose infusion increases hepatic glycogen deposition in conscious unrestrained rats J Appl Physiol 1999 87 1470 1475 10517780
Parkes JL Grieninger G Insulin, not glucose, controls hepatocellular glycogen deposition. A re-evaluation of the role of both agents in cultured liver cells J Biol Chem 1985 260 8090 8097 3924911
Gustafson LA Neeft M Reijngoud DJ Kuipers F Sauerwein HP Romijn JA Herling AW Burger HJ Meijer AJ Fatty acid and amino acid modulation of glucose cycling in isolated rat hepatocytes Biochem J 2001 358 665 671 11535127 10.1042/0264-6021:3580665
Gallis JL Delmas-Beauvieux MC Biran M Rousse N Durand T Canioni P Is cellular integrity responsible for the partial NMR invisibility of ATP in isolated ischemic rat liver? NMR Biomed 1991 4 279 285 1816806
|
16300674
|
PMC1315323
|
CC BY
|
2021-01-04 16:37:46
|
no
|
Nutr Metab (Lond). 2005 Nov 21; 2:32
|
utf-8
|
Nutr Metab (Lond)
| 2,005 |
10.1186/1743-7075-2-32
|
oa_comm
|
==== Front
Nutr Metab (Lond)Nutrition & Metabolism1743-7075BioMed Central London 1743-7075-2-321630067410.1186/1743-7075-2-32ResearchInsulin induces a positive relationship between the rates of ATP and glycogen changes in isolated rat liver in presence of glucose; a 31P and 13C NMR study Baillet-Blanco Laurence [email protected] Marie-Christine [email protected] Henri [email protected] Vincent [email protected] Jean-Louis [email protected] Service de Diabétologie-Nutrition, Hôpital du Haut-Lévêque, Avenue de Magellan, F-33604 Pessac, France2 Centre de Résonance Magnétique des Systèmes Biologiques, UMR 5536 CNRS-Université Bordeaux 2, 146 rue Léo Saignat, F-33076 Bordeaux Cedex, France2005 21 11 2005 2 32 32 23 8 2005 21 11 2005 Copyright © 2005 Baillet-Blanco et al; licensee BioMed Central Ltd.2005Baillet-Blanco 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
There is an emerging theory suggesting that insulin, which is known to be the predominant postprandial anabolic hormone, is also a major regulator of mitochondrial oxidative phosphorylation in human skeletal muscle. However, little is known about its effects in the liver. Since there is a theoretical relationship between glycogen metabolism and energy status, a simultaneous and continuous investigation of hepatic ATP and glycogen content was performed in intact and isolated perfused liver by 31P and 13C nuclear magnetic resonance (NMR) The hepatic rates of ATP and glycogen changes were evaluated with different concentrations of insulin and glucose during continuous and short-term supply.
Results
Liver from rats fed ad libitum were perfused with Krebs-Henseleit Buffer (KHB)(controls) or KHB containing 6 mM glucose, 30 mM glucose, insulin alone, insulin + 6 mM glucose, insulin + 30 mM glucose. In the control, glycogenolysis occurred at a rate of -0.53 ± 0.021 %·min-1 and ATP content decreased at a rate of -0.28 ± 0.029 %·min-1. In the absence of insulin, there was a close proportional relationship between the glycogen flux and the glucose concentration, whereas ATP rates never varied. With insulin + glucose, both glycogen and ATP rates were strongly related to the glucose concentration; the magnitude of net glycogen flux was linearly correlated to the magnitude of net ATP flux: fluxglycogen = 72.543(fluxATP) + 172.08, R2 = 0.98.
Conclusion
Only the co-infusion of 30 mM glucose and insulin led to (i) a net glycogen synthesis, (ii) the maintenance of the hepatic ATP content, and a strong positive correlation between their net fluxes. This has never previously been reported. The specific effect of insulin on ATP change is likely related to a rapid stimulation of the hepatic mitochondrial oxidative phosphorylation. We propose that variations in the correlation between rates of ATP and glycogen changes could be a probe for insulin resistance due to the action of substrates, drugs or pathologic situations. Consequently, any work evaluating insulin resistance on isolated organs or in vivo should determine both ATP and glycogen fluxes.
insulinATPglycogenoxidative phosphorylationliver
==== Body
Background
In the metformin treatment of insulin resistance-related complications, the mitochondrial effects of the drug are probably crucial in explaining its unique efficacy [1]. Mitochondrial dysfunctions have been reported in the muscle in type 2 diabetes [2] and in age-related insulin resistance [3], suggesting a link between insulin action and oxidative capacity in humans [4]. Thus, in healthy humans, it has been demonstrated that high physiological insulin sustained stimulated muscle protein synthesis and mitochondrial ATP production rate for 8 hr [5]. However, a rapid stimulatory action of insulin on ATP production was not shown.
Owing to its strong capacity for glucose production and utilization, the liver is a key regulator of glucose homeostasis. One of its major functions is to store glucose as glycogen after meals (glycogen synthesis) and to release glucose from this glycogen (glycogenolysis) at the post-absorptive state, which accounts for most endogenous glucose production. Disturbance of this function is thought to play a major role in the hyperglycemia of type 2 diabetes and in other insulin-resistant states. Despite much work on the issue, the effect of insulin on hepatic glycogenosynthesis remains controversial: insulin is known to activate glycogen synthase in vitro [6], but hepatic glycogenesis in vivo seems to need an increase in both insulin and plasma glucose levels.
Glucose is the main energy substrate and its hepatic metabolism can lead to ATP production during glycogenolysis (by cytosolic glycolysis and mitochondrial oxidative phosphorylation) or direct ATP consumption during glycogen synthesis (or indirectly from gluconeogenesis). Despite this strong link, no study to our knowledge has simultaneously addressed the effects of insulin and glucose on the rates of changes of hepatic glycogen and ATP contents. This knowledge gap is particularly regrettable as some studies have reported abnormal hepatic ATP contents in insulin-resistant states as obesity [7] and nonalcoholic steatohepatitis [8].
Nuclear magnetic resonance spectroscopy (NMR) is the only technique that allows non-invasive repetitive and simultaneous quantitation of both glycogen and ATP contents in the isolated liver. Monitoring in real-time makes it possible to calculate the rates of change in metabolites. We chose to work on isolated whole liver, because it is not subject to hormonal or nutritional parameters (in contrast to in vivo models). Moreover, the isolated whole liver is a closer model to physiological conditions than isolated hepatocytes. The initial presence of hepatic glycogen is the prerequisite for the kinetic study of its rate of change. As the glycogen level is very low in the fasting state, this study was performed on isolated livers from fed rats.
The purpose of this work was to explore in the liver the relationship between the glycogen pathway and energy metabolism and their dependence on insulin and/or glucose supply.
Results
(1) Rates of change of liver glycogen content (Table 1)
Table 1 Rates of hepatic glycogen changes and ATP changes (nmol·min-1·g-1 of liver ww).
Experimental conditions
Control (KHB)
Group A,
n = 7 G 6 mM
Group B,
n = 8 G 30 mM
Group C,
n = 8 Insulin
Group D,
n = 5 Insulin + G 6 mM
Group E,
n = 5 Insulin + G 30 mM
Group F,
n = 5
GLYCOGEN -387 ± 73 (a) -64 ± 109*
p = 0.02 vs control
(b) -474 ± 102
NS vs control (a) +52 ± 95**
p = 0.003 vs control
(b) -277 ± 80
NS vs control -431 ± 29
NS vs control -29 ± 73
p < 0.003 vs insulin NS vs G6(a) +139 ± 47
p < 0.00001 vs insulin
p = 0.015 vs G30(a)
ATP -7.28 ± 0.76 -10.14 ± 1.61
NS vs control -5.98 ± 1.22
NS vs control -8.06 ± 1.22
NS vs control -3.38 ± 0.17
p = 0.03 vs insulin
p = 0.01 vs G6 -0.10 ± 0.42
p = 0.003 vs insulin
p = 0.006 vs G30
100% glycogen = 73 ± 8.5 μmol glycosyl units·g-1 of liver ww; 100% ATP = 2.60 ± 0.65 μmol·g-1 of liver ww. The rates of glycogen changes were biphasic in group B and C; the first phase (a) lasted *20 min or **60 min. Mean ± SEM.
Since glycogen has been shown to be nearly 100% NMR-visible [9], changes in the peak area of the C1 resonance of glycogen can be used to monitor and calculate the rate of change in glycogen content in real time. In the absence of glucose in the isotonic perfusion medium (KHB, control group A), a linear decrease in glycogen content occurred without any lag time and its rate of disappearance was -0.53 ± 0.021 %·min-1 (100% being considered as the initial content = 73 ± 8.5 μmol glycosyl units·g-1 liver wet weight) (fig. 1).
Figure 1 Natural abundance 13C NMR spectra of two typical experiments at the beginning (A1 and B1) and after 115 min of perfusion (A2 and B2). A1 and A2 were 13C NMR spectra from a control rat perfused with Krebs-Henseleit buffer alone and B1 and B2 were 13C NMR spectra from rats perfused with both 30 mM glucose and insulin. Major resonances are assigned to (a,g) fatty acid, (b) C-1 glycogen, (c) glucose and glycogen (C-3β, C-5β glucose, glycogen; C-2 glucose, C-3α glucose; C2, C-5α glucose, C-5 glycogen; C-4αβ glucose, glycogen), (d) C-6 glucose, glycogen, (e) choline, (f) ethanolamine. Note the C-1 glycogen (b) decrease in the KHB group (A1 and A2) and the relative stability of the glycogen content of the liver during perfusion of both 30 mM glucose and insulin (B1 and B2).
In the presence of glucose in the KHB medium (Group B and C), the glycogen content tended to remain near the initial level for 20 min (6 mM glucose) and for 50 min with 30 mM glucose. During this step, a slight decrease in glycogen content with a rate of -0.087 ± 0.29 %·min-1 (p = 0.022 vs. control) occurred in the 6 mM glucose group and an increase in glycogen content was observed (+0.071 ± 0.13%·min-1, n = 4, p = 0.002 vs. KHB perfusion) in the 30 mM glucose group (fig 2). This positive balance between glycogenolysis and glycogenosynthesis was about 52 nmoles glucosyl units·min-1·g-1. After this relatively stable initial period, the apparent rate of glycogenolysis was similar in all the groups (-0.65 ± 0.14 and -0.38 ± 0.11%·min-1, in 6 mM and 30 mM glucose groups respectively; NS vs. KHB perfusion).
Figure 2 Rates of liver glycogen and ATP changes in absence of insulin in the perfusate. The results are expressed in %·min-1 (mean ± SEM; 100% initial glycogen content = 73 ± 8.5 μmol glycosyl units·g-1 liver wet weight; 100% initial ATP content = 2.60 ± 0.65 μmol/g liver wet weight). Only the rate of glycogen content increased with the perfusate glucose concentration.
With insulin alone in the KHB at different concentrations (Group D) (60 mU/L, 120 mU/L or 600 mU/L), the decrease in glycogen content was similar to that in the KHB group. The results are expressed with different insulin doses in the same group D for purposes of comparison with the other liver groups (-0.59 ± 0.04%·min-1, n = 9) (fig. 3)
Figure 3 Rates of liver glycogen and ATP changes in presence of insulin in the perfusate. The results are expressed in %·min-1 (mean ± SEM; 100% initial glycogen content = 73 ± 8.5 μmol glycosyl units·g-1 liver wet weight; 100% initial ATP content = 2.60 ± 0.65 μmol/g liver wet weight). The rates of both ATP and glycogen contents increased with the perfusate glucose concentration.
In the presence of both 6 mM glucose and insulin (Group E) the initial liver glycogen content was maintained for a period ranging from 45 to 90 min (slight decrease in glycogen content of -0.04 ± 0.38%·min-1, p = 0.01 vs. insulin group (group D) and NS vs. 6 mM glucose alone (group B). In the presence of both 30 mM glucose and insulin in the medium (group F), we observed throughout perfusion an increase in the rate of glycogen change in the isolated liver (+0.19 ± 0.065%·min-1, p = 0.0001 vs. KHB perfusion; p < 0.0001 vs. insulin alone; NS vs. 30 mM glucose alone) (fig. 3 and 4), leading to a glycogen content 2.6-fold higher than in glucose 30 mM. This positive balance was about 139 nmoles glycosyl units·min-1·g-1.
Figure 4 Time course of the change in hepatic glycogen content throughout the entire protocol perfusion in presence of insulin. The results are presented for three typical individual experiments: insulin alone (group D), insulin + 6 mM glucose (group E) and insulin + 30 mM glucose (group F). The results are expressed as the percentage of initial glycogen content, 100% being the value at the onset of the experiment. Note that the decrease in glycogen content observed with insulin alone was similar to that in the KHB control group.
(2) Rates of Change of liver ATP content (Table 1)
In the absence of insulin, the liver ATP content in the control group (0 glucose) decreased at a rate of -0.28 ± 0.029 %·min-1, near -7.3 nmol. min-1·g-1, 100% ATP corresponding to 2.60 ± 0.65 μmol/g liver wet weight (fig. 2 and 5). Similar rates were observed with 6 mM glucose (group B; -0.39 ± 0.062) or with 30 mM glucose (group C; -0.23 ± 0.047 %·min-1) (fig 2).
Figure 5 31P NMR spectra of 2 typical experiments at the beginning (A1 and B1) and after 115 min of perfusion (A2 and B2). A1 and A2 were 31P NMR spectra from a control rat perfused with Krebs-Henseleit buffer alone and B1 and B2 were 31P NMR spectra from rat perfused with both 30 mM glucose and insulin. Major resonances are assigned to (a) phosphomonoesters, (b) phosphocholine, (c) intracellular inorganic phosphate, (d) glycerol-3-phosphorylcholine, (f) nucleoside-5'-triphosphates (γNTP) and diphosphates (βNDP), (g) α-NTP and α-NDP, (h,i) nicotinamide adenine dinucleotide and uridine-5'-diphosphoglucose, (j) βNTP. The external reference is not shown (18.40 ppm). Note the βATP (j) decrease in the KHB group (A1 and A2) and the relative stability of the ATP content of the liver during the perfusion of both 30 mM glucose and insulin (B1 and B2).
In the presence of insulin alone in the medium (group D), there was also no significant difference compared to the control group. At all insulin concentrations, no difference was observed in ATP evolution so the results are pooled in group D (-0.31 ± 0.047%·min-1, n = 9).
In the presence of both insulin and glucose (group E and F), two different responses were observed: (i) with 6 mM glucose (group E) a decrease in the ATP content (-0.13 ± 0.006 %·min-1) similar to that in the control group (ii) with 30 mM glucose (group F) the rate of ATP decrease was almost zero; -0.004 ± 0.016 %·min-1) (fig 3), so 100% of the initial content was maintained during the first 90 min (fig. 6). This latter rate was significantly lower than with 30 mM glucose alone (p = 0.031) or insulin alone (p < 0.001).
Figure 6 Time course of the change in hepatic ATP content throughout the entire protocol perfusion in presence of insulin. The results are presented for three typical individual experiments: insulin alone (group D), insulin + 6 mM glucose (group E) and insulin + 30 mM glucose (group F). The results are expressed as the percentage of initial ATP content, 100% being the value at the onset of the experiment. Note that in the control KHB group, the 6 mM glucose group and the 30 mM glucose group, the ATP content decreased at a similar rate to the insulin group.
(3) Relationship between rates of hepatic glycogen and ATP changes
Simultaneous analysis of ATP and glycogen contents under the same experimental conditions showed differences in the evolution of the net metabolic flux that was closely dependent on the presence or absence of insulin. In the absence of insulin, there was a close proportional relationship between the glycogen flux and the concentration of glucose supply, whereas the ATP rates were independent of the glucose concentration (fig 2). Conversely, in the presence of both glucose and insulin, glycogen and ATP rates were both strongly related to the perfusate glucose concentration (fig 3). The magnitude of net glycogen flux was thus correlated with the net ATP flux : fluxglycogen = 72.543(fluxATP) + 172.08, R2 = 0.98 (fig 7), and the final glycogen content was significantly higher than in the presence of glucose alone.
Figure 7 Correlation between the rates of ATP and glycogen changes in presence of insulin according to the perfusate glucose concentration. The rates are expressed in nmol·min-1·g-1. The equation for the linear correlation is fluxglycogen = 72.543(fluxATP) + 172.08, R2 = 0.98. To obtain a positive net flux of glycogen (net synthesis), the net ATP consumption must not exceed 2.4 nmol·min-1·g-1.
Discussion
Recent evidence supports the hypothesis that insulin, the key postprandial hormone involved in fuel metabolism, is also a major regulator of mitochondrial oxidative phosphorylation in human skeletal muscle [4,5]. However, little is known about its effects in the liver.
Given the theoretical relationship between glycogen metabolism and energy status, hepatic ATP and glycogen content were investigated in rat liver. Since methods based on extraction of glycogen and/or ATP from liver cannot monitor their evolution in real time in the same intact liver, NMR spectroscopy was used to study the appearance or disappearance of metabolites in the same organ. This work is the first to investigate in real time the rapid effects of glucose and/or insulin on the rates of changes in glycogen and ATP content in the isolated liver of fed rats.
Insulin alone had no effect on the natural decrease in hepatic glycogen and ATP contents during the 2-hr perfusion. The increase in the concentration from 0 to 30 mM of glucose alone in the perfusate maintained liver glycogen in a dose-dependent manner. Moreover, glucose had a glycogen-sparing effect, in agreement with other data [10,11], but had no effect on basal ATP consumption, particularly at 30 mM. Indeed it is known that the glucose concentration must be around 20–30 mM to induce both a decrease in the activity of phosphorylase a below 10% (of total activity) and a subsequent activation of the glycogen synthase [6,12].
Only co-perfusion with insulin and 30 mM glucose led to a glycogen synthesis 2.6-fold (260%) higher than with 30 mM glucose alone, and this positive balance was in agreement with previous results obtained in fasted rats in conditions of neoglucogenesis (addition of alanine and/or lactate) [9] or with fructose [11]. This isolated liver model clearly demonstrates that insulin acts directly on the cells, thereby invalidating the hypothesis of an insensitivity of the liver to insulin developed elsewhere [13]. Since glycogen synthesis implies some UTP (and ATP) use, one might expect an increase in ATP consumption resulting in a reduction of liver ATP content. In fact, the presence of both 30 mM glucose and insulin led to (i) the maintenance of the hepatic ATP content, demonstrating that ATP consumption equaled ATP synthesis, and (ii) a net glycogen synthesis reflecting a positive balance between glycogenolysis and glycogen synthesis. More significant is the evidence of a strong linear correlation between the net fluxes of these two metabolites. We did not differentiate between the reduction of glycogenolysis and the enhancement of glycogenosis, or between the separate rates of ATP production and utilization that led to modifying hepatic glycogen and ATP contents. However, our global kinetic study of the balance between the synthesis and lysis pathways clearly underlines for the first time that it is only in the presence of insulin that the net flux of ATP correlates with the net flux of hepatic glycogen synthesis from exogenous glucose in the isolated liver.
To date, the role of insulin in the correlation of the net hepatic flux of ATP and glycogen has not been described, although previous reports point to a link between carbohydrate metabolism and ATP pathways. Insulin and glucose perfusions during resuscitation of rats from hemorrhagic shock increase the hepatic ATP content [14]. Both liver glycogen and ATP contents are decreased by fasting and exercise [15]. Cortez-Pinto et al [8] reported that recovery from hepatic ATP depletion was reduced with a body mass increase in healthy humans and was severely impaired in subjects with nonalcoholic steatohepatitis, a condition associated with hepatic insulin resistance. This result is corroborated by an inverse correlation observed between hepatic ATP content and Body Mass Index [7].
In 48 hr-cultured hepatocytes in the presence of insulin, a previous study showed an increase in the ATP concentration from 12 hr with a concomitant close correlation between the increases in exogenous glucose transport rate and glycogen content [16]. However, the authors underlined that no clear effect of insulin on glucose transport was found in the isolated 2 hr-perfused liver system. Moreover, the main role of insulin in maintaining the metabolic competence of cultured hepatocytes could be linked to the absolute ATP concentration rather than to the energy charge [17]. In rat hepatocytes and muscle, insulin, which otherwise has no significant effect on respiration, acts specifically on the mitochondrial Krebs cycle, within 30 sec, to stimulate by 30% the oxidation of only carbons 2 and 3 of pyruvate (or acetate) to CO2 [18]. Other authors [19] showed in the isolated hepatic mitochondria of rats treated with insulin for 9 weeks that the hormone improved the function of oxidative phosphorylation by increasing (i) ATP synthase activity, (ii) the ADP/O ratio and (iii) the respiratory control ratio. It may therefore be hypothesized that the rapid action of insulin (in the presence of exogenous glucose) on liver glycogen flux is exerted through the increase in ATP turnover, via (i) an enhancement in glycolytic ATP production and/or (ii) an increase in ATP flux originating from oxidative phosphorylation related to the activation of Krebs cycle and/or (iii) an increase in the oxidative phosphorylation yield.
Conclusion
Beside extending numerous NMR studies using non-physiological conditions (such as clamps to maintain insulin and glucose levels; review in [20]) or expensive 13C-labeling [21], this kinetic study performed by natural abundance NMR in isolated liver describes a simple reliable method to analyze the link between ATP and glycogen.
The main finding was a close metabolic link between hepatic ATP and glycogen induced by insulin in the presence of glucose. Such a major role for energy metabolism is in agreement with the emergent hypothesis suggesting a defect in mitochondrial oxidative phosphorylation in insulin-resistance [22]. We propose that variations in the correlation between rates of ATP and glycogen changes could be a probe for insulin resistance action of substrates, drugs or pathologic situations. Using NMR surface coil techniques, it should be possible to address this question in a similar fashion in vivo in the rat and in clinical investigations in human. Any work evaluating insulin resistance on isolated organs or in vivo should determine both ATP and glycogen fluxes.
Methods
Animal perfusion conditions
Male Wistar rats weighing between 80 and 120 g were used. Rats were obtained from the existing colony in the animal unit of Bordeaux 2 University. They were maintained on standard rat chow and were housed in an environmentally controlled room (temperature, humidity and airflow conditions) with a 12-hour light/dark cycle. The standard non-purified diet contained by weight (g/100 g): 63 starch (corn, sorghum, wheat, oats, barley), 22 protein, 3.5 fat, 6 fiber, 1 vitamin mixture and 4 salt mixture (Ralston, Purina, St. Louis, MO). The rats were fed ad libitum to facilitate glycogen synthesis and visualization of glycogen liver store at the beginning of the experiment. The rats were anesthetized with intraperitoneal barbiturate injection (pentobarbital sodium: 50 mg/kg of body weight). The rat liver was commonly used to investigate energetic metabolism [23] and glycogen turn-over [9]. The liver (4–6 g) from rats was perfused in situ through the portal vein (anterograde perfusion technique) as previously described [23]. The bile duct was cannulated in order to avoid mixing of bile with the surrounding medium. The perfusion was performed at 37°C with 95% O2–5% CO2 gassed Krebs-Henseleit buffer (KHB) as a flow rate (5 ml/min·g wet weight (ww)). The KHB composition was (mmol/L): 120 NaCl, 4.70 KCl, 1.20 MgSO4, 25 NaHCO3, 1.20 K2HPO4-KH2PO4, 1.30 CaCl2, 0.3 mM Na-pyruvate and 2.10 mM Na-lactate (pH = 7.35 at 37°C). The temperature of the perfusion circuit was regulated with a thermostatic bath. The liver (perfused in a non-recirculating mode) was then excised from the rat abdomen and transferred to a 20-mm diameter NMR cell. The perfusate temperature and pH were monitored before entering and after leaving the liver by continuous flow pH electrodes and temperature probes. High-grade chemicals were purchased from Sigma Chemicals (Saint Louis MO) except where otherwise specified.
This study complied with NIH guidelines (national research 1985).
Experimental procedures
Different groups of liver were perfused for 150 min under various medium concentrations of glucose and insulin (Actrapid, Novo Nordisk A/S 2880 Bagsvaerd Danemark). Insulin and glucose concentrations in this work reflect the portal concentration in in vivo conditions [24-26]. 60 mU/L of insulin is known to be the physiological level in the portal vein in the fasting state but this concentration probably increases in the fed state [24]. To study the effect of insulin on the liver, we chose to use (i) the physiological concentration, and (ii) 2- and 10-fold this concentration to mimic the fed portal state. Plasma glucose concentration in rats is around 6 mmol/L but the level in the portal vein during the fed state is higher (around 30 mmol/L) [25,26].
- Group A: control group (KHB alone) (n = 7)
- Group B: D-glucose 6 mM (n = 8)
- Group C: D-glucose 30 mM (n = 8)
- Group D: Insulin 60 mU/L (n = 2)
Insulin 120 mU/L (n = 3)
Insulin 600 mU/L (n = 4)
- Group E: D-glucose 6 mM + insulin 60 mU/L (n = 5)
- Group F: D-glucose 30 mM + insulin 60 mU/L (n = 5)
In the absence of a dose effect of insulin in group D, we performed the last experiments (insulin + glucose i.e. E and F groups) with 60 mU/L insulin concentration. Moreover, in some experiments performed with glucose 30 mM, no difference was observed when insulin concentration was 120 mU/L instead of 60 mU/L (data not shown).
Perfusion lines were saturated by KHB containing insulin at least 30 min before the beginning of perfusion to avoid absorption phenomena during the experiment as reported by others [25].
NMR methodology
The spectra were obtained using a 31P/13C double tuned 20 mm probe operating at 9.4T. Liver ATP content was followed by 31P NMR and carbohydrate content in natural abundance was assessed by 13C NMR. It was therefore possible to calculate the rate of changes of glycogen and ATP content. 31P and 13C NMR spectra were recorded at 161.9 and 100.6 MHz respectively on a DPX400 spectrometer. The magnetic field was adjusted on the water proton signal. 31P NMR spectra were obtained every 2 minutes (240 free induction decays (FID)) without proton decoupling. Radiofrequency pulses (70° flip angle) and 10,000 Hz spectral width were used for data acquisition. 13C NMR spectra were proton-decoupled using a gated bilevel mode. 13C NMR spectra were obtained every 8 min (500FID) resulting from a 66° radiofrequency pulse repeated every second (25,000 Hz spectral width).
Lorentzian line broadening of 15 Hz was applied before Fourier transformation for both 31P and 13C NMR spectra. Chemical shift of phosphorylated metabolites was expressed relative to the position of resonance in the frequency scale of an internal reference set (glycerophosphoryl-choline) at 0.47 ppm.13C chemical shifts were expressed from an external silicone reference (1.45 ppm). During the initial perfusion period, any liver showing an increase in the intensity of inorganic phosphate resonance occurring with a concomitant decrease in NTP signals probably reflecting some partial lobe ischemia was discarded.
Analysis
ATP and glycogen levels were estimated from peak areas and expressed as a percentage of the initial value. Relative changes in ATP levels, reflecting the dynamic changes in ATP liver content, were estimated from changes in the area of the spectral peaks of the β-phosphate of nucleoside triphosphates. βATP represents at least 80% of the NTP (around -18 ppm) [27]. The glycogen signal was characterized by a narrow signal at 100.5 ppm (C-1 resonance of glycosyl subunit in glycogen). The difference between the observed change in C1 resonance glycogen represents an estimate for the amount of glycogen that was broken down or synthesized during the time interval. 13C NMR quantification of glycogen content at the beginning of KHB perfusion was performed from a calibration curve established with oyster glycogen (0 to 185 mM glycosyl units).
Rate was expressed as %·min-1 or nmol. min-1·g-1 and can be calculated from the change in the metabolite signal area, as the linewidth remained constant throughout the experiment:
Rate = 0 : Synthesis = lysis
Rate>0 : Synthesis>lysis
Rate<0 : Synthesis<lysis
Data in the text, table and figures are given as mean ± SEM. Statistical analysis was performed with the software package Excel. A t test was used after a one-way ANOVA to identify significant differences between different perfusate compositions. Probability values of P < 0.05 were considered to be significant.
Abbreviations
ADP, adenosyl di phosphate; ATP, adenosyl tri phosphate; FID, free induction decay; FFA, free fatty acids; G6P, glucose 6 phosphate; KHB, Krebs-Henseleit buffer; NEFA, non-esterified fatty acids; NMR, nuclear magnetic resonance; NTP, nucleoside tri phosphate; UDPG, uridine diphosphoglucose.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
Laurence Baillet-Blanco: contribution to conception and design; acquisition of data; analysis of data and contribution to interpretation of data.
Marie-Christine Beauvieux: contribution to design, acquisition and analysis of data; involved in drafting the manuscript.
Vincent Rigalleau: critical revision.
Henri Gin: contribution to study design; involved in drafting the manuscript.
Jean-Louis Gallis: contribution to study design; analysis of data and large contribution to interpretation of data; involved in drafting the manuscript.
Acknowledgements
The authors are grateful to Ray Cook (Département des Langues Vivantes, Université Bordeaux 2) for revising the manuscript.
==== Refs
Leverve XM Guigas B Detaille D Batandier C Koceir EA Chauvin C Fontaine E Wiernsperger NF Mitochondrial metabolism and type-2 diabetes: a specific target of metformin Diabetes Metab 2003 29 6S88 94 14502105
Kelley DE He J Menshikova EV Ritov VB Dysfunction of mitochondria in human skeletal muscle in type 2 diabetes Diabetes 2002 51 2944 2950 12351431
Petersen KF Befroy D Dufour S Dziura J Ariyan C Rothman DL DiPietro L Cline GW Shulman GI Mitochondrial dysfunction in the elderly: possible role in insulin resistance Science 2003 300 1140 1142 12750520 10.1126/science.1082889
Boirie Y Insulin regulation of mitochondrial proteins and oxidative phosphorylation in human muscle TRENDS Endocrinol Metab 2003 14 393 394 14580754 10.1016/j.tem.2003.09.002
Stump CS Short KR Bigelow ML Schimke JM Nair KS Effect of insulin on human skeletal muscle mitochondrial ATP production, protein synthesis, and mRNA transcripts Proc Natl Acad Sci USA 2003 100 7996 8001 12808136 10.1073/pnas.1332551100
Hers HG The control of glycogen metabolism in the liver Annu Rev Biochem 1976 45 167 190 183599 10.1146/annurev.bi.45.070176.001123
Nair S Chacko VP Arnold C Diehl AM Hepatic ATP reserve and efficiency of replenishing: comparison between obese and nonobese normal individuals Am J Gastroenterol 2003 98 466 471 12591070
Cortez-Pinto H Chatham J Chacko VP Arnold C Rashid A Diehl AM Alterations in liver ATP homeostasis in human nonalcoholic steatohepatitis: a pilot study JAMA 1999 282 1659 1664 10553793 10.1001/jama.282.17.1659
Shulman GI Rothman DL Chung Y Rossetti L Petit WA JrBarrett EJ Shulman RG 13C NMR studies of glycogen turnover in the perfused rat liver J Biol Chem 1988 263 5027 5029 3128534
Parniak M Kalant N Incorporation of glucose into glycogen in primary cultures of rat hepatocytes an J Biochem Cell Biol 1985 63 C333 340
Youn JH Youn MS Bergman RN Synergism of glucose and fructose in net glycogen synthesis in perfused rat livers J Biol Chem 1986 261 15960 15969 3023336
Stalmans W The role of the liver in the homeostasis of blood glucose Curr Top Cellul Reg 1976 11 51 97
Bergman RN Non-esterified fatty acids and the liver: why is insulin secreted into the portal vein? Diabetologia 2000 43 946 952 10952470 10.1007/s001250051474
Chang CG Van Way CW Dhar A Helling T JrHahn Y The use of insulin and glucose during resuscitation from hemorrhagic shock increases hepatic ATP J Surg Res 2000 92 171 176 10896818 10.1006/jsre.2000.5857
Ghanbari-Niaki A Bergeron R Latour MG Lavoie J Effects of physical exercise on liver ATP levels in fasted and phosphate-injected rats Arch Physiol Biochem 1999 107 393 402 10916167 10.1076/1381-3455(199912)107:05;1-5;FT393
Jeejeebhoy KN Ho J Mehra R Bruce-Robertson A Hepatotrophic effects of insulin on glucose, glycogen and adenine nucleotides in hepatocytes isolated from fed adult rats Can J Biochem 1980 58 1004 1011 7006753
Gallo G Mazzei M Voci A Fugassa E Effects of insulin and dexamethasone on adenine nucleotide levels in cultured hepatocytes from adult rat Cell Biochem Funct 1988 6 101 105 3288371 10.1002/cbf.290060204
Bessman SP Mohan C Insulin as a probe of mitochondrial metabolism in situ Mol Cell Biochem 1997 174 91 96 9309671 10.1023/A:1006834408181
Huang Q Shao L Jiang H Miao ZC Shi QD Liu SS Effect of insulin on oxygen free radicals and oxidative phosphorylation in liver mitochondria of diabetic rats Acta Pharmacol Sin 2001 22 455 458 11743896
Shulman GI Cellular mechanisms of insulin resistance in humans Am J Cardiol 1999 84 3J 10J 10418851 10.1016/S0002-9149(99)00350-1
Shulman GI Rothman DL Smith D Johnson CM Blair JB Shulman RG DeFronzo RA Mechanism of liver glycogen repletion in vivo by nuclear magnetic resonance spectroscopy J Clin Invest 1985 76 1229 36 4044833
Petersen KF Dufour S Befroy D Garcia R Shulman GI Impaired mitochondrial activity in the insulin-resistant offspring of patients with type 2 diabetes N Engl J Med 2004 350 664 671 14960743 10.1056/NEJMoa031314
Delmas-Beauvieux MC Gallis JL Rousse N Clerc M Canioni P Phosphorus-31 nuclear magnetic resonance of isolated rat liver during hypothermic ischemia and subsequent normothermic perfusion J Hepatol 1992 15 192 201 1506639 10.1016/0168-8278(92)90035-N
Cardin S Emshwiller M Jackson PA Snead WL Hastings J Edgerton DS Cherrington AD Portal glucose infusion increases hepatic glycogen deposition in conscious unrestrained rats J Appl Physiol 1999 87 1470 1475 10517780
Parkes JL Grieninger G Insulin, not glucose, controls hepatocellular glycogen deposition. A re-evaluation of the role of both agents in cultured liver cells J Biol Chem 1985 260 8090 8097 3924911
Gustafson LA Neeft M Reijngoud DJ Kuipers F Sauerwein HP Romijn JA Herling AW Burger HJ Meijer AJ Fatty acid and amino acid modulation of glucose cycling in isolated rat hepatocytes Biochem J 2001 358 665 671 11535127 10.1042/0264-6021:3580665
Gallis JL Delmas-Beauvieux MC Biran M Rousse N Durand T Canioni P Is cellular integrity responsible for the partial NMR invisibility of ATP in isolated ischemic rat liver? NMR Biomed 1991 4 279 285 1816806
|
16285887
|
PMC1315324
|
CC BY
|
2021-01-04 16:39:03
|
no
|
World J Surg Oncol. 2005 Nov 15; 3:74
|
latin-1
|
World J Surg Oncol
| 2,005 |
10.1186/1477-7819-3-74
|
oa_comm
|
==== Front
BMC Infect DisBMC Infectious Diseases1471-2334BioMed Central London 1471-2334-5-1081632115210.1186/1471-2334-5-108Research ArticleHigh frequency of Human Cytomegalovirus DNA in the Liver of Infants with Extrahepatic Neonatal Cholestasis De Tommaso Adriana MA [email protected] Paula D [email protected] Sandra CB [email protected] Cecília AF [email protected] Gabriel [email protected] Department of Pediatrics, Faculty of Medical Sciences, State University of Campinas (UNICAMP), Campinas-SP, Brazil2 Department of Internal Medicine, Faculty of Medical Sciences, State University of Campinas (UNICAMP), Campinas-SP, Brazil3 Department of Pathology, Faculty of Medical Sciences, State University of Campinas (UNICAMP), Campinas-SP, Brazil2005 1 12 2005 5 108 108 7 3 2005 1 12 2005 Copyright © 2005 De Tommaso et al; licensee BioMed Central Ltd.2005De Tommaso 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
Biliary atresia (BA) is the most severe hepatic disorder in newborns and its etiopathogenesis remains unknown. Viral involvement has been proposed, including the human cytomegalovirus (HCMV). The aims of the study were to use the polymerase chain reaction (PCR) to screen the liver tissue of infants with extrahepatic cholestasis for HCMV and to correlate the results with serological antibodies against HCMV and histological findings.
Methods
A retrospective study in a tertiary care setting included 35 patients (31 BA, 1 BA associated with a choledochal cyst, 2 congenital stenosis of the distal common bile duct and 1 hepatic cyst). HCMV serology was determined by ELISA. Liver and porta hepatis were examined histologically. Liver samples from infants and a control group were screened for HCMV DNA.
Results
Twelve patients had HCMV negative serology, 9 were positive for IgG antibodies and 14 were positive for IgG and IgM. Nine liver and seven porta hepatis samples were positive for HCMV DNA but none of the control group were positive (general frequency of positivity was 34.3% – 12/35). There was no correlation between HCMV positivity by PCR and the histological findings. The accuracy of serology for detecting HCMV antibodies was low.
Conclusion
These results indicate an elevated frequency of HCMV in pediatric patients with extrahepatic neonatal cholestasis. They also show the low accuracy of serological tests for detecting active HCMV infection and the lack of correlation between HCMV positivity by PCR and the histopathological changes.
==== Body
Background
The most common cause of extrahepatic neonatal cholestasis is biliary atresia (BA). BA is an obstructive cholangiopathy and the end result of a destructive inflammatory process that affects the biliary ductal system. This disease is more common in girls and involves 0.8–1/10000 live births [1]. The etiology of BA remains unknown but viral infections have been implicated with several viruses studied [2-12], including human cytomegalovirus (HCMV) [13-17]. The etiology of the other causes of extrahepatic neonatal cholestasis remains unknown.
HCMV is a slow replicating virus that belongs to the herpesvirus family (betaherpesvirus subfamily). Congenital infection by HCMV occurs in approximately 1% of all neonates. Of this 1%, only 5–10% have a typical cytomegalic inclusion disease with hepatosplenomegaly, jaundice and petechias. Another 10% will have a subclinical congenital infection [18], and the remaining 80–85% will be asymptomatic. Virus detection in the urine, saliva or tissues of neonates during the first three weeks of life defines congenital infection [19]. In contrast to infections such as rubella and congenital toxoplasmosis, maternal immunity against HCMV does not always provide adequate protection against intrauterine transmission [19]. Hepatic involvement is frequent and clinical evidence of hepatitis is occasionally found [20-22].
Perinatal infection is diagnosed in children who do not eliminate the virus in urine at birth but who begin to show the virus between the 4th and 12th week of life. The infection results from viral transmission during delivery, through maternal milk or by blood transfusions, the first two of these being the most important. The vast majority of infants remain asymptomatic [19].
The purpose of this study was to screen the liver tissue of infants with extrahepatic neonatal cholestasis for the presence of HCMV using PCR, and to correlate the results with the HCMV serologies (ELISA systems) and histopathological findings in these patients.
Methods
Thirty-five patients (13 males and 22 females) with extrahepatic neonatal cholestasis were evaluated upon admission to the Pediatric Gastroenterology Service of the university hospital at UNICAMP, from September 1992 to July 2000. The median infant age at the time of the first visit was 82.5 days (range 25–239 days). Neonatal cholestasis was secondary to BA in 31 infants and BA was associated with a choledochal cyst in one, to distal choledochal stenosis in two, and to hepatic cyst (measuring 3.2 × 3.4 cm, close to the porta hepatis, compressing the biliary tree and leading to obstructive cholestasis) in one.
The diagnosis of BA was confirmed by intraoperative cholangiogram. The diagnosis of choledochal cyst was made by ultrasound and the distal choledochal stenosis was diagnosed by intraoperative cholangiogram.
Liver tissue samples of nine infants were included as controls and had presented the following diagnoses: 1 had drug-related hepatotoxicity, 2 had alpha-1-antitrypsin deficiency, 2 had galactosemia, 1 had cystic fibrosis, 1 had congenital hepatic fibrosis, 1 had hepatoblastoma and 1 had a metabolic disease of undefined cause. The median infant age was 46 days (range 28–180 days).
Serological investigation
The patients were divided into three groups, based on the CMV serology results (by ELISA commercial system: SORIN BIOMEDICA, Italy): 1) ELISA IgG-/IgM-; 2). ELISA IgG+/IgM-, and 3) ELISA IgG+/IgM+. IgM antibodies were detected by ELISA-capture assay.
Histological Analysis
All histological analyses were done by the same pathologist. Fifty three samples (33 liver tissues and 20 porta hepatis fragments) were evaluated by PCR. Of the 33 liver tissues evaluated, 10 were collected by percutaneous biopsies, while the others were surgical biopsies. The following characteristics were evaluated: portal fibrosis (0 = absent, 1 = slight, 2 = moderate, 3 = severe), septa (0 = absent, 1 = mild, 2 = moderate, 3 = severe), nodules (P = present, A = absent), cholestasis and cholangitis (0 = absent, 1 = mild, 2 = moderate, 3 = severe), giant cell transformation (P = present, A = absent), eosinophils (P = present, A = absent), myeloid metaplasia (P = present, A = absent), siderosis (0 = absent, I = degree I, II = degree II, III = degree III), cytomegalic inclusion (P = present, A = absent), microabscesses (P = present, A = absent), ductal proliferation (0 = absent, 1 = slight, 2 = moderate, 3 = intense), porta hepatis (normal, chronic inflammatory process with predominance of lymphomononuclear cells or chronically active with predominance of polymorphonuclear neutrophils or macrophages) and inflammatory process of the porta hepatis (1 = mild, 2 = moderate, 3 = severe).
DNA extraction/amplification
DNA was extracted from fresh material (4 patients) according to Rogers et al. [23] and from formalin-fixed paraffin-embbeded fragments by phenol-chloroform procedure described by Latchman [24]. PCR conditions for HCMV and β-globin were the same. Five sections of 10 μm thick were obtained for DNA amplification [25]. The human β-globin gene was amplified according SAIKI et al. [26]. HCMV was detected by PCR and nested-PCR, according Saiki et al. [26], Shibata et al. [27] and Demmler et al. [28].
The amplifications were done in a DNA thermocycler (Robocycler 40 – Stratagene) using 35 cycles of 94°C for 45 s, 55°C for 45 s and 72°C for 1 min. The cycles were preceded by an initial denaturation at 94°C for 5 min and were followed by a final extension for 7 min at 72°C. The primers for the human β-globin gene were: PCO3+ (5' CCTCTGACACAACTGTGTTCACTAGC 3') and PCO4+ (5' TCACCACCAACTTCATCCACGTTCACC 3'). The primers for HCMV DNA were: MIE4 (CCA AGC GGC CTC TGA TAA CCA AGC C), MIE5 (CAG CAC CAT CCT CCT CTT CCT CTG G), IE1 (CCA CCC GTG GTG CCA GCT CC) and IE2 (CCC GCT CCT CCT GAG CAC CC). The MIE4 and MIE5 primers were used for the first reaction (PCR) and IE1 and IE2 for the nested-PCR.
The reaction mixture consisted of 0.5 – 1.0 μl of DNA in a total volume of 20 μl containing 50 mM KCl, 10 mM Tris (pH 8.4), 3 mM MgCl, 0.1 mM of each primer, 200 mM of each deoxyribonucleotide triphosphate (dATP, dGTP, dCTP, dTTP) and 2–4 units Taq polymerase. Water was used to complete the total reaction volume. The mixture was covered with a drop of mineral oil.
The reaction products were separeted by electrophoresis in 2% agarose gels, stained with ethidium bromide and visualized under ultraviolet light. For each CMV reaction, fibroblast fluid containing strain AD 169 was used as a positive control and the reaction mixture without the strain was used as a negative control. Recommended procedures were used to avoid contamination [23].
The statistical analysis consisted of a descriptive analysis with a frequency table, Fisher's exact test, the Kappa coefficient and accuracy measurements. The significance level was set at 5%. This study was approved by the ethics committee of the Faculty of Medical Sciences, UNICAMP, and all infant's parents provided their informed consent prior to enrolling in the study.
Results
Active HCMV infection was defined if one or both of the following conditions are present: positive result for HCMV DNA by PCR and a positive test for IgM HCMV antibodies.
Twelve patients had HCMV negative serology (group I), 9 were positive for IgG antibodies (group II) and 14 were positive for IgG and IgM (group III). In group I, 11 patients had BA (associated with a situs inversus in one patient) and one patient had a distal choledochal stenosis. All of the patients in group II had BA and in group III, 12 children had BA (associated with a choledochal cyst in one of them), one had a distal choledochal stenosis and another had a hepatic cyst. The general characteristics of the patients are show in Table 1. All were full term neonates. The median time between HCMV serology and liver biopsy was 8 days.
Table 1 Gender and median of birth weigh, alanine aminotransferase (ALT), gamma-glutamyl transpeptidase (GGT) and direct bilirubin (DB) in the patients.
CHARACTERISTICS/DIAGNOSIS GENDER BIRTH WEIGH ALTi GGTi DBi (mg%)
BA* (10) 5F/5M 3180 g 4 12.2 22.7
Hepatic cyst* (1) F 2970 g 2.9 3.5 19.7
Stenosis of the distal common bile duct *(1) F ? 6.6 5.4 31
BA (21) 13F/8M 2950 g 5.3 16 23.7
BA + Choledocal cyst (1) F 3100 g 2.5 27.5 15.2
Stenosis of the distal common bile duct (1) F 3100 g 2.3 11.8 11.7
* = pcr positive patients, M = male; F = female; ALTi = ALT divided by the upper reference value; GGTi = GGT divided by the upper reference value; DBi = DB divided by the upper reference value;
Of 33 liver samples screened by PCR, 9 (27.3%) were positive for HCMV DNA and 7 out of 20 (35%) porta hepatis fragments were positive (figure 1). There was adequate agreement between these results (Kappa coefficient = 0.51), albeit some divergences. Two patients had HCMV DNA in the liver but not in porta hepatic. Three children were positive for HCMV DNA in the porta hepatis but negative for HCMV in the liver tissue. The overall frequency of HCMV DNA was 34.3% (12/35) based on a positive result in one of the two tissue samples. Based on the PCR distribution among the groups there were three positive patients in group I, three in group II and six in group III.
Figure 1 HCMV DNA results in some patients. 1 = Ladder 2 = Positive control 3 = Porta-hepatis fragment from patient A (CMV-) 4 = Liver tissue from patient A (CMV+) 5 = Liver tissue from patient B (CMV+) 6 = Urine from patient B (CMV+) 7 = Empty lane 8 = Negative control
The histological characteristics of HCMV positive patients are show in Table 2 and of HCMV negative patients in Table 3. The pathologist did not observe the presence of cytomegalic cells in any case. No significant correlation was observed between any of the findings.
Table 2 Histological characteristics of HCMV positive patients.
Diagnosis PHF Liver tissue
PF GCT Ductal proliferation Septs Nodules Cholestasis Cholangitis Eosinophils
BA MoCIP + + + + + + + +
BA# no surgery + - + + + + - -
BA normal + + + + - + - +
BA MoCAP + + + + - + + +
BA MCIP + + + + + + + +
BA MCIP + + + + + + + +
BA MCIP + + + + + + + +
BA ? ? ? ? ? ? ? ? ?
BA MCAP + - + + + + - -
BA normal + + + + - + - -
DCS No + - + + + + + +
HC No + + + + + + - +
BA = biliary atresia, DCS = distal choledocal stenosis, HC = hepatis cyst, PHF = porta hepatis fragment, # = situs inversus PF = portal fibrosis, GCT = giant cell transformation, MCIP = mild chronic inflammatory process, MoCIP = moderate chronic inflammatory process, MCPA = mild chronic active inflammatory process, MoCAP = moderate chronic active inflammatory process.
Table 3 Histological characteristics of HCMV negative patients.
Diagnosis PHF Liver tissue
PF GCT Ductal proliferation Septs Nodules Cholestasis Cholangitis Eosinophils
BA MoCAP + + + + + + - +
BA SCAP + + + + + + - -
BA MoCAP + + + + - + - -
BA no surgery + - + + - + - -
BA MCAP + + + + + + - +
BA MCIP + + + + + + - +
BA no surgery + - + + + + - +
BA MCIP + + + + - + + -
BA SCAP + + - - - + - -
BA MCAP + + + + - + + +
BA no surgery + - + - - + - -
BA MoCAP + + + + - + - +
BA MoCIP + + + + + + - +
BA no surgery + + + + + - +
BA no surgery + + + + + + - +
BA MoCAP + + + + + + - +
BA MoCAP + + + + + + - +
BA MoCAP + - + + - + - +
BA no surgery + + + + + + - +
BA ? + - + + + + - -
BA+CC MoCAP + - + + + + - +
CC MoCIP + + + + + + + +
DCS No + - + + + + _ +
BA = biliary atresia, DCS = distal choledocal stenosis, CC = choledocal cyst, PHF = porta hepatis fragment, PF = portal fibrosis, GCT = giant cell transformation, MCIP = mild chronic inflammatory process, MoCIP = moderate chronic inflammatory process, MCAP = mild chronic active inflammatory process, MoCAP = moderate chronic active inflammatory process, SCAP = severe chronic active inflammatory process.
Compared to PCR, serology had a low accuracy (59%) to detect active HCMV infection, with 54% sensitivity and 61% specificity. The patients' specimens from the control group had no HCMV DNA detected by PCR analysis. The table 4 shows the results of serology and PCR in all patients.
Table 4 Results of serology and PCR for detecting HCMV.
PATIENT SEROLOGY PCR – LIVER PCR – PORTA HEPATIS
AFS IgM-/IgG+ POSITIVE POSITIVE
ADS NEGATIVE POSITIVE NR
ADPS IgM+/IgG+ NEGATIVE NEGATIVE
AHL IgM+/IgG+ NR NEGATIVE
AGP NEGATIVE NEGATIVE NEGATIVE
ACHA NEGATIVE NEGATIVE NR
ACNG IgM+/IgG+ NEGATIVE NR
CT IgM+/IgG+ POSITIVE NR
CAAM IgM+/IgG+ NEGATIVE NR
CCA NEGATIVE POSITIVE POSITIVE
DAMG NEGATIVE NEGATIVE NR
DRA IgM+/IgG+ NEGATIVE POSITIVE
ESC IgM-/IgG+ NEGATIVE NEGATIVE
FRNP NEGATIVE NEGATIVE NEGATIVE
GDS IgM+/IgG+ NEGATIVE NR
GFO IgM-/IgG+ POSITIVE POSITIVE
JG NEGATIVE NEGATIVE NEGATIVE
JSO IgM-/IgG+ NEGATIVE NEGATIVE
LaGS IgM-/IgG+ POSITIVE NEGATIVE
LiGS NEGATIVE NEGATIVE NEGATIVE
LS IgM-/IgG+ NEGATIVE NR
MCC IgM+/IgG+ NEGATIVE NEGATIVE
MFRF NEGATIVE NEGATIVE NR
MFV IgM+/IgG+ POSITIVE NEGATIVE
NFAM IgM-/IgG+ NEGATIVE NR
PAL IgM+/IgG+ POSITIVE NR
RCM IgM+/IgG+ NEGATIVE POSITIVE
RRF IgM-/IgG+ NEGATIVE NR
RRX IgM-/IgG+ NEGATIVE NR
SCF IgM+/IgG+ NR POSITIVE
TDF NEGATIVE POSITIVE POSITIVE
TFMF IgM+/IgG+ NEGATIVE NEGATIVE
YSR NEGATIVE NEGATIVE NEGATIVE
WROS IgM+/IgG+ NEGATIVE NR
WGSS NEGATIVE NEGATIVE NR
NR = no results
Discussion
Progress has been made in the diagnosis, treatment and prognosis of BA and other causes of extrahepatic cholestasis. The etiopathogenic origin of the disease still remains unknown, although viral involvement has been suggested. Based on several diagnostic methods, including PCR, cytomegalovirus has been found in association with BA [13,14,16,17]. However, other studies have found no such correlation [12,15]. Of the other causes of obstructive neonatal cholestasis, only choledochal cysts have been associated with viral infections.
Sokol et al. [10] found no positivity for reovirus type 3 (29 patients with BA, 1 with a choledochal cyst) and Tyler et al. [15] found positivity in 7 of 9 children with choledochal cysts. Our study is the first to investigate the ocurrence of HCMV in other causes of extrahepatic neonatal cholestasis. HCMV DNA was found in patients with a hepatic cyst and a distal choledochal stenosis.
Serological screening revealed a seroprevalence of HCMV infection of 65.7% [23/35]. Previous studies made in Brazil have evaluated the incidence of congenital infection and prevalence of antibodies in mothers. Pannuti et al. [29] showed a prevalence of HCMV antibodies in mothers of 84.4% and the incidence of congenital infection was 0.98% in the low socioeconomic population. In the middle socioeconomic cohort, the maternal prevalence and incidence of congenital infections were 66.5% and 0.39%, respectively [29]. Yamamoto et al. [30] showed a prevalence of congenital infection of 2.6% with maternal seropositivity of 95%. On the other hand, the acquisition risk of perinatal infection was 30.9% [31]. Unfortunately, we don't have data regarding the serologic pattern of our patient's mothers since this test is not part of our institution's routine clinical practice.
The accuracy of serology to detected active HCMV infection compared with PCR in this study was low. The correlation between serology and PCR results suggests the need for caution in concluding, based on positive serology, that HCMV is responsible for the observed cholestasis. When the 14 patients with positive IgM serology were analyzed by PCR in biopsies, only 6 were positive. Thus, although 8 patients had an active HCMV infection detected by serology, the cholestasis was not secondary to this infection and its cause requires further investigation.
In several cases, a positive serology delayed referral of the patient to our service since they were considered to have hepatitis secondary to viral infection. This is important since most pediatricians begin the investigation of cholestasis by determining that infectious agents are involved, and suspend further analysis when a positive serology is found. Another factor to be considered in patients with positive IgM serology and negative PCR is the possibility of a false-positive result for IgM. False-positives can be secondary to the presence of rheumatoid factor (IgM of the neonate against maternal IgG), or to cross-reaction with other herpesviruses [22]. Until the screening of IgM antibody is improved, physicians should use more than one method in their diagnosis [22].
Of the 9 IgG+/IgM- patients, 3 were positive for HCMV DNA. The presence of IgG may represent the transplacental passage of maternal IgG (the infant being infected or not). With such a serological profile, the physician should be concerned not only with completing the investigation, but also with repeating the serology because of the possibility of IgM appearing in a subsequent collection, especially when the infant is infected. Another possibility would be to measure IgG levels as well, since they remain high in congenital or perinatal infection [32]. In patients with a positive PCR, the absence of IgM antibodies could represent a false-negative result. False-negative results may be secondary to the competition between the high levels of maternal IgG antibodies and the relatively low levels of fetal IgM [22]. In the post-natal period, when facing an antigenic stimulus, IgM levels rise rapidly during the first month of life and increase gradually. At around one year of age, these values reach approximately 60% of the adult levels [33].
Of the 12 ELISA-negative patients, 3 were positive by PCR. Sample contamination cannot be excluded, despite the low probability of this occurring in view of the care taken during extraction and PCR, including the use of nested PCR.
There was good agreement between liver and porta hepatis tissues when analysed separately. However, two patients with HCMV DNA in liver tissue were negative in the porta hepatis. In both, the band representing β-globin DNA in the porta hepatis fragment was considered weakly positive, while the corresponding hepatic band was considered strongly positive. This may indicate a false-negative result, judging by the sample DNA quality. Two patients were positive for HCMV DNA in the porta hepatis and negative in the liver. They had a positive serology (IgG+/IgM+) and very intense β-globin band. Since the HCMV was not genotyped, we cannot exclude the possibility of the existence of strains with different affinity for ducts or hepatocytes, and of strains with differences in their virulence. Further studies are needed to clarify this question.
In the absence of in situ hybridization it is unclear whether the detected HCMV was within the hepatocytes and/or biliary duct cells, or whether it was simply circulating in the plasma or in leucocytes.
In this study the pathologist did not observed the presence of cytomegalic cells or microabscess in any case. Similar results have been reported by others [9,14,16].
The 34.3% positivity for HCMV DNA found in infants with extrahepatic cholestasis contrast with the lack of positivity in any of the controls. These findings, and those of Fischler et al. [13] and Domiati-Saad et al. [11], indicate that cytomegalovirus may be involved in the etiopathogenesis of BA. This hypothesis could be tested with an experimental model of BA induced by HCMV in animals using the methodology described by Petersen et al. [34].
In contrast, Chang et al. [9] and Jevon et al. [14] found no HCMV involvement in the etiopathogenesis of BA. In the study by Jevon et al. [14], no HCMV was found in 12 infants with BA. A probable explanation for this finding could be the low prevalence of the infection in a Canadian population, which would not exclude viral participation in cholestasis. Chang et al. [9] examined the occurrence of HCMV in 50 infants with neonatal hepatitis (26 with BA) and 30 controls, and found positivity in two cases with BA, in 23 with hepatitis, but in none of the controls. A possible explanation for this finding could be the strain of virus involved.
The hypothesis that biliary epithelial cells function as antigen presenters was considered. In normal liver, MHC (major histocompatibility complex) class I antigens are expressed on the sinusoidal cells, large vessel and sinusoidal endothelium, biliary epithelium, and dendritic cells. Class II antigens are found only on the sinusoidal cells, capillary endothelium and dendritic cells. The ductal epithelium of BA patients has an aberrant expression of class II antigens [35]. Recently, specific morphological alterations in the biliary ducts has been described [36,37]. Injury to the epithelium secondary to an immune modulated response may be triggered by a viral infection.
If a viral infection contributes to this injury, then antiviral therapy in the infants with HCMV infection can be considered as a possible intervention and consequently may improve outcomes. Further controlled randomized studies are needed to clarify this aspect.
Conclusion
In conclusion, the frequency of HCMV detected by PCR was high (34.3%). The accuracy of serology for detecting active HCMV infection was low compared to PCR, and there was no correlation between the presence of HCMV and the typical histopathological alterations (microabscess and cytomegalic cells) or the inflammatory process in porta hepatis in our patients.
Authors' contributions
AMADT had primary responsibility for protocol development, patient screening, enrolment, outcome assessment, laboratory investigation, preliminary data analysis and writing the manuscript.
PDA participated in the development of the protocol, laboratory investigation and contribut to the writing of the manuscript.
SCBC participated in the development of the protocol and analytic framework for the study, and contribut to the writing of the manuscript.
CAFE was responsible for histological analysis and contributed to the writing of the manuscript.
GH supervised the design and execution of the study, performed the final data analysis and contributed to the writing of the manuscript.
All authors read and approved the final manuscript.
Pre-publication history
The pre-publication history for this paper can be accessed here:
Acknowledgements
The authors thank the FAPESP and CAPES for the financial of this study.
==== Refs
Howard ER Extrahepatic biliary atresia: a review of current management Br J Surg 1983 70 193 197 6338994
Bobo L Ojeh C Chiu D Machado A Colombani P Schwarz K Lack of evidence for rotavirus by polymerase chain reaction/enzyme immunoassay of hepatobiliary samples from children with biliary atresia Pediatr Res 1997 41 229 234 9029644
Drut R Gomez MA Drut RM Cueto Rua E Lojo M Virus Papiloma Humano, hepatitis gigantocelular neonatal y atresia de vias biliares Acta Gastroent Latinoamer 1998 28 27 31
Glaser JH Balistreri WF Morecki R Role of reovirus type 3 in persistent infantile cholestasis J Pediatr 1984 105 912 915 6502341
Morecki R Glaser JH Cho S Balistreri WF Horwitz MS Biliary atresia and reovirus type 3 infection N Engl J Med 1982 307 481 484 6285193
Petersen C Grasshoff S Luciano L Diverse morphology of biliary atresia in an animal model J Hepatol 1998 28 603 607 9566828 10.1016/S0168-8278(98)80283-3
Riepenhoff-Talty M Gouvea V Evans MJ Syensson L Hoffenberg E Sokol RJ Uhnoo I Greenberg SJ Schael K Zhaori G Fitzgerald J Chong S El-Yousef M Nemeth A Brown M Piccoli D Hyams J Ruffin D Rossi T Detection of group C Rotavirus in infants with extrahepatic biliary atresia J Infect Dis 1996 174 8 15 8656017
Riepenhoff-Talty M Schaekel K Clark F Mueller W Uhnoo I Rossi T Fisher J Ogra PL Group A Rotaviruses produce extrahepatic biliary obstruction in orally inoculated newborn mice Pediatr Res 1993 33 394 399 8386833
Chang M-H Huang H-H Huang E-S Kao C-L Hsu H-Y Lee C-Y Polymerase Chain Reaction to detect human cytomegalovirus in livers of infants with neonatal hepatitis Gastroenterology 1992 103 1022 1025 1323495
Sokol RJ Endo N Ohara J Lilly J Levin M Silverman A Brown WR Reovirus 3 RNA not detected by polymerase chain reaction in tissues from infants with extrahepatic biliary atresia [Abstract] Hepatology 1991 14 125A
Domiati-Saad R Dawson DB Margraf LR Finegold MJ Weinberg AG Rogers BB Cytomegalovirus and human herpesvirus 6, but not human papillomavirus, are present in neonatal giant cell hepatitis and extrahepatic biliary atresia Pediatr Dev Pathol 2000 3 367 373 10890252 10.1007/s100240010045
Steele MI Marshall CM Lloyd RE Randolph VE Reovirus 3 not detected by reverse transcriptase-mediated polymerase chain reaction analysis of preserved tissue from infants with cholestatic liver disease Hepatology 1995 21 697 702 7533124 10.1016/0270-9139(95)90520-0
FischlerR B Ehrnst A Forsgren M Orvell C Nemeth A The viral association of neonatal cholestasis in Sweden: a possible link between cytomegalovirus infection and extrahepatic biliary atresia J Pediatr Gastroenterol Nutr 1998 27 57 64 9669727 10.1097/00005176-199807000-00010
Jevon GP Dimmick JE Biliary atresia and cytomegalovirus infection: a DNA study Pediatr Dev Pathol 1999 2 11 14 9841700 10.1007/s100249900083
Tyler KL Sokol RJ Oberhaus SM Le M Karrer FM Narkewicz MR Tyson RW Murphy JR Low R Brown WR Detection of reovirus RNA in hepatobiliary tissues from patients with extrahepatic biliary atresia and choledochal cysts Hepatology 1998 27 1475 1482 9620316 10.1002/hep.510270603
Tarr PI Haas JE Christie DL Biliary atresia, cytomegalovirus and age at referral Pediatrics 1996 97 828 831 8657522
Lurie M Elmalach I Schuger L Weintraub Liver findings in infantile cytomegalovirus infection: similarity to extrahepatic biliary obstruction Histopathol 1987 11 1171 1180
Boppana SB Pass RF Britt WJ Stagno S Alford CA Symptomatic congenital cytomegalovirus infection: neonatal morbidity and mortality Pediatr Inf Dis 1992 11 93 99
Stagno S Pass RF Dworsky ME Alford CA Congenital and perinatal cytomegalovirus infections Sem Perinatol 1983 7 31 42
Alix D Castel Y Gouedard H Hepatic calcification in congenital cytomegalic inclusion disease J Pediatr 1978 92 856 205645
Ansari BM Davis DB Jones MR Calcification in liver associated with congenital cytomegalic inclusion disease J Pediatr 1977 90 661 662 190372
Stagno S Cytomegalovirus Infectious diseases of the fetus and newborn infant 1995 4 W.B. Saunders Company. Philadelphia 312 353
Rogers BB Alpert LC Hine EAS Buffone GF Analysis of DNA in fresh and fixed tissue by the polymerase chain reaction Am J Pathol 1990 136 2156429
Latchman DS The basics of molecular biology PCR: Applications in Pathology Principles and practice 1995 Oxford 6 53
Kwok S Higuchi R Avoiding false positives with PCR Nature 1989 339 237 238 2716852 10.1038/339237a0
Saiki RK Scharf S Faloona F Mullis KB Horn GT Erlich HÁ Arnheim N Enzymatic amplification of β-globin genomic sequences and restriction site analysis for diagnosis of sickle cell anemia Science 1985 230 1350 1354 2999980
Shibata D Martin J Appleman MD Causey DM Leedom JM Arnheim N Detection of cytomegalovirus DNA in peripheral blood of patients infected with human immunodeficiency virus J Infect Dis 1988 158 1185 1192 2848898
Demmler GJ Buffone GJ Schimbor CM May RA Detection of cytomegalovirus in urine from newborns by using polymerase chain reaction DNA amplification J Infect Dis 1988 158 1177 1184 2848897
Pannuti CS Vilas-Boas LS Angelo MJO Carvalho RPS Segre CM Congenital cytomegalovirus infection. Occurrence in two socioeconomically distinct populations of a developing country Rev Inst Med Trop São Paulo 1985 27 105 107
Yamamoto AY Figueiredo LTM Mussi-Pinhata MM Prevalence and clinical aspects of congenital cytomegalovirus infection J Pediatr (Rio J) 1999 75 23 28 14685559
Machado CM Fink MCDS Vilas-Boas LS Sumita LM Weinberg A Shiguematsu K Souza IC Casanova LD Pannuti CS Infecção perinatal pelo cytomegalovirus em hospital público do município de São Paulo: estudo prospectivo Rev Inst Med Trop São Paulo 1991 33 159 166
Santos PH Iazzetti AV Citomegalia congênita e perinatal Pediatria Moderna 2000 36 805 826
Lewis DB Wilson CB Developmental immunology and role of host defenses in neonatal susceptibility to infection Infectious diseases of the fetus and newborn infant 1995 4 W.B. Saunders Company. Philadelphia 20 98
Petersen C Biermanns D Kuske M Schäel K Meyer-Junghäel L Mildenberg H New aspects in a murine model for extrahepatic biliary atresia J Pediatr Surg 1997 32 1190 1195 9269968 10.1016/S0022-3468(97)90680-1
Kobayashi H Puri P O'briain DS Surana R Miyano T Hepatic overexpression of MHC class II antigens and macrophage-associated antigens (CD68) in patients with biliary atresia of poor prognosis J Ped Surg 1997 32 590 593 10.1016/S0022-3468(97)90714-4
Broomé U Nemeth A Hultcrantz R Scheynius A Different expression of HLA-DR and ICAM-1 in livers from patients with biliary atresia and Byler's disease J Hepatol 1997 26 857 862 9126800 10.1016/S0168-8278(97)80253-X
Davenport M Gonde C Redkar R Koukoulis G Tredger M Mieli-Vergani G Portmann B Howard ER Immunohistochemistry of the liver and biliary tree in extrahepatic biliary atresia J Pediatr Surg 2001 36 1017 1025 11431768 10.1053/jpsu.2001.24730
|
16321152
|
PMC1315325
|
CC BY
|
2021-01-04 16:28:14
|
no
|
BMC Infect Dis. 2005 Dec 1; 5:108
|
utf-8
|
BMC Infect Dis
| 2,005 |
10.1186/1471-2334-5-108
|
oa_comm
|
==== Front
Behav Brain FunctBehavioral and brain functions : BBF1744-9081BioMed Central London 1744-9081-1-201628750410.1186/1744-9081-1-20ResearchEEG correlates of verbal and nonverbal working memory Hwang Grace [email protected] Joshua [email protected] Aaron [email protected] Jared [email protected] Robert [email protected] Michael J [email protected] Volen Center for Complex Systems, Brandeis University, Waltham, MA, 02454, USA2 Department of Psychology, University of Pennsylvania, Philadelphia, PA, 19104, USA2005 15 11 2005 1 20 20 1 7 2005 15 11 2005 Copyright © 2005 Hwang et al; licensee BioMed Central Ltd.2005Hwang 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
Distinct cognitive processes support verbal and nonverbal working memory, with verbal memory depending specifically on the subvocal rehearsal of items.
Methods
We recorded scalp EEG while subjects performed a Sternberg task. In each trial, subjects judged whether a probe item was one of the three items in a study list. Lists were composed of stimuli from one of five pools whose items either were verbally rehearsable (letters, words, pictures of common objects) or resistant to verbal rehearsal (sinusoidal grating patterns, single dot locations).
Results
We found oscillatory correlates unique to verbal stimuli in the θ (4–8 Hz), α (9–12 Hz), β (14–28 Hz), and γ (30–50 Hz) frequency bands. Verbal stimuli generally elicited greater power than did nonverbal stimuli. Enhanced verbal power was found bilaterally in the θ band, over frontal and occipital areas in the α and β bands, and centrally in the γ band. When we looked specifically for cases where oscillatory power in the time interval between item presentations was greater than oscillatory power during item presentation, we found enhanced β activity in the frontal and occipital regions.
Conclusion
These results implicate stimulus-induced oscillatory activity in verbal working memory and β activity in the process of subvocal rehearsal.
==== Body
Background
Evidence from studies of working memory has long indicated that the rate at which items are encoded into memory and subsequently recognized varies with stimulus type [1-3], that different processes are involved in verbal and nonverbal working memory [4,5], and that verbal memory depends specifically on the subvocal rehearsal of items [6,7]. Evidence from magnetoencephalography (MEG), scalp EEG, and intracranial EEG studies has implicated oscillations in both verbal and nonverbal working memory [8-27]. In some studies, oscillations were sustained from the time an item was presented until it was later tested [12,17]. In other studies, oscillations were linked to the retention interval, with oscillatory amplitude or coherence increasing with memory load [10,11]. Across these studies, oscillations associated with working memory were reported in the 4–8 Hz θ band [10,14,15,17,19,20,25], the 9–12 Hz α band [9,11,16,18], the 14–28 Hz β band [20,21,23] and the 30–60 Hz γ band [12,19,23,24,26,27]. Building on the abundance of research linking oscillations and working memory, we attempted to identify oscillatory correlates of verbal memory, and particularly the correlates of subvocal rehearsal. We addressed this aim by examining the oscillatory correlates of working memory for a range of stimuli that were designed to vary widely in the extent to which they could be verbally rehearsed.
EEGs were recorded from subjects while they performed a variant of the Sternberg task [28], a prototypical and widely used test of working memory. Subjects were shown a set of 3 study items to hold in memory over a short retention interval. This retention interval was followed by a probe item. Subjects were asked to indicate, as quickly and accurately as possible, whether the probe item was a target (an item present in the study list) or a lure (an item absent from the study list). Study lists were composed of stimuli drawn from each of five pools. As illustrated in Figure 1, these stimuli were letters, words, namable objects, single dot (i.e., spatial) positions, and sinusoidal grating patterns. Whereas subjects could easily rehearse lists of letters, words, and namable objects, spatial positions and sinusoidal grating patterns were difficult for subjects to name and rehearse. Hereafter, we refer to the letter, word, and object stimuli collectively as the verbal stimulus type, and we refer to the spatial and grating stimuli collectively as the nonverbal stimulus type. We hypothesized that studying differential neuroelectrical responses across these stimulus types and across successive phases of the viewing cycle would isolate distinct neural processes that correlate with subvocal rehearsal.
Figure 1 Sternberg Paradigm. Schematic illustrating one trial of each stimulus pool in the Sternberg task: letter, word, object, spatial, grating. Also shown are stimulus class definitions. The time interval from the moment the subject depressed the advance key until the onset of the fixation stimulus was 400 ms; we therefore used this 400-ms interval as our baseline stimulus class (not labeled). The fixation stimulus class (labeled Fixation) covers the time interval starting at the presentation of an asterisk and ending just before the presentation of the first study item. The encoding stimulus class (labeled Encoding) is the time interval covering the presentation of the 3 study items (labeled 1st, 2nd, 3rd) including the 500-ms interval following the disappearance of the third study item (labeled Retention). The time interval covering the start of the presentation of the first item and ending with the presentation of the third item is referred to as the study interval. The encoding stimulus class therefore contains two intervals: the study interval followed by the retention interval. The probe stimulus class (labeled Probe) is the time interval starting with the onset of the probe stimulus and ending with the subject's response. The response stimulus class started with the onset of subject response and ended 1 sec later, well before the start of the next trial (not labeled).
Our central question was whether differences in oscillatory power reflect the verbal rehearsal process. Although this question was difficult to answer, because there was no way to determine precisely when subjects were rehearsing, we could make certain inferences about subvocal rehearsal from our understanding of the stimuli themselves. First, we had reason to expect that subvocal rehearsal would be greater for verbal, namable items than for nonverbal items, such as dot locations or sinusoidal gratings. This enabled us to predict that there would be a systematic difference in overall oscillatory power between verbal and nonverbal stimulus types (Criterion 1). Second, we expected that rehearsal of verbal stimuli would be more likely to occur during interstimulus intervals (ISI) – that is, between item presentations or between the last item and the test probe – than during stimulus-presentation intervals (SPI) when stimuli are in view. Consistent with previous observations that showed increases in power to be correlated temporally with encoding [12,15,17] and retention [17,23], we predicted that power due to verbal rehearsal during ISI would be greater than power during SPI (Criterion 2). Combining these two criteria, we investigated oscillatory power across stimulus types and across successive phases of the viewing cycle (i.e., ISI vs. SPI).
Results
Behavioral Data
As shown in Figure 2, mean accuracy was highest for the verbal stimuli (i.e., letter, word, and namable objects) and lowest for the sinusoidal gratings. As expected based on previous studies [29,30], accuracy for the verbal and grating stimuli was highest for targets that replicated the most recently presented list item. Consistent with these trends, a 5 (stimulus pool) × 4 (probe position) ANOVA on accuracy revealed statistically significant main effects of stimulus pool (F(4, 44) = 99, MSe = 0.0021, p < 0.001) and probe position (F(3,33) = 9.4, MSe = 0.0048, p < 0.001), as well as a significant interaction between these factors (F(12,132) = 4.9, MSe = 0.0027, p < 0.001). Reaction times (RTs) across stimuli also varied, with verbal stimuli eliciting the shortest RTs and nonverbal stimuli eliciting the longest RTs. Since we did not vary the length of study lists, accuracy and RTs were not analyzed as a function of list length.
Figure 2 Performance Accuracy. Performance accuracy is plotted for each of the five stimulus pools as a function of probe position across all subjects (n = 12). Probes that did not appear in the study list are labeled Lure, while probes that were part of the study list are labeled by their position (i.e., Pos1 refers to probes that appeared first in a study list).
Electrophysiological Data
Figure 3 shows the time course of mean Z-transformed oscillatory power at electrode Pz across fixation, encoding, retention, and probe intervals. From left to right, the upper panels show time courses at each of the four frequency bands. Within each panel of Figure 3 (top row), time courses of oscillatory power are seen to be quite similar across the three verbal stimuli (letter, word, object denoted by red symbols) but dissimilar from both the spatial (green) and grating (blue) stimuli. Because we did not observe any systematic differences among the three verbal stimuli, and because analyses of other electrodes exhibited qualitatively similar time courses to those shown in Figure 3 (upper panel), we combined data across verbal stimuli for all subsequent analyses. We treated the two nonverbal stimuli separately because they exhibited systematic differences in several of our analyses, as described in greater detail below. The middle and lower panels of Figure 3 compare the time courses of oscillatory power for verbal-vs.-spatial stimuli and verbal-vs.-grating stimuli, respectively. Significant differences for each of these comparisons are denoted by a thick horizontal bar on the bottom of each panel (t(11) = 2.2, p < 0.05).
Figure 3 Power Time Curves. Z-transformed wavelet power in the θ, α, β, and γ frequency bands are illustrated in the parietal brain region (from electrode Pz) for 12 subjects. The fixation asterisk appeared at t = 0 sec, as indicated by the vertical dashed line. The 3 study items appeared at average times of 1.2 sec, 2.2 sec, and 3.2 sec as indicated by vertical dotted lines. The retention interval started at average time t = 3.9 sec. The probe item appeared at average time t = 4.4 sec (vertical dotted line), and the response was made on average at t = 5 sec (vertical dashed line). The first row illustrates power for each of the five stimulus pools. The second and third rows illustrate power in the verbal-vs-grating and in the verbal-vs-spatial tasks, respectively. Verbal refers to the mean power from the letter, word, and object tasks. Significant differences between verbal and each of the nonverbal tasks (t(11) = 2.2, p < 0.05) are denoted by a thick horizontal bar at the bottom of each panel in the second and third rows.
During the encoding interval, the amplitude of power in θ, α, and β exhibited responses that were induced by the onset of the stimulus (see Figure 3, columns 1–3). θ showed stimulus-induced enhancement: that is, θ power increased and decreased sharply in each SPI (Figure 3, column 1). α and β showed stimulus-induced reduction: that is, α and β power decreased during the SPI and increased again during the ISI (Figure 3, columns 2 and 3). During the probe interval, α and β power declined steadily to a minimum and immediately returned to above-baseline levels following the responses. γ power showed a slight systematic decline that was insensitive to the timing of study-item presentations, with a rapid return to baseline levels immediately following the responses (Figure 3, column 4). Power in all bands was generally lower for grating stimuli than for spatial stimuli, both of which were exceeded by verbal stimuli, with the largest difference in power seen in θ and β (Figure 3, columns 1 and 3). This consistent pattern of higher oscillatory power elicited by verbal stimuli indicates that oscillatory activity increases during the processing of verbally rehearsable stimuli. In addition, the overall decline in α, β, and γ power over the course of the trial for all stimulus types, suggests a generalized desynchronization of oscillations with increased memory load.
Criterion 1. Verbal versus Nonverbal Stimulus Types
To determine whether power elicited by verbal stimuli differed significantly from power elicited by spatial and grating stimuli, we conducted a topographic analysis, as shown in Figures 4a and 4b. We refer to instances where verbal tasks elicited significantly greater power than did nonverbal tasks as enhanced verbal power (EVP; randomization test, p < 0.001, df = 11). Similarly, instances where nonverbal tasks elicited significantly greater power than did verbal tasks are referred to as diminished verbal power (DVP; randomization test, p < 0.001, df = 11). Figures 4a and 4b illustrate the topographic distribution of EVP and DVP for the comparison of verbal-vs.-grating (VG) stimuli and the comparison of verbal-vs.-spatial (VS) stimuli, respectively. Within each panel, EVP and DVP are illustrated separately for the fixation (top row), study (middle row), and retention intervals (bottom row).
Figure 4 Topographic Analysis of Stimulus Types. Brain regions that show enhanced verbal power and enhanced nonverbal power are illustrated in (a) Verbal-vs.-Grating, and (b) Verbal-vs.-Spatial during the fixation, study, and retention intervals. Red means the verbal stimuli elicited significantly more mean power than did nonverbal stimuli during each of the respective intervals. Blue means nonverbal stimuli elicited significantly more mean power than did verbal stimuli during each of the respective intervals. Significance was determined by a randomization test on matched pairs, with a Type I error rate of 0.001, which corresponds to 0.24 electrodes that could have shown significance by chance across all subjects (randomization test, p < 0.001, df = 11). Multiple comparisons for the number of electrodes and frequency bands were accounted for with a resampling method on matched pairs (see section on Randomization Test Procedure). Topographic visualization was done with the topoplot function in the EEGLAB Matlab toolbox [56].
In the fixation interval, EVP was sparsely distributed in all frequency bands in the VG comparison (Figure 4a, top row); EVP was observed centrally in θ, and in frontal and parietal areas in α and β in the VS comparison (Figure 4b, top row). Sparsely distributed DVP was observed in γ in only the VS comparison (Figure 4b, top row, column 4).
In the study interval, EVP occurred bilaterally in θ, α, and β, and centrally in γ in the VG comparison (Figure 4a, middle row). For the VS comparison, we found EVP in θ distributed in frontal and occipital areas, EVP in α localized to one parieto-occipital site, EVP in β distributed in frontal and parietal areas (Figure 4b, middle row), and DVP in γ localized to one left frontal site (Figure 4b, middle row, column 4).
During the retention interval, the VG comparison revealed that EVP in θ was distributed bilaterally in the parieto-occipital regions (Figure 4a, bottom row, column 1), EVP in α occurred sparsely in bilateral regions (Figure 4a, bottom row, column 2), EVP in β occurred bilaterally along the midline (Figure 4a, bottom row, column 3), and EVP in γ occurred centrally (Figure 4a, bottom row, column 4). The VS comparison showed that EVP in θ was distributed bilaterally in the centro-parietal regions (Figure 4b, bottom row, column 1), EVP in α was localized to one parieto-occipital site (POz; Figure 4b, bottom row, column 2), and EVP in β occurred in the frontal and parieto-occipital areas (Figure 4b, bottom row, column 3). We did not observe EVP in the γ band.
To understand the relation between topographic significance during the task interval and consecutive time bins at which EVP was present, we further explored the temporal characteristics of EVP. Hence we present time-frequency plots from frontal and parietal regions for both the VG and VS comparisons, first from a right centro-frontal site (FC2) and then from a midline parietal site (Pz). For the VG comparison, Figure 5a shows a temporally persistent EVP in β at electrode FC2 during the retention interval (which started at approximately 3.9 sec and ended at approximately 4.4 sec). Similarly, Figure 5b shows a continuously persistent EVP in θ, β, and 32-Hz γ at electrode Pz during the retention interval. Both Figures 5a and 5b fail to show EVP in α during the retention interval, which is consistent with the topographic analysis shown in Figure 4a (bottom row, column 2). For the VS comparison, Figure 5c shows temporally persistent EVP in β at electrode FC2 during the retention interval. Similarly, Figure 5d shows EVP in β at electrode Pz during the retention interval. Both Figures 5c and 5d fail to show EVP in α and γ during the retention interval, which is consistent with the topographic results shown in Figure 4b (bottom row, columns 2 and 4).
Figure 5 Time-Frequency Spectrogram. Time-frequency representations of p values in the verbal-vs.-grating comparison are shown in (a) from a frontal site (electrode FC2), and in (b) from a parietal site (electrode Pz). Time-frequency representations of p values in the verbal-vs.-spatial comparison are shown in (c) from a frontal site (electrode FC2), and in (d) from a parietal site (electrode Pz). Frequency bands of interest in the retention interval are boxed in black. Red means the verbal stimuli elicited significantly more power than did nonverbal stimuli. Blue means nonverbal stimuli elicited significantly more power than did verbal stimuli. Significance was determined by a randomization test on matched pairs, with a Type I error rate of 0.01, which corresponds to 53 time bins that could have shown significance by chance across all subjects (randomization test, p < 0.01, df = 11). Multiple comparisons for the number of electrodes and frequency bands were accounted for with a resampling method on matched pairs (see section on Randomization Test Procedure).
Criterion 2. Comparison of Viewing-Cycle Phases (ISI vs. SPI)
We used a randomization procedure to determine whether ISI power was greater than SPI power in the verbal, spatial, and grating tasks (see Figures 6a–c). As shown in Figure 6a (column 3), only the β band from verbal tasks elicited significantly more ISI power than SPI power over the frontal and occipital areas (randomization test, p < 0.001, df = 11). In contrast, nonverbal tasks did not elicit more ISI power than SPI power at any electrode or frequency band (Figures 6b and 6c). Figures 6a–c (column 1) also show that power during SPI was greater than power during ISI in θ for both verbal and nonverbal stimulus types (randomization test, p < 0.001, df = 11).
Figure 6 Topographic Analysis of Phases of the Viewing Cycle: stimulus presentation interval (SPI) versus (ISI) interstimulus presentation interval. Brain regions that differ significantly in power levels between SPI and ISI are illustrated. (a) Verbal (mean of letter, word, and object), (b) Spatial, and (c) Grating. Red means that power during ISI was significantly greater than power during SPI; blue means that power during SPI was significantly greater than power during ISI. Significance was determined by a randomization test on matched pairs, with a Type I error rate of 0.001, which corresponds to 0.24 electrodes that could have shown significance by chance across all subjects (randomization test, p < 0.001, df = 11). Multiple comparisons for the number of electrodes and frequency bands were accounted for with a resampling method on matched pairs (see section on Randomization Test Procedure).
Discussion
Summary of Criterion 1. Verbal versus Nonverbal Stimulus Types
Overall, we observed EVP across the fixation, study and retention intervals (see Figure 4). Because we employed a blocked paradigm, subjects were aware of which stimulus pool to expect in successive trials. Therefore, EVP during the fixation interval may reflect preparatory processes that are unique to (or relatively more prominent in) the verbal tasks. EVP during the study and retention intervals may reflect any aspect of information processing that differentiates verbal and non-verbal stimuli. One candidate process is subvocal rehearsal, which subjects actively use for verbal materials but which is almost impossible with the nonverbal stimuli used in our study. It is also possible, however, that EVP reflects the greater ease of encoding and retaining verbal items. We will return to this issue later in the discussion.
The differing EVP topographies between the VG and VS comparisons during the study and retention intervals indicate that grating stimuli attenuated α oscillations more than spatial stimuli did (see Figures 4a and 4b, second and third rows, column 2). We offer two possibilities that could give rise to these different α EVP topographies. First, we suggest that α oscillations could play differential roles in the spatial and grating tasks. α oscillations could be enhanced by the strategies that subjects used to spatially encode items that appeared close to their peripheral field-of-view while they simultaneously attempted to suppress eye movements. An alternative explanation can be cast in terms of attentional demands: in ratings along a continuum of task difficulty obtained from our strategy questionnaire, the verbal tasks rated much easier than the spatial task, and the spatial task was in turn easier than the grating task. If we take the common view that α oscillations index task-related attentional demands [13], and if we posit that subjects allocated the most attention to the hardest tasks, then the EVP observed in the VG comparison could directly reflect differences in attentional demand between the verbal and grating tasks. Similarly, the lack of EVP observed in the VS comparison could suggest that the difference in attentional demand, as reflected by the amplitude of α, did not reach significance.
Summary of Criterion 2. Comparison of Viewing-Cycle Phases
All stimulus types elicited significantly more SPI θ power than ISI θ power (see Figures 6a–c, column 1); this finding was consistent with the patterns of stimulus-induced θ increase observed in Figure 3 (column 1). These observations suggest that θ may be involved with perceptual and cognitive processes common to both verbal and nonverbal tasks while showing more power in response to the verbal tasks. In contrast, verbal tasks elicited significantly more ISI power than SPI power exclusively in the β frequency band (Figure 6a, column 3), thus showing that β frequency activity dissociates working memory for verbal and noverbal stimuli. Subvocal rehearsal is one candidate process that is responsible for this dissociation. Alternatively, differences in β may be a consequence of the greater difficulty associated with encoding and maintaining the nonverbal items.
Roles of β Oscillations in Working Memory
Recalling that EVP in β was consistently observed during the retention interval in the VS and VG comparisons along the midline frontal and parietal areas (Figures 4a and 4b), and that ISI power was greater than SPI power in β (Figure 6, first row, column 3), we reason that these data provide functional evidence that β oscillations support an important process in subvocal rehearsal. Although β oscillations have often been associated with muscle movement artifacts [31], β effects observed in our dataset were not localized along the temporal regions (on the scalp) known to be involved in finger movements. Furthermore, prior research associated sustained β oscillations with the retention of stimuli [23,24,32]. In a delayed-match-to-sample study that used stimuli composed of abstract shapes, β oscillations increased and remained elevated during the retention interval [23] at a frontal (Fz) and an occipital site (POz) along the midline. However, these authors did not assess the effect of stimulus-rehearsability on β activity. We therefore can only speculate, extrapolating from observations in our study, that β oscillations induced by verbal stimuli, if verbal stimuli had been tested, might also have been sustained during the retention interval, as were those elicited by abstract shapes. In our study, verbal stimuli elicited larger and more sustained β oscillations than did nonverbal stimuli during the retention interval (as depicted in Figure 3, column 3).
Roles of Oscillations in Verbal and Nonverbal Memory
To identify oscillatory and topographical correlates of rehearsal, a number of studies have specifically compared verbal and nonverbal stimulus types. In a stimulus-reproduction task, oscillations in the θ and γ bands were shown to exhibit levels of increased synchrony between the posterior association cortex and the prefrontal cortex during the retention interval [19]. The degree of synchrony did not distinguish abstract stimuli from verbal stimuli, however. In a Sternberg task, increased power in θ and γ during the retention interval was shown to be synchronous in the left frontal area between electrode pairs Fz and Fpl for verbal stimuli but not for irregular rectangular stimuli [20]. This finding was consistent with the idea that the left prefrontal cortex is involved in verbal processing. An n-back task, which required subjects to identify both letters (verbal condition) and their spatial position (visual condition), revealed that the visual condition attenuated the amplitude of upper α oscillations in the right hemisphere, while the verbal condition did not [9]. This finding suggests that working memory for verbal information is less dependent on the right posterior cortex than working memory required for visual information, a distinction that is consistent with the idea that visual processing occurs in the right hemisphere. Most of these studies [9,20] have successfully used attributes of oscillations (i.e., levels of synchrony, fluctuation in amplitude) to link verbal processes with the left hemisphere and visual processes with the right hemisphere. We did not find a hemispheric dissociation between verbal and nonverbal memory; instead, we found oscillatory effects of stimulus types distributed bilaterally in both θ and β frequency bands.
Neuroimaging Evidence in Verbal Working Memory
Evidence from PET and fMRI studies, coupled with support from lesion studies [33-35], strongly suggests that distinct neural substrates subserve each of the two subsidiary systems of working memory: the phonological loop (verbal memory) and the visuospatial sketchpad (visual memory) [4,5]. In these studies, mental processes that require subvocal rehearsal preferentially activated the left prefrontal cortex, the bilateral occipital cortex [36], Broca's region, the premotor cortex, the supplemental motor areas, the left posterior parietal cortex [37], and the cerebellum [38,39]. In contrast, processes requiring the use of visual memory preferentially activated the right dorsal prefrontal cortex, the right parietal cortex, and the right middle frontal gyrus [36,40].
We observe caution in making claims about a possible correlation between our results based on scalp EEG signals and any results obtained by neuroimaging. In this study, we found that verbal stimuli generally elicited greater oscillatory power than did nonverbal stimuli during the study, retention and retrieval phases of the Sternberg task. When we further compared oscillatory power in the stimulus-presentation and interstimulus intervals, we found that β power was significantly higher during the stimulus-presentation interval, but only for the verbal tasks. This pattern, which was not found at other frequencies, indicates that β covaries either with rehearsal or some other aspect of information processing that differs between verbal and nonverbal stimuli. Wherease this β effect was seen bilaterally, several neuroimaging studies have linked the left prefrontal area to the process of subvocal rehearsal [5,41]. It should be noted, however, that numerous studies employing verbal tasks have shown activations bilaterally in both hemispheres [39,41-43] during the task-maintenance interval (akin to our retention interval). One study attributed frontal right-hemispheric activations to the process of updating the central executive [44,45] with verbal information [46]. Because the scalp EEG technique lacks fine spatial resolution, it was not possible in our experiment to disentangle subvocal rehearsal from other aspects of executive function. At the present time, the relation between scalp-recorded oscillations and hemodynamic responses measured using fMRI remains largely unknown, however recent evidence from intracranial recordings suggests a link between locally-generated gamma oscillations and increased hemodynamic activity [47]. Future studies that combine scalp EEG and fMRI recordings during working memory tasks may help to resolve these open issues.
Conclusion
We investigated oscillatory power across stimulus types and across phases of the viewing cycle, considering only those brain regions satisfying Criterion 1 (verbal power differs from nonverbal power) and Criterion 2 (ISI power is greater than SPI power) to be correlated with subvocal rehearsal. Our results indicate that stimulus-induced oscillatory activity is involved in verbal working memory, with verbal stimuli eliciting more power than nonverbal stimuli in θ. The mean amplitude of θ power during ISI was not greater than the mean θ power during SPI, thus suggesting that θ oscillations are involved with perceptual and memory encoding processing common to both verbal and nonverbal tasks. In contrast, β oscillations simultaneously satisfied Criteria 1 and 2 along the midline at frontal and parietal brain regions, thereby distinguishing verbal stimuli from nonverbal stimuli across stimulus types and across phases of the viewing cycle. These results thus implicate β oscillations in the subvocal rehearsal process of verbalizable items.
Methods
Subjects
Subjects were 12 right-handed volunteers ranging in age from 19 to 29. Eight were male and four were female. All subjects had normal or corrected-to-normal vision. All subjects gave informed consent to a protocol reviewed and approved by the Brandeis University Committee for the Protection of Human Subjects. Subjects were given a base payment plus bonus payments proportional to their performance. Subjects participated in a total of five sessions, each of which was conducted on a different day.
Procedure
Five pools of different stimuli were employed: single letters, pictures of objects from the Snodgrass [48] picture set, one-syllable words corresponding to the object pool, dots that appeared at different (spatial) positions on the monitor, and sinusoidal patterns (grating). Each stimulus-type pool contained 16 stimuli. We considered a stimulus verbal if it was amenable to subvocal rehearsal, and nonverbal if it resisted subvocal rehearsal. Hence, letters, words, and objects were all considered verbal because of the ready availability of tags for rehearsal (e.g., the word car for the corresponding object), whereas spatial and grating stimuli were considered nonverbal. A sample of each stimulus pool is shown in Figure 1.
Each subject was tested with stimuli from each of the five different stimulus pools (letter, word, object, spatial, grating). Stimuli were shown on a computer monitor positioned 57 cm away from the subject. The approximate visual angle for the letter, word, object, and grating stimuli was 5°; for the spatial stimuli, it was 10°. Every trial of the experiment was self-paced; subjects initiated a trial by pressing an advance key; this was followed by a 400-ms interval before the onset of a fixation cue (an asterisk). Each trial started with a fixation cue centered in the middle of the computer monitor for a duration of 1 sec (± 200 ms jitter), followed by a study set of 3 stimuli. Each stimulus was shown on the monitor for 700 ms, followed by a 275 ms ± 75 ms ISI; the ISI was randomly jittered serving to decorrelate physiological responses which may occur due to successive stimuli presentations. After the offset of the third stimulus, a short retention interval (500 ms ± 75 ms jitter) was followed by a fourth item (probe) for 750 ms. Subjects were preinstructed to determine as quickly and accurately as possible after probe onset whether the probe item was part of the preceding study list. In order to obtain the fastest reaction times possible, they were also preinstructed to respond to a target by pressing the right control key (dominant hand); they were to respond to a lure by pressing the left control key (nondominant hand). A blink break followed each response, and subjects were prompted to continue at their own pace by pressing the down-arrow key. To prevent potential interference between a completed trial and the start of the next trial, a minimum of 1.5 sec was preprogrammed to pass before the next trial could begin. Feedback on accuracy and response time was given at the end of each block of trials.
Stimulus Description
The letter pool contained the following letters: b, c, d, f, g, h, j, k, 1, m, n, p, q, r, t, and v. The object pool contained pictures of the following nouns: ball, bat, bed, bell, cake, car, chair, dog, ear, fly, fork, hat, heart, key, kite, and shoe. Each word in the word pool corresponded exactly to each object in the object pool. The spatial pool contained presentations of a solid white circle 1 cm in diameter (dot) at 16 nonoverlapping locations along the circumference of an invisible circle (10 cm in diameter) that was centered on the computer monitor. This configuration ensured that all dots fell close to the subjects' peripheral vision. Because there were 16 fixed dot positions, these positions were not easily encoded as clock-face positions (e.g., three o'clock). The grating pool contained two-dimensional textures, similar to those used in prior studies [49,50]. Each stimulus was a superposition of one horizontal and one vertical sinusoidal luminance grating, generating a luminance profile described by
where Lavg represents mean luminance (L); f and g represent the spatial frequency of the vertical and horizontal components, respectively; and A is defined by
Parameters used to generate the 16 grating stimuli were A = 0.25, . The luminance of the monitor was linearized by means of calibration routines from Brainard & Pelli's Psychtoolbox [51].
Stimulus-presentation Constraints
There were 10 blocks per session, 2 blocks for each of the 5 stimulus pool, for a total of 300 trials per session. Across all 5 sessions there were thus 1500 trials, 300 trials per stimulus pool. Each block comprised 15 targets and 15 lures. The target trials, in turn, probed each study position with equal probability. Study items in the current trial could not be from study items in the preceding two trials. Similarly, probe items in the current trial could not be from the preceding two trials. Further, a study item was allowed to be a lure only once in each block. Finally, the sequence of target trials and lure trials was randomized.
EEG Recording
During EEG recording, subjects were instructed to remain silent and to minimize all body and eye movements (particularly blinking). Because the study was self-paced, subjects were also encouraged to take as many breaks between trials as they needed to maintain their concentration and optimize their performance. During the entire study, an experimenter quietly monitored the session from the back of the testing room. Lighting in the room was maintained at a constant level at all times. Recordings were obtained from 60 tin electrodes located in standard electrode positions embedded in an elastic cap (ElectroCap). EEG signals were amplified 10,000 times (Sensorium EPA6, 1 GΩ input impedance) with band limits between 0.03 and 50 Hz (12 dB/octave). Analog-to-digital signal conversion was implemented with a 12-bit data acquisition card (National Instrument PCI-6071E) with ± 5 V dynamic range. The overall system resolution was therefore 0.24 μV/bit. Raw data was digitized at 256 Hz, well above the Nyquist minimum data sampling limit for our frequency region of interest (i.e., 2–50 Hz). Amplified signals were then digitally notch-filtered between 59 and 61 Hz to minimize 60-Hz line noise.
Electrode impedances were brought to < 50 kΩ, and interelectrode impedances were within 20 kΩ; skin impedances (ground and reference) were kept below 10 kΩ. Any electrodes that exhibited poor electrical characteristics were disconnected. All EEG signals were recorded referentially using the right mastoid (or right ear lobe). EEG signals were digitally re-referenced to the average EEG signal recorded from all electrically sound electrodes. Only signals recorded from low-impedance (< 50 kΩ) and electrically sound electrodes were included in the re-referencing. On any given session, no more than 5% of all electrodes had poor electrical contact and/or high impedances.
Six tin disc-electrodes were used to monitor electro-oculogram (EOG) activity. Vertical eye movements were isolated with electrodes positioned above and below each eye. Horizontal eye movements were isolated with electrodes placed at the lateral canthus of each eye. Each pair of EOGs was recorded bipolarly. Raw signals from EOG were used to detect blinking and automatic eye movements following Net Station's™ weighted running-average algorithm [52]. If any one pair of EOG exceeded the combined rejection threshold of |100 μvolts|, the event (e.g., first study item presentation) that corresponded to the EOG spike was excluded from analysis. Fewer than 6% of all trials were excluded.
Subject Questionnaire
At the conclusion of the final study session, subjects were asked to complete a strategy questionnaire. In section 1, subjects reported the strategies they used across study sessions as well as the relative effectiveness of those strategies. In section 2, subjects rated, on a 5-point scale, how often they used visual imagery and verbal labels to complete each of the five tasks (1 = never, 5 = always). Finally, in section 3, subjects rated the difficulty of each task relative to the other tasks on a continuum from 1 to 5 (1 = easiest, 5 = hardest).
Behavioral Data Analysis
Accuracy and reaction times (RTs) were recorded for each trial. Trials with RTs shorter than 200 ms and greater than 1300 ms were excluded from analysis; this amounted to 2.1% of all trials.
Oscillatory Power Analysis
Analyses were done separately for each stimulus pool and all stimulus classes (see Figure 1 for stimulus class definition). For each correctly answered trial (200 ms ≤ RT ≤ 1300 ms), oscillatory power was calculated by transforming the raw EEG signals with a 6-cycle Morlet wavelet [53] in logarithmically spaced frequencies between 2 and 50 Hz (38 intervals: 2x/8 Hz, for x ∊ {8...45}). Due to random temporal jitter in the ISI, stimulus class durations differed slightly across trials. To overcome this temporal variability, we applied a classical binning technique that accounted for intersubject and intertrial variance [54] by resampling stimulus classes into 20 equal length time bins; because the encoding stimulus class spanned the presentation of 3 study items plus the retention interval, the encoding class was resampled into 60 time bins. To minimize the variations in power across subjects and across trials, Z-transformed log10 wavelet power was computed for each frequency band, stimulus pool, subject, and stimulus class, following
where Z denotes Z-transformed power (Zpower, unitless); P denotes log10 wavelet power (dB); t denotes time bin (bin); B denotes time bins in the baseline stimulus class, which corresponds to 400 ms before and to the onset of the fixation stimulus class, |B| denotes the number of items in vector B; σNB denotes standard deviation across trials calculated using the mean power from the baseline stimulus class (dB), and N denotes the total number of trials selected in a given condition. For a given frequency band, the Z-transform indicates how many standard deviations the average signal changed with respect to the baseline interval.
Randomization Test Procedure
Signals in our EEG dataset were not independent; therefore, in order to account for multiple comparisons (across electrodes and frequency bands) and deviations from normality, we used a nonparametric randomization approach. Because we had equal electrode coverage across subjects and employed a repeated measures design, we were able to perform a randomization procedure on paired samples [55] at every electrode. That is, when we searched for stimulus-type-specific effects (e.g., verbal vs. grating), the first condition referred to the mean power in all verbal trials, the second condition referred to the mean power in all grating trials, and both conditions were calculated from the same time interval (e.g., encoding). In contrast, when we searched for time-interval-specific effects, the first condition referred to the mean ISI power, and the second condition referred to the mean SPI power, and both conditions were calculated from the same stimulus type (e.g., verbal). To account for multiple comparisons, we first generated one empirical distribution, using difference values across conditions at every electrode and frequency band, and then we combined data from all frequency bands and all electrodes into one large dataset. We used this this dataset to create 10,000 pseudodistributions by assigning opposite signs (+1, -1) to each difference value with equal probability over the 10,000 iterations. This was followed by a paired t test for every iteration of the pseudodistribution and then sorting the 10,000 t scores in order of increasing magnitude. Finally, we set the Type I error rate to 0.001, which equated to 0.24 electrodes that could have been significant by chance (0.001 × × 4 frequency bands = 0.24 electrodes). The t score from the pseudopopulation that corresponded to this Type I error rate was compared to that obtained from the original dataset at a given electrode and frequency band; if the t score from the original dataset was more significant than the t score generated by the pseudopopulation, we considered the original dataset to be significant (randomization test, p < 0.001, df = 11).
Similarly, when we generated time-frequency plots, in order to correct for multiple comparisons (across electrodes, frequency bands, and time bins) and deviations from normality, we used the same randomization procedure described above with two modifications. We added one additional dimension to represent time bin, and we randomized power at the level of discrete frequencies instead of taking the mean power across frequency band. Then we set the Type I error rate to 0.01, which equated to 53 time bins that could have been significant by chance (0.01 × × 38 frequency bins = 53 time bins). The t score from the pseudopopulation that corresponded to this Type I error rate was compared to that obtained from the original dataset at a given electrode and frequency band; if the t score from the original dataset was more significant than the t score generated by the pseudopopulation, we considered the original dataset to be significant (randomization test, p < 0.01, df = 11).
Declaration of Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
GH designed and assembled the EEG system, collected data, performed all analyses, developed analysis techniques (e.g., Z-transform to baseline and paired randomization test procedure), and wrote the manuscript. JJ performed preliminary data analyses and wrote the hardware driver code for realtime interface. AG wrote the C code that generated the Sternberg experiment and the grating stimuli, and assisted with data collection. JD assisted with data collection and preliminary behavioral analysis. RS provided technical input in the design of the grating task. MJK conceived the experiment and contributed important insights to the manuscript. All authors read, provided feedback, and approved the manuscript.
Acknowledgements
We wish to thank John Lisman and Sri Raghavachari for useful discussions that lead to the generation of this manuscript. GH would like to thank Polly Johnsen and Hal Grodzins for editorial assistance. GH would also like to thank Joe Monaco and Marieke van Vugt for thoughtful comments on the manuscript. Support by NIH grants MH068404 and MH61975.
==== Refs
Cavanagh JP Relationship between the immediate memory span and the memory search rate Psychol Rev 1972 79 525 530
Sternberg S Memory-scanning: mental processes revealed by reaction-time experiments Am Sci 1969 57 421 457 5360276
Sternberg S Memory scanning: new findings and current controversies Q J Exp Psychol 1975 27 1 32
Baddeley A Working memory: looking back and forward Nat Rev Neurosci 2003 4 829 839 14523382 10.1038/nrn1201
Cabeza R Nyberg L Neural bases of learning and memory: functional neuroimaging evidence Curr Opin Neurobiol 2000 13 415 421 10.1097/00019052-200008000-00008
Murray D Vocalization-at-presentation and immediate recall, with varying presentation-rates Q J Exp Psychol 1965 17 47 56 14255465
Gruber O Effects of domain-specific interference on brain activation associated with verbal working memory task performance Cereb Cortex 2001 11 1047 1055 11590114 10.1093/cercor/11.11.1047
Bastiaansen M Hagoort P Event-induced theta responses as a window on the dynamics of memory Cortex 2003 39 967 992 14584562
Gevins A Smith ME McEvoy L Yu D High-resolution EEG mapping of cortical activation related to working memory: effects of task difficulty, type of processing, and practice Cereb Cortex 1997 7 374 385 9177767 10.1093/cercor/7.4.374
Jensen O Tesche CD Frontal theta activity in humans increases with memory load in a working memory task Eur J Neurosci 2002 15 1395 11994134 10.1046/j.1460-9568.2002.01975.x
Jensen O Gelfand J Kounios J Lisman JE Oscillations in the alpha band (9–12 Hz) increase with memory load during retention in a short-term memory task Cereb Cortex 2002 12 877 882 12122036 10.1093/cercor/12.8.877
Howard M Rizzuto D Caplan J Madsen J Lisman J Aschenbrenner-Scheibe R Schulze-Bonhage A Kahana M Gamma oscillations correlate with working memory load in humans Cereb Cortex 2003 13 1369 1374 14615302 10.1093/cercor/bhg084
Klimesch W EEG alpha and theta oscillations reflect cognitive and memory performance: a review and analysis Brain Res Rev 1999 29 169 195 10209231 10.1016/S0165-0173(98)00056-3
Klimesch W Doppelmayr M Russegger H Pachinger T Theta band power in the human scalp EEG and the encoding of new information Neuroreport 1996 7 1235 1240 8817539
Klimesch W Doppelmayr M Stadler W Pollhuber D Sauseng P Rohm D Episodic retrieval is reflected by a process specific increase in human electroencephalographic theta activity Neurosci Lett 2001 302 49 52 11278109 10.1016/S0304-3940(01)01656-1
Klimesch W Doppelmayr M Pachinger T Ripper B Brain oscillations and human memory: EEG correlates in the upper alpha and theta band Neurosci Lett 1997 238 9 12 9464642 10.1016/S0304-3940(97)00771-4
Raghavachari S Kahana MJ Rizzuto DS Caplan JB Kirschen MP Bourgeois B Madsen JR Lisman JE Gating of human theta oscillations by a working memory task J Neurosci 2001 21 3175 3183 11312302
Rizzuto DS Madsen JR Biomfield EB Schulze-Bonhage A Seelig D Aschenbrenner-Scheibe R Kahana MJ Reset of human neocortical oscillations during a working memory task Proc Natl Acad Sci USA 2003 100 7931 7936 12792019 10.1073/pnas.0732061100
Sarnthein J Petsche H Rappelsberger P Shaw GL von Stein A Synchronization between prefrontal and posterior association cortex during human working memory Proc Natl Acad Sci USA 1998 95 7092 7096 9618544 10.1073/pnas.95.12.7092
Schack B Vath N Petsche H Geissler HG Möller E Phase-coupling of theta-gamma EEG rhythms during short-term memory processing Int J Psychophysiol 2002 44 143 163 11909647 10.1016/S0167-8760(01)00199-4
von Stein A Rappelsberger P Sarnthein J Petsche H Synchronization between temporal and parietal cortex during multimodal object processing in man Cereb Cortex 1999 9 137 150 10220226 10.1093/cercor/9.2.137
Tallon-Baudry C Bertrand O Oscillatory gamma activity in humans and its role in object representation Trends Cogn Sci 1999 3 151 162 10322469 10.1016/S1364-6613(99)01299-1
Tallon-Baudry C Kreiter A Bertrand O Sustained and transient oscillatory responses in the gamma and beta bands in a visual short-term memory task in humans Vis Neurosc 1999 16 449 459 10.1017/S0952523899163065
Tallon-Baudry C Bertrand O Fischer C Oscillatory synchrony between human extrastriate areas during visual short-term memory maintenance J Neurosci 2001 21 177
Tesche CD Karhu J Theta oscillations index human hippocampal activation during a working memory task Proc Natl Acad Sci USA 2000 97 919 924 10639180 10.1073/pnas.97.2.919
Gruber T Muller MM Oscillatory brain activity dissociates between associative stimulus content in a repetition priming task in the human EEG Cereb Cortex 2005 15 109 116 15238442 10.1093/cercor/bhh113
Kaiser J Lutzenberger W Human gamma-band activity: a window to cognitive processing Neuroreport 2005 16 207 211 15706221 10.1097/00001756-200502280-00001
Sternberg S High speed scanning in human memory Science 1966 153 652 654 5939936
Patterson JV Pratt H Starr A Event-related potential correlates of the serial position effect in short-term memory Electroencephalogr Clin Neurophysiol 1991 78 424 437 1712278 10.1016/0013-4694(91)90060-H
Crites SL Delgado P Devine JV Lozano DI Immediate and delayed stimulus repetitions evoke different ERPs in a serial-probe recognition task Psychophysiology 2000 37 850 11117465 10.1017/S0048577200991261
Pfurtscheller G Lopes da Silva FH Event-related EEG/MEG synchronization and desynchronization: basic principles Clin Neurophysiol 1999 110 1842 1857 10576479 10.1016/S1388-2457(99)00141-8
Tallon-Baudry C Bertrand O Peronnet F Pernier J Induced gamma-band activity during the delay of a visual short-term memory task in humans J Neurosci 1998 18 4244 4254 9592102
Shallice T Warrington E Independent functioning of verbal memory stores: a neuropsychological study Q J Exp Psychol 1970 22 261 273 5431401
Shallice T Vallar G The impairment of auditory-verbal short-term storage Neuropsychological Impairments of Short-Term Memory 1990 New York: Cambridge University Press 11 53
Stepankova K Fenton AA Pastalkova E Kalina M Bohbot VD Object-location memory impairment in patients with thermal lesions to the right or left hippocampus Neuropsychologia 2004 42 1017 1028 15093141 10.1016/j.neuropsychologia.2004.01.002
Walter H Bretschneider V Grön G Zurowski B Wunderlich AP Tomczak R Spitzer M Evidence for quantitative domain dominance for verbal and spatial working memory in frontal and parietal cortex Cortex 2003 39 897 911 14584558
Smith EE Jonides J Neuroimaging analyses of human working memory Proc Natl Acad Sci USA 1998 95 12061 12068 9751790 10.1073/pnas.95.20.12061
Desmond JE Gabrieli JD Wagner AD Ginier BL Glover GH Lobular patterns of cerebellar activation in verbal working-memory and finger-tapping tasks as revealed by functional MRI J Neurosci 1997 17 9675 9685 9391022
Chen SHA Desmond JE Cerebrocerebellar networks during articulatory rehearsal and verbal working memory tasks Neuroimage 2005 24 332 338 15627576 10.1016/j.neuroimage.2004.08.032
Jonides J Smith E Rugg MD The architecture of working memory Cognitive Neuroscience 1997 Cambridge, Massachusetts: The MIT Press 243 276
Wager TD Smith EE Neuroimaging studies of working memory: a meta-analysis Cogn Affect Behav Neurosci 2003 3 255 74 15040547
Davachi L Maril A Wagner AD When keeping in mind supports later bringing to mind: neural markers of phonological rehearsal predict subsequent remembering J Cogn Neurosci 2001 13 1059 1070 11784444 10.1162/089892901753294356
Cairo TA Liddle PF Woodward TS Ngan ETC The influence of working memory load on phase specific patterns of cortical activity Brain Res Cogn Brain Res 2004 21 377 387 15511653 10.1016/j.cogbrainres.2004.06.014
Baddeley A Working Memory 1986 No. 11 in Oxford Psychology Series, Oxford [Oxfordshire]: Clarendon Press; New York; Oxford University Press
Baddeley A Working memory Science 1992 255 556 559 1736359
Clark CR Egan GF McFarlane AC Morris P Weber D Sonkkilla C Marcina J Tochon-Danguy HJ Updating working memory for words: a PET activation study Hum Brain Mapp 2000 9 42 54 10643729 10.1002/(SICI)1097-0193(2000)9:1<42::AID-HBM5>3.0.CO;2-6
Mukamel R Gelbard H Arieli A Hasson U Fried I Malach R Coupling Between Neuronal Firing, Field Potentials, and fMRI in Human Auditory Cortex Science 2005 309 951 954 16081741 10.1126/science.1110913
Snodgrass JG Vanderwart M A standardized set of 260 pictures: norms for name agreement, image agreement, familiarity, and visual complexity J Exp Psychol [Hum Learn] 1980 6 174 215 7373248
Zhou F Kahana MJ Sekuler R Short-term episodic memory for visual textures: a roving probe gathers some memory Psychol Sci 2004 15 112 118 14738518 10.1111/j.0963-7214.2004.01502007.x
Kahana MJ Sekuler R Recognizing spatial patterns: a noisy exemplar approach Vision Res 2002 42 2177 2192 12207978 10.1016/S0042-6989(02)00118-9
Pelli D The VideoToolbox software for visual psychophysics: transforming numbers into movies Spat Vis 1997 10 437 442 9176953
EGI Net Station Waveform Tools Technical Manual. S-MAN-200-WFTR-001 2003 Electrical Geodesies, Inc., Eugene, OR, USA
Tallon-Baudry C Bertrand O Delpuech C Pernier J Stimulus specificity of phase-locked and non-phase-locked 40 Hz visual responses in humans J Neurosci 1996 16 4240 4249 8753885
Workman W Adams O Suggestions for simplifying the computation of vincent curves J Gen Psychol 1950 43 147 151
Howell DC Statistical Methods for Psychology 2002 Pacific Grove, CA: Duxbury Thomson Learning
Delorme A Makeig S EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis J Neurosci Methods 2004 134 9 21 15102499 10.1016/j.jneumeth.2003.10.009
|
16287504
|
PMC1315326
|
CC BY
|
2021-01-04 16:39:20
|
no
|
Behav Brain Funct. 2005 Nov 15; 1:20
|
utf-8
|
Behav Brain Funct
| 2,005 |
10.1186/1744-9081-1-20
|
oa_comm
|
==== Front
Reprod Biol EndocrinolReproductive biology and endocrinology : RB&E1477-7827BioMed Central London 1477-7827-3-661628866010.1186/1477-7827-3-66ReviewBiology and physiology of Calbindin-D9k in female reproductive tissues: Involvement of steroids and endocrine disruptors Choi Kyung-Chul [email protected] Peter CK [email protected] Eui-Bae [email protected] Department of Obstetrics and Gynecology, British Columbia Children's and Women's Hospital, Child and Family Research Institute, University of British Columbia, Vancouver, BC V6H 3V5 Canada2 Laboratory of Veterinary Biochemistry and Molecular Biology, College of Veterinary Medicine and Research Institute of Veterinary Medicine, Chungbuk National University, Cheongju, Chungbuk, 361-763 Republic of Korea2005 16 11 2005 3 66 66 2 9 2005 16 11 2005 Copyright © 2005 Choi et al; licensee BioMed Central Ltd.2005Choi 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.
Although Calbindin-D9k (CaBP-9k), a cytosolic calcium binding protein which has calcium binding sites, is expressed in various tissues, i.e., intestine, uterus, and placenta, potential roles of this gene and its protein are not clearly understood. Uterine CaBP-9k may be involved in controlling myometrial activity related with intracellular calcium level and is not under the control of vitamin D despite the presence of vitamin D receptors. But, it is under the control of the sex steroid hormones, estrogen (E2) and progesterone (P4), in female reproductive systems including the uterus and placenta. Thus, in this review, we summarize recent research literature in regards to the expression and regulation of CaBP-9k in mammals and introduce the research data of recent studies by us and others.
==== Body
Introduction
A 9-kilodalton cytosolic calcium-binding protein termed as Calbindin-D9k (CaBP-9k) belongs to a family of intracellular proteins which have high affinities for calcium, and has two calcium binding domains [1]. The full-length cDNA encoding the human CaBP-9k has been cloned using reverse transcription/PCR, which includes coding region of 79 amino acids, 57 nucleotides 5'- and 159 nucleotides 3'-non-coding region, and a poly(A) tail (total 600 nucleotides in length) [2]. Further, our study revealed that this gene spans about 5.5-kb and is localized on the X-chromosome, consists of three exons and carries four Alu repeats [3]. In addition to its genomic structure, a sequence of 50 nucleotides downstream from the promoter showed an extensive homology to the estrogen response element (ERE) at the same location within the rat calbindin-D9k gene, suggesting that a two-nucleotide change within this region in human causes the gene to lack expression in human uterus and placenta [3].
It has been demonstrated that CaBP-9k is expressed in diverse mammalian tissues, i.e., intestine, uterus, kidney, and bone [4-7]. The functional role of CaBP-9k is involved in intestinal calcium absorption and its gene is regulated at the transcriptional or post-transcriptional level by 1,25-dihydroxyvitamin D3 (1,25-(OH)2D3), a hormonal form of vitamin D [8,9]. This hormonal form caused a parallel increase in CaBP-9k mRNA and intestinal absorption of calcium in rats [10]. In addition, uterine CaBP-9k may be involved in controlling myometrial activity related with intracellular calcium level [6], but an exact role of CaBP-9k in the uterus is still under investigation by us and a few of other research groups. Recently, we demonstrated that uterine CaBP-9k is responsive to exogenous estrogen (E2) and can be a biomarker for environmental estrogenic chemicals, so called as "endocrine disruptors" in rat models [11-15]. Thus, in this review, we summarize recent research literature in regards to the expression and regulation of CaBP-9k in mammals and introduce updated research results by us and others.
Uterine expression of CaBP-9k
It has been demonstrated that CaBP-9k is mainly expressed in the endometrial stroma and myometrium of the uterus in non-pregnant rats [16,17], whereas this gene is translocated into the epithelium of the uterus in pregnant rats [18]. However, it has been shown that the CaBP-9k is only expressed in the luminal and glandular epithelium of the endometrium, not in the myometrium and in the stromal cells of the endometrium in non-pregnant cows [19]. In contrast to the regulation of CaBP-9k in the intestine, CaBP-9k gene is not under the control of vitamin D in the uterus despite the presence of vitamin D receptors in this tissue. This gene appears to be under the control of the sex steroid hormones [17,20,21].
There is a strong body of evidence that CaBP-9k is regulated by sex steroid hormones in the uterus of rats. Treatment of 21-day-old rats with E2 resulted in an increase in the expression of CaBP-9k mRNA up to 300-fold and its mRNA was shown to fluctuate in the uterus of rats during estrous cycle, where serum E2 level was also under fluctuation [22]. Although the expression of CaBP-9k mRNA is not detectable at diestrus when E2 level is at the lowest, this mRNA increases at proestrus and reaches the highest level at estrus in response to the rise in plasma E2 and then decreases at metestrus in the uterus of rats [20,22]. In addition, E2-dependent regulation of CaBP-9k gene was demonstrated. i.e., CaBP-9k synthesis decreased drastically in the uterus of ovariectomized rats, whereas it was greatly enhanced by low physiological doses of E2 in a dose-dependent manner by CaBP-radioimmunoassay (RIA) [17]. This E2-dependent regulation of CaBP-9k gene was also approved in the uterus of mature ovariectomized and immature rats by slot and Northern blot analysis [23,24]. In estrogen-primed ovariectomized rats, progesterone (P4) inhibited E2-induced CaBP-9k gene expression, which was completely abolished by co-administration of RU486, a P4 antagonist [20]. In the pregnant rats, P4 was shown to be responsible for down-regulation of CaBP-9k gene in the uterus during early pregnancy [25]. In the ovariectomized (OVX) gilts, E2 treatment induced an increase in CaBP-9k mRNA level, whereas P4 administration to ovariectomized pigs decreased CaBP-9k mRNA levels [26]. Recently, we demonstrated that CaBP-9k mRNA and protein are dominantly expressed during luteal phase, indicating that P4 may play an important role in the up-regulation of CaBP-9k gene in the porcine uterus during luteal phase, which is unlike the condition in the rat uterus (Fig. 1) [27]. In this study, the porcine CaBP-9k may be dominantly expressed in the epithelium and glandular structure of the porcine uterus during luteal phase, suggesting that CaBP-9k gene may also be differentially regulated during this cycle presumably by steroid hormones, especially up-regulated P4 level in this tissue [27]. In addition, a positive role of P4 on the expression of CaBP-9k has been demonstrated in the bovine uterus, which indicates that the expression of CaBP-9k was greatest during P4-dominated luteal phase of the bovine estrous cycle [19].
Figure 1 Porcine CaBP-9k mRNA expression in the endometrium and whole uterus during an estrus cycle. To investigate a role of CaBP-9k in the tissue compartment of the pig uterus, the expression levels of CaBP-9k mRNA (A) and protein (B) were analyzed by Northern blot and immunoblot analyses. [Reproduced with permission from Yun S-M, Choi KC, Kim IH, An BS, Lee GS, Hong EJ, Son JH, Oh GT, Jeung E-B 2004 Calbindin-D9k mRNA expression and regulation during estrus cycle in the pig uterus. Mol Reprod Dev 67: 251\endash 256]
In contrast to the uterus of rats, the expression of CaBP-9k gene is not under strict E2 regulation in the uterus of mice. The uterus of mice has been demonstrated to express CaBP-9k and its level increases in this tissue during early pregnancy and implantation [28,29]. CaBP-9k mRNA is expressed in the endometrial epithelia, both luminal and glandular, in the uterus at the time of implantation, and in the luminal, but not in the glandular, epithelia on early pregnancy (day 5 of pregnancy). P4 enhanced CaBP-9k mRNA expression in the uterus, whereas E2 did not in the oophorectomized adult mice [28]. A higher expression of CaBP-9k mRNA was observed in the uterus of mice at diestrus and metestrus, whereas only basal level of its expression at proestrus and estrus, and E2 alone did not induce uterine CaBP-9k mRNA in this study [29]. Taken together, these results suggest the complex hormonal regulation of CaBP-9k in the uterus of different species. To date, there was no evidence that CaBP-9k gene may be regulated by E2 in female reproductive tract of mice. In the recent study, we demonstrated that RU486, a P4 antagonist, induced a significant decrease in CaBP-9k mRNA expression, whereas tamoxifen and ICI 182,780, an E2 antagonists, had no effect on CaBP-9k mRNA expression, suggesting that P4, not E2, is a key regulator of CaBP-9k mRNA expression during late pregnancy and lactation in the uterus of mice [30].
The mechanism involved in the regulation of uterine CaBP-9k gene by steroids is relatively well understood in rats. In the uterus of rats, estrogen is known to up-regulate and progesterone down-regulate the expression of CaBP-9k gene during estrous cycle and early pregnancy [20,22,24,25]. The expression level of CaBP-9k mRNA fluctuates during the estrous cycle, but shows very different expression pattern in the uterus of between rats and mice as abovementioned [29]. The mechanism of distinct regulation of CaBP-9k gene between the rat and mouse is not clear at the moment. The effect of E2 on the regulation of CaBP-9k appears to be mediated through an imperfect estrogen-responsive-like element (ERE) identified in the intron A of mouse CaBP-9k gene [8,24]. The regulation of this gene is known to be mediated by an E2 response element located at the first intron of rat CaBP-9k gene [31]. Recently, cloning of intron A of the mouse CaBP-9k gene have revealed single-base difference in the ERE compared to that of the rat [32]. This may partially explain the observed difference in the hormonal regulation of CaBP-9k gene in the uterus of between rats and mice. In addition, a distinct regulation of porcine CaBP-9k gene in the uterus is explained by no presence of a functional ERE within intron A region [26]. However, we can not rule out the possibility of involvement of other unknown cell-, tissue-, and species-specific factors in the CaBP-9k gene expression. This idea is supported by the finding that E2 regulation of CaBP-9k gene in rats was only uterine-specific and this does not occur in the intestine [17]. Although the putative ERE failed to bind to the estrogen receptor (ER) from the mouse uterus, we isolated mouse genomic clones of the CaBP-9k gene and analyzed their expression in the mouse uterus [33]. In addition, we found a promoter region of CaBP-9k gene containing the putative progesterone response element (PRE) and its expression was stimulated by P4, suggesting that the mouse uterine CaBP-9k gene is expressed under the control of a PRE [33]. In the recent study, we demonstrated that P4 and PR may be a dominant factor in the regulation of CaBP-9k and E2 and ERα can also influence the expression of CaBP-9k gene via an indirect pathway in the uterus of immature mice [34]. Currently, endocrine disruptor-induced expression of CaBP-9k mRNA and protein was reversed or abolished by pretreatment with RU486 or ICI 182,780, suggesting that these synthetic chemicals may have both progestogenic and estrogenic properties by acting through PR or ER in the induction of uterine CaBP-9k mRNA and protein in the uterus of immature mice [35].
Uterine induction of CaBP-9k as a biomarker for endocrine disruption
Endocrine disruptors (EDs) are environmental chemicals that interfere with physiological systems, adversely affecting hormone balance (endocrine system) or disrupting normal function in the organs that hormones regulate or modulate, for example, female and male reproductive system [36]. Representative example of suspected environmental estrogenic EDs includes the drugs which have been specifically designed to treat hormone imbalance in humans. These estrogenic compounds, including octylphenol (OP), nonylphenol (NP), bisphenol A (BPA), and methoxychlor (MXC), can also be transferred through the placenta to the fetus and through breast milk to infants [11,13,37]. Screening methods to detect endocrine disrupters have been evaluated by many groups, i.e. the receptor binding assay, reporter gene assay, and immature rat uterotrophic assay. The reporter gene assay has many benefits as a promising prescreening procedure, because this assay could be performed as a high throughput screening process to detect an endocrine disruptor from thousands of chemicals and no use of experimental animals is required [38]. To screen estrogenic chemicals in the induction of endocrine disruption, genetically sensitive animal models, mice and rats, should be developed. Efforts to identify the mechanisms of endocrine disruption by estrogenic chemicals need to be supported for optimal test methods for thousands of potential chemicals in reproductive development and function [39]. Thus, we have recently established a sensitive method to detect CaBP-9k mRNA and protein using immature rats, which can be used as a biomarker for endocrine disruptors, thus, we introduce our current results in regard to estrogenic effect of endocrine disruptors in the uterus of immature rats [12]. Among the assays for the estrogenic activity of chemicals, an assay to detect an endogenous gene expression that measures estrogen-induced changes either in cultured cells or in selected tissues from exposed animals has been proposed and is widely being used. In our previous study, we demonstrated that phthalate esters exhibit a weak estrogenic activity in vitro assay at high concentrations. Although phthalates resulted in an increase in MCF-7 cell proliferation by estrogenic effect, they could not induce CaBP-9k expression in vivo system following oral treatments, assuming that these phthalates are easily metabolized to inactive forms in vivo system. These results suggest that a conflict may exist in estrogenic effect by various phthalates between in vitro and in vivo models related to the expression of CaBP-9k [40].
The expression levels of CaBP-9k mRNA and protein are strongly up-regulated by estrogenic compounds (OP, NP and BPA) and E2 itself in the uterus of immature rats (Fig. 2), indicating that CaBP-9k can be a useful biomarker for detection of the estrogenicity of putative estrogenic compounds. Thus, regarding to risk assessment, we propose that CaBP-9k mRNA and protein assay in the immature rat uterus can be a very sensitive and powerful tool to identify compounds with estrogenic activity when used in combination with classical assays [11,23]. Treatments of dams with OP, NP and BPA resulted in an increase of CaBP-9k mRNA and protein in maternal and fetal uteri (Fig. 3) of immature rats [13]. These results demonstrate that maternally injected estrogenic compounds resulted in an increase of CaBP-9k mRNA and/or protein in the maternal tissues (uterus and placenta) and fetal uterus during late pregnancy, suggesting that placenta may not be a reliable barrier against these estrogenic compounds for fetal health [13]. The uterus is a highly estrogen-responsive tissue, which can be measured through changes in CaBP-9k expression. In addition, we investigated the potential for estrogenic compounds, OP, NP, BPA, diethylstilbestrol (DES) and E2 to be transferred through breast milk from the dam to the neonate during lactation via measuring the induction of CaBP-9k in uterine tissue [14]. These results indicate that these compounds have an estrogenic effect on the maternal uterus during the lactation period, as shown by the induction of both CaBP-9k mRNA and protein. There was a significant increase in CaBP-9k mRNA in neonatal uterus when the dams were treated with high doses of estrogenic compounds, but protein levels of CaBP-9k were undetectable (Fig. 4). Taken together, these findings suggest that maternally injected estrogenic compounds may be transferred to neonates through breast milk and thus, affect uterine function, as shown by the induction of CaBP-9k gene expression in neonatal uterus [14]. In addition, we recently examined the effect of OP, NP and BPA on the expression of CaBP-9k following maternal exposures during late pregnancy in maternal and fetal uterus [15,41]. The expression of CaBP-9k mRNA was also induced following treatment with a high dose (600 mg/kg BW) of OP transferred from mother exposed to fetuses during late pregnancy and persisted in 5 day of lactation (Fig. 5). In parallel with mRNA level, the expression level of CaBP-9k protein was significantly induced by treatment with a high dose of OP and NP in the maternal uterus by immunohistochemistry (Fig. 6). In conclusion, the maternal exposures to OP, NP and BPA during late pregnancy increased the expression levels of CaBP-9k mRNA and protein in maternal or neonatal uteri, suggesting that the absorption and distribution of environmental estrogenic compounds in maternal and neonatal uteri are extremely rapid, and these chemicals can easily pass though placenta during pregnancy to affect functions of neonatal reproductive tissues [15,41]. In addition, a novel in vivo model was introduced to detect both estrogenic and progestogenic activities of EDs in the induction of CaBP-9k mRNA and protein in the uterus of immature mice [35].
Figure 2 Induction of uterine CaBP-9k protein by estrogenic compounds, i.e., OP, NP and BPA was assessed in a dose-dependent manner at 24 h after final injection by immunoblot analysis. Data are presented as the mean ± SD. a; significantly different compared with vehicle at P < 0.05. [Reproduced with permission from An B-S, Choi KC, Kang SK, Hwang WS, Jeung EB 2003 Novel Calbindin-D9k protein as a useful biomarker for environmental estrogenic compounds in the uterus of immature rats. Reprod Toxicol 17: 311–319]
Figure 3 Induction of CaBP-9k mRNA expression in fetal uterus by estrogenic compounds. The values represent means ± SD. a, P < 0.05 vs. vehicle. [Reproduced with permission from Hong EJ, Choi KC, Jeung E-B 2003 Maternal-fetal transfer of endocrine disruptors in the induction of Calbindin-D9k mRNA and protein during pregnancy in rat model. Mol Cell Endocrinol 212: 63–72]
Figure 4 Effect of endocrine disruptors on the induction of CaBP-9k mRNA in neonatal uterus. RT-PCR/Southern blot analysis for CaBP-9k mRNA on day 6 of lactation was performed as previously described. The values represent means ± SD (n = 5). a, P < 0.05 vs. vehicle (VE). [Reproduced with permission from Hong E-J, Choi KC, Jung YW, Leung PCK, Jeung E-B 2004 Transfer of maternally injected endocrine disruptors through breast milk during lactation induces neonatal Calbindin-D9k in the rat model. Reprod Toxicol 18: 661–668]
Figure 5 Effect of OP and NP on the induction of CaBP-9k mRNA in neonatal uterus. RT-PCR/Southern blot assay was performed during lactation period in neonatal uterus. The values represent means ± SD. a, P < 0.05 vs. vehicle. [Reproduced with permission from Hong EJ, Choi KC, Jeung EB 2004 Induction of Calbindin-D9k mRNA and protein by maternal exposure to alkylphenols during late pregnancy in maternal and postnatal uteri of rats. Biol Reprod 71: 669–675]
Figure 6 Localization of CaBP-9k protein by immunohistochemicalstaining in maternal uterus. Immuno-reactivity of CaBP-9k protein expression following treatment with OP and NP was investigated in endometrium and smooth myometrial fibers dose-dependently. Especially, these proteins that are widely spaced through the stromal cells in endometrium s, stroma cells; le, Luminal epithelial cell; ge, glandular epithelial cell. Magnification × 100. [Reproduced with permission from Hong EJ, Choi KC, Jeung EB 2004 Induction of Calbindin-D9k mRNA and protein by maternal exposure to alkylphenols during late pregnancy in maternal and postnatal uteri of rats. Biol Reprod 71: 669–675]
Genistein, a phytoestrogen, has been shown to have relatively 20-fold higher binding affinity to the ERβ than ER by a solid-phase binding assay [42]. To determine which ER is involved in the induction of CaBP-9k gene, we employed genistein as a potent ERβ agonist to clarify its effect on uterine CaBP-9k regulation [43]. Both CaBP-9k mRNA and protein levels were significantly induced by genistein in the uterus of immature rats. It is of interest that the pre-treatment of immature rats with ICI 182,780 (ICI), followed by genistein or E2, completely blocked genistein- and E2-induced CaBP-9k protein in this tissue as demonstrated in Fig. 7. In addition, genistein was demonstrated to induce ERα protein, but not ERβ or PR mRNA, an E2-responsive gene, in this tissue. These results imply that genistein, an ERβ ligand, may regulate CaBP-9k gene through ERα pathway and ERα may be a key mediator in the induction of uterine CaBP-9k gene in immature rats [43]. To support an involvement of ERα-dependent pathway by EDs, we demonstrated that uterine CaBP-9k gene expression is mainly mediated by propyl pyrazole triol (PPT), an ERα-selective ligand, in a dose- and time-dependent manner, in the uterus of immature rats [44]. In contrast, no significant alteration in the uterine CaBP-9k gene was observed after diarylpropionitrile (DPN), an ERβ-selective ligand. In addition, an estrogenicity of PPT in inducing CaBP-9k expression was completely blocked by ICI 182,780; which suggests that uterine CaBP-9k is solely enhanced though ERα. Taken together, these results indicate that uterine CaBP-9k is induced by E2 and endocrine disrupting chemicals via ERα pathway, but not ERβ, in the uterus of immature rats [44].
Figure 7 Effect of ICI on genistein-induced uterine CaBP-9k protein expression. Immature rats were injected SC with ICI at 30 min prior to genistein (40 mg/kg BW per day) or E2, and euthanized 24 h after final injection. The level of CaBP-9k protein was analyzed by immunoblot analysis. The values represent means ± SD. a, P < 0.05 vs. vehicle; b, P < 0.05 vs. genistein or E2 treatment only [Reproduced with permission from Lee GS, Choi KC, Kim HJ, Jeung EB 2004 Effect of genistein on the expression of Calbindin-D9k as a potential estrogenic compound in the uterus of immature rats through estrogen receptors. Toxicol Sci 82: 451–457]
Placental expression of CaBP-9k
Transport of Ca2+ through placenta is responsible for developing fetus, and CaBP-9k appears to play an important role in the regulation of Ca2+ from the mother to the fetus during pregnancy. However, the role of CaBP-9k is unknown to date in placenta during pregnancy. A recent study demonstrated that CaBP-9k transcript is present in cytotrophoblast cells and syncytiotrophoblasts of human term placenta, with a lower expression in cytotrophoblast cells as compared to syncytiotrophoblasts [45]. In addition, CaBP-9k protein was present in cytotrophoblast and syncytiotrophoblast placental tissue sections as well as in cultured cells, indicating that CaBP-9k is unequivocally expressed by trophoblast cells from human term placenta [45]. The expression of CaBP-9k gene has been investigated in the placenta of other species [26,46-48]. The high level of CaBP-9k has been localized to epithelial cells of the yolk sac and endodermal cells of the placenta [16]. The expression of CaBP-9k mRNA was not detectable by Northern blot analysis, while this transcript was detected in porcine myometrium and placenta by RT-PCR [26]. As previously described, CaBP-9k mRNA has been also localized in the trophoblasts in various species. It is hypothesized that CaBP-9k plays a role in calcium transfer and fetal growth by parallel gestational changes in placental CaBP-9k which reflects the fetal accumulation of calcium. An increased level of CaBP-9k gene in the caruncular epithelium during the last trimester is in response to the increasing need for calcium to supply the fetal skeleton with mineralization, suggesting that CaBP9k may play a role in transporting calcium across the placenta in cattle [48]. In the placenta of mice, the distinct regulation of CaBP-9k has been demonstrated in the placenta compared to other tissues such as intestine and kidney, indicating that the expression of this gene is not dependent on Vitamin D receptor [49]. Recently we demonstrated the expression of CaBP-9k for the first time in mouse placenta and extra-embryonic membrane separately, and CaBP-9k mRNA may be regulated by sex steroid hormones (E2 and P4) and their receptors through complex pathway in these tissues [50].
Concluding Remarks
Although CaBP-9k is mainly expressed in female reproductive tissues, i.e., uterus and placenta of various species, the role of CaBP-9k remains unknown. It can be hypothesized that uterine CaBP-9k may be involved in controlling myometrial activity related with intracellular calcium level and placental CaBP-9k plays a role in calcium transfer from the mother to the fetus for fetal growth. It appears that CaBP-9k gene is not under the control of vitamin D in the uterus despite the presence of vitamin D receptors in this tissue; instead it is under the control of the sex steroid hormones. The hormonal mechanism controlling uterine CaBP-9k gene is relatively well understood in the rat. In the uterus of rats, estrogen is known to up-regulate and progesterone down-regulate the expression of CaBP-9k gene during estrous cycle and pregnancy. However, the recent studies demonstrated that CaBP-9k is mainly regulated by progesterone, not estrogen in the uterus of mice because of lack of responsiveness caused by a single-base difference in the ERE of mouse CaBP-9k gene compared to that of rats. Until now, a few studies demonstrated the expression and regulation of CaBP-9k gene in the placenta of various species. It appears that CaBP-9k mRNA may be regulated by sex steroid hormones (E2 and P4) and their receptors through complex pathway in these tissues. We assumed that ERα may be a key mediator in uterine CaBP-9k gene induction in immature rats. The elucidation of other factors that regulate CaBP-9k mRNA will further provide insight into the understanding of regulation of CaBP-9k in these tissues, and its roles in the control of reproductive functions.
The uterus is a highly estrogen-responsive tissue, which can be measured through changes in CaBP-9k expression. We demonstrated that the expression levels of CaBP-9k mRNA and protein are induced by estrogenic chemicals, so called "endocrine disruptors", in the uterus of immature rats. In addition, maternally injected estrogenic compounds resulted in an increase of CaBP-9k mRNA and/or protein in the fetal uterus during late pregnancy, suggesting that placenta may not be a reliable barrier against these estrogenic compounds for fetal health. It is of interest that maternally injected estrogenic compounds may be transferred to neonates through breast milk and thus affect uterine function, as shown by the induction of CaBP-9k gene expression in neonatal uterus. The expression of CaBP-9k mRNA and/or protein is an excellent biomarker to detect an estrogenic chemical in the uterus of immature rats which we developed and established. Availability of this gene using immature rats will provide an insight of risk assessment for estrogenic and progestogenic chemicals in our environment.
Acknowledgements
This work was supported by the research grant of the Chungbuk National University in 2005. The authors would like to thank Mr. Geun-Shik LEE (College of Veterinary Medicine, Chungbuk National University) for editing the figures.
==== Refs
Christakos S Gabrielides C Rhoten WB Vitamin D-dependent calcium binding proteins: chemistry, distribution, functional considerations, and molecular biology Endocr Rev 1989 10 3 26 2666110
Jeung EB Krisinger J Dann JL Leung PC Molecular cloning of the full-length cDNA encoding the human calbindin-D9k FEBS Lett 1992 307 224 228 1379540 10.1016/0014-5793(92)80772-9
Jeung EB Leung PC Krisinger J The human calbindin-D9k gene. Complete structure and implications on steroid hormone regulation J Mol Biol 1994 235 1231 1238 8308886 10.1006/jmbi.1994.1076
Armbrecht HJ Boltz M Strong R Richardson A Bruns ME Christakos S Expression of calbindin-D decreases with age in intestine and kidney Endocrinology 1989 125 2950 2956 2583050
Mathieu CL Burnett SH Mills SE Overpeck JG Bruns DE Bruns ME Gestational changes in calbindin-D9k in rat uterus, yolk sac, and placenta: implications for maternal-fetal calcium transport and uterine muscle function Proc Natl Acad Sci U S A 1989 86 3433 3437 2717621
Mathieu CL Mills SE Burnett SH Cloney DL Bruns DE Bruns ME The presence and estrogen control of immunoreactive calbindin-D9k in the fallopian tube of the rat Endocrinology 1989 125 2745 2750 2676491
Seifert MF Gray RW Bruns ME Elevated levels of vitamin D-dependent calcium-binding protein (calbindin-D9k) in the osteosclerotic (oc) mouse Endocrinology 1988 122 1067 1073 3342744
Darwish HM DeLuca HF Identification of a 1,25-dihydroxyvitamin D3-response element in the 5'-flanking region of the rat calbindin D-9k gene Proc Natl Acad Sci U S A 1992 89 603 607 1309950
Wasserman RH Fullmer CS On the molecular mechanism of intestinal calcium transport Adv Exp Med Biol 1989 249 45 65 2543194
Krisinger J Strom M Darwish HD Perlman K Smith C DeLuca HF Induction of calbindin-D 9k mRNA but not calcium transport in rat intestine by 1,25-dihydroxyvitamin D3 24-homologs J Biol Chem 1991 266 1910 1913 1988453
An BS Choi KC Kang SK Hwang WS Jeung EB Novel Calbindin-D(9k) protein as a useful biomarker for environmental estrogenic compounds in the uterus of immature rats Reprod Toxicol 2003 17 311 319 12759100 10.1016/S0890-6238(03)00003-0
Choi KC Jeung EB The biomarker and endocrine disruptors in mammals J Reprod Dev 2003 49 337 345 14967909 10.1262/jrd.49.337
Hong EJ Choi KC Jeung EB Maternal-fetal transfer of endocrine disruptors in the induction of Calbindin-D9k mRNA and protein during pregnancy in rat model Mol Cell Endocrinol 2003 212 63 72 14654251 10.1016/j.mce.2003.08.011
Hong EJ Choi KC Jung YW Leung PC Jeung EB Transfer of maternally injected endocrine disruptors through breast milk during lactation induces neonatal Calbindin-D9k in the rat model Reprod Toxicol 2004 18 661 668 15219628 10.1016/j.reprotox.2004.03.005
Hong EJ Choi KC Jeung EB Induction of Calbindin-D9k mRNA and protein by maternal exposure to alkylphenols during late pregnancy in maternal and neonatal uteri of rats Biol Reprod 2004 71 669 675 15070829 10.1095/biolreprod.103.026146
Bruns ME Kleeman E Mills SE Bruns DE Herr JC Immunochemical localization of vitamin D-dependent calcium-binding protein in mouse placenta and yolk sac Anat Rec 1985 213 514 7, 532-5 4083532 10.1002/ar.1092130406
Delorme AC Danan JL Acker MG Ripoche MA Mathieu H In rat uterus 17 beta-estradiol stimulates a calcium-binding protein similar to the duodenal vitamin D-dependent calcium-binding protein Endocrinology 1983 113 1340 1347 6688582
Warembourg M Perret C Thomasset M Analysis and in situ detection of cholecalcin messenger RNA (9000 Mr CaBP) in the uterus of the pregnant rat Cell Tissue Res 1987 247 51 57 3829120 10.1007/BF00216546
Inpanbutr N Miller EK Petroff BK Iacopino AM CaBP9K levels during the luteal and follicular phases of the estrous cycle in the bovine uterus Biol Reprod 1994 50 561 571 8167228 10.1095/biolreprod50.3.561
L'Horset F Blin C Brehier A Thomasset M Perret C Estrogen-induced calbindin-D 9k gene expression in the rat uterus during the estrous cycle: late antagonistic effect of progesterone Endocrinology 1993 132 489 495 8425470 10.1210/en.132.2.489
L'Horset F Perret C Brehier A Thomasset M 17 beta-estradiol stimulates the calbindin-D9k (CaBP9k) gene expression at the transcriptional and posttranscriptional levels in the rat uterus Endocrinology 1990 127 2891 2897 2249631
Krisinger J Dann JL Currie WD Jeung EB Leung PC Calbindin-D9k mRNA is tightly regulated during the estrous cycle in the rat uterus Mol Cell Endocrinol 1992 86 119 123 1511778 10.1016/0303-7207(92)90182-6
An BS Kang SK Shin JH Jeung EB Stimulation of calbindin-D(9k) mRNA expression in the rat uterus by octyl-phenol, nonylphenol and bisphenol Mol Cell Endocrinol 2002 191 177 186 12062901 10.1016/S0303-7207(02)00042-4
L'Horset F Blin C Colnot S Lambert M Thomasset M Perret C Calbindin-D9k gene expression in the uterus: study of the two messenger ribonucleic acid species and analysis of an imperfect estrogen-responsive element Endocrinology 1994 134 11 18 7506202 10.1210/en.134.1.11
Krisinger J Setoyama T Leung PC Expression of calbindin-D9k in the early pregnant rat uterus: effects of RU 486 and correlation to estrogen receptor mRNA Mol Cell Endocrinol 1994 102 15 22 7926266 10.1016/0303-7207(94)90092-2
Krisinger J Jeung EB Simmen RC Leung PC Porcine calbindin-D9k gene: expression in endometrium, myometrium, and placenta in the absence of a functional estrogen response element in intron A Biol Reprod 1995 52 115 123 7711170 10.1095/biolreprod52.1.115
Yun SM Choi KC Kim IH An BS Lee GS Hong EJ Oh GT Jeung EB Dominant expression of porcine Calbindin-D9k in the uterus during a luteal phase Mol Reprod Dev 2004 67 251 256 14735485 10.1002/mrd.20019
Tatsumi K Higuchi T Fujiwara H Nakayama T Itoh K Mori T Fujii S Fujita J Expression of calcium binding protein D-9k messenger RNA in the mouse uterine endometrium during implantation Mol Hum Reprod 1999 5 153 161 10065871 10.1093/molehr/5.2.153
Nie GY Li Y Wang J Minoura H Findlay JK Salamonsen LA Complex regulation of calcium-binding protein D9k (calbindin-D(9k)) in the mouse uterus during early pregnancy and at the site of embryo implantation Biol Reprod 2000 62 27 36 10611064 10.1095/biolreprod62.1.27
An BS Choi KC Kang SK Lee GS Hong EJ Hwang WS Jeung EB Mouse calbindin-D(9k) gene expression in the uterus during late pregnancy and lactation Mol Cell Endocrinol 2003 205 79 88 12890569 10.1016/S0303-7207(03)00203-X
Darwish H Krisinger J Furlow JD Smith C Murdoch FE DeLuca HF An estrogen-responsive element mediates the transcriptional regulation of calbindin D-9K gene in rat uterus J Biol Chem 1991 266 551 558 1985915
Krisinger J Dann JL Applegarth O Currie WD Jeung EB Staun M Leung PC Calbindin-D9k gene expression during the perinatal period in the rat: correlation to estrogen receptor expression in uterus Mol Cell Endocrinol 1993 97 61 69 8143906 10.1016/0303-7207(93)90211-2
Lee KY Oh GT Kang JH Shin SM Heo BE Yun YW Paik SG Krisinger J Leung PC Jeung EB Transcriptional regulation of the mouse calbindin-D9k gene by the ovarian sex hormone Mol Cells 2003 16 48 53 14503844
An BS Choi KC Hong EJ Jung YW Manabe N Jeung EB Differential transcriptional and translational regulations of calbindin-D9k by steroid hormones and their receptors in the uterus of immature mice J Reprod Dev 2004 50 445 453 15329476 10.1262/jrd.50.445
Jung YW Hong EJ Choi KC Jeung EB Novel progestogenic activity of environmental endocrine disruptors in the upregulation of calbindin-D9k in an immature mouse model Toxicol Sci 2005 83 78 88 15509668 10.1093/toxsci/kfi015
Daston GP Cook JC Kavlock RJ Uncertainties for endocrine disrupters: our view on progress Toxicol Sci 2003 74 245 252 12730617 10.1093/toxsci/kfg105
Laws SC Carey SA Ferrell JM Bodman GJ Cooper RL Estrogenic activity of octylphenol, nonylphenol, bisphenol A and methoxychlor in rats Toxicol Sci 2000 54 154 167 10746942 10.1093/toxsci/54.1.154
Takeyoshi M Yamasaki K Sawaki M Nakai M Noda S Takatsuki M The efficacy of endocrine disruptor screening tests in detecting anti-estrogenic effects downstream of receptor-ligand interactions Toxicol Lett 2002 126 91 98 11751013 10.1016/S0378-4274(01)00446-5
Spearow JL Barkley M Reassessment of models used to test xenobiotics for oestrogenic potency is overdue Hum Reprod 2001 16 1027 1029 11331655 10.1093/humrep/16.5.1027
Hong EJ Ji YK Choi KC Manabe N Jeung EB Conflict of estrogenic activity by various phthalates between in vitro and in vivo models related to the expression of Calbindin-D9k J Reprod Dev 2005 51 253 263 15883486 10.1262/jrd.16075
Hong EJ Choi KC Jeung EB Maternal Exposure to Bisphenol A during Late Pregnancy Resulted in an Increase of Calbindin-D9k mRNA and Protein in Maternal and Postnatal Rat Uteri J Reprod Dev 2005 In press
Kuiper GG Lemmen JG Carlsson B Corton JC Safe SH van der Saag PT van der Burg B Gustafsson JA Interaction of estrogenic chemicals and phytoestrogens with estrogen receptor beta Endocrinology 1998 139 4252 4263 9751507 10.1210/en.139.10.4252
Lee GS Choi KC Kim HJ Jeung EB Effect of genistein as a selective estrogen receptor beta agonist on the expression of Calbindin-D9k in the uterus of immature rats Toxicol Sci 2004 82 451 457 15456916 10.1093/toxsci/kfh296
Lee GS Kim HJ Jung YW Choi KC Jeung EB Estrogen receptor alpha pathway is involved in the regulation of Calbindin-D9k in the uterus of immature rats Toxicol Sci 2005 84 270 277 15635152 10.1093/toxsci/kfi072
Belkacemi L Gariepy G Mounier C Simoneau L Lafond J Calbindin-D9k (CaBP9k) localization and levels of expression in trophoblast cells from human term placenta Cell Tissue Res 2004 315 107 117 14586690 10.1007/s00441-003-0811-4
Jeung EB Krisinger J Dann JL Leung PC Cloning of the porcine Calbindin-D9k complementary deoxyribonucleic acid by anchored polymerase chain reaction technique Biol Reprod 1992 47 503 508 1391336 10.1095/biolreprod47.4.503
Jeung EB Fan NC Leung PC Herr JC Freemerman A Krisinger J The baboon expresses the calbindin-D9k gene in intestine but not in uterus and placenta: implication for conservation of the gene in primates Mol Reprod Dev 1995 40 400 407 7598905 10.1002/mrd.1080400403
Reiswig JD Frazer GS Inpanbutr N Calbindin-D9k expression in the pregnant cow uterus and placenta Histochem Cell Biol 1995 104 169 174 8536074 10.1007/BF01451576
Shamley DR Veale G Pettifor JM Buffenstein R Trophoblastic giant cells of the mouse placenta contain calbindin-D9k but not the vitamin D receptor J Endocrinol 1996 150 25 32 8708559 10.1677/joe.0.1500025
An BS Choi KC Lee GS Leung PC Jeung EB Complex regulation of Calbindin-D9k in the mouse placenta and extra-embryonic membrane during mid- and late pregnancy Mol Cell Endocrinol 2004 214 39 52 15062543 10.1016/j.mce.2003.11.029
|
16288660
|
PMC1315327
|
CC BY
|
2021-01-04 16:37:13
|
no
|
Reprod Biol Endocrinol. 2005 Nov 16; 3:66
|
utf-8
|
Reprod Biol Endocrinol
| 2,005 |
10.1186/1477-7827-3-66
|
oa_comm
|
==== Front
Respir ResRespiratory Research1465-99211465-993XBioMed Central London 1465-9921-6-1401630954810.1186/1465-9921-6-140ResearchEffects of cigarette smoke condensate on proliferation and wound closure of bronchial epithelial cells in vitro: role of glutathione Luppi Fabrizio [email protected] Jamil [email protected] Wetering Sandra [email protected] Irfan [email protected] Boer Willem I [email protected] Klaus F [email protected] Pieter S [email protected] Department of Pulmonology, Leiden University Medical Center, P.O. Box 9600, 2300RC, Leiden, The Netherlands2 Department of Environmental Medicine, Division of Lung Biology and Disease, University of Rochester Medical Center, Rochester, NY 14642, USA2005 25 11 2005 6 1 140 140 13 5 2005 25 11 2005 Copyright © 2005 Luppi et al; licensee BioMed Central Ltd.2005Luppi 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
Increased airway epithelial proliferation is frequently observed in smokers. To elucidate the molecular mechanisms leading to these epithelial changes, we studied the effect of cigarette smoke condensate (CSC) on cell proliferation, wound closure and mitogen activated protein kinase (MAPK) activation. We also studied whether modulation of intracellular glutathione/thiol levels could attenuate CSC-induced cell proliferation.
Methods
Cells of the bronchial epithelial cell line NCI-H292 and subcultures of primary bronchial epithelial cells were used for the present study. The effect of CSC on epithelial proliferation was assessed using 5-bromo-2-deoxyuridine (BrdU) incorporation. Modulation of epithelial wound repair was studied by analysis of closure of 3 mm circular scrape wounds during 72 hours of culture. Wound closure was calculated from digital images obtained at 24 h intervals. Activation of mitogen-activated protein kinases was assessed by Western blotting using phospho-specific antibodies.
Results
At low concentrations CSC increased proliferation of NCI-H292 cells, whereas high concentrations were inhibitory as a result of cytotoxicity. Low concentrations of CSC also increased epithelial wound closure of both NCI-H292 and PBEC, whereas at high concentrations closure was inhibited. At low, mitogenic concentrations, CSC caused persistent activation of ERK1/2, a MAPK involved in cell proliferation. Inhibition of cell proliferation by high concentrations of CSC was associated with activation of the pro-apoptotic MAP kinases p38 and JNK. Modulation of intracellular glutathione (GSH)/thiol levels using N-acetyl-L-cysteine, GSH or buthionine sulphoximine (BSO), demonstrated that both the stimulatory and the inhibitory effects of CSC were regulated in part by intracellular GSH levels.
Conclusion
These results indicate that CSC may increase cell proliferation and wound closure dependent on the local concentration of cigarette smoke and the anti-oxidant status. These findings are consistent with increased epithelial proliferation in smokers, and may provide further insight in the development of lung cancer.
==== Body
Background
Cigarette smoke, the major risk factor for COPD and lung cancer, contains over 4,500 chemical compounds, including free radicals and oxidants. These compounds are present in both the gas and the tar phase [1] and have been shown to cause epithelial lung injury [2,3]. Epithelial integrity is normally restored by a repair process, that may also result in squamous cell metaplasia and/or goblet cell hyperplasia, especially after repeated injury [4]. This altered composition of the airway epithelium can be observed in smokers [5]. Although these epithelial changes have been observed both for smokers with and without airflow obstruction, some of these epithelial features are more pronounced in COPD patients than in asymptomatic smokers [6,7]. Furthermore, analysis of bronchial biopsies from smokers with chronic bronchitis showed an increased epithelial cell proliferation [8], and studies in both current and former smokers revealed epithelial cell proliferation at sites of metaplasia [9]. These studies indicate that epithelial cell proliferation is a key feature of the epithelial changes observed in smoking-induced lung disease.
Oxidative stress is considered to play a main role in the pathogenesis of inflammatory lung disease, including chronic obstructive pulmonary disease (COPD) [3]. In smokers, this oxidative stress may result both from cigarette smoke itself, and from oxidants released by inflammatory cells that are recruited as a result of smoke-induced injury. The potential importance of oxidative stress in COPD is supported by various studies such as those showing an increase in markers of oxidative stress in patients with COPD [3]. The airway epithelium is a main target for exogenous oxidants such as those present in cigarette smoke. Oxidative stress not only induces cell injury, but also appears to play a central role in e.g. gene expression and cell proliferation. An efficient anti-oxidant system that is present in the lung provides protection against these oxidants, and glutathione (GSH) is considered to be a main antioxidant molecule [10].
Epithelial cell proliferation, as well as various other cellular processes in epithelial cells, is regulated at least in part by Epidermal Growth Factor (EGF)-like factors and the EGF receptor (EGFR). Analysis of the expression of EGF-like growth factors and EGFR in human lung disease has provided evidence for a role of these factors in epithelial remodeling. Kurie et al. observed that EGFR expression was increased in metaplastic bronchial epithelium, and reversal of bronchial metaplasia was associated with decreased EGFR expression [11]. Furthermore, Vignola et al. observed that EGF expression was significantly increased in chronic bronchitis patients in comparison with healthy non-smokers [12]. Both these studies suggest a role for EGFR and its ligands in the epithelial pathological features observed in smokers with and without COPD.
Downstream signaling pathways that are activated via the EGFR and regulate cell survival and proliferation include phosphorylation of mitogen activated protein kinases (MAPK) and Akt/PI-3 kinase pathways. Activation of the MAPK extracellular-regulated kinase (ERK) 1/2 has been associated with cell survival and proliferation, whereas c-jun N-terminal kinases (JNK) and p38 MAPK are linked to induction of apoptosis [13]. In addition to ligands of the EGFR, oxidants have been shown to cause activation of EGFR [14]. Therefore, oxidants may not only cause direct killing of epithelial cells, but also activate specific signaling pathways. Anti-oxidants such as N-acetyl-L-cysteine (NAC) have been found to be an important tool to study the cellular consequences of oxidative stress. Such studies have shown that the increase in cellular GSH/thiol provided by NAC protects cells against oxidative stress.
The aim of the present study was to analyze the effect of cigarette smoke on cell proliferation and wound repair using an in vitro cell culture model. The underlying mechanisms were explored by analyzing the role of MAPK activation and the contribution of an oxidant/antioxidant imbalance in these cellular functions.
Materials and methods
Preparation of cigarette smoke condensate (CSC)
Commercial (Caballero, British American Tobacco Group) and standard research cigarettes (Research cigarettes produced for the University of Kentucky Research Foundation, Reference cigarette: code 1R3, date 3/74) were used in this study. CSC was prepared immediately before use essentially as described by Kim JK et al. [15]. Briefly, cigarette smoke derived from one cigarette was drawn slowly into a 50 ml glass syringe and bubbled into a tube containing 1 ml of phosphate-buffered saline (PBS), at room temperature. Each cigarette was completely burned after an average of 8 draws of the syringe, with each individual draw taking approximately 10 seconds to complete. The pH of the CSC solution was between pH 7.0 and 7.4. Subsequently, the CSC was filtered through a 0.22 μm pore filter (Schleicher & Schuell GmbH, Dassel, Germany). To prevent possible inactivation of compounds present in the CSC, the CSC was kept in the dark. The concentration of CSC in the solution was calculated by measuring the OD value of the 100-fold diluted solution at a wavelength at which the maximal absorbance (ODmax) was detected. In the CSC solution this ODmax was achieved between OD 270–280. The pattern of absorbance observed showed very little difference between different batches of CSC. The concentration, expressed as arbitrary units (AU) per ml, was calculated based on the following formula: ODmax × 2 × dilution factor. The CSC was further diluted to the required concentration in culture medium. Ten AU/ml was found to correspond to a mean of 5 % (vol/vol) CSC. Bronchial epithelial cells were exposed to various concentrations of CSC within 30 min after CSC preparation.
Cell culture
NCI-H292 cells, a human pulmonary muco-epidermoid carcinoma cell line, were obtained from the American Type Culture Collection (ATCC, Manassas, VA). The cells were routinely cultured in RPMI 1640 (Gibco, Grand Island, NY) medium containing 2 mM L-glutamine, 20 U/ml penicillin, 20 μg/ml streptomycin (all from Bio Whittaker, Walkersville, MD), and 10% heat-inactivated FCS (Gibco) at 37°C in a humidified 5% CO2 atmosphere. Cells were passaged weekly using Trypsin Versene (Bio Whittaker, Walkersville, MD), and starved for growth factors by overnight incubation in serum-free medium before exposure to CSC.
Subcultures from primary bronchial epithelial cells (PBEC) were derived from bronchial tissue that was obtained from resected lungs, derived from patients that underwent lung surgery for lung cancer at the Leiden University Medical Center (Leiden, The Netherlands). In this study, we used cells obtained from seven smokers: four without airflow limitation (FEV1 > 81% of the predicted value) and three with airflow limitation (FEV1 < 70% of the predicted value). PBEC were isolated from bronchial rings using enzymatic digestion of tissue as previously described [16]. For the experiments, cells from passage two were cultured in DMEM/Ham F12 (1:1) medium (Gibco) supplemented with 10 ng/ml recombinant EGF (Sigma), 2% (v/v) Ultroser G (Gibco), 1 μM isoproterenol, 1 μM insulin (Sigma), 1 μM hydrocortisone (Sigma), 2 mM L-glutamine, 1 mM Hepes (Gibco), 20 U/ml penicillin and 20 μg/ml streptomycin. PBEC were cultured in tissue culture plates precoated with 10 μg/ml fibronectin (isolated from human plasma), 30 μg/ml Vitrogen (Cohesion technologies Inc., Palo Alto, CA) and 10 μg/ml bovine serum albumin (Sigma Chemical Co.).
Prior to the experiments, PBEC were starved for growth factors by overnight incubation in DMEM/HamF12 medium without UltroSer and EGF.
Cell proliferation
Cell proliferation was assessed using 5-bromo-2-deoxyuridine (BrdU) incorporation as previously described [17]. Briefly, after stimulation, cells were incubated with BrdU (Sigma) for 20 or 24 hours in the presence of the stimulus in starvation medium. Cells were washed twice in PBS and fixed in ethanol 70% (v/v) for at least 1 hour. Cells were then permeabilized with 1 M hydrochloric acid followed by subsequent washes with 0.1 M sodium tetraborate and PBS. BrdU incorporation was demonstrated by incubation with a mouse anti-BrdU mAb followed by incubation with a peroxidase-labeled rabbit anti-mouse polyclonal antibody (both Dako, Glostrup, Denmark). BrdU incorporation was visualized using Nova RED (Vector Laboratories, Burlingame, CA) and the percentage BrdU-positive nuclei was calculated. The percentage BrdU positive nuclei was determined from images that were collected using a digital camera and Axiovision (Carl Zeiss Vision GmbH, München-Hallbermoos, Germany) and Adobe Photoshop (Adobe Systems Incorporated, San Jose, CA) software.
To study the role of oxidants and GSH in the effects of CSC, cells were exposed to N-acetyl cysteine (NAC; Sigma) at 1 mM and CSC. In other experiments, cells were preincubated with DL-buthionine sulphoximine (BSO, Sigma) for 12 hours at a concentration of 10 μM before addition of CSC; BSO was also present during CSC exposure.
Wound closure model
Epithelial wound closure was studied essentially as described by Aarbiou et al. [18]. Both NCI-H292 and PBEC were cultured to confluence. After overnight starvation for growth factors, three circular wounds of 3 mm in diameter were prepared in each well using a Pasteur pipette with sharpened silicone tube. After washing with PBS to eliminate debris, cells were allowed to recover for one hour in starvation medium and subsequently incubated in starvation medium in presence or absence of CSC or TGF-α. In experiments using BSO, cells were pretreated for 12 hours prior to stimulation. NAC was replaced every 12 hours. Images of wounded areas were collected using a digital camera and Axiovision software (Carl Zeiss Vision GmbH, Munchen-Hallbermoos, Germany) at the start of the experiments and at various time points as indicated. Images were used to determine the percentage remaining wound area as compared to the start of the experiment (t = 0) using the Axiovision interactive measurement module (Carl Zeiss Vision).
Immunoblotting for ERK1/2
PBEC and NCI-H292 cells were cultured to confluence, starved overnight and subsequently stimulated with transforming growth factor (TGF)-α (20 ng/ml) or various concentrations of CSC for 15 minutes, 1, 6 or 24 hours. After washing with washing buffer (5 mM Tris, pH 6.4, 100 mM NaCl, 1 mM CaCl2, 1 mM MgCl2), cells were lysed in ice-cold lysis buffer (0.5% [v/v] Triton X-100, 0.1 M Tris-HCl pH 7.4, 100 mM NaCl, 1 mM MgCl2, 1 mM CaCl2 1 mM Na3VO4, mini complete protease inhibitor cocktail [Roche, Basel, Switzerland]). Following incubation for 10 minutes on ice, cell lysates were centrifuged at 13,000 rpm for 5 minutes at 4°C to remove insoluble debris. Aliquots of the samples containing equal amounts of protein were suspended in reducing SDS-PAGE sample buffer and boiled for 5 minutes. Proteins were separated by 10% SDS-PAGE and transferred to polyvinylidene difluoride (PVDF) membranes using the Mini-transblot system (both Biorad, Hercules, CA). These membranes were incubated in blocking buffer (0.05% Tween-20 in PBS containing 0.5% (w/v) casein) for one hour, followed by overnight incubation with rabbit antibodies directed against total (t) ERK1/2 and phosphorylated (p) ERK1/2 at 4°C (New England Biolabs, Beverly, MA). After incubation with a secondary horseradish peroxidase (HRP) conjugated goat anti-rabbit polyclonal antibody (BD Transduction Laboratories, Franklin Lake, NJ), immunoreactivity was detected by electrochemiluminescent (ECL) detection system (Amersham Pharmacia Biotech, Uppsala, Sweden). In selected experiments, cells were preincubated with the inhibitor of EGFR tyrosine kinase activity AG1478 (Calbiochem, La Jolla, CA).
Immunoblotting for p38 and JNK
Subconfluent cell cultures were starved overnight and stimulated with TGF-α (20 ng/ml) and various concentrations of CSC for 15 minutes, 1, 6 and 24 hours in RPMI 1640 medium containing glutamine, penicillin and streptomycin. After washing with ice-cold PBS, stimulated cells were lysed with reducing sample buffer and incubated for 10 minutes on ice. Proteins were separated by SDS-PAGE using 10% acrylamide gels and proteins were then transferred to nitrocellulose membrane (Schleicher & Schuell GmbH, Dassel, Germany). These were incubated with 0.05% Tween-20 in Tris Buffered Saline (TBST) containing 5% (w/v) skimmed milk (ELK, Campina, Zoetermeer, The Netherlands) for at least one hour, followed by incubation with antibodies directed against total and phosphorylated p38 and JNK at 4°C (New England Biolabs, Beverly, MA), diluted in TBST. After incubation with horseradish peroxidase (HRP) conjugated donkey anti-rabbit polyclonal antibody (Amersham Pharmacia Biotech, UK), immunoreactivity was visualized as described above.
Measurement of cellular GSH content
GSH content of epithelial cells was assessed in cellular lysates that were prepared after washing the cells with ice-cold PBS [19]. Briefly, washed cells were lysed by adding ice-cold lysis buffer (0.6 % [w/v] sulfosalicylic acid, 0.1 % [v/v] Triton X-100, 5 mM EDTA in 0.1 M potassium phosphate buffer, pH 7.5) and incubation for 10 min on ice. Lysates were harvested and cell pellets, obtained after centrifugation, were disrupted using a Teflon pestle followed by vortexing. This solution was cleared by centrifugation, and the GSH content of the supernatant was assessed using the method of Tietze [20]. GSH content was calculated using a standard curve, and expressed as nmol/mg protein. The protein content of the lysates was determined using the bicinchonic acid (BCA) method (Pierce Chemical Co, Rockford, IL). In the experiments where the effect of NAC was assessed, cells were preincubated for 16 hours with NAC.
Statistical analysis
The data are expressed as mean ± SEM. Statistical analysis was performed with Student's t test for paired samples following analysis of variance. Differences were considered statistically significant when p < 0.05.
Results
Effect of cigarette smoke condensate (CSC) on cell proliferation
The effect of CSC on cell proliferation was studied using BrdU incorporation in subconfluent cultures of NCI-H292 bronchial epithelial cells. CSC caused a dose-dependent increase in cell proliferation at low concentrations (0.25 – 1 AU/ml) after 24 hours, whereas higher concentrations decreased cell proliferation (Fig. 1A). Comparable results were obtained using the tetrazolium salt MTT [3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide] assay to assess viable cells by determining mitochondrial activity (data not shown). Preincubation of the cells with the antioxidant NAC (1 mM) markedly reduced the mitogenic effect of low concentrations of CSC (Fig 1B), whereas it partially restored cell proliferation in the presence of 5 or 10 AU/ml CSC (data not shown). In line with this finding, NAC also prevented CSC-induced cytotoxicity, as demonstrated by trypan blue exclusion (data not shown). In contrast, when intracellular GSH was depleted using buthionine sulphoximine (BSO; 10 μM), an inhibitor of glutamate cysteine ligase, GCL (formerly known as γ-glutamylcysteine synthetase, γ-GCS), cell proliferation at mitogenic concentrations of CSC (1 and 0.5 AU/ml) was markedly reduced, whereas proliferation in cells incubated with submitogenic concentrations of CSC (0.125 and 0.25 AU/ml) was increased (Figure 1C). These results demonstrate that modulation of intracellular GSH affects both the mitogenic and the toxic effects of CSC.
Figure 1 Effect of CSC and its modulation by BSO and NAC on cell proliferation in NCI-H292. Subconfluent cultures of NCI-H292 cells were incubated for 24 h with various concentrations of CSC (A), CSC and NAC (B) or preincubated for 16 h with BSO (10 μM), followed by the addition of freshly prepared CSC (C). Next, BrdU was added and the cells were incubated for another 4 h and subsequently washed and fixed. BrdU incorporation was detected by immunocytochemistry (for details see material and methods). Horizontal bars indicate BrdU incorporation observed with TGF-α or medium alone. The results are mean ± SEM of 3 independent experiments, each performed in duplicate. Note the difference in scaling of the x-axes. * p < 0.05; ** p < 0.001; *** p < 0.001 vs medium-treated cells (Fig. 1A) or vs cells exposed to the same concentration of CSC in absence of NAC (Fig. 1B) or BSO (Fig. 1C).
Effect of CSC on epithelial wound repair
Since epithelial cell proliferation plays a central role in epithelial wound closure, we next assessed whether the effects of CSC on epithelial cell proliferation in subconfluent cultures were also reflected by similar findings in a model of epithelial wound closure. Therefore we used a model that we recently developed [18], in which closure of wounds prepared by scraping a defined circular wound in a confluent layer of NCI-H292 cells or PBEC is studied.
In NCI-H292 cells, TGF-α caused a marked increase in wound closure at all time points studied (Figure 2A). At 5 AU/ml, CSC completely inhibited wound closure as a result of cytotoxicity (demonstrated using trypan blue exclusion; data not shown). In contrast, CSC at 1 AU/ml caused a limited, but significant (p = 0.05, at 24 and 48 h) increase in wound closure. Essentially similar results were obtained when studying wound closure in PBEC (Figure 2B), although wounds prepared in PBEC cultures closed faster. Also in PBEC, TGF-α increased wound closure at all time points studied. Whereas at 5 AU/ml, CSC inhibited wound closure (p < 0.002 vs. medium control), at 1 AU/ml a significant increase in wound closure was observed at all time points (p < 0.02 vs. medium control) (Figure 2B).
Figure 2 Effect of CSC on epithelial wound closure in NCI-H292 cells and PBEC. Mechanical wounds were prepared in monolayers of NCI-H292 (panel A) or PBEC (panel B) and the area of the wound was determined at various time points as indicated and used to calculate the % wound closure. Following wounding, the cultures were incubated with freshly prepared CSC (5 or 1 AU/ml; ■, △), TGF-α (▼; 20 ng/ml) or medium alone (◆). The results in panel A (NCI-H292 cells) are mean ± SEM of 6 independent experiments, each performed in triplicate. The results in panel B (PBEC) are mean ± SEM of PBEC cultures derived from 7 different donors. ** P < 0.004; * P < 0.05 vs medium alone
Whereas NAC significantly restored wound closure in both NCI-H292 and PBEC treated with 5 AU/ml CSC, it did not affect wound closure in presence of 1 AU/ml (Figure 3). The same results were obtained after incubating cells with GSH (1.25 – 5 mM) instead of NAC (data not shown). In contrast, in NCI-H292 cells but not in PBEC, depleting GSH using BSO resulted in a full inhibition of wound repair in presence of 1 AU/ml that was accompanied by cytotoxicity (Figure 4). Also higher concentrations of BSO (150 μM) did not affect wound closure in presence of 1 AU/ml CSC (data not shown). These results demonstrate the crucial involvement of oxidants/free radicals in the inhibitory effects of CSC on wound closure, and illustrate the differential sensitivity of NCI-H292 and PBEC to oxidative stress induced by CSC.
Figure 3 Effect of N-acetylcysteine (NAC) on CSC-induced epithelial wound repair in NCI-H292 and in PBEC. Mechanical wounds prepared in monolayers of NCI-H292 (panel A) or PBEC (panel B). Following wounding, the cultures were incubated for 24 h (NCI-H292; panel A) or 8 h (PBEC; panel B) with freshly prepared CSC (5 or 1 AU/ml), TGF-α (20 ng/ml), medium alone, NAC (1 mM) (alone or in combination with CSC). Next the residual wound area was determined and used to calculate the % wound repair. Similar results were obtained when analyzing wound closure at 72 h (data not shown). The results in NCI-H292 cells are mean ± SEM of 3 independent experiments, each performed in triplicate. The results in panel B are mean ± SEM of PBEC cultures derived from 5 different donors. The cultures from the different donors were performed on different days, and each experiment was performed in triplicate. * p < 0.05 vs. medium alone; + p < 0.05 vs. CSC 5 AU/ml
Figure 4 Effect of buthionine sulphoximine (BSO) on CSC-induced epithelial wound closure in NCI-H292 and in PBEC. Mechanical wounds prepared in monolayers of NCI-H292 (panel A) or PBEC (panel B). Before wounding, the cultures were preincubated for 16 hours with BSO, followed by incubation for 72 h (NCI-H292; panel A) or 24 h (PBEC; panel B) with freshly prepared CSC (△; 1 AU/ml), TGF-α (◆; 20 ng/ml), medium alone (●), BSO (alone [▽] or in combination with 1 AU/ml of CSC [■]). Next the residual wound area was determined and used to calculate the % wound repair. The results of both the experiments are mean ± SEM of 3 independent experiments, each performed in triplicate. P < 0.03 vs CSC 1 AU/ml.
When studying epithelial wound repair, essentially no difference was observed between the effect of CSC prepared from commercial brand cigarettes and that from University of Kentucky standard research cigarettes (data not shown). Therefore, all experiments were performed using CSC prepared from commercial brand cigarettes.
Effect of CSC on cell proliferation during epithelial wound closure
The stimulatory effects of CSC on wound closure were less pronounced than their effect on cell proliferation in subconfluent cultures. Since epithelial wound closure is mediated by both cell migration and proliferation, and since it has been described that CSC inhibits epithelial migration [21], we next investigated the effect of CSC on cell proliferation using BrdU incorporation in cells present in the original wound area at different phases of the repair process.
The results revealed that CSC had similar effects on cell proliferation in the original wound area in the wound closure model in NCI-H292 (Table 1), as observed with subconfluent cultures of NCI-H292 (Figure 1). First, the percentage of BrdU positive cells was higher in the wound area compared to cells outside the wound area, irrespective of the conditions tested. Second, TGF-α and 1 AU/ml CSC caused an increase in BrdU incorporation in cells present within and outside the original wound area already after 24 hours (Table 1). No BrdU incorporation was observed in cultures incubated with 5 AU/ml.
Table 1 Cell proliferation in mechanically wounded NCI-H292 cell monolayers: effect of CSC.
% BrdU positive nucleia
Stimuli BrdU incubation period
24–48 h 48–72 h
wound intact wound intact
Medium 21.5 ± 0.7 17.0 ± 1.6 20.9 ± 1.0 12.4 ± 1.4
TGFα 47.3 ± 3.9* 34.9 ± 1.5* 27.5 ± 0.5* 16.4 ± 2.2
NAC 22.5 ± 2.9 16.3 ± 1.8 20.5 ± 1.1 14.4 ± 1.3
CSC 1 AU/ml 39.8 ± 1.0* 29.5 ± 3.8* 42.2 ± 1.6* 24.0 ± 2.3
CSC 1 AU/ml+NAC 25.7 ± 0.7ˆ 17.4 ± 2.7ˆ 21.8 ± 2.5ˆ 16.0 ± 1.8ˆ
aPercentage BrdU-positive nuclei within a distance of 10 cells from the wound edge (i.e. within the original wound area) and in intact areas. Data are mean ± SEM of 3 independent experiments. *: p < 0.05 versus serum-free medium-treated cells. ˆ:p < 0.05 vs CSC 1 AU/ml
In wounded PBEC layers, the percentage of proliferating cells was lower (< 10 % BrdU positive nuclei) than observed in wounded NCI-H292 monolayers. Within one well, the percentage of proliferating PBEC was lower inside the original wound area as compared to outside this area (data not shown). As it appears that cell proliferation does not markedly contribute to wound closure in the PBEC model, the effect of CSC on proliferation in this model was not further explored.
In summary, these data indicate that CSC has dual effects on epithelial cell proliferation by increasing proliferation at low, and decreasing proliferation at high concentrations both in subconfluent layers of epithelial cells and during wound closure.
Effect of CSC on epithelial GSH content
To investigate the effect of CSC on intracellular GSH in the epithelial cells used, the intracellular GSH content of NCI-H292 cells was assessed at different times after exposure to CSC. The results show that CSC causes a time and dose-dependent decrease in GSH, that was partly prevented by preincubation with NAC (Figure 5).
Figure 5 Intracellular GSH in NCI-H292 after exposure to CSC. NCI-H292 cells were exposed to various concentrations of CSC (5, 2.5 and 1.25 AU/ml) in presence or absence of NAC (1 mM), or NAC alone for 1 (open bars), 3 (hatched bars) or 5 h (filled bars). Intracellular GSH content was measured in cellular lysates and expressed as mean percentage ± SEM of that of medium-treated cells. p < 0.05 vs medium-treated cells; GSH content of cells exposed to 5 AU/ml CSC alone or with NAC differed significantly at 3 h (P = 0.04).
Effect of CSC on MAPK activation
Because various members of the MAPK family play different roles in the regulation of cell fate, the effect of CSC on activation of the MAPK ERK1/2, p38 and JNK was explored (Figure 6 and 7 and data summary in table 2). In these experiments, TNF-α was included as a positive control for activation of p38 and JNK. Whereas ERK1/2 activation was observed at all concentrations of CSC both in NCI-H292 (Figure 6A) and PBEC (Figure 6B), activation of p38 and JNK was only observed at the higher concentrations (5 and 10 AU/ml). Furthermore, ERK1/2 activation was already observed at 15 min and persisted for 24 hours, while p38 and JNK activation was maximal at 6 h; no activation was observed after 24 hours (Figure 7). These results indicate that CSC exerts a concentration- and time-dependent effect on the ratio between activated ERK1/2 and p38/JNK, suggesting a shift towards a predominance of p38/JNK activation at higher CSC concentrations following prolonged incubation. In addition, the persistent activation of ERK1/2 in the absence of notable p38/JNK activation observed with 1 AU/ml, is in line with the proposed role of this MAP kinase in cell proliferation. In the presence of NAC, CSC did not induce ERK1/2 activation, indicating that this process involves the action of oxidants/free radicals (Figure 6C). Our observation that the EGFR tyrosine kinase inhibitor AG1478 blocks CSC-induced phosphorylation of ERK1/2 is in line with studies showing a role of EGFR in cigarette smoke induced epithelial cell activation (Figure 6D).
Table 2 Summary of the data on analysis of MAPK ERK1/2, p38 and JNK in NCI-H292 cells following exposure to CSC.
Treatmentb
MAPK Timea medium TGF-α TNF-α CSC (AU/ml)
10 5 1
ERK1/2 15 min - + ± + + +
1 hour - + ± + + +
6 hours - + - + + +
24 hours - + - - - +
p38 10 min - - + + + -
1 hour - - ± + + -
6 hours - - - ++ ++ -
24 hours - - - - - -
JNK 10 min - - - - - -
1 hour - - - + ± -
6 hours - - - ++ + -
24 hours - - - - - -
a Cells were exposed to the various treatments for the indicated time periods. b The degree of phosphorylation of the three MAPK studied is indicated by the symbol.
Figure 6 CSC-induced phosphorylation of Extracellular signal Regulated Kinases (ERK1/2) 1/2. NCI-H292 cells (A and C) were stimulated with medium, TGF-α, and CSC (10, 5, 2 and 1 AU/ml) for 15 minutes, 6 and 24 hours (A) or with medium, TGF-α, and 2 and 1 AU/ml of CSC in presence (C) or absence (A) of NAC. PBEC were stimulated with medium, TGF-α and CSC (10, 5, 2 and 1 AU/ml) for 15 minutes and 24 hours (B). Next cellular lysates were prepared, and used to detect the level of total (t-ERK1/2) and phosphorylated ERK1/2 (p-ERK1/2) using Western blot analysis. The results shown in each panel are from one experiment that was repeated three times with similar results. To assess the role of EGFR in CSC-induced ERK1/2 phosphorylation, NCI-H292 cells were preincubated for 1 hour with 1 μM AG1478, and incubated for 15 minutes or 6 hours with medium alone or with CSC (10 AU/ml) (D).
Figure 7 CSC-induced phosphorylation of p38 and JNK. NCI-H292 cells were stimulated with medium, TGF-α, TNF-α or CSC (10, 5 and 1 AU/ml) for 10 minutes, 1 and 6 hours. Next cellular lysates were prepared and used to detect the level of total (t-p38 and t-JNK) and phosphorylated (p-38 and t-JNK) p38 and JNK using Western blot analysis. The results shown in each panel are from one experiment that was repeated three times with similar results.
Discussion
The results from the present study show a dose-dependent and dual effect of cigarette smoke on bronchial epithelial cell proliferation and wound repair. In cultures of the bronchial epithelial cell line NCI-H292, proliferation was inhibited at high and stimulated at low concentrations of CSC. Similar effects of CSC on epithelial wound closure in NCI-H292 or PBEC supported these results. Experiments using NAC, GSH and BSO to modify the intracellular thiol status, revealed a critical role of oxidants/free radicals in mediating these effects of CSC. Activation of ERK1/2, a MAPK involved in cell proliferation and survival, was increased by various concentrations of CSC, and sustained up to 24 h only at mitogenic concentrations of CSC (1 AU/ml). Higher, cytotoxic concentrations of CSC resulted in activation of the pro-apoptotic MAPK p38 and JNK. These results suggest an involvement of different MAP kinases in CSC-induced cell proliferation and cytotoxicity.
Various studies have demonstrated marked effects of cigarette smoke and its aqueous extracts on epithelial cell behavior, including proliferation and wound repair. However, most of these studies focused on high, cytotoxic concentrations that were found to inhibit proliferation and wound repair in bronchial epithelial cells [22]. In contrast to our observations, Lannan et al. did not observe any increase in proliferation of alveolar A549 epithelial lung adenocarcinoma cells by CSC using 1–10% of CSC [2], which may be a specific feature of these cells or the CSC concentration used. Our results are however in line with the observation that short term exposure of rats to cigarette smoke condensate results in an increase in cell proliferation in the bronchiolar epithelium and the pulmonary vasculature [23]. Furthermore, our conclusions are also supported by studies showing that a broad range of oxidants, other than those present in cigarette smoke, can stimulate epithelial cell proliferation [24-26].
The importance of an oxidant/antioxidant imbalance in regulating both CSC-induced cell death, inhibition of wound repair and mitogenesis was demonstrated in studies using NAC, GSH and BSO. Previous to our study, the importance of GSH in cellular defense against CSC was demonstrated by the ability of both GSH and NAC to protect cells from CSC-induced cell death [2,27]. The inhibitory action of NAC on the effects of CSC may have been the result of both the extracellular scavenging action of NAC or its ability to increase cellular GSH since NAC was present in the culture medium during exposure to CSC. NAC prevented the CSC-induced decrease in GSH, but did not increase GSH in the absence of CSC. This finding is in line with previous observations, showing that NAC does increase total non-protein thiols but does not increase GSH [28,29]. More direct evidence for the involvement of cellular GSH came from our observation that pharmacological inhibition of GCL by BSO increases the sensitivity of epithelial cells to oxidative stress resulting from CSC exposure. This observed effect of BSO may also be relevant for our insight into the way TGFβ expression may alter the response to CSC in the (susceptible) smoker. TGFβ is a pleiotropic cytokine and its expression is higher in smokers with COPD when compared to those without COPD [30]. For the present study it is interesting to note that TGFβ blocks GCL synthesis in cultured epithelial cells [31], and thereby – like BSO, a chemical inhibitor of GCL – may decrease intracellular GSH levels. It needs to be noted that GSH and glutathione S-transferases (GSTs) not only protect cells from the action of oxidants, but also play a more general role in the detoxification of electrophilic components that are present in CSC [32]. Therefore the observed depletion of GSH and modulatory effects of NAC and BSO do not provide definitive proof for an exclusive role of oxidants in the observed effects of CSC. GSH can also directly scavenge the electrophilic compounds present in CSC [19].
What is the mechanism involved in the increase in epithelial wound repair and the proliferative response following exposure to low concentrations of CSC? It is known that oxidants and CSC are able to cause ligand independent transactivation of the epidermal growth factor receptor (EGFR) [14]. Reactive oxygen species may employ EGFR phosphorylation to activate MAPK such as extracellular-signal regulated kinase (ERK) 1/2, that in turn may induce shedding of the EGF receptor ligands such as TGFα and thereby lead to further activation of the EGFR [33]. Our finding that the EGFR tyrosine kinase inhibitor AG1478 blocks CSC-induced ERK1/2 activation is in accordance with these observations. In line with this, reports demonstrated the involvement of ADAM 17 (tumor necrosis factor α converting enzyme; TACE) in shedding of TGFα and amphiregulin from bronchial epithelial cells exposed to suspended smoke particles [34,35]. The present study confirms the stimulatory effect of CSC on epithelial cell proliferation, and links this effect to an imbalance between oxidants and antioxidants and to MAPK activation. Our observation that CSC at mitogenic concentrations induces phosphorylation of ERK1/2 suggests a potential role of this MAPK pathway in the development of both epithelial hyperplasia and metaplasia in smokers, features that may predispose to the development of lung cancer [36,37]. Furthermore, Richter et al. demonstrated that CSC not only induces release but also increases expression of selected EGFR ligands from airway epithelial cells [38]. Taken together these data and our observation on prolonged activation of ERK1/2 following exposure to subtoxic concentrations of CSC may provide a mechanistic basis for the observed stimulatory effect of CSC on cell proliferation and epithelial wound repair in vitro. Future studies are needed to define a functional involvement of ERK1/2 activation in the observed effects of CSC on proliferation and wound closure. Our results are in agreement with a recent study showing that two compounds of cigarette smoke, nicotine and 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone, activate the serine/threonine kinase Akt leading to increased cell survival and tumorigenesis in human airway epithelial cells [39], and a study showing that nicotine induces cell proliferation in neoplastic epithelial cells [40]. The relative contribution of 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone and nicotine or other reactive aldehydes to the effects of CSC observed in the present study is not known, also because it is not clear whether intracellular thiols block the stimulatory effect of these components. Furthermore, we have not explored the role of the serine/threonine kinase Akt in the mitogenic effects of CSC observed in the present study. In addition to the MAPK pathway, the Akt/PI-3 kinase pathway may play an important role in CSC-induced epithelial cell proliferation. Finally, a stimulatory effect of aged suspended smoke particles on cultured human bronchial epithelial cells was recently described [34].
At higher concentrations, CSC has been shown to cause cell death, which seems to be due to high concentrations of oxidants and other radicals [27] and this study). Moderate oxidative stress induces apoptosis, whereas necrosis occurs when cells are exposed to a higher dose of oxidants [41,42]. It has been demonstrated that p38 and JNK are involved in oxidant-induced cell death [43], and cell fate is regulated by a balance between all three MAP kinase pathways [44]. Interestingly, at the higher concentrations tested (10 and 5 AU/ml) we observed activation of both pro-apoptotic (p38 and JNK) and pro-survival/proliferation (ERK1/2) MAPK pathways. Based on the observation that these high concentrations of CSC induce cell death, it appears that activation of pro-apoptotic signals predominates. Studies using inhibition of these separate MAPK signaling pathways are needed to delineate their role in the cellular effects of CSC observed in the present study.
In our epithelial wound repair model, we observed differences between NCI-H292 cells and PBEC that are relevant to the interpretation of the effects of CSC on repair. Following injury in NCI-H292 cells, we observed marked proliferation in the cells that covered the original wound area, indicating a contribution of proliferation to the repair process. In contrast, in PBEC cultures very few proliferating cells were present in the wound area, suggesting that in PBEC repair of wounds of the size used in the present study is mainly mediated by migration or cell spreading. Therefore, a stimulatory effect of CSC on proliferation of PBEC could not be observed. Nevertheless, a small but significant effect of low concentrations of CSC on wound closure of both NCI-H292 and PBEC was observed. This may indicate an effect of CSC on cell migration that is dependent on the concentration of CSC used and the cell type investigated. Previously, an inhibitory effect of toxic concentrations of CSC on epithelial migration was reported [21]. We also observed that PBEC appear more resistant to the cytotoxic effects of CSC than NCI-H292. Because of the role of intracellular GSH in mediating cellular defense against oxidative stress resulting from CSC exposure, we hypothesized that GSH levels may differ between PBEC and NCI-H292. Furthermore, since all PBEC cultures used in the present study were derived from smokers, the possibility that bronchial epithelial cells from smokers may display increased GSH levels needs to be considered [3]. In addition, further studies are required to delineate differences between the effects of CSC on NCI-H292 and PBEC. In our study, we have used an aqueous extract of cigarette smoke to gain insight into the effect of cigarette smoke on epithelial cells. Much of our knowledge on the cellular effects of smoke is based on studies using such smoke extracts instead of smoke. Notwithstanding the inherent limitations of this model, it can be argued that epithelial cells in the lung – like those in our cultures – are also exposed to smoke components that have been extracted into a fluid, i.e. the epithelial lining fluid [45]. The use of the terms "low" and "high" to describe the CSC concentration does not imply any comparison with actual levels of cigarette smoke compounds in lungs. Since the pulmonary levels of individual compounds of cigarette smoke are unknown, comparisons between the in vivo and in vitro situation are difficult to make. However, the CSC concentrations used in this study are comparable to the concentrations used in other in vitro studies.
In our study we have focussed on the effects of cigarette smoke on proliferation and wound closure in cultures of bronchial epithelial cells. These in vitro results may add new elements to our insight into the pathogenesis of smoking-induced lung injury, and more specifically to the epithelial changes that may accompany COPD and chronic bronchitis. Our observation on CSC-induced epithelial cell proliferation suggests that cigarette smoke alone may partly explain the increased amount of proliferating cells observed in bronchial biopsies obtained from smokers [8], and may be relevant for our understanding of mechanisms involved in the epithelial hyperplasia and metaplasia that is frequently observed in smokers with and without airflow limitation [7]. Our findings may also be relevant for our insight in the development of smoking-induced lung cancer, since epithelial metaplasia and hyperplasia induced by cigarette smoke is considered a precancerous lesion [46]. In this respect it is interesting to note that constitutive activation of ERK1/2 may suffice to cause transformation [47], and that carcinoma cells often demonstrate high basal levels of ERK1/2 activation [48]. Therefore, a better understanding of the mechanisms by which cigarette smoke in particular by redox signaling affects wound repair may lead to improved therapeutic interventions for the prevention and treatment of smoking-induced lung disease.
Conclusion
Our study revealed dual effects of cigarette smoke on epithelial cell behavior. Our observation that low concentrations of CSC induce cell proliferation may be relevant for our understanding of epithelial changes in chronic bronchitis and COPD. Furthermore, our observation that different patterns of MAPK activation are associated with CSC-induced proliferation and cell death provides insight into the cellular response to cigarette smoke, and its putative modulation using pharmacological inhibition of the different MAPK pathways.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
FL carried out the cell culture experiments, analysis of MAP kinase activation and wrote the manuscript. JA, SW and IR introduced techniques used in the present study. WIB, KFR and PSH were involved in the design, supervision and writing of the manuscript. All authors have participated in the study design and evaluation, and have read, contributed and approved the manuscript.
Acknowledgements
The authors thank Mrs. Marianne van Sterkenburg (Dept. of Pulmonology) for culturing of human bronchial epithelial cells, Sylvia Lazeroms for performing the GSH studies and our colleagues at the Department of Pathology for their help in collecting bronchial tissue. The study was supported by grants from the Netherlands Asthma Foundation (grants 98.12, 97.55 and 01.27) and a Long Term Research Fellowship from the European Respiratory Society (grant LTRF-2001-026), and the Environmental Health Sciences Center support ES01247 (to I.R.).
==== Refs
Church DF Pryor WA Free-radical chemistry of cigarette smoke and its toxicological implications Environ Health Perspect 1985 64 111 126 3007083
Lannan S Donaldson K Brown D MacNee W Effect of cigarette smoke and its condensates on alveolar epithelial cell injury in vitro Am J Physiol 1994 266 1 Pt 1 L92 100 8304473
MacNee W Rahman I Is oxidative stress central to the pathogenesis of chronic obstructive pulmonary disease? Trends Mol Med 2001 7 2 55 62 11286755 10.1016/S1471-4914(01)01912-8
Rennard SI Inflammation and repair processes in chronic obstructive pulmonary disease Am J Respir Crit Care Med 1999 160 5 Pt 2 S12 16 10556162
Jeffery PK Remodeling in asthma and chronic obstructive lung disease Am J Respir Crit Care Med 2001 164 10 Pt 2 S28 38 11734464
Saetta M Turato G Baraldo S Zanin A Braccioni F Mapp CE Maestrelli P Cavallesco G Papi A Fabbri LM Goblet cell hyperplasia and epithelial inflammation in peripheral airways of smokers with both symptoms of chronic bronchitis and chronic airflow limitation Am J Respir Crit Care Med 2000 161 3 Pt 1 1016 1021 10712357
Trevisani L Sartori S Bovolenta MR Mazzoni M Pazzi P Putinati S Potena A Structural characterization of the bronchial epithelium of subjects with chronic bronchitis and in asymptomatic smokers Respiration 1992 59 3 136 144 1439224
Demoly P Simony-Lafontaine J Chanez P Pujol JL Lequeux N Michel FB Bousquet J Cell proliferation in the bronchial mucosa of asthmatics and chronic bronchitics Am J Respir Crit Care Med 1994 150 1 214 217 7912988
Lee JJ Liu D Lee JS Kurie JM Khuri FR Ibarguen H Morice RC Walsh G Ro JY Broxson A Long-term impact of smoking on lung epithelial proliferation in current and former smokers J Natl Cancer Inst 2001 93 14 1081 1088 11459869 10.1093/jnci/93.14.1081
Rahman I MacNee W Lung glutathione and oxidative stress: implications in cigarette smoke-induced airway disease Am J Physiol 1999 277 6 Pt 1 L1067 1088 10600876
Kurie JM Shin HJ Lee JS Morice RC Ro JY Lippman SM Hittelman WN Yu R Lee JJ Hong WK Increased epidermal growth factor receptor expression in metaplastic bronchial epithelium Clin Cancer Res 1996 2 10 1787 1793 9816131
Vignola AM Chanez P Chiappara G Merendino A Pace E Rizzo A la Rocca AM Bellia V Bonsignore G Bousquet J Transforming growth factor-beta expression in mucosal biopsies in asthma and chronic bronchitis Am J Respir Crit Care Med 1997 156 2 Pt 1 591 599 9279245
Chang L Karin M Mammalian MAP kinase signalling cascades Nature 2001 410 6824 37 40 11242034 10.1038/35065000
Takeyama K Jung B Shim JJ Burgel PR Dao-Pick T Ueki IF Protin U Kroschel P Nadel JA Activation of epidermal growth factor receptors is responsible for mucin synthesis induced by cigarette smoke Am J Physiol Lung Cell Mol Physiol 2001 280 1 L165 172 11133506
Kim HJ Liu X Wang H Kohyama T Kobayashi T Wen FQ Romberger DJ Abe S MacNee W Rahman I Glutathione prevents inhibition of fibroblast-mediated collagen gel contraction by cigarette smoke Am J Physiol Lung Cell Mol Physiol 2002 283 2 L409 417 12114203
van Wetering S van der Linden AC van Sterkenburg MA de Boer WI Kuijpers AL Schalkwijk J Hiemstra PS Regulation of SLPI and elafin release from bronchial epithelial cells by neutrophil defensins Am J Physiol Lung Cell Mol Physiol 2000 278 1 L51 58 10645890
Aarbiou J Ertmann M van Wetering S van Noort P Rook D Rabe KF Litvinov SV van Krieken JH de Boer WI Hiemstra PS Human neutrophil defensins induce lung epithelial cell proliferation in vitro J Leukoc Biol 2002 72 1 167 174 12101277
Aarbiou J Verhoosel RM Van Wetering S De Boer WI Van Krieken JH Litvinov SV Rabe KF Hiemstra PS Neutrophil defensins enhance lung epithelial wound closure and mucin gene expression in vitro Am J Respir Cell Mol Biol 2004 30 2 193 201 12871849 10.1165/rcmb.2002-0267OC
Rahman I Li XY Donaldson K Harrison DJ MacNee W Glutathione homeostasis in alveolar epithelial cells in vitro and lung in vivo under oxidative stress Am J Physiol 1995 269 3 Pt 1 L285 292 7573460
Tietze F Enzymic method for quantitative determination of nanogram amounts of total and oxidized glutathione: applications to mammalian blood and other tissues Anal Biochem 1969 27 3 502 522 4388022 10.1016/0003-2697(69)90064-5
Cantral DE Sisson JH Veys T Rennard SI Spurzem JR Effects of cigarette smoke extract on bovine bronchial epithelial cell attachment and migration Am J Physiol 1995 268 5 Pt 1 L723 728 7762674
Wang H Liu X Umino T Skold CM Zhu Y Kohyama T Spurzem JR Romberger DJ Rennard SI Cigarette smoke inhibits human bronchial epithelial cell repair processes Am J Respir Cell Mol Biol 2001 25 6 772 779 11726404
Sekhon HS Wright JL Churg A Cigarette smoke causes rapid cell proliferation in small airways and associated pulmonary arteries Am J Physiol 1994 267 5 Pt 1 L557 563 7977766
Ohguro N Fukuda M Sasabe T Tano Y Concentration dependent effects of hydrogen peroxide on lens epithelial cells Br J Ophthalmol 1999 83 9 1064 1068 10460777
Dal-Pizzol F Klamt F Dalmolin RJ Bernard EA Moreira JC Mitogenic signaling mediated by oxidants in retinol treated Sertoli cells Free Radic Res 2001 35 6 749 755 11811526
Gotoh Y Noda T Iwakiri R Fujimoto K Rhoads CA Aw TY Lipid peroxide-induced redox imbalance differentially mediates CaCo-2 cell proliferation and growth arrest Cell Prolif 2002 35 4 221 235 12153614 10.1046/j.1365-2184.2002.00241.x
Hoshino Y Mio T Nagai S Miki H Ito I Izumi T Cytotoxic effects of cigarette smoke extract on an alveolar type II cell-derived cell line Am J Physiol Lung Cell Mol Physiol 2001 281 2 L509 516 11435227
Rahman I Mulier B Gilmour PS Watchorn T Donaldson K Jeffery PK MacNee W Oxidant-mediated lung epithelial cell tolerance: the role of intracellular glutathione and nuclear factor-kappaB Biochem Pharmacol 2001 62 6 787 794 11551525 10.1016/S0006-2952(01)00702-X
Mulier B Rahman I Watchorn T Donaldson K MacNee W Jeffery PK Hydrogen peroxide-induced epithelial injury: the protective role of intracellular nonprotein thiols (NPSH) Eur Respir J 1998 11 2 384 391 9551743 10.1183/09031936.98.11020384
de Boer WI van Schadewijk A Sont JK Sharma HS Stolk J Hiemstra PS van Krieken JH Transforming growth factor beta1 and recruitment of macrophages and mast cells in airways in chronic obstructive pulmonary disease Am J Respir Crit Care Med 1998 158 6 1951 1957 9847291
Jardine H MacNee W Donaldson K Rahman I Molecular mechanism of transforming growth factor (TGF)-beta1-induced glutathione depletion in alveolar epithelial cells. Involvement of AP-1/ARE and Fra-1 J Biol Chem 2002 277 24 21158 21166 11912197 10.1074/jbc.M112145200
Reddy S Finkelstein EI Wong PS Phung A Cross CE van der Vliet A Identification of glutathione modifications by cigarette smoke Free Radic Biol Med 2002 33 11 1490 1498 12446206 10.1016/S0891-5849(02)01079-1
Fan H Derynck R Ectodomain shedding of TGF-alpha and other transmembrane proteins is induced by receptor tyrosine kinase activation and MAP kinase signaling cascades Embo J 1999 18 24 6962 6972 10601018 10.1093/emboj/18.24.6962
Lemjabbar H Li D Gallup M Sidhu S Drori E Basbaum C Tobacco smoke-induced lung cell proliferation mediated by tumor necrosis factor alpha-converting enzyme and amphiregulin J Biol Chem 2003 278 28 26202 26207 12711607 10.1074/jbc.M207018200
Shao MX Nakanaga T Nadel JA Cigarette smoke induces MUC5AC mucin overproduction via tumor necrosis factor-alpha-converting enzyme in human airway epithelial (NCI-H292) cells Am J Physiol Lung Cell Mol Physiol 2004 287 2 L420 427 15121636 10.1152/ajplung.00019.2004
Papi A Casoni G Caramori G Guzzinati I Boschetto P Ravenna F Calia N Petruzzelli S Corbetta L Cavallesco G COPD increases the risk of squamous histological subtype in smokers who develop non-small cell lung carcinoma Thorax 2004 59 8 679 681 15282388 10.1136/thx.2003.018291
Skillrud DM Offord KP Miller RD Higher risk of lung cancer in chronic obstructive pulmonary disease. A prospective, matched, controlled study Ann Intern Med 1986 105 4 503 507 3752756
Richter A O'Donnell RA Powell RM Sanders MW Holgate ST Djukanovic R Davies DE Autocrine ligands for the epidermal growth factor receptor mediate interleukin-8 release from bronchial epithelial cells in response to cigarette smoke Am J Respir Cell Mol Biol 2002 27 1 85 90 12091250
West KA Brognard J Clark AS Linnoila IR Yang X Swain SM Harris C Belinsky S Dennis PA Rapid Akt activation by nicotine and a tobacco carcinogen modulates the phenotype of normal human airway epithelial cells J Clin Invest 2003 111 1 81 90 12511591 10.1172/JCI200316147
Cattaneo MG Codignola A Vicentini LM Clementi F Sher E Nicotine stimulates a serotonergic autocrine loop in human small-cell lung carcinoma Cancer Res 1993 53 22 5566 5568 8221698
Wickenden JA Clarke MC Rossi AG Rahman I Faux SP Donaldson K MacNee W Cigarette smoke prevents apoptosis through inhibition of caspase activation and induces necrosis Am J Respir Cell Mol Biol 2003 29 5 562 570 12748058 10.1165/rcmb.2002-0235OC
Dypbukt JM Ankarcrona M Burkitt M Sjoholm A Strom K Orrenius S Nicotera P Different prooxidant levels stimulate growth, trigger apoptosis, or produce necrosis of insulin-secreting RINm5F cells. The role of intracellular polyamines J Biol Chem 1994 269 48 30553 30560 7982974
Moreno-Manzano V Ishikawa Y Lucio-Cazana J Kitamura M Suppression of apoptosis by all-trans-retinoic acid. Dual intervention in the c-Jun n-terminal kinase-AP-1 pathway J Biol Chem 1999 274 29 20251 20258 10400643 10.1074/jbc.274.29.20251
Makin G Dive C Apoptosis and cancer chemotherapy Trends Cell Biol 2001 11 11 S22 26 11684438 10.1016/S0962-8924(01)02124-9
Rennard SI Cigarette smoke in research Am J Respir Cell Mol Biol 2004 31 5 479 480 15494467 10.1165/rcmb.F284
Trichopoulos D Mollo F Tomatis L Agapitos E Delsedime L Zavitsanos X Kalandidi A Katsouyanni K Riboli E Saracci R Active and passive smoking and pathological indicators of lung cancer risk in an autopsy study Jama 1992 268 13 1697 1701 1527879 10.1001/jama.268.13.1697
Mansour SJ Matten WT Hermann AS Candia JM Rong S Fukasawa K Vande Woude GF Ahn NG Transformation of mammalian cells by constitutively active MAP kinase kinase Science 1994 265 5174 966 970 8052857
Brognard J Dennis PA Variable apoptotic response of NSCLC cells to inhibition of the MEK/ERK pathway by small molecules or dominant negative mutants Cell Death Differ 2002 9 9 893 904 12181740 10.1038/sj.cdd.4401054
|
16309548
|
PMC1315328
|
CC BY
|
2021-01-04 16:36:25
|
no
|
Respir Res. 2005 Nov 25; 6(1):140
|
utf-8
|
Respir Res
| 2,005 |
10.1186/1465-9921-6-140
|
oa_comm
|
==== Front
Saline Syst
Saline Syst
Saline Systems
1746-1448
BioMed Central
1746-1448-1-10
16297237
10.1186/1746-1448-1-10
Review
Saline systems of the Great Plains of western Canada: an overview of the limnogeology and paleolimnology
Last William M [email protected]
Ginn Fawn M [email protected]
1 Department of Geological Sciences, University of Manitoba Winnipeg, R3T 2N2 Canada
2 Department of Microbiology, University of Manitoba, R3T 2N2 Canada
2005
18 11 2005
1 1010
6 7 2005
18 11 2005
Copyright © 2005 Last and Ginn; licensee BioMed Central Ltd.
2005
Last and Ginn; licensee BioMed Central Ltd.
https://creativecommons.org/licenses/by/2.0/ This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
In much of the northern Great Plains, saline and hypersaline lacustrine brines are the only surface waters present. As a group, the lakes of this region are unique: there is no other area in the world that can match the concentration and diversity of saline lake environments exhibited in the prairie region of Canada and northern United States. The immense number of individual salt lakes and saline wetlands in this region of North America is staggering. Estimates vary from about one million to greater than 10 million, with densities in some areas being as high as 120 lakes/km2.
Despite over a century of scientific investigation of these salt lakes, we have only in the last twenty years advanced far enough to appreciate the wide spectrum of lake types, water chemistries, and limnological processes that are operating in the modern settings. Hydrochemical data are available for about 800 of the lake brines in the region. Composition, textural, and geochemical information on the modern bottom sediments has been collected for just over 150 of these lakes. Characterization of the biological and ecological features of these lakes is based on even fewer investigations, and the stratigraphic records of only twenty basins have been examined.
The lake waters show a considerable range in ionic composition and concentration. Early investigators, concentrating on the most saline brines, emphasized a strong predominance of Na+ and SO4-2 in the lakes. It is now realized, however, that not only is there a complete spectrum of salinities from less than 1 ppt TDS to nearly 400 ppt, but also virtually every water chemistry type is represented in lakes of the region. With such a vast array of compositions, it is difficult to generalize. Nonetheless, the paucity of Cl-rich lakes makes the northern Great Plains basins somewhat unusual compared with salt lakes in many other areas of the world (e.g., Australia, western United States). Compilations of the lake water chemistries show distinct spatial trends and regional variations controlled by groundwater input, climate, and geomorphology. Short-term temporal variations in the brine composition, which can have significant effects on the composition of the modern sediments, have also been well documented in several individual basins.
From a sedimentological and mineralogical perspective, the wide range of water chemistries exhibited by the lakes leads to an unusually large diversity of modern sediment composition. Over 40 species of endogenic precipitates and authigenic minerals have been identified in the lacustrine sediments. The most common non-detrital components of the modern sediments include: calcium and calcium-magnesium carbonates (magnesian calcite, aragonite, dolomite), and sodium, magnesium, and sodium-magnesium sulfates (mirabilite, thenardite, bloedite, epsomite). Many of the basins whose brines have very high Mg/Ca ratios also have hydromagnesite, magnesite, and nesquehonite. Unlike salt lakes in many other areas of the world, halite, gypsum, and calcite are relatively rare endogenic precipitates in the Great Plains lakes. The detrital fraction of the lacustrine sediments is normally dominated by clay minerals, carbonate minerals, quartz, and feldspars.
Sediment accumulation in these salt lakes is controlled and modified by a wide variety of physical, chemical, and biological processes. Although the details of these modern sedimentary processes can be exceedingly complex and difficult to discuss in isolation, in broad terms, the processes operating in the salt lakes of the Great Plains are ultimately controlled by three basic factors or conditions of the basin: (a) basin morphology; (b) basin hydrology; and (c) water salinity and composition. Combinations of these parameters interact to control nearly all aspects of modern sedimentation in these salt lakes and give rise to four 'end member' types of modern saline lacustrine settings in the Great Plains: (a) clastics-dominated playas; (b) salt-dominated playas; (c) deep water, non-stratified lakes; and (d) deep water, "permanently" stratified lakes.
==== Body
pmcIntroduction
"The scientific exploration of North American salt lakes was relatively slow off the mark." [1]
The geoenvironmental examination of lakes had its origin during the latter part of the nineteenth century with classic works on the Great Basin lakes of western United States [2-4] and central Europe [5,6]. Although these early landmark studies were directly responsible for the formulation and development of many of our currently-accepted ideas about such well known geological processes as turbidity flow, evaporative concentration and mineral precipitation, and deltaic/shoreline sedimentation, progress on the study of lacustrine sediments, as a distinct sedimentary entity, throughout most of the twentieth century was slow. As recent as just a few decades ago, the status of lake sediment research was equated to that of the hole in a doughnut [7].
Interest in lacustrine geological processes and lake deposits increased dramatically beginning in the 1970's [8-10]. This geoscientific involvement with lakes is attributed to two factors: (i) the recognition that lake sediments provide a source of valuable industrial minerals and fossil fuels [11-14], and (ii) the increased use of inorganic components of lake sediments to monitor pollution, decipher environmental changes, and deduce past climatic and hydrological conditions [15-18].
Geolimnology, a term introduced by Professor J. T. Teller [19,10], is the study and interpretation of physical, biological, geochemical, and hydrogeological processes in lakes and the sedimentological records of lacustrine basins. During the past fifteen years there has been rapid advance in our understanding of the physical and chemical processes operating in lakes and how these processes apply to the stratigraphic sequences preserved in lacustrine basins [20-23]. It is now generally accepted that probably no other continental setting has as much to offer in terms of potential significant contributions to the Earth sciences as the lake environment [24-27]. Paralleling this dramatic increase in growth in geolimnology, the subdiscipline of paleolimnology has also seen an explosion of interest and widespread application [28-30]
In this paper we wish to introduce and provide an overview of recent advances in our geolimnological understanding of the wide assortment of modern lacustrine environments in the northern Great Plains of western Canada (Figure 1). We hope this overview will also help establish a framework for future limnological, limnogeological, and paleolimnological research efforts on the Holocene sedimentary records preserved in lakes in this large geographic region of North America. Within the space limitations of this paper it is important to note that emphasis is placed on saline lakes and saline lake sediments, and on extant lakes and their sediment records. Comprehensive reviews of the biological aspects of the lakes in this region are provided elsewhere [31-39]. Likewise, the sedimentology, chronology, history and development of extinct (mainly glacial and proglacial) lakes and wetlands are summarized in numerous other publications [e.g., [40-45]]. Unfortunately, with the notable exception of Lake Winnipeg [46-48], there has been little geolimnological research on the few but interesting freshwater basins.
Figure 1 Northern Great Plains, western Canada. Location map of the northern Great Plains of western Canada (yellow shading) and the provinces of Manitoba, Saskatchewan, and Alberta.
Setting and physical background for geolimnology on the northern Great Plains
"...in the central part of the continent there is a region, desert, or semi-desert in character, which can never be expected to become occupied by settlers..." [49].
The northern Great Plains is a land of many contrasts: rolling prairies, deeply incised river valleys, flat, featureless lake plains, sand hills and dunes, hummocky and hilly topography, well-treed uplands and barren, shade-less, wind-swept plains. Although the physical, geological, and climatic setting of the region has been summarized in many other papers and volumes (e.g., [50-60]), it is useful to emphasize here several of the most important characteristics of the region that impart the distinct and often unique geolimnological features, namely geomorphology, climate, and geology.
Geomorphology
"A great untimbered, level, dried-up sea of land." [61]
The northern Great Plains physiographic province of Canada stretches from the Precambrian Shield immediately east of Winnipeg, Manitoba, westward for about 1600 km to the Foothills of the Rocky Mountains, and northward some 500 km from the United States-Canada border (Figures 1, 2, and 3). This region is characterized by hummocky to gently rolling topography interspersed with numerous deep, often terraced valleys that have been cut by glacial meltwater.
Figure 2 Eastern portion of northern Great Plains, western Canada. Shaded relief topography map of the eastern portion of the northern Great Plains of western Canada showing the major geomorphic features and lakes identified and discussed within the text.
Figure 3 Western portion of northern Great Plains, western Canada. Map of the western portion of the northern Great Plains of western Canada. The numbers refer to salt lakes identified or discussed within the text. 1: Ingebright Lake; 2: Deadmoose Lake; 3: Crater Lake; 4: Howe Lake; 5: Muskiki Lake; 6: Bitter Lake; 7: Freefight Lake; 8: Little Manitou Lake; 9: Ceylon Lake; 10: Waldsea Lake; 11: Chappice Lake; 12: Snakehole Lake; 13: Whiteshore Lake; 14: Vincent Lake; 15: Verlo Lake; 16: Corral Lake; 17: Lydden Lake; 18: Metiskow Lake. The brown shaded area (excluding the Cypress Hills) is the driest part of the region, defined by the Brown Chernozemic Soil type. The red line delineates the Palliser Triangle, a region defined by John Palliser during his 19th century expeditions and approximately coinciding with the transition zone between grassland (prairie) and aspen parkland vegetation.
Within the region, locally separate and distinct sub zones or smaller geomorphological units can be identified. The low (<300 m elevation), often swampy and wetland area of the Manitoba Lowland extends westward from the Precambrian Shield to the Manitoba Escarpment. This area contains the largest lakes in the Great Plains: Lakes Winnipeg, Manitoba, Winnipegosis, Cedar, and Dauphin (Figure 2). With their large drainage basins, generally high sedimentation rates, and early association with extinct giant proglacial lakes, the sediments in these basins reveal long but exceedingly complex developmental histories that are affected by tectonics, evolving landscapes, variable fluvial inputs and regional climate fluctuations.
Extending westward from about the Manitoba border is the Saskatchewan Plains region or sometimes referred to as the Second Prairie Level. This is a large area of gentle relief between about 450 and 700 m elevation containing a very large number of mainly small and shallow lakes (Figure 4). Estimates range from 4 to 10 million lakes and wetlands in this region [62,39] and densities as high as 90–120 lakes km-2 in some localities [63,64]. The Saskatchewan River Lowlands in the east give way to the greater topographic diversity of the Central Saskatchewan Plains and the various Uplands areas.
Figure 4 Examples of high density of saline lake basins in the northern Great Plains. Top: An aerial image of southern Saskatchewan (area shown approximately 135 km2) showing the large number of small lake basins. Bottom: An aerial image of a section of the Dirt Hills area in southern Saskatchewan showing a large number of lakes associated with the hummocky topography of this ice-thrust moraine feature.
Separating the Second Prairie Level from the western portion of the geomorphic province is the Missouri Coteau and its eastward-facing edge, the Missouri Escarpment. The Missouri Coteau is a distinct, 50 to 100 km wide band of knob and kettle topography that extends for over 1200 km through the Great Plains from central South Dakota northwestward into west-central Saskatchewan [65]. The terrain in this region is rougher and the elevation (550 to 1400 m) greater than in the Central Saskatchewan Plains to the east. In some areas "badlands" topography has developed with local relief being as much as 150 m. The Coteau is an important geomorphic feature of the Plains and contains many brackish and to hypersaline lakes (Figure 5).
Figure 5 Landsat image of southern Saskatchewan. Landsat image of a portion of southern Saskatchewan showing the Missouri Coteau immediately west of glacial Lake Regina. Nearly all lakes show in this image are saline. Note the abundance of lakes west of the Coteau and the strong dominance of riverine basin morphologies in this area of the northern Great Plains.
Finally, the Alberta Plains (or Third Prairie Level) continue westward from western Saskatchewan to meet the Foothills of the Rocky Mountains. Although this western part of the Great Plains contains relatively fewer lake basins, nonetheless, several of the best-studied Holocene lacustrine stratigraphic sequences occur here.
Climate
"The Great Plains are a region of temperature and precipitation extremes: a decidedly continental climate that nearly defies generalization." [52]
Long-term (>100-year) temperature, precipitation, and other climatic records for the region exist for Winnipeg, Brandon, Indian Head, Regina, Swift Current, Medicine Hat Edmonton, and Calgary, with shorter records for some 200 other climatic stations in the northern Great Plains. In general terms, the northern Great Plains experience a cold continental, sub-humid to semi-arid steppe climate. Stable, high pressure continental and Arctic air masses dominate during the winter months giving the region its characteristic cold, clear weather. Most of the region is south of the mean path of winter low pressure systems, but pressure and temperature gradients associated with these systems often lead to the area being influenced by high winds. Continental Arctic and Polar air masses dominate the summer weather, resulting in generally warm and dry conditions. Mean daily temperature during January over most of the region is about -18°C and during July it is 19°C; the mean annual temperature shows a narrow range from 1.1°C to 2.9°C [66]. However, the most important characteristic of the region in terms of temperature is its extreme variability. There are wide variations in temperature between seasons, between years, and between day and night. This temperature variability has a significant impact on many chemical and physical aspects and processes of the lakes in the region.
In addition to temperature, another important climatic factor influencing the geolimnology of the region is the high evaporation to precipitation ratio. The region receives about 40 cm of precipitation per year, whereas as much as 1.5 m of water can be lost annually through evaporation from open water bodies [66]. This annual moisture deficit is one of the major variables that help to impart and control the characteristically high salinities of water in most of the lakes.
Wind is also important in dictating the processes operating in the lakes. Although most (but not all) of the lakes form a winter ice cover, during much of the ice-free season, the average wind speed is moderate to high and mainly from the west and southwest directions. In addition to greatly aiding evaporation of water from the lakes, wind plays a major role in current and wave generation and, hence, sediment deposition and erosion [e.g., [67]]. Wind has also been shown to cause significant local variation in sedimentary facies patterns within basins, and can be an important agent of transport of clastic sediment and salts into or out of the lakes [68-70].
Hydrology and Geology
"We passed during the day many salt lakes, fringed round the edges with thick encrustations of salt, highly indicative of the rapid evaporation that takes place in these arid regions. (J. Palliser, October, 1857, on the area north and west of present-day Lake Diefenbaker) [71]
Probably the most significant factor influencing the nature, distribution, and sedimentary characteristics of the lakes is the presence of large areas of internal drainage in the northern Great Plains. Because of its lack of integrated drainage, the Missouri Coteau contains many individual closed basins. Further large areas of internal drainage also exist east of Saskatoon, west and north of Swift Current and in eastern and central Alberta (Figure 6). Together, these basins comprise one of the largest and best studied areas of endorheic drainage in North America [72-77].
Figure 6 Closed basins, northern Great Plains. Map showing major drainage systems of the northern Great Plains of western Canada and the areas of closed (internal) drainage.
The lack of integrated drainage patterns throughout such a large region makes precise definition of the various watersheds and drainage divides somewhat difficult. The areas of the northern Great Plains that are not characterized by closed basins are drained by three major river systems. The Saskatchewan system originates in western Alberta and, together with the Qu'Appelle-Assiniboine system, drains southern Saskatchewan to the east into Lakes Manitoba and Winnipeg, and, ultimately, north into Hudson Bay. Runoff in the southern-most parts of Saskatchewan and Alberta is directed south into the Missouri River system and onward to the Gulf of Mexico.
The Canadian Plains region is underlain by nearly horizontal Phanerozoic sedimentary rocks of thicknesses up to 5000 m. The Paleozoic section consists mainly of a series of stacked carbonate-evaporite cycles, whereas the overlying Mesozoic and Cenozoic bedrock is dominantly a sand-shale sequence. Dissolution of the highly soluble Paleozoic evaporites by groundwater has modified the relatively simple structural relationships of the flat-lying formations and has created collapse structures over much of the area [78,79]. Several authors have suggested this evaporite dissolution has provided a source of ions for the many salt lakes of the region [80-85].
The bedrock surface has also been strongly modified by preglacial erosion. By the start of the Quaternary Period a mature, dendritic drainage pattern had been established over much of the northern Great Plains [86-88]. The most important of these fluvial channels are the Hatfield, Tyner, Battleford, Swift Current and Estevan Valleys. In general, today's streams reflect this ancestral pattern, except that much of the upper Missouri River flowed northeastward into Hudson Bay rather than into the Mississippi River basin [89,90]. This ancient dendritic drainage morphology is important in helping localize and channel groundwater and therefore is a major factor in salt lake occurrence in the Prairies.
The bedrock is mantled by unconsolidated Quaternary sediment, which is over 300 m thick in places [41]. These deposits consist of till, fluvial sands and gravels, and lacustrine silts and clays. During deglaciation, meltwater from the retreating glacier carved numerous ice-marginal channels and spillways in this sediment [40,91,92]. Although now abandoned or buried under more recent sediment, these valleys often form modern lake basins. The hydrodynamic properties of the Quaternary fill in the valleys influence to a major degree the character and composition of Holocene sediments in these lacustrine basins by controlling the direction of flow and quantity of groundwater discharge [93-97].
Groundwaters play a pivotal role in the geolimnology of this region. As summarized elsewhere [10,98-108], subsurface water compositions in the region are of several main types. Most of the groundwater in unconsolidated "surficial" aquifers is of low to moderate salinity (<3000 ppm total dissolved solids) and dominated by Ca, Mg, and HCO3 ions. In the areas of lowest precipitation, shallow drift groundwater is usually dominated by the SO4 ion rather than HCO3. The shallow bedrock aquifers (Upper Cretaceous and younger rocks) are mainly Na-HCO3 in southern Alberta, Ca-Mg-Na-SO4 in Saskatchewan, and Ca-Mg-Na-HCO3 in western Manitoba. The deeper Paleozoic and Cenozoic bedrock contains higher salinity water (up to 300 ppt TDS) that is usually dominated by Na and Cl ions. The variable input of groundwater from these sources is one of the most significant factors in dictating the brine composition of the lakes at the surface [109].
Origin of the lake basins
"Lakes arise from phenomena that are almost entirely geologic in nature. Once formed, they are doomed. Because of the concave nature of basins, there is a compulsory trend toward obliteration as they fill with sediments... Enormous and deep lakes may be far from death as a result of shoaling, but climatic changes or geologic events leading to desiccation or drainage eventually mark their ends." [110].
Like the vast majority of lacustrine basins in north-temperate regions of the continent, a glacial origin for most of the lakes of the Canadian Great Plains is evident considering the fact almost the entire area was glaciated during ~23–14 ka. Although only a few lake basins can be attributed to gouging, scraping or scouring action of the glacial ice, many have their origins intimately associated with deglaciation processes. A complex but reasonably clear picture of Late Wisconsinan ice retreat in the Prairie region has emerged over the past several decades [111,41,42]. Ponding of meltwater against the retreating ice margin, due to the regional northward slope and differential isostatic depression, lead to the formation of large proglacial lakes such as Lake Regina, Lake Hind, Lake Saskatchewan, and Lake Agassiz. While few in number, present-day remnants of these proglacial lacustrine basins, such as Lakes Manitoba and Winnipeg in Manitoba (Figure 7) and the Quill Lakes complex in Saskatchewan (Figure 2), are clearly important sources of late Pleistocene and early Holocene paleoenvironmental information. These remnant basins, often simply large but shallow low spots in the glacial deposits, are usually surrounded by old strandlines and glaciolacustrine sediments.
Figure 7 Large lakes of the Great Plains. Landsat image showing the Manitoba Lowlands and the presence of large lakes at the extreme eastern side of the northern Great Plains. The Great Plains extend some 1500 km westward from Lake Winnipeg to the foothills of the Rocky Mountains. The large lake basins shown in the image are remnants of glacial Lake Agassiz.
As noteworthy as these extensive ice-marginal lakes were, however, only a small number of the millions of extant lakes in the region have been shown to be direct remnants. Instead, the majority of lacustrine basins in the northern Great Plains are the result of stagnant or dead-ice glacial processes which were not sedimentologically directly related to ice-contact precursor lakes. Slow melting of ice buried beneath a thick superglacial drift blanket resulted in creation of a variety of irregular depressions, ice disintegration trenches, kettles, sinkholes, and donuts, as well as the poorly integrated (i.e., topographically closed) drainage which characterizes much of the region [112,10,65,114]. These ice-stagnation basins tend to be small and circular but some have great thicknesses of Holocene and late Pleistocene lacustrine clastics and salts. For example, Ingebright Lake in southwestern Saskatchewan (Figure 8), today a hypersaline playa basin of less than 1 km2 area, contains in excess of 40 m of Holocene salts and clays [115,116]. Similarly, the Deadmoose Lake basin, east of Saskatoon, contains several anomalous troughs and circular depressions up to 50 m deep, which are likely due to ice-block meltout [117,59]
Figure 8 Ingebright Lake, Saskatchewan. Aerial photograph of Ingebright Lake, Saskatchewan. This lake host the largest NaSO4 deposit in North America (see also Figure 11) and contains in excess of 40 meters of continuous Holocene salt. The depths in meters refer to thickness of salt. North is toward the top of the photo.
In addition to these hollows on the landscape created by stagnant ice melting, many lake basins in the Great Plains have an obvious fluvial origin as evidenced by their long, linear, riverine morphologies. Like the meltout structures, most of these river-carved basins were created during late Pleistocene deglaciation between about 15 and 11 ka, but some have been shown to occupy older drainage valleys [115].
A number of lake basins in the Plains region owe their existence to several unusual origins. Seismic geophysics and drilling [118,119] confirm that Crater Lake, a small, circular lake about 6 m deep located near Yorkton, Saskatchewan, occupies a collapse chimney created by dissolution of the Prairie Evaporite, some 900 m below the surface. Howe Lake near Wynyard, Saskatchewan, and several other basins and wetlands in North Dakota originated by the process of "hydrodynamic blowout" [[118,120] referenced in [121-123]]. These basins were created when meltwater from the retreating glacier was able to over pressure a shallow groundwater aquifer. The high pressure artesian water exited through a small opening to the surface, in the case of Howe Lake probably a fracture system related to a salt solution-collapse structure. Initially, the extreme pressure was sufficient to expel particles and excavate a basin. Clearly the limnology, sediment composition and overall ecology of the lake in these types of basins are controlled, to a major degree, by the dynamics of the continued groundwater flow and the composition of the groundwater solution. In Howe Lake for example, the groundwater aquifer (sandstones of the Cretaceous Mannville Group) contains freshwater, thus Howe Lake is anomalously fresh despite its closed basin and high evaporation/precipitation ratios. In the case of Kelley Slough, Salt Lake, and Lake Ardoch in North Dakota, the groundwater aquifer contains saline brines and these lakes are therefore anomalously saline, in this case in spite of a relatively humid climatic setting.
Finally, tectonic features associated with glacial thrusting of previously deposited glacial sediments and bedrock are well known in the northern Great Plains. Lake basins can be created both within the irregular ridge and furrow topography of the deformed thrust blocks (see, for example, Figure 13 in [124] and Figure 3 in [125]).
Economic geology
"The brine...is ladled into the kettles, and the salt scooped out as it forms, and allowed to remain for a short time to drain before it is packed in birch bark roggins for transportation to Red river, where it commands twelve shilling sterling a bushel.... The brine is very strong. From one kettle two bushels of salt can be made in one day in dry weather." [126].
The lakes and wetlands of the northern Great Plains serve a great variety of uses. Many studies have documented the importance of these terrestrial environments on surface runoff and flow stabilization, erosion control, waste assimilation, agriculture, irrigation, and wildlife habitat [121,38,39,128-130]. One of the most important economic aspects of the lakes is they are a source of valuable industrial materials, minerals, and compounds (Figure 9).
Figure 9 Salt lake industrial minerals, northern Great Plains. Map showing the locations and size of lacustrine industrial mineral reserves of the northern Great Plains. The category of reserves listed as 'other' includes halite, silicate, clay, gypsum, and aggregate resources.
Exploitation of the lakes in western Canada probably started well before the arrival of Europeans. Journals and diaries of nineteenth century European settlers commonly refer to Aboriginal use of the lacustrine salts and brines for medicinal purposes, tanning, and food preservation. These salts also provided the basis for several of earliest commercial industrial efforts on the northern Great Plains [126,131]. Large-scale mineral production from the lakes began in 1918 (Figure 10) with the extraction of magnesium and sodium sulfates and carbonates from Muskiki Lake near Saskatoon [80]. Production of anhydrous sodium sulfate (salt cake) from some 20 different lakes (Figure 11) gradually increased over the next five decades to a high of approximately 700,000 tonnes in 1973. Today, the region supplies nearly 50% of the total North American demand for sodium sulfate, with the rest coming from deposits in southwestern United States and as artificial by products from various manufacturing processes. A large increase in the price of salt cake during 1973–1975 (from $15 to $48 tonne-1) and again during 1980–83 (from $62 to $108 tonne-1) saw a renewed interest in leasing and mining activities in the region during these periods. Despite softening markets and production declines during the last several years, price stabilization at about $90 tonne-1 has lead to an average of about $30,000,000 worth of sodium sulfate produced annually from the lakes [132,332].
Figure 10 Sodium sulfate production from salt lakes in the northern Great Plains of western Canada. Historical changes in value of Na2SO4 (in Canadian dollars) (upper part of figure) and tonnage (lower part of figure) mined from saline lakes in Alberta and Saskatchewan from 1906 to 2000 [59].
Figure 11 Na2SO4 reserves. Reserves of anhydrous sodium sulfate in the twelve largest lacustrine deposits in the northern Great Plains of western Canada. Ingebright and North Ingebright deposits have been separated but geologically are one contiguous deposit.
Historically, the two largest uses of sodium sulfate have been in producing kraft paper and allied products, and in the manufacture of detergents [133,85]. More recently, however, the energy industry has been consuming larger amounts of the salt by its use as a conditioner to facilitate fly ash suppression in coal burning power plants [134]. Another new use of salt cake is in the manufacture of potassium sulfate by the reaction of Na2SO4 with KCl [135-137]. Other potentially significant applications include use in glass, ceramic, and paint manufacture, and in solar energy collectors.
The sodium sulfate industry is based on reserves of three basic types [138,332]: (a) the sodium sulfate that is dissolved in the lake water, (b) the hydrated sodium sulfate mineral mirabilite (Na2SO4 10H2O) which occurs seasonally on the floor of the salt lakes due to precipitation within the overlying water column, and (c) the bedded salts composed mainly of mirabilite and thenardite (Na2SO4) that make up the Holocene sedimentary fill in some basins. Although each of these three sources has been exploited during the past eighty years, most production today is from the hypersaline lake waters, which are pumped from the basin into holding reservoirs (Figure 12). Upon further concentration by evaporation during summer and then cooling during the fall season, mirabilite (known to the miners as Glauber's salt) is precipitated from the solution. The overlying brine is then drained back into the lake basin, and the salt is removed to stockpiles. Solution mining using hot water and dredge mining of the permanent salt beds have also been used [139].
Figure 12 The lacustrine sodium sulfate industry of the northern Great Plains. Cartoon showing the industrial processing of lacustrine salts (mainly sodium and magnesium sulfates) from the northern Great Plains of western Canada.
Figure 13 Salt lake morphology. Size classification of saline lake basins in the Great Plains of western Canada.
This harvested Glauber's salt (Na2SO4·10H2O) must be dehydrated prior to marketing. The methods for this processing vary considerably [140,133,94]. Most producers simply raise the temperature of the salt to above its fusion point (about 32°C), and then either continue heating to evaporate the water of crystallization or remove the solid anhydrous precipitate from the slurry.
In addition to sodium sulfate, some of the lakes contain marketable amounts of magnesium (in the form of both magnesium sulfates and carbonates [141]), sodium bicarbonate (baking soda [80,142,143,131]), and sodium chloride [131]. Finally, coarse clastics (sands, gravels) deposited on beaches, along shorelines, and in deltas of both proglacial and modern lakes are utilized by many of the urban communities in the region [144-146].
Size and shape of the lakes
"The lakes I suppose are not unusual except in numbers alone but if you were able to stand on a great height wherever you are and able to see all the water at once it would still be difficult to find words describing anything but quantity!" [147]
Basin morphology reveals much about a lake's origin [148] and also exerts a profound influence on the sedimentary processes and resulting spatial distribution of detrital sediments within the lake [149-152,22]. The shape characteristics of the basin similarly help control the distribution of chemical precipitates in lakes [153,154,8].
The lakes of the Plains region exhibit a great range in basin size (from small, less than 1 km2 prairie potholes and kettles to several of the continent's largest basins), shape (from nearly perfectly circular to linear troughs to highly irregular shorelines), and depth (from playas to mean depths of over 25 m). Morphometric details (area, maximum depth, mean depth, shoreline length, shoreline development, volume, and volume development) for over 50 saline lakes in southern Saskatchewan have been described [155,36]. Detailed morphometric information on about 40 lakes in the prairie region of Alberta has been reported in [156]. The saline lacustrine basins of the entire region of the northern Great Plains have also been classified on the basis of size, depth, and degree of permanence [157,158]. Figure 13 shows this classification using an expanded database of about 800 prairie lakes.
Most of the lakes in the northern Great Plains are small and shallow. There are very few basins in the region that can be classified as large (greater than 100 km2 surface area). Four of the six largest lakes are located in the Manitoba Lowland area in the eastern part of the prairies (Figure 7). Lake Winnipeg, the seventh largest lacustrine basin in North America, is also the largest lake in western Canada located off the Precambrian Shield. Lake Manitoba is North America's thirteenth largest lake and Canada's largest saline lake. Of the lakes in southern Saskatchewan and Alberta, two of the largest, Big Quill and Old Wives, are among the six largest inland saline bodies of water on the continent.
Most of the small lakes (less than 100 km2) are also shallow. Only about 8% of the basins have mean depths greater than 3 m. However, these small, deep basins are very important. They are attractive targets for paleoenvironmental research because their sediment records often contain undisturbed, finely laminated sequences. It is generally held that deposits in these deeper water basins may be less susceptible to wind and current redistribution and to diagenetic changes brought about by either subaerial exposure or subaqueous chemical fluctuations [159,160]. Several of these small, deep basins are also meromictic, which further enhances their appeal for paleolimnological study [161-163].
Of the many small, shallow salt lakes, one further distinction can be made based on the degree of permanence of the water body. The word playa, meaning "beach", "shore" or "coast" in Spanish, has many definitions, synonyms, and, in some cases, contradictory connotations. Most previous researchers on the Great Plains considered a playa to be any low, seasonally flooded but intermittently dry basinal area, and deemed the term to be generally synonymous with ephemeral lake, slough, or wetland. Some also recommend that the term playa be restricted to continental basin settings characterized by an annual net negative water balance (i.e., more water is lost through evaporation and discharge/seepage than is received through all incoming sources) and in which the capillary fringe is close enough to the surface such that evaporation will cause groundwater to discharge to the surface [164,165]. In the northern Great Plains, about half the lakes with mean depths less than 3 m exhibit playa characteristics in that they fill with water during the spring and summer and usually dry completely by late summer. However, not all of these shallow, ephemeral lakes exhibit groundwater discharge.
Lake water chemistry and properties
"In this region, there are numerous ponds and small lakes in the hollows among the hills, most of them being more or less brackish or nauseous to the taste from the presence of sulfates of magnesia and soda and other salts. During the dry season of autumn, the water evaporates completely from many of these ponds leaving their beds covered by the dry white salts, which look like snow and are blown about in the wind. Around all the ponds, except those which become completely dry, there is a rank growth of reeds, sedges and grasses, the deep green colour of which forms a strong contrast to the dull grey appearance of the stunted and scanty grass of the hills, which indeed, in many places are almost bare." [166]
As indicated above, the Great Plains of western Canada contain millions of lakes. Most of these lakes are saline. Throughout much of the region ponded saline and hypersaline brines are the only surface waters present. Salinity has a significant impact on the emergent vegetation of the lake's littoral zone and, thus, influences the value of the area as a waterfowl nesting and staging ground. In fact, these lakes, collectively, act as a major breeding ground for 80% of North America's ducks [167,128]. In addition to the importance of this surface water within the realm of wildlife conservation, future agricultural, industrial, and urban development in the Great Plains will likely lead to conflicts and potential environmental problems associated with the lakes in the region [168,169]. Thus, since the first analyses were reported from these lakes over 120 years ago [166], considerable effort has been made to collect water composition data and information relevant to the management of water resources. Historically, both Federal and Provincial agencies have monitored many of the lakes in this region. Numerous compilations of these data, on a regional basis, exist in the literature [170,31,73,156,117,178]. Other important sources of regional water composition information exist for central and eastern Alberta [179,180], southern and central Saskatchewan [181-183], and the Riding Mountain area of western Manitoba [185,186].
We now have chemical data from more than 800 of the lakes in the Canadian Great Plains (Figure 14). Although most of these data represent analyses of single samples, some are averages of numerous samples collected over a period of months or years. In general, the larger lakes (e.g., lakes Winnipeg, Manitoba, Quill, etc.) have the longest temporal records, in some cases dating back to the early twentieth century. However, no lake in the Canadian prairies has a continuous monitoring record of more than four decades in duration. Of the 800+ lakes for which there are data, 10% are located in Manitoba, 72% in Saskatchewan, and 18% in Alberta.
Figure 14 Salt lake database. Maps showing the locations of lakes with water chemistry data (upper part of figure) and modern sediment data (lower part of figure); modified from [10, 59, 117, 178].
Salinity and composition
"The lakes which fill the hollows are nearly all salt, and even as early in the season of the year, the soil is whitened with salty efflorescence." [71]
Although a very simple concept, discussion about water salinity can be confusing due to the large variety of methods used to measure this basic parameter [187,10] and a plethora of nomenclature applied (e.g., psu, TDS, ppt, conductivity, molality, g L-1, etc.). Biological limnologists often use conductivity as a measure of salinity. Conductivity, or specific conductance, is a measure of the ability of the solution to carry an electrical current: in general, the greater the salinity, the greater the conductivity. However, this salinity-conductivity relationship is not straightforward. It is a function of the specific ions present in the solution, as well as the level of concentration of the ions. For example, a conductivity value of 126 mS20 in Bitter Lake, Saskatchewan, which is dominated by sodium and chloride ions, is equivalent to ~100 parts per thousand (ppt or ‰) total dissolved solids (TDS). However, this same conductivity would be measured in a brine having only 75 ppt TDS that was dominated by magnesium and sulfate ions [10,188]. Clearly, the use of conductivity to estimate salinity in lakes having diverse chemical compositions, such as these basins in western Canada, should be avoided despite the inherent convenience of this method.
The abundance of individual ionic components in the water is often reported as weight concentration of the ion in solution (i.e., weight of the dissolved ion in g or mg per kg of solution). This is preferred to using the weight per litre of solution because of the large increase in density of the solution at high salinities. However, care must be exercised in evaluating and comparing published reports using g L-1 units. Unless the actual density of the solution is taken into account by the analyst, the difference between weight/weight versus weight/volume can be significant at elevated salinities [189]. For example, the bottom water of Freefight Lake, which is a deep, meromictic lake in Saskatchewan, is 275 ppt (weight/weight) or 340 g L-1.
Investigators involved with appraising the saline lake waters from the perspective of thermodynamics of the aqueous/mineral reactions usually report water composition using traditional chemical concentration nomenclature: molal units (m; moles per kilogram of solvent) or molar units (M; moles per litre of solution) and in equivalents. Equivalents, equivalent weight, or sometimes called combining weight, is the formula weight of the dissolved component divided by its valence charge. Equivalent weights are particularly useful when comparing ionic ratios of one water with another.
In most non-saline aqueous systems, molal and molar units are essential the same. In concentrated solutions, such as those in the lakes of western Canada, differences become significant. Discussions dealing with mineral precipitation/dissolution in the lakes usually evaluate thermodynamic activity of the ions (γ) and ionic strength (I, which is one half the sum of the product of molality times the square of the valence of each ion). In lake waters having less than about 30 ppt TDS, there is relatively little difference between thermodynamic activity and laboratory-derived molality. However, at higher concentrations the electrostatic interactions between the ions greatly reduce their thermodynamic concentration and ionic strength is considerably smaller than molality. Furthermore, like conductivity, γ is also greatly dependent on the specific ions in solution: a 1 m NaCl solution has an ionic strength of 1, whereas I of a solution having the same analytical concentration but dominated by NaSO4 is 3 [190].
Nomenclature and terminology also vary widely for the various levels of salinity [191,192,36,10]. Biological limnologists often use the classification scheme of: fresh water (less than 1‰), subsaline (1–3‰), hyposaline (3–20‰), mesosaline (20–50‰), and hypersaline (greater than 50‰). Groundwater researchers usually refer to fresh water as less than 1‰, brackish water as 1–10‰, saline water as 10–100‰, and brine as greater than 100‰. Most geoscientific literature uses: fresh water (less than 3‰), saline (3–35‰), and hypersaline (greater than 35‰).
Even though nearly all of the lakes in the Great Plains have similar origins and most are relatively small and shallow, the waters show considerable diversity in terms ionic composition and concentration (Table 1). Early investigators, mainly economic geologists concentrating on the most saline brines, emphasized a strong predominance of Na and SO4 in the lakes [80,193]. This importance of sodium and sulfate components in the lakes was similarly recognized by later researchers [73,175]. However, it was not until near the end of the 20th century that the compositional range and degree of diversity of the lakes on a regional basis became obvious. It is now clear that not only is there a complete spectrum of salinities (Figure 15), but also virtually every water chemistry type is represented in lakes of the region [63,109].
Table 1 Mean brine composition of saline and hypersaline lakes in selected geographic areas of the northern Great Plains (ionic concentrations in mmol/l; TDS in ppt); modified from [63].
Geographic Area Ca Mg Na K HCO3 CO3 Cl SO4 TDS
Eastern Prairies 4 24 4 1 6 1 2 24 3
Central Saskatchewan 19 149 193 5 7 3 54 251 22
SW Saskatchewan/SE Alberta 12 93 1088 4 96 36 29 1073 80
West-central Saskatchewan and east-central Alberta 3 144 1362 10 268 44 107 1125 102
Figure 15 Brine composition. Ternary diagrams showing the range of cation and anion composition (eq%) of saline lake waters from the Great Plains of western Canada; modified from [10, 109].
It is not surprising that the lake waters of the northern Great Plains show such a considerable range in ionic composition and concentration, considering the enormous geographic area and the varying hydrologic, geomorphic, and climatic settings. The lakes range in salinity from relatively dilute water (0.1 ppt TDS) to brines more than an order of magnitude greater than normal sea water (Figure 16). Although it is obviously misleading to generalize by quoting means and averages, the "average" lake water has about 40 ppt TDS and shows: Na>Mg>Ca>K and SO4>Cl>HCO3>CO3.
Figure 16 Salinity of lakes. Histogram of water salinity of lakes from the Great Plains region of western Canada. Note logarithmic salinity scale; modified from [10, 109].
With such a vast range of salinities, it follows that the concentrations of the individual ionic components also vary greatly. The frequency distributions of Mg, Na+K, Cl, and SO4 concentrations in the lake waters tend to be multimodal as opposed to the Ca and HCO3 ions which show a much narrower distribution pattern.
Sulfate and carbonate-rich lakes clearly dominate the Great Plains [63,109], comprising over 95% of the total lakes. This paucity of Cl-rich lakes makes the region unusual compared with many other areas of the world (e.g., Australia, western United States). The cation ratios are considerably more diverse, with the abundance of all three major types showing approximately subequal proportions.
As would be expected, most of the solutes in the lake waters increase in concentration with increasing total salinity (Figure 17). Sulfate and sodium + potassium ions show the most statistically significant correlations with TDS, whereas calcium and carbonate concentrations are less directly related to salinity.
Figure 17 Ion composition trends with salinity. Best-fit trend lines showing the change in ionic concentration and ionic proportion with increasing salinity in lakes from the Great Plains region of western Canada. Note logarithmic salinity scale. Modified from [10, 109].
The proportions of some of the solutes also show a systematic change with salinity. Sulfate increases in relative ionic proportion from less than 30% equivalents in dilute lakes to generally more than 90% in lakes with more than 10 ppt TDS. Calcium and bicarbonate + carbonate proportions show an inverse relationship with salinity, decreasing from over 70% equivalents in the dilute waters to nearly 5% in lakes with more than 25 ppt TDS.
Spatial variation
"Little Manitou lake, near the town of Watrous, contains a salt of slightly different type. The lake is deep and does not evaporate in summer. Therapeutic value is claimed for the waters and it is much used as a summer resort." [194]
The relatively uniform distribution of lakes in the Great Plains for which water chemistry data exists permits examination of the ionic contents on a spatial basis similar to that undertaken in the United States portion of the prairies [195,196]. The regional changes in TDS are shown in Figure 18; the results of these regional trend analyses of individual ions are summarized elsewhere: [157,177,84,109]. Lakes with highest Na+K, Mg, and SO4 concentrations generally occur in the east-central Alberta, west-central and southern Saskatchewan area, whereas lakes with high HCO3+CO3 and Cl contents are found in central Alberta and western Saskatchewan. Lakes with relatively low proportions of Ca and Mg occur in the northern and central parts of the Plains.
Figure 18 Regional salinity changes. Map showing the variation in total dissolved solids of lakes in the northern Great Plains of western Canada. Contour interval is logarithmic; values in ppt TDS; modified from [59, 109].
Water chemistry controlling factors: a statistical approach
"Geology differs from the experimental sciences in that most geological data are fragmentary and are derived from surface manifestation of natural processes that are uncontrolled by the investigator.... So the geologist must make do with the data available, which are seldom those with which he would prefer to work. It is here that statistics may enable him to plan data collection and deduce inferences that are not readily discernable from the raw fragmentary observations that he collects." [197]
Ion composition and concentration of these lake waters are the result of: (a) a complex interaction between unconsolidated glacial and nonglacial sediments, bedrock, and precipitation/meltwater in the drainage basin, (b) the composition and quantity of groundwater recharge (and discharge) and streamflow in each basin, and (c) a variety of other physical, chemical, and biological processes operating within the water column itself. In general, several different approaches have been taken to help understand the interplay of these major factors controlling surface water chemistry. These geochemical investigative methods include mass balance calculations, thermodynamic equilibrium considerations, and statistical evaluations [198,199,190]. In western Canada specifically, both mass balance and thermodynamic calculations have proved valuable in deciphering many of the intrinsic (i.e., within the drainage basin) processes important in water composition on a local scale [200,157,204]. In contrast to these local studies, on a regional scale, numerous statistical techniques have been successfully applied to help understand the relationships between the water chemistry and extrinsic environmental factors (e.g., climate, bedrock type, geomorphology, till composition [205,206,109,10]). It should be noted that these statistical approaches lack the ability to resolve the often-important local conditions and processes; however, they are essential to our overall understanding of the lacustrine geochemical setting of the region as a whole.
One of the most clear-cut ways to analyse the interrelationships within any data set is to examine the linear correlations that exist among the analytical parameters. Not surprisingly, the concentrations of Na, Ca and Mg in the brines of these lakes are all significantly positively correlated, as are SO4 and Cl. In addition, the ion pairs of Mg-SO4, Mg-Cl, and Na-Cl tend to strongly covary. Importantly, the concentrations of Na and SO4 do not show statistically significant linear correlation, suggesting different suites of processes affect the abundance of each of these ions. The proportions of Ca and HCO3 exhibit significant positive covariation, whereas the proportions of Mg and Na, and HCO3 and SO4 are inversely related.
Using a Q-mode cluster analysis (statistical associations among the lakes), it is possible to subdivide the lakes into two major categories [109]: a group of relatively high salinity (greater than 20 ppt TDS) lakes and a group characterized by lower TDS values. In contrast, R-mode analysis (statistical associations among the geochemical parameters) indicated the following groups of variables: (a) TDS, Na, and SO4; (b) K and Cl; (c) Ca and Mg; and (d) HCO3 and CO3.
By combining morphological (basin area, maximum depth), geological (bedrock type, depth to bedrock, till type), hydrological (drainage basin area, number of streams entering lake, elevation, groundwater composition), and climatic (mean annual precipitation, evaporation, temperature) variables with the 800-lake water chemistry database, R-mode factor analysis can identify a set of seven statistical factors that explained over 90% of the variance in the data [84,177,109]. The interpretation of these statistical factors is that the most important controls of water composition and concentration on a regional basis are: (a) composition of inflowing groundwater, (b) evaporation/precipitation, and (c) elevation or position of the basin within the drainage basin. Variables related to bedrock type, glacial drift composition, fluvial input, and lake morphology are statistically less important.
Short-term temporal variation in water chemistry
A major complicating factor in characterizing the geochemistry of the lakes of the northern Great Plains is that many of the basins exhibit playa characteristics. This strong seasonality of water levels gives rise to dramatic changes in both ion concentrations and ratios, as demonstrated by numerous studies [207,175,77,211,68]. For example, Ceylon Lake, a salt-dominated playa in southern Saskatchewan, annually undergoes changes in concentration from about 30 ppt TDS to greater than 300 ppt. This lake also exhibits dramatic fluctuations in ionic ratios on a seasonal basis from a Na-(Mg)-SO4-HCO3 type in early spring to a Mg-(Na)-Cl-SO4 composition by fall. Unfortunately, only a few basins in the northern Great Plains have undergone periodic detailed sampling over a period of years.
Source of salts
"Many observations about the occurrence of the salts are valid, but interpretations of their genesis have generally been needlessly complex and unsubstantiated by the observed facts." [212]
The origin and ultimate source of the major ions in the lakes of the northern Great Plains have been topics of considerable discussion in the scientific literature [85]. Some of the early research suggested that dissolution of the deeply buried Paleozoic evaporites could be a possible source for the dissolved components in the lakes. The occurrence of lacustrine basins (and other geomorphic features) whose origin may be ascribed to collapse of salt-solution chimneys [78,83,79] supports this contention. For example, it has been suggested there is a correlation between the occurrence of the lacustrine sodium sulfate deposits at the surface and the presence and trends of various salt units in the Devonian Prairie Formation in the region [83]. However, others maintain that dissolution of the Prairie Evaporite, assuming no other ionic evolution mechanism, could not contribute waters having ionic ratios compatible with the majority of prairie lakes [157,138].
Indeed, as is clear from mapping of the major sodium sulfate bodies [213,63,178,59] and the brine compositions [84,109] there is no consistent spatial correspondence between the surface features and chemistries and the subsurface Paleozoic rocks in most of the Great Plains region. Nonetheless, it is evident that hypersaline groundwaters from the Paleozoic sequence are important contributors to the salinity in the eastern part of the Plains [214,203,215].
In contrast, relatively shallow Cretaceous and Tertiary bedrock, as opposed to the deep Paleozoic sequence, has been implicated as the source of at least some of the dissolved components in the lakes [80,81,216,201]. Furthermore, many researches [93,94,115,100] have stressed the close association of the more saline lacustrine brines with buried preglacial channels, and have concluded that these buried valleys, often filled with relatively porous and permeable sediments, act as conduits for groundwater supplying dissolved material to the lakes.
Finally, rather than invoking bedrock sources, there is considerable support for the source of the ions being largely the Quaternary deposits within which the lakes are immediately situated [73,212,200,217-221], Although simply defining flow patterns and groundwater dynamics in a natural system characterized by poorly integrated surface drainage, undulating surface relief, and multiple discontinuous permeable zones is exceedingly complex [206,222], on a local basis there is some suggestion that the glacial till is a large soluble salt reservoir which can provide salinity to the shallow groundwater systems and therefore to the lakes [223]. A variety of physicochemical and biochemical reactions, including cation exchange, dissolution of feldspars and detrital carbonates, oxidation of reduced-S mineral species and organic matter, and precipitation of authigenic sulfate, carbonate, and silicate phases in the tills can be documented which support this latter hypothesis. Once a convincing ion source can be attributed to the salt budget of a basin, and assuming an unchanging source water chemistry, it is possible to use the level of salinity in the lake to determine the amount of time required to accumulate the dissolved components in selected closed-basin lakes and wetlands in the region [74-76,224]. From this it is possible to identify the episodes during the Holocene at which the now-closed basins overflowed [76].
Modern lacustrine sediments and sedimentary processes
Sediment types
"The study of sedimentary rocks, as compared with igneous rocks, is somewhat hampered by the poverty of terminology." [225]
Very few of the millions of saline lake basins in the northern Great Plains have been examined from either a sedimentological or stratigraphic perspective. Of the approximately 800 lakes for which water chemistry has been documented, we have knowledge of the sediments in only about 20% of these. Similar to the early brine composition work, initial sedimentological efforts on these lakes stressed the dominance of sodium sulfate salts, and were directed mainly toward basins with large reserves of economically important industrial minerals. It is now realized that the lakes in the Canadian Great Plains exhibit a complete spectrum of sediment types (Figure 19), from basins dominated by allogenic or clastic material to those in which relatively pure, clastic-free evaporite minerals are forming [157,10].
Figure 19 Sediment type versus morphology. Range and examples of salt lake types according to morphology and sediment type. Modified from [10, 59 117].
The modern sediments in most of these lakes consist of mixtures of coarse to finely crystalline salts and organic-rich, silty clays/clayey silts in the offshore portions of the basins, grading to somewhat coarser clastics (sands and silts) in the nearshore areas. Due to their small size and negligible fetch distances, most of the lakes have only a very narrow margin of shoreline/nearshore coarse silts and sands. However, in the larger basins, such as Lake Manitoba or the Quill Lakes, and in lakes with long wind-fetch distances, such as characterize the many linear, riverine lakes, coarse clastics can extend farther out into the basin [226,227]. Because of the paucity of fluvial input to most of the lakes, deltas are rare except in the larger basins such as Lakes Manitoba and Winnipeg. Away from the relatively nearshore areas, however, there is usually little variation in grain size in any of the basins [228,75,46,77], [229-231].
Generally the organic content of the modern sediments is low, although in some basins organic matter can be as much as 45% (by weight) of the total sediment. The organic content of sediments from playa basins and shallow lakes is not significantly different from that of most of the perennial deep water basins. However, the chemical composition of this organic matter can vary significantly. Playa lake sediments in the region show consistently much lower Rock-Eval hydrogen indices (HI) and somewhat higher oxygen indices (OH) relative to both shallow and deep perennial lakes (Figure 20), reflecting a more terrestrial source for the organics in the playas [232,233,10].
Figure 20 Rock-Eval analysis of salt lake sediments. Plot of Hydrogen Index (HI) versus Oxygen Index (OI) values for sediment from saline lakes of the northern Great Plains [modified from [10, 59, 117, 178]. Red = Redberry Lake offshore sediment; FF = Freefight Lake offshore sediment, Elk = Elk Lake (Minnesota [232]) offshore sediment; Lm = Little Manitou Lake offshore sediment; Dm = Deadmoose Lake offshore sediment; Wald = Waldsea Lake offshore sediment; Man = Lake Manitou offshore sediment; green shaded area = samples from playa lakes; Minnesota and the Dakotas [232]; orange shaded area = samples from pedogenic zones in Lake Manitoba [233].
Detrital versus chemical sediments
"The union of alkaline earths with carbonate and calcium with sulfate are unlikely in Saskatchewan." [76]
Sedimentologists have long recognized three basic genetic types of sediment in most lacustrine basins; (a) allogenic (i.e., detrital): that material derived from weathering and erosion of the soils and bedrock of the watershed and transported to the lake by fluvial, sheetwash, gravity, or aeolian processes; (b) endogenic: sediment originating from biological or inorganic processes occurring entirely within the water column of the lake; and (c) authigenic (i.e., diagenetic): material resulting from mainly chemical and biological processes occurring once the sediment has been deposited. The endogenic and authigenic fractions are sometimes collectively referred to in a nongenetic sense as 'chemical sediment', as opposed to 'clastic sediment', which is usually dominated by the allogenic fraction. However, use of this latter designation should be avoided when discussing the deposits of saline lakes because of the presence of such things as 'clastic' sediment comprised of fragmented endogenic components, or 'chemical' sediment derived largely by post-depositional diagenetic processes.
The allogenic component in most of the lakes of western Canada consists mainly of a mixture of (in approximate decreasing order of abundance): clay minerals, quartz, carbonates, feldspars, and ferromagnesian minerals. The relative proportion of each of these detrital mineral groups is generally similar to that of the surrounding glacial debris, except sorting by the transporting medium and diagenetic processes operating within the watershed soils can also influence the final detrital mineralogy. The less chemically stable mineral components, such as detrital carbonates and feldspars, are particularly susceptible to loss by weathering processes in the drainage basin [234-237]. Superimposed on this element of chemical stability of the detrital components is the inherent size fractionation of various minerals caused due to physical abrasion and breakage of grains by the action of glacial ice. For example, because the terminal grade for glacier-comminuted dolomite is medium to coarse silt [238,239], the modern sedimentary facies of the larger lakes in the region show a gradation from dolomite-enriched sediment nearshore to dolomite-depleted deposits offshore [46,240,231].
Although the sediments of only a small number of lakes have been examined for detailed clay mineralogy, the most common layered silicates are kaolinite, illite, smectite, and chlorite [228,237,77,230,241,162,233]. A variety of mixed-layer clays, most commonly illite-smectite and illite-chlorite species, have also been reported, but at least some of these interstratifed minerals have been attributed to possible authigenic processes within the lake basins [204,242,233]. In lakes west of the Manitoba Lowland, the sediments are almost completely dominated by smectitic clays, undoubtedly a reflection of the bentonitic Cretaceous shales, which underlie most of the area west of the Manitoba Escarpment. The abundances of kaolinite, illite and smectite in the eastern part of the region tend to be approximately subequal.
The only detrital carbonate minerals identified in the lakes of the Great Plains are dolomite and calcite. In bulk (i.e., non-size fractionated) samples, CaMg(CO3)2 (dolomite) is usually considerably more abundant than CaCO3 (calcite). This is most likely a reflection of the relative abundance of these two minerals in the tills and bedrock of the region, as well as the higher solubility and, hence, lower weathering stability of calcite relative to dolomite [239,229,243,244,231]. Similarly, potassic feldspars are usually more abundant than plagioclase in the lakes due to the partial loss of the relatively unstable Na and Ca feldspars by hydrolysis in the tills and watershed soils.
Considering the great range of water compositions in lakes of the Great Plains, it is not surprising there is an equally significant breadth of endogenic and authigenic minerals found in these lakes. In the playas of the Great Plains, there are two main types of endogenic precipitates in the modern sediments: (a) very soluble salts, comprising mainly sodium and magnesium sulfates and carbonates, and (b) sparingly soluble precipitates, including mainly carbonates, sulfates, and silicates (Figure 21). There have now been close to fifty non-detrital minerals identified from the lakes (Table 2). These endogenic and authigenic minerals can also be subdivided according to their genesis and occurrence within the lake: (a) surficial efflorescent crusts and hardgrounds, usually occupying nearshore and seasonally flooded areas; (b) massive and bedded precipitates, most often found blanketing the floor of the basins from shallow marginal zones down to deep central offshore areas; and (c) accumulations of salts and precipitates associated with either subaqueous or subaerial springs.
Figure 21 Endogenic and authigenic salts. Examples of modern chemical precipitates from the salt lakes of the northern Great Plains of western Canada; modified from [10, 59, 117]. A. Shoreline accumulation of rounded accretionary grains of mirabilite; B. Floating crust of thenardite; depth of brine under the crust is ~40 cm; water temperature is 34°C; C. Large dendritic accumulation of bloedite crystals; D. Large subaqueous accumulation of mirabilite crystals; E. Accumulation of halite crystals; F. Acicular thenardite crystals that have been pseudomorphologically replaced by halite; G. Large aggregate of epsomite crystals; H. Dogtooth crystals of mirabilite overlain by flat-lying bloedite crystals; I. and J. Rounded accretionary grains of mirabilite, termed mirabolites; K. Spring-orifice accumulation of bloedite and mirabilite.
Table 2 Endogenic and authigenic minerals in saline lakes of the northern Great Plains; modified from [10, 59, 116].
Mineral Name Composition Occurrence
Carbonate Minerals
Aragonite CaCO3 very common
Artinite Mg2CO3(OH)23H2O very rare
Ankerite Ca(Fe,Mg)(CO3)2 rare
Benstonite Ca7Ba6(CO3)13 very rare
Calcite CaCO3 very common
Gaylussite Na2Ca(CO3)2 5H2O rare
Kutnohorite Ca(Mn,Mg)(CO3)2 very rare
Magnesite1 MgCO3 common
Magnesian Calcite (MgxCa1-x)CO3 very common
Minrecordite CaZn(CO3)2 very rare
Protodolomite CaMg(CO3)2 common
Zemkorite Na2Ca(CO3)2 very rare
Sulfate Minerals
Arcanite (K,NH4)2SO4 rare
Bloedite Na2Mg(SO4)2 4H2O very common
Despujolsite Ca3Mn(SO4)2(OH)3(H2O) rare
Epsomite MgSO4 7H2O common
Eugsterite Na2Ca(SO4)3 2H2O very rare
Gypsum CaSO4 2H2O very common
Hexahydrite MgSO4 4H2O common
Kieserite MgSO4 H2O common
Krausite Fe2(SO4)3 2H2O very rare
Mercallite KHSO4 very rare
Mirabilite Na2SO4 10H2O very common
Potassium alum KAl(SO4)212H2O very rare
Thenardite Na2SO4 very common
Wattevilleite Na2Ca(SO4)2 4H2O very rare
Carbonate-Sulfate, Carbonate-Sulfate-Chloride, and Carbonate-Phosphate Minerals
Bonshtedite Na3Fe(PO4)(CO3) very rare
Bradleyite Na3Mg(PO4)(CO3) very rare
Burkeite Na4(SO4)CO3 very rare
Rapidcreekite Ca2(CO3)SO4 4H2O very rare
Tychite Na6Mg2SO4(CO3) very rare
Chloride Minerals
Bischofite MgCl2 6H2O Very rare
Halite NaCl rare
Nitrate and Borate Minerals
Inderborite CaMgB6O11H2O rare
Niter KNO3 rare
Soda Niter NaNO3 rare
Nitrobarite Ba(NO3)2 rare
Other
Pyrite FeS2 common
Ranciete Ca0.75Mn4O9·3(H2O) very rare
Sepiolite Mg4Si6O15(OH)26H2O very rare
1Includes hydromagnesite and psuedohydromagnesite.
Distinguishing endogenic lacustrine minerals from those formed by diagenetic processes after the sediment has been deposited (i.e., authigenic) is often an exceedingly difficult task. Many times, detailed petrographic and geochemical studies are required to convincingly demonstrate the precise origin for the specific minerals. To date, few lakes and lacustrine sequences in western Canada have been examined in this amount of detail. Although a complete discussion of the genesis (viz. endogenic versus authigenic) of the minerals listed in Table 2 is beyond the scope of this paper, some of the more common or noteworthy occurrences will be summarized briefly.
Among the endogenic and authigenic carbonate minerals identified in these lakes, aragonite, magnesian calcite (i.e., hi-Mg calcite), and dolomite are the most common. Inorganic precipitation of carbonate minerals due to thermodynamic supersaturation is very common in lakes of all types on a global basis [245,246,8-10]; the Great Plains region is no exception. Nearly all of the 131 lakes that were surveyed in the early 1980's were found to be strongly supersaturated with respect to these carbonate minerals [157]. Supersaturation and precipitation of carbonates can take place for a variety of reasons, including photosynthetic uptake of CO2 and consequent increase in pH, concentration changes brought about by evaporation or dilution, temperature changes, and mixing of brines of different compositions. In most of the lakes of the Great Plains, carbonate mineral supersaturation is due mainly to, but not exclusively, the seasonal uptake of CO2 by primary organic productivity.
The specific carbonate mineral to be precipitated from the supersaturated solution is controlled mainly by the cations in solution (in particular the ratio of Mg to Ca in the water [247,239]). The elevated Mg/Ca ratios that characterize the lake waters of the Great Plains give rise to a carbonate mineral assemblage dominated by aragonite (the orthorhombic form of CaCO3) and Mg-bearing carbonates, such as dolomite, magnesite, huntite and Mg-calcite.
The discovery of non-detrital dolomite in the lakes of the Great Plains has been an important contribution to our understanding of the genesis of this economically important mineral [249]. Dolomite formation in the sedimentary realm is a subject of long-standing interest and study. Probably no other mineral or sedimentary rock has attracted as much speculation regarding its genesis as dolomite [248-250]. The "dolomite problem" has been summarized in many recent reviews [251-254] and in most sedimentary geology textbooks. In essence, the dolomite problem is the enigma that the mineral dolomite (CaMg(CO3)2) is a very common component of ancient rocks, but is very rare in modern and Holocene sediment. It does not appear to occur as a primary precipitate from marine water of normal salinity and composition, nor cannot be readily synthesized in the laboratory at low temperatures and pressures.
The occurrence of very early diagenetic (i.e., penecontemporaneous) dolomite in the surficial offshore sediments of Devils Lake in northeastern North Dakota [255,256] was one of the earliest documented examples of lacustrine dolomite formation. This occurrence of dolomite also emphasized that dolomitization could take place in solutions of moderate salinities (~20 ppt TDS) and in water with high sulfate contents. Others, working mainly in highly evaporitic marginal marine settings, have suggested that dolomite formation is favoured by elevated salinities (greater than 35 ppt TDS) and greatly reduced SO4 levels [257-259].
Modern primary and early diagenetic dolomite occurs in numerous lakes in the Canadian Great Plains [249]. Waldsea and Deadmoose Lakes are two adjacent, meromictic saline lakes located in central Saskatchewan. Two distinct types of non-detrital dolomite have been identified in these lakes [260,161,163]. Most is very finely crystalline with subhedral to anhedral crystallites forming aggregates of about 2 μm in diameter. This dolomite, like most Holocene dolomite, is poorly crystalline and poorly ordered (ordering refers to the regular alteration of the layers of cations and anions within the crystal structure), but has a composition close to that of the ideal, stoichiometric mineral (i.e., Ca=Mg). The other type is also poorly crystalline, but is considerably enriched with calcium: Ca(Ca0.2Mg0.8)(CO3)2. Like that of Devils Lake, the dolomite in both Waldsea and Deadmoose is forming subaqueously as a primary inorganic precipitate in sulfate-rich waters.
Another important occurrence of dolomite is in Freefight Lake, a 25 m deep, hypersaline meromictic lake in southwestern Saskatchewan. Preliminary sedimentological examination of this lake [261,204,242] indicates a considerable range of formative environments and dolomitization processes. Calcian (i.e., Ca-enriched) dolomite occurs in the deep-water, offshore sediments and also in the subaerially exposed mudflats surrounding the basin. Within the mudflat sediments, it is often associated with aragonite and Mg-calcite, and petrographic evidence suggests it may be an early diagenetic product forming in response to the strong evaporative flux experienced on the mudflats during the ice-free season. In contrast, in the organic-rich, highly reducing offshore sediments, dolomite is associated with a variety of soluble sulfate salts and pyrite. It is forming at the sediment-water interface as a primary precipitate in response to the increased alkalinity brought about by sulfate reduction in the anoxic monimolimnion of the lake.
A final example showing the diversity of Mg-carbonate mineral genesis in these lakes is that of Chappice Lake in southeastern Alberta. Chappice Lake is a small, hypersaline, groundwater-fed playa whose brine is dominated by Na and SO4 [262-265]. Although endogenic carbonates are not forming in the lake today, the 7000 year long stratigraphic record preserved in the basin contains thick sequences of very finely laminated carbonates [266]. The carbonate crystals and crystal aggregates making up these laminae are euhedral and contain no petrographic evidence of reworking, abrasion, corrosion, or diagenetic alteration, thus suggesting the laminae were generated by inorganic precipitation from within the water column. The lack of detrital grains in these laminae and the absence of rhythmicity indicate relatively rapid and non-annual precipitation events. As shown in Figure 22, it is clear from the detailed mineralogical composition of the sequence that the brine underwent striking compositional changes in both carbonate alkalinity and cation ratios.
Figure 22 Chappice Lake laminated carbonates. Photograph of a section of core from Chappice Lake showing the fine lamination and the complex carbonate mineral assemblage present in this otherwise relatively simple playa basin; modified from [10, 59].
The mineralogy of the endogenic and authigenic sulfates in the Great Plains' lakes is diverse and the genesis of these inorganic components complex. Not only is the stable sulfate mineral suite controlled by the ionic composition and cation ratios in the brine, but also the temperature at which the precipitation occurs is also very important (Figure 23). For example, in a simple binary salt system (e.g., Na-SO4-H2O, Mg-SO4-H2O, or Na-Cl-H2O) the solubility of mineral phases such as mirabilite (Na2SO4 10H20), and epsomite (MgSO4 7H2O) show enormous ranges over temperatures normally encountered by the brine on a diurnal and seasonal basis. In contrast, common non-sulfate salts, such as halite or bischofite (MgCl2 6H2O), exhibit comparatively little solubility change over these near-surface temperature ranges. In more complex aqueous systems, such as Na-Mg-Cl-SO4-H2O, prediction of the stable phase(s) becomes somewhat more complicated, but the influence of temperature remains important.
Figure 23 Temperature-solubility relationships of binary salt systems. Relationship between temperature and solubility of commonly occurring evaporite minerals in the lakes of the northern Great Plains. Modified from [63].
The effect of temperature on the mineral suite in these lakes is most obvious in the playas and shallow lakes/wetlands [219,267,268] where the annual cycle of sediment accumulation (precipitation) and dissolution is readily apparent (Figure 24). However, the deep perennial salt lakes are also strongly influenced by the annual (and diurnal) temperature regime. For example, in Freefight Lake, the surficial waters become supersaturated with respect to mirabilite and epsomite during the winter due to low temperatures of the upper part of the mixolimnion (-10 to -15°C depending on salinity) and the formation of an ice cover which increases the brine concentration in the upper metre. The precipitated Na2SO4 10H2O and MgSO4 7H2O crystals undergo corrosion and dissolution as they settle through the understaturated (-5 to 5°C) mixolimnion below the ice cover. Dissolution of these sulfate salts in the mixolimnion is of sufficient magnitude to, in turn, affect the temperature of the water. The dissolution of mirabilite is a strongly endothermic reaction. When 1 mol of mirabilite dissolves the reaction absorbs over 4 MJ of heat. This heat must be acquired from the surrounding water, thereby lowering the temperature. However, at the chemocline sulfate salt precipitation begins anew, despite a still higher temperature, due to the greater salinity. Gypsum shows a similar pattern of precipitation from the surficial water; dissolution within the mixolimnion, and precipitation through the monimolimnion, except the supersaturation at the surface is usually brought about during the ice-free season due to evaporative concentration of the brine. Thus, the petrography of the resulting deep-water sulfate salts in this basin records a complex series of multiple precipitation and dissolution events that are controlled largely by seasonal temperature fluctuations within the water column.
Figure 24 Sedimentary cycles in cold saline playa lakes of the northern Great Plains. Schematic illustration showing the major cycles (diurnal, annual, evolutionary) characteristic of playas in the northern Great Plains; modified from [10, 59].
A spectrum of lacustrine sedimentary processes
"I often say that if you measure that of which you speak, you know something of your subject; but if you cannot measure it, your knowledge is meager and unsatisfactory." (Lord Kelvin)
The lakes of the northern Great Plains offer an excellent opportunity to examine the processes of lacustrine sedimentation on both a local and regional scale. As is the case with most other depositional settings, these lakes exhibit a continuous spectrum of sedimentary regimes. This continuum exists in terms of nearly every parameter that can be evaluated for the basins: morphology, hydrology, degree of permanence, brine salinity and ionic composition, sediment character, composition, etc. This great diversity presents somewhat of a problem in studying and discussing the lakes.
In order to subdivide this continuum, several "end member" types of lakes can be recognized (Figure 19). A basic sedimentological distinction must be made between the basins whose brines are shallow enough to permit periodic drying and those whose brines are deep enough to maintain a relatively permanent water body. The suite of processes operating in shallow intermittent basins (i.e., playas) includes: cyclic flooding and desiccation of the playa surface, formation of salt crusts, efflorescent crusts, hardgrounds, spring deposits, and intrasedimentary salts, formation of solution pits and chimneys, and periodic detrital sedimentation by sheet flow and wind (Table 3; Figure 24). Most previously published summaries of these basic playa processes deal with lacustrine systems occurring in warm and dry climatic regimes [10,27,199]. The five to six months of subfreezing temperatures experienced in western Canada, as well as the occurrence of snow and high seasonal runoffs associated with snowmelt, dictate some modifications and additional processes be considered when studying the playas of the northern Great Plains.
Table 3 Lacustrine sedimentary processes operating in saline lakes of the northern Great Plains; after [10, 117, 158, 178].
Shallow Water-Shallow Basin & Shallow Water-Deep Basin Lakes
Cyclic flooding & desiccation
Wind setup
Wind-controlled localization of salt ppt
Aeolian influx
Deflation
Pedogenesis
Intrasedimentary salt precipitation
Formation of subsurface salt cements & hardgrounds
Formation of carbonate crusts
Precipitation of salts at air-water interface
Formation of crystal rafts & aggregates
Evaporative pumping
Formation of efflorescent crusts
Microbialite formation
Formation of vegetation mats
Development of meromixis
Subaqueous cumulate & bottom salt ppt
Formation of salt spring deposits
Formation of rounded accretionary salt grains
Reworking/redistribution of clastic salts
Temperature-induced mineral transformations
Phase changes
Formation of salt cements
Salt karsting
Mud diapirism
Sediment disruption by freeze-thaw
Remobilization & reworking of fine-grained clastics
Deep Water-Deep Basin Lakes
Development of meromixis
Bio-mediated carbonate precipitation
Evaporite carbonate precipitation
Formation of subaqueous salt cumulates
Solute concentration by formation of ice
Freeze-out precipitation of salts
Sulfate reduction, sulfide mineral ppt
Cyclic and rhythmic sedimentation
Clay mineral authigenesis
Development of thermal stratification
Turbidity flow, interflow
Flocculation of fine-grained material
Shoreline erosion
Delta sedimentation
The large number of individual playa basins in the Great Plains, the tremendous diversity of chemical and hydrologic settings, and the variable impact of humans are the main factors that have given rise to the complex suite of interrelated sedimentary processes in these lakes. To better understand the relative importance of the many and diverse processes, Q-mode cluster analyses has been applied to help quantitatively group various classes of playas in the region [77]. This analysis identified five groups of playa basins. One group of lakes, typified by the Chaplin, Bigstick, and the Quill Lakes complex, are characterized by having large areas, moderate salinities but high Cl contents, and high detrital to endogenic sediment ratios. Mudflat and sand flat sedimentary processes dominate. In contrast, basins such as Snakehole, Whiteshore, and Muskiki are marked by brines with very high salinities that are almost completely dominated by sodium and sulfate ions. Sedimentation in this group of lakes is controlled largely by chemical processes and the basins contain thick sequences of bedded to massive salts. Chemical sedimentation processes also dominate in lakes such as Ceylon, Ingebright, and Vincent, but they exemplify a third group of playas distinguished by their relatively small areas, and high shoreline length and development indices (i.e., very irregular shorelines or very long, linear basin morphology). Lakes belonging to the final two groups include those whose brines are characterized by calcium, magnesium, and bicarbonate ions rather than sodium and sulfate. Sedimentation is controlled by either mixed chemical/physical processes (e.g., Verlo, Corral, Lydden) or mainly chemical (e.g., Metiskow Lake).
Despite the range of sediment types and hydrology of these lakes, the processes in these basins create a discrete suite of modern sedimentary facies. Although significant variations in the development of these facies do exist from basin to basin, a basic facies pattern common to most of the playas in western Canada can be recognized. The outer shoreline/nearshore complex comprises colluvium, mudflat/sand flat, and beach facies, and grades basinward into a salt pan complex. This basic pattern shares many of the same features recognized in saline lakes in British Columbia [269-272] and is broadly similar to the facies distributions in playas in Spain [273,274].
The colluvium facies consists of an often chaotic mixture of coarse to fine detrital material derived from the adjacent glacial deposits by mass wasting and creep. The mudflats are characterized by a mixture of sandy to silty detrital sediments often capped by thick efflorescent crusts and cemented hardgrounds. Unlike efflorescent crusts on mudflats in playas elsewhere [199,275,276], these crusts are not monomineralic. They usually consist of a complex mineral assemblage of both hydrated and anhydrous Na, Na+Mg, and Ca sulfate and sulfate-chloride salts [277,278,63]. In some basins, carbonate hardgrounds and crusts are also present. The mudflats are areas of active beach processes during seasonally high water levels, but for most of the ice-free season evaporation is the dominant process. Evaporation during the warm summer months results in upward capillary movement of groundwater, ionic concentration, and intrasedimentary mineral precipitation. Once efflorescent crusts have formed, further evaporation is curtailed, and the marginal areas of the lake can maintain high moisture contents throughout the dry season.
The mudflats can be colonized by extensive areas of algae and cyanobacterial mats (Figure 25). The sediments in these areas are distinctively laminated [266], organic-rich, and are usually sites of biogenic carbonate mineral genesis and diagenesis. Many of the more exotic carbonate mineral species found in the Great Plains (e.g., hydromagnesite, psuedohydromagnesite, kutnahorite, siderite, tychite, huntite) have been identified in these biolaminated algal flat sediments. To date, the biology and biogeochemistry of these mats have undergone only cursory examination [207,204,279].
Figure 25 Microbial mats and other sedimentary-biological features. Examples of modern bio-sedimentary components of the salt lakes of the northern Great Plains of western Canada. A and B. Modern bio-sedimentary facies in Freefight Lake. Colluvium and mudflats/salt flats grade basinward into a 50 meter wide zone of microbial mats and then into deepwater basinal facies. C. Example of microbial mat material ripped up by wave agitation. D, E, and F. Examples of microbial and vegetation mats from saline lakes in the Great Plains. Note the laminated nature of the sediment immediately below the modern mat in F. Carbonate mineral diagenesis in this environment has created a considerable range and diversity of authigenic species (see also Figure 22). G. Hopper shaped mirabilite crystal incorporating Artemia sp. H. Bio-induced early diagenetic pyrite filling a diatom frustrule. I. Hopper shaped salt crystals floating at the surface of a brine pool. The red colouration of the water is due to the high abundance of Artemia sp.
The salt pan complex usually comprises over 50% of a typical playa basin. It can be covered by brine for much of the ice-free period, but is often exposed by late summer. This facies is usually characterized by elevated endogenic to detrital sediment ratios; however, a complete gradation exists from salt-dominated pans to mud-dominated basins. The most distinctive feature of the salt pan facies is its distinctive annual cycle [208,77,68,10]. During spring, relatively dilute inflow from melting snow and rainfall dissolves much of the very soluble salt that may have been precipitated the pan during the previous dry episode. This dissolution results in a significant increase in brine salinity and usually dramatic changes in ion ratios. These brine compositional changes at this time drive many of the penecontemporaneous diagenetic reactions that take place during the initial seasonal flooding of the playa [237,280]. Throughout the rest of the ice-free period brine salinities continue to increase due to evaporation. As discussed above, even minor diurnal temperature changes can cause massive salt precipitation or dissolution. If brine remains in the basin after the onset of freezing conditions, considerable thicknesses of salts can occur due to freeze-out precipitation.
The playa pans can be further subdivided into several subfacies on the basis of mineralogy and crystal morphology [68]. Near the margins of the pan where the salts interfinger with the mudflat facies, a zone of large dog-tooth mirabilite crystals develop, which grow displacively downward into the soft, water-saturated mudflat sediments (Figure 21H). Further out into the center of the basin, the floor of the salt pan usually consists of a mosaic of large, interlocking, bladed crystals. The specific mineralogy in the salt pan is controlled by the ionic content of the brine and the diurnal and seasonal temperature fluctuations experienced. Elevated temperatures favour minerals such as thenardite, hexahydrite, anyhydrite, and thermonatrite whereas cooler temperatures encourage mirabilite, epsomite, natron, and gypsum.
Superposed on these salt pan evaporites or clastics can be zones of mirabolites and/or sediments associated with spring openings. Mirabolites are rounded accretionary grains (analogous to carbonate pisolites) composed of Na2SO4 10H2O or MgSO4 7H2O that form in the shallow, supersaturated, and wind agitated brine (Figure 21J). These mirabolites can be moulded into bedforms, shoals, and ridges on the pan surface or form beach deposits on the marginal mudflats. Spring openings can be sites of massive salt precipitation because these are areas where water of different temperatures and compositions mix (Figure 21K). Frequently the springs continue to discharge onto the playa floor after the pan has been completely desiccated. This continued discharge can form large, low mounds and ridges.
In summary, present-day sedimentation in the playa basins of the Canadian Plains region is controlled by the interplay of: (a) flooding of the playa which causes dissolution of soluble minerals of the salt pan facies and efflorescent crusts of the mudflat and sand flat facies; (b) evaporative concentration of the brine which results in supersaturated conditions and precipitation of various soluble and sparingly soluble salts; (c) detrital influx by streamflow, wind, and spring discharge; and (d) organic productivity. In addition, evaporative pumping of shallow groundwater in the mudflats and sand flats causes growth of intrasedimentary displacive and poikilitic salt crystals in the near-surface clastic sediment.
This suite of interrelated sedimentary mechanisms operating in playa basins of the Great Plains is in contrast to the deposition in the region's perennial lakes, which appears to be controlled by a much smaller array of processes (Table 3). The majority of processes, such as shoreline erosion, beach formation, wave/current distribution of sediments, deltaic sedimentation, and turbidity flow and interflow, are basic sedimentological processes common to any perennial basin. Numerous papers and compilation volumes already provide excellent summaries of these processes in perennial lakes [281,15,282,22], as do most current introductory geoscience textbooks. However, several processes are unique to the perennial basins of western Canada or are of such fundamental importance that they will be introduced here briefly.
Probably the single most critical process operating in the perennial saline lakes is development of stratification of the water column. The influence of temperature stratification on seasonal carbonate mineral saturation and equilibria in the lakes is well known. However, the superposition of temperature stratification on an already chemically stratified water column complicates many mineral precipitation and dissolution reactions. In Deadmoose Lake, for example, both evaporitic and biologically-induced carbonate precipitation occurs in the well-mixed epilimnion. Most of these endogenic carbonates, however, are dissolved upon passing through the thermocline into the anoxic, lower pH water of the hypolimnion, such that the modern lake only contains carbonate-rich muds at water depths of less than 8 m [261,163,117]. In contrast, endogenic CaSO4 2H2O precipitation occurs below about 12 m depth. This evaporitic mineral is found only in the modern sediments of the monimolimnion. A further complication in the gypsum genesis in this meromictic lake is that the surface waters become supersaturated with respect to CaSO4 2H2O during the winter and gypsum precipitation occurs. However, sediment trap data indicate that this gypsum originating in mixolimnion is re-dissolved before reaching the chemocline. The same complexity arises for mirabilite: precipitation of this sulfate mineral occurs both at the chemocline and in the surface water during winter. However, it is only preserved in the sediments of the monimolimnion at water depths greater than 20 m. Although this inhomogeneity can be relatively easily modelled once sufficient details are known about the chemical budgets of the stratified brine, any interpretation of the stratigraphic record is made much more obscure because even subtle chemical and/or temperature changes can result in substantial changes in mineral precipitation and preservation.
Two other important aspects of deposition in these perennial lakes of the Great Plains are the generation and accumulation of deep water-soluble salts, and the extraordinarily high rates of sedimentation experienced by some basins. Many of the non-playa lakes in the region are characterized by either mixed endogenic-allogenic sediments or entirely allogenic deposits [157]. However, it has been shown that in some basins soluble and sparingly soluble salts are forming in deep water, offshore environments [283,204,261], [284-286,227,10]. Sedimentologists have recognized for some time that there are very few examples of modern sedimentary environments in which deep water evaporite mineral formation is occurring. This lack of modern deep water settings in which evaporites occur is problematic since many ancient evaporitic sequences have been interpreted as forming in deep water [287-289]. Consequently, the perennial lakes in the Great Plains in which deep water salts are forming today provide a critical analogue in helping to understand the sedimentology, geochemistry, and stratigraphy of these ancient deposits.
To date, three basins in the Great Plains have been identified in which soluble evaporite minerals are forming and accumulating in their deep, offshore areas: Deadmoose, Little Manitou, and Freefight Lakes. It must be emphasized that there are probably numerous such basins among the many perennial lakes of the region, but our knowledge of the deep basinal facies of most of the basins is limited. Furthermore, subaqueous salt precipitation by freeze-out mechanisms have been reported from numerous other lakes in the region [80,175,157] but presumably these salts are seasonal and re-dissolve during the ice-free period.
Both Deadmoose and Freefight Lakes are hypersaline and meromictic, whereas Little Manitou experiences 'temporary' meromixis. All three lakes are dominated by sodium, magnesium and sulfate ions but the mineralogy of the modern offshore precipitates is somewhat different in each lake as controlled by the specific ionic ratios in the brines. As outlined above, the diurnal and seasonal temperature fluctuations of the mixolimnions in Freefight and Deadmoose Lakes and the elevated salinities of the monimolimnions in these basins provide a complex multi-site source for the endogenic evaporite precipitates within the water columns. In Little Manitou, it appears that brine mixing associated with subaqueous spring discharge and the irregular breakdown of the weak meromixis initiates precipitation.
In addition to the formation of these deep water salts, the rate at which they are accumulating is noteworthy. For example, sedimentation rates at the sediment-water interface in the deepest part of Freefight Lake average nearly 30 kg m-2 yr-1. While these sediment trap data do not represent net accumulation, nonetheless the stratigraphic sequence recovered in the offshore areas of the basin suggest linear accumulation rates in excess of 2 cm yr-1, values that are entirely consistent with the extraordinary mass/year sedimentation rates. As discussed elsewhere [290-292], such high rates of chemical sedimentation should be expected in an evaporitic regime. However, until the discovery of these salt lakes in western Canada, such rates were not adequately documented in modern deep water environments.
Northern Great Plains paleolimnology
"It has long been my feeling that when a geologist gets into trouble, he changes the climate." [293]
Until recently, there has been relatively little research effort directed toward investigating the paleohydrology, paleolimnology and paleoenvironmental settings of the saline lakes of the northern Great Plains [294]. There are many reasons for this paucity of study, including difficult chronology problems associated with the widespread abundance of 'old' carbon and its incorporation within the lacustrine deposits [295], the common occurrence of drying/desiccation horizons and incipient soil profiles within the lacustrine stratigraphies [240], and ubiquitous diagenesis and post-depositional physical/chemical alternation of the sediments in the basins [280,158]. However, beginning in the 1990's, the recognition of global change and the necessity to better understand the past climate of the region [296] caused an explosion of interest in the Holocene and late Pleistocene records of the lakes in the Great Plains. The major objective of many of these projects was to decipher the timing and severity of postglacial climatic changes and their geomorphic impact on the prairie landscape [58,297-302]. Clearly, investigation of the Holocene and late Pleistocene stratigraphic records preserved in the lakes of the region forms a pivotal role in accomplishing this objective. Because of their great diversity in morphology, chemistry, and depositional processes, these basins offer an outstanding opportunity to examine past environmental conditions.
Furthermore, the large number of lacustrine basins and their presence over such a broad area will ultimately allow paleolimnologists to integrate numerous sites in a variety of hydrologic and geomorphic settings in order to gain a regional perspective of environmental changes.
Of the approximately 150 salt lakes for which modern sedimentological data are available, the stratigraphic records of fewer than two dozen have been investigated in any type of detail [294,303,59]. The two deep, meromictic basins east of Saskatoon mentioned previously (Waldsea and Deadmoose Lakes) have received considerable paleolimnological attention. The mid to late Holocene records in these two basins suggest dramatic fluctuations in water levels, organic productivity, and chemical composition [304,305]. Similar changes in brine chemistry and hydrology were interpreted from the stratigraphy of Ceylon Lake in southern Saskatchewan [68,306,307]. The salts deposited in this playa lake suggest that the basin evolved from a relatively low salinity, riverine lake to one in which initially Na-rich and then Mg-rich hypersaline brines dominated. Lake Manitoba, the largest saline lake in the Great Plains, has also undergone intensive paleoenvironmental study [160,236,241,239,233]. Significant changes in water levels during the 12,000-year long history of this lake are associated with brine chemistry changes (particularly with respect to the Mg/Ca ratio of the lake water) and organic productivity fluctuations (Figure 26).
Figure 26 Lake Manitoba paleolimnology. Summary diagram showing the paleolimnological interpretations of Lake Manitoba. Modified from [233].
Paleoenvironmental interpretations of salt lake sediments are not without pitfalls. Saline and hypersaline lacustrine environments are amongst the least understood depositional regimes in sedimentary geology. Thus, interpretations of the preserved stratigraphic records are hampered by this incomplete understanding of the modern depositional and diagenetic processes.
Although the relatively few lacustrine sedimentary records that have been examined clearly indicate dramatic changes in water levels and related brine chemistry, the causal mechanism(s) of these temporal changes are still largely unknown. Climate has certainly controlled the sedimentation and geochemistry of nearly all of the basins. However, the precise roles of other factors, such as fluctuating groundwater hydrology and hydrochemistry, or postdepositional alteration of the sediments, still remain to be evaluated.
Despite the sensitivity of these deposits to environmental change, interpreting the records in terms of paleoclimate, hydrology, and chemistry is fraught with difficulty. Factors that complicate these interpretations include: diagenesis of the evaporites, post-depositional physical disruption of the sediments, and a lack of proper understanding of the depositional processes operating in lakes of this type. Furthermore, an active and growing industrial minerals industry based on the deposits of the salt lakes has obliterated, and will likely continue to adversely affect, the stratigraphic records of some of the basins with the greatest research potential. Notwithstanding these problems, the sediments of the salt lakes provide the best and, in some cases, only record of past environmental conditions in this semi-arid region. Paleolimnology in the northern Great Plains is poised for a rapid expansion, fuelled by the combination of significant technological breakthroughs, improvements in methodology, and a more positive view of the importance of paleolimnological research in environmental management.
Biological processes
"The absence and incomplete gathering of physiochemical habitat data to associate with biological collections have impeded progress in the classification of distinct ecological communities in salt lakes"[308]
Biological processes in salt lakes of the Great Plains are overall quite similar to those in shallow, fresh, standing waters, notwithstanding their physical and chemical extremes. Processes such as photosynthesis cause a rise in pH as CO2 is utilized by the flora, which, in turn, creates favourable conditions for carbonate precipitation [36]. The biota, however, differ between fresh and saline lakes [309]; at low salinities the species composition of salt lakes is comparable to that of their freshwater counterparts [303]. As salinity increases the diversity of species declines [37], and as salinities reach extremely high values, species diversity becomes very low and the lake is usually dominated only by halotolerant organisms [37]. Ionic composition also affects species diversity. Certain taxa are found in hypersaline waters dominated by a particular solute; anions including chloride, bicarbonate-carbonate and sulfate are important controlling factors in the species composition of salt lakes [308]. Because the brines in many of these lakes are sulfate rich, most of the meromictic lakes in the Great Plains region have a plate of purple sulfate reducing bacteria at the chemocline [175]. The importance and role of these organisms in the saline lake ecosystem requires further investigation, especially with respect to their potentially important role in the formation of some carbonate minerals.
Fossil remains of some organisms of the Great Plains salt lakes have been evaluated for use as paleoindicators. Research on the feasibility of siliceous algae, fossil pigments, as well as ostracodes for use as paleoindicators of changes in salinity and climate have been carried out on some of these lakes [310-312].
Saline springs
"...at this rate [of spring discharge] it would have taken only about 5.7 million years to remove all the missing salt from the Devonian in Saskatchewan and Manitoba." [214]
Springs and spring deposits (travertine, tufa, sinter, etc.) are common in many terrestrial settings in the Great Plains of North America. Most of these occur in association with freshwater lakes and wetlands [313,156,314], although saline springs may be an important and recognizable component of the hydrologic budgets of the salt lakes also [80,315,316,69,227,265]. The springs associated with the saline lakes often form impressive platform and pinnacle accumulations of salts up to several meters high, particularly during the winter when the cold temperatures of the lake water cause rapid and massive precipitation of many of the dissolved Mg and Na sulfates [80,176,77,63].
Saline spring systems in the Great Plains that are not associated with salt lakes have generally not received much scientific study. Saline springs are common in the Pasquia Hills region of east-central Saskatchewan, in southern and southeastern Alberta, in the Turtle Mountain area of southern Manitoba, and in the Lake Winnipegosis area of west-central Manitoba [317,132,214,319]. The hydrology, geochemistry, and biological characteristics and processes of the brine springs of the Winnipegosis area of Manitoba have been studied in detail [214,320,322-325]. The chemical composition of the brines, dominated by Na and Cl, is strikingly different than that of the saline lakes of the Great Plains and clearly reflects the dissolution of deeply buried Devonian evaporites by groundwater. The occurrence of collapse breccias and structures both within the Paleozoic bedrock in the eastern part of the Prairies and in the overlying Pleistocene sediments further confirms this groundwater dissolution hypothesis. Unfortunately, the composition, genesis and diagenesis of the tufa and sinter deposits of these springs have not been examined in detail, although the diversity of mineralogy and biologically-mediated carbonate precipitation processes are evident (Figure 27).
Figure 27 Saline springs, Lake Winnipegosis region, Manitoba. A, D: Examples of partially desiccated microbial mat surrounding a saline spring. The brain-like morphology of the mat is coated with evaporitic minerals. B: Fragmented iron-stained carbonate (high-Mg calcite) tufa deposit. C: Example of living microbial mat near a saline spring; mats can reach several centimeters in thickness and are often laminated.
In addition to their chemistry, the saline pools, marshes and saltpans associated with the brine springs in the Winnipegosis area are noteworthy because they have been found to harbour a variety of marine organisms [320,321,326,327]. The occurrence, and in some cases, dominance, of marine organisms in saline systems thousands of kilometres from the marine environment requires further investigation into dispersal mechanisms and the paleoenvironmental importance of such factors as avian transport and colonization of non-marine habitats by marine organisms.
Conclusions and future directions
The scientific study of saline lakes in the northern Great Plains is rich with unique research opportunities. Even though the lakes of the region have been studied for over a century, there are still many un-answered questions about source of the salts and the genesis and diagenesis of the inorganic components in these basins. Some of the more noteworthy sedimentary processes and concepts that have not been adequately investigated include: the role of aeolian transport in terms of sediment budgets; anthropogenic activity and its affect on brine composition, water column structure and ecology; and the impact of extreme weather events and climate change on the physical and chemical setting of the basins. Finally, there has been very little application of stable isotope techniques to better understand the water and sediment budgets of these lakes.
Likewise, since the 1930's there has been considerable effort made to document the biological characteristics and processes operating in selected basins, nonetheless there is still a major gap in our knowledge about the ecological communities of these salt lakes. Similarly, the precise role that the microbiota play in the formation of endogenic and authigenic minerals in the lakes remain largely unexplored. Of particular importance over the coming years is to better decipher how sulfate reducing bacteria and cyanobacteria, which are commonly found in both the deeper and chemically stratified lakes, as well as the microbial mats found in many of these lakes, interact with the inorganic components to help control penecontemporaneous diagenesis and authigenic mineral formation.
Because of the great diversity in basin types, brine chemistries, and depositional processes, the sediments in these lakes offer a tremendous opportunity to examine past hydrological and environmental conditions in the region. The past decade has witnessed considerable growth in interest and research on the Holocene stratigraphic records in these saline basins. It is now recognized that these lakes provide nearly the only source of detailed, high resolution, physical, biological and chemical paleoenvironmental information for the Holocene of the region. Despite the sensitivity of these deposits to environmental change, interpreting the records in terms of paleohydrology, chemistry, and climate is fraught with difficulty. Factors that complicate these interpretations include: diagenesis of the evaporites, post-depositional physical disruption of the sediments, and a lack of proper understanding of the depositional processes operating in lakes of this type.
Acknowledgements
"In comparing various authors with one another, I have discovered that some of the gravest and latest writers have transcribed, word for word, from former works, without making acknowledgment." (Pliny the Elder; Natural History. Book i. Dedication, Sect. 22)
Portions of this manuscript was prepared for and presented as parts of various guidebooks for field trips run in conjunction with conferences and research projects [117,178,55]. We thank the organizers of these conferences for permission to use this material. We are also indebted to the many students and co-researchers we have worked with on these lakes over the years and who have provided much of the data and observations summarized here. This manuscript benefited considerably from the reviews of three anonymous referees.
==== Refs
Williams WD De Deckker P, Williams WD Limnology, the study of inland waters: a comment on perceptions of studies of salt lakes, past and present Limnology in Australia 1986 Melbourne: CSIRO Australia 471 496
Gilbert GK The topographic features of lake shores United States Geological Survey Annual Report 5 1885 75 123
Gilbert GK Lake Bonneville United States Geological Survey, Monograph 1 1890 438
Russel IC Geological History of Lake Lahontan, A Quaternary Lake of Northwestern Nevada United States Geological Survey Monograph 11 1885
Forel FA Les ravins sous-lacustre des fleuves glaciares C R Acad Sci Paris 1885 101 725 728
Forel FA Le ravins sous-lacustre du Rhône dans le lac Léman Bull Soc Vaudoise Sci Nat 1887 23 85 107
Collinson JD Reading HG Lakes Sedimentary Environments and Facies 1978 Boston: Blackwell Scientific Publications 61 79
Eugster HP Kelts K Goudie AS, Pye K Lacustrine chemical sediments Chemical Sediments and Geomorphology 1983 London: Academic Press 321 368
Dean WE Fouch TD Scholle PA, Bebout DG, Moore CH Chapter 2: Lacustrine Carbonate Depositional Environments. American Association of Petroleum Geologists Memoir 33 1983 98 130
Last WM Geolimnology of salt lakes Geosciences Journal 2002 6 347 369
Reeves CC Matter A, Tucker ME Economic significance of playa lake deposits Modern and Ancient Lake Sediments 1978 Oxford: Blackwell Scientific Publications 279 290
Picard MD High LR Ethridge FG, Flores RM Physical stratigraphy of ancient lacustrine deposits Recent and Ancient Nonmarine Depositional Environments: Models for Exploration. SEPM Special Publication No 31 1981 233 259
Katz BJ (Ed) Lacustrine Basin Exploration – Case Studies and Modern Analogs American Association of Petroleum Geologists Memoir 50, Tulsa 1990
Sladen CP Gierlowski-Kordesch E, Kelts K Key elements during the search for hydrocarbons in lake systems Global Geological Record of Lake Basins 1994 1 New York :Cambridge University Press 3 17
Håkanson L Jansson M Principles of Lake Sedimentology 1983 New York : Springer-Verlag
Haworth EY Lund JWG (Eds) Lake Sediments and Environmental History 1984 Minneapolis: University of Minnesota Press
Berglund BE (Ed) Handbook of Holocene Palaeoecology and Palaeohydrology 1986 New York: John Wiley & Sons
Chivas AR Torgersen T Bowler JM (Eds) Palaeoenvironments of Salt Lakes Palaeogeography, Palaeoclimatology, Palaeoecology 1986 54 1 328 10.1016/0031-0182(86)90113-6
University of Manitoba The University of Manitoba 1972–1973 General Calendar. Winnipeg, Canada 1972
Anadón P Cabrera Ll Kelts K (Eds) Lacustrine Facies Analysis 1991 Oxford: Blackwell Scientific Publications
Gierlowski-Kordesch E Kelts K (Eds) Global Geological Record of Lake Basins, Volume 1 1994 New York: Cambridge University Press
Talbot MR Allen PA Reading HG Lakes Sedimentary Environments: Processes, Facies and Stratigraphy 1996 Oxford: Blackwell Science Ltd
Gierlowski-Kordesch EH Kelts KR (Eds) Lake Basins Through Space and Time AAPG Studies in Geology #46 2000
Davis RB (Ed) Paleolimnology and the Reconstruction of Ancient Environments 1990 Dordrecht, The Netherlands: Kluwer Academic Publishers
Saarnisto M Kahra A (Eds) INQUA Commission for the Study of the Holocene – Working Group on Laminated Sediments Proceedings of Workshop at Lammi Biological Station. Geological Survey of Finland, Special Paper 14 1992
Williams MAJ Dunkerley DL De Deckker P Kershaw AP Stokes T Quaternary Environments 1993 London: Edward Arnold
Rosen MR (Ed) Paleoclimate and Basin Evolution of Playa Systems Geological Society of America Special Paper 289 1994
Last WM Smol JP Last WM, Smol JP An Introduction to Basin Analysis, Coring and Chronological Techniques Used in Paleolimnology Tracking Environmental Change Using Lake Sediments. Volume 1: Physical, Coring, and Basin Analysis Techniques 2001 Dordrecht, The Netherlands: Kluwer Academic Publishers
Last WM Smol JP Last WM, Smol JP An Introduction to Physical and Geochemical Techniques Used in Paleolimnology Tracking Environmental Change Using Lake Sediments. Volume 2: Physical and Geochemical Techniques 2001 Dordrecht, The Netherlands: Kluwer Academic Publishers
Smol JP Pollution of Lakes and Rivers: A Paleolimnological Perspective 2002 London: Arnold
Rawson DS Moore GE The saline lakes of Saskatchewan Canadian Journal of Research (Series D) 1944 22 141 201
Northcote TG Larkin PA Frey DG Chapter 16: Western Canada Limnology in North America 1963 Madison: University of Wisconsin Press 451 485
Haynes RC Hammer UT The saline lakes of Saskatchewan. IV Primary production of phytoplankton in selected saline ecosystems Int Rev ges Hydrobiol 1978 63 337 351
Hammer UT Primary production in saline lakes. A review Hydrobiologia 1981 66 701 743
Hammer UT Taub FB The saline lakes of Canada Lakes and Reservoirs, Ecosystems of the World 23 1984 Amsterdam: Elsevier
Hammer UT Saline Lake Ecosystems of the World 1986 Dordrecht, The Netherlands Dr W. Junk Publ
Hammer UT Shamess J Haynes RC The distribution and abundance of algae in saline lakes of Saskatchewan, Canada Hydrobiologia 1983 105 1 26 10.1007/BF00025173
Waite DT (Ed) Evaluating Saline Waters in a Plains Environment Canadian Plains Proceedings 17 1986 Regina: Canada Plains Research Center
Adams GD Wetlands of the Prairies of Canada National Wetlands Working Group, Canada Committee on Ecological Land Classification, Wetlands of Canada. Ecological Land Classification Series 1988 24 155 198
Christiansen EA The Wisconsinan deglaciation of southern Saskatchewan and adjacent areas Canadian Journal of Earth Sciences 1979 16 913 938
Klassen RW Fulton RJ Quaternary geology of the southern Canadian Interior Plains Quaternary Geology of Canada and Greenland, Geological Survey of Canada Geology of Canada 1989 1 138 173
Klassen RW Late Wisconsinan and Holocene history of southwestern Saskatchewan Canadian Journal of Earth Sciences 1994 31 1822 1937
Teller JT Kehew AE (Eds) Late Glacial History of Large Proglacial Lakes and Meltwater Runoff Along the Laurentide Ice Sheet Quaternary Science Reviews 1994 13 795 981 10.1016/0277-3791(94)90001-9
Beaudoin AB A compendium and evaluation of postglacial pollen records in Alberta Canadian Journal of Archeology 1993 17 92 112
Beaudoin AB Lemmen DS Vance RE Paleoenvironmental records of postglacial climate change in the Prairie ecozone Plenary presentation at the 3rd National EMAN Meeting, Saskatoon 1997 http://www.eman-rese.ca/eman/reports/publications/nm97_paleo/intro.html
Brunskill GJ Graham BW The Offshore Sediments of Lake Winnipeg Canadian Fisheries and Marine Services. Manuscript Report Number 1540 1979
Todd BJ Lewis CFM Thorleifson LH Nielsen E Lake Winnipeg Project: Cruise Report and Scientific Results Geological Survey of Canada Open File Report 3113 1996
Todd BJ Lewis CFM Thorleifson LH Nielsen E Last WM Paleolimnology of Lake Winnipeg: Introduction and overview Journal of Paleolimnology 1998 19 211 213 10.1023/A:1007984407573
Palliser J Journals, detailed reports and observations relative to the exploration by Captain Palliser of a portion of British North America; Exploration of British North America. London: Report presented to both houses of Parliament 1862
Hammer UT The saline lakes of Saskatchewan. I Background and rationale for saline lakes research Int Rev ges Hydrobiol 1978 63 173 177
Barendregt RW (Ed) Symposium on the Paleoenvironmental Reconstruction of the Late Wisconsin Deglaciation and the Holocene, Program-with-Abstracts and Field Guide 1985 Lethbridge, Alberta; University of Lethbridge
Wilson MC Dijks IJ Barendrgt RW, Wilson MC, Jankunis FJ Introductory essay – Land of no quarter, the Palliser Triangle as an environmental-cultural pump The Palliser Triangle, A Region in Space and Time 1993 University of Lethbridge
Sauchyn DJ (Ed) Quaternary and Late Tertiary Landscapes of Southwestern Saskatchewan and Adjacent Areas Canadian Plains Proceedings 25 1993 2
Mossop G Shetsen I (Eds) Geological Atlas of Western Canada Sedimentary Basin 1994 Calgary, Alberta: Canadian Society of Petroleum Geologists and Alberta Research Council
Lemmen DS (Ed) Landscapes of the Palliser Triangle, Guidebook for the Canadian Geomorphology Research Group Field Trip, Saskatoon, Canada Canadian Association of Geographers 1996 Annual Meeting 1996
Teller JT Thorleifson LH Matile G Brisbin WC Nielsen E Sedimentology, Geomorphology, and History of the Central Lake Agassiz Basin. Fieldtrip B2 Guidebook, Winnipeg, Canada Geological Association of Canada Annual Meeting 1996
Wellsted JE Everitt JC Stadel C (Eds) The Geography of Manitoba: Its Land and Its People 1996 Winnipeg, Canada: University of Manitoba Press
Lemmen DS Dyke LD Edlund SA The Geological Survey of Canada's Ingetrated Research and Monitoring Area (IRMA) Projects: a contribution to Canadian global change research Journal of Paleolimnology 1993 9 77 83 10.1007/BF00680038
Last WM Kelley L Salt Lakes of the Northern Great Plains Field Trip Guidebook B2 2002 Saskatoon: GAC-MAC
Wolfe S Aitken A Dyck I Walker E Holocene geomorphology, archaeology, and environmental change in south-central Saskatchewan Field Trip Guidebook.A4 2002 Saskatoon: GAC-MAC
Wilby T A Motor Tour Through Canada 1914 Toronto: Bell & Cockburn
Gollop JB Wetland inventories in western Canada Transactions, 6th Congress, International Union of Game Biologists. Bournemouth, England 1963 249 264
Last WM Continental brines and evaporites of the northern Great Plains of Canada Sedimentary Geology 1989 64 207 221 10.1016/0037-0738(89)90049-3
Adams GD Didiuk AB Pochailo D Eskowich K Robarts RD, Bothwell ML Impact of drought and land use on Prairie potholes in the Antler Municipality, Saskatchewan Aquatic Ecosystems in Semi-Arid Regions, NHRI Symposium No 7 1992 27 40
Clayton L Freers TF Clayton L, Freers TF Road Log Glacial Geology of the Missouri Coteau and Adjacent Areas. North Dakota Geological Survey Miscellaneous Series 30 1967 1 19
CNC/IHD (Canadian National Committee for the International Hydrologic Decade) Hydrologic Atlas of Canada 1978 Ottawa: Fisheries and Environment Canada
Kenny BC Sediment resuspension and currents in Lake Manitoba Journal of Great Lakes Research 1985 11 85 96
Last WM Sedimentology of a saline playa in the northern Great Plains, Canada Sedimentology 1989 36 109 123
Donovan JJ Renaut R, Last WM On the measurement of reactive mass fluxes in evaporative groundwater-source lakes Sedimentology and Geochemistry of Modern and Ancient Saline Lakes. SEPM Special Publication No 50 1994 33 50
Wood WW Sanford WE Eolian transport, saline lake basins, and groundwater solutes Water Resources Research 1995 31 3121 3129 10.1029/95WR02572
Spry IM (Ed) The papers of the Palliser Expedition, 1857–1860 1968 Toronto :Champlain Society
Langham EJ Interim report and program review of the Big Quill Lake study 1970 Saskatoon: Saskatchewan Research Council Report E 70-12
Rutherford AA Water quality survey of Saskatchewan surface waters 1970 Saskatoon: Saskatchewan Research Council Report C 70-1
Whiting JM Closed drainage lakes are a climatic change indicator 1974 Saskatoon: Saskatchewan Research Council Report E74-10
Whiting JM The hydrologic and chemical balance of the Big Quill Lake basin 1977 Saskatoon: Saskatchewan Research Council Publication E77-12
Whiting JM The effects of climate on prairie lakes Proceedings of the Canadian Hydrology Symposium No 16, Drought: The Impending Crisis? 1986 Associate Committee on Hydrology
Last WM Sedimentology of playa lakes of the northern Great Plains Canadian Journal of Earth Sciences 1984 21 107 125
Christiansen EA Collapse structures near Saskatoon, Saskatchewan Canadian Journal of Earth Sciences 1967 4 757 767
Christiansen EA Geology of the Crater Lake collapse structure in southeastern Saskatchewan Canadian Journal of Earth Sciences 1971 8 1505 1513
Cole LH Sodium sulfate of western Canada. Occurrence, uses and technology Canadian Department of Mines Publication Number 646 1926
Sahinen UM Preliminary report on sodium sulfate in Montana Bureau of Mines and Geology, Montana School of Mines Report 1948
Grossman IG The sodium sulfate deposits of western North Dakota North Dakota Geological Survey Report of Investigations No 1 1949
Grossman IG Origin of sodium sulfate deposits of the Northern Great Plains of Canada and the United States United States Geological Survey Professional Paper 600-B 1968 104 109
Last WM Nicholaichuk W, Steppuhn H Salt lakes of western Canada: a spatial and temporal geochemical perspective Proceedings of Symposium on Water Management Affecting the Wet-to-Dry Transition 1988 Saskatoon, Saskatchwan: Water Studies Institute 99 113
Murphy EC The Sodium Sulfate Deposits of Northwestern North Dakota North Dakota Geological Survey Report of Investigation No 99 1996
Stalker A Buried valleys in central and southern Alberta Geological Survey of Canada Paper 60-32 1961
Christiansen EA Preglacial valleys in southern Saskatchewan Saskatchewan Research Council Map No 3 1967
Bird JB The Natural Landscapes of Canada, A Study in Regional Earth Science 1980 2 Toronto: John Wiley & Sons Canada Limited
Meneley WA Christiansen EA Kupsch WO Preglacial Missouri River in Saskatchewan Journal of Geology 1957 65 441 447
Bluemle JP The Face of North Dakota North Dakota Geological Survey, Education Series 21 1991
Kehew AE Lord ML Glacial-lake spillways of the central interior plains, Canada-U.S.A Canadian Geographer, Canadian Landform Examples – 12 1989 33 274 277
Kehew AE Teller JT History of late glacial runoff along the southwestern margin of the Laurentide ice sheet Quaternary Science Reviews 1994 13 859 877 10.1016/0277-3791(94)90006-X
Witkind IJ The localization of sodium sulfate deposits in northeastern Montana and northwestern North Dakota American Journal of Science 1952 250 667 676
Rueffel PG Rau JL, Dellwig LF Natural sodium sulfate in North America Third Symposium on Salt, Northern Ohio Geological Society 1968 429 451
Wood WW Sanford WE Groundwater control of evaporite deposition Economic Geology 1990 85 1226 1235
Donovan JJ Geochemical and hydrological dynamics of evaporative groundwater-dominated lakes of Montana and North Dakota PhD Thesis 1992 Pennsylvania State University
Donovan JJ Rose AW Geochemical evolution of lacustrine brines from variable-scale groundwater circulation Journal of Hydrology 1994 154 35 62 10.1016/0022-1694(94)90211-9
Rutherford AA Water Quality Survey of Saskatchewan Groundwaters Saskatoon: Saskatchewan Research Council, Report C-66-1 1967
Brown IC Groundwater in Canada Economic Geology Report Number 24 1967 Geological Survey of Canada
Freeze RA Regional groundwater flow – Old Wives Lake drainage basin, Saskatchewan Canadian Inland Waters Directorate, Scientific Series No 5 1969
Tokarsky O Hydrogeologic Profile Alberta-Saskatchewan Boundary Prairie Province Water Board 1985
Tokarsky O Hydrogeologic Profile Saskatchewan-Manitoba Boundary Prairie Province Water Board Report No 79 1986
Lennox DH Maathuis H Pederson D Back W, Rosenshein JS, Seaber PR Region 13, Western glaciated plains The Geology of North America, volume O-2, Hydrogeology 1988 The Geological Society of America
Pupp C Maathuis H Grove G Groundwater Quality in Saskatchewan: Hydrogeology, Quality Concerns, Management National Hydrology Research Institute Contribution No CS-91028 1981
Betcher R Grove G Pupp C Groundwater in Manitoba: Hydrogeology, Quality Concerns, Management National Hydrology Research Institute Contribution No CS-93017 1995
Remenda VH Birks SJ Lemmen DS, Vance RE Groundwater in the Palliser Triangle: an overview of its vulnerability and potential to archive climate information Holocene Climate and Environmental Changes in the Palliser Triangle, Southern Canadian Prairies: Geological Survey of Canada Bulletin 534 1999 23 55
Birks SJ Remenda VH Lemmen DS, Vance RE Groundwater inputs to a closed-basin saline lake, Chappice Lake, Alberta Holocene Climate and Environmental Changes in the Palliser Triangle, Southern Canadian Prairies: Geological Survey of Canada Bulletin 534 1999 23 55
Campbell ID Last WM Campbell C Clair S McAndrews JH Edwards TWD The Late Holocene Paleohydrology of Pine Lake, Alberta: a Comparison of Proxy Types Journal of Paleolimnology 2000 25 427 441 10.1023/A:1008155622656
Last WM Chemical composition of saline and subsaline lakes of the northern Great Plains, western Canada International Journal of Salt Lake Research 1992 1 47 76
Cole GA Textbook of Limnology 1975 St. Louis: Mosby
Dyke AS Prest VK Late Wisconsinan and Holocene retreat of the Laurentide Ice Sheet Geological Survey of Canada, Map 1702A 1987
Gravenor CP Kupsch WO Ice-disintegration features in western Canada Journal of Geology 1959 67 48 67
Clayton L Cherry JA Clayton L, Freers TF Pleistocene superglacial and ice-walled lakes of west-central North America Glacial Geology of the Missouri Coteau and Adjacent Areas. North Dakota Geological Survey Miscellaneous Series 30 1967 47 52
Christiansen EA Quaternary Geology and its application to engineering practice in the Saskatoon-Regina-Watrous area, Saskatchewan Geological Association of Canada Guidebook, Field trip C 1973
Rueffel PG Development of the largest sodium sulfate deposit in Canada Can Min Met Bull 1968 61 1217 1228
Shang Y Paleolimnology, paleochemistry and paleohydrology of North Ingebright Lake, southwestern Saskatchewan, Canada PhD Thesis 1999 University of Manitoba
Last WM Sedimentology, Geochemistry, and Evolution of Saline Lakes of the Saskatoon Area Within-Conference Fieldtrip Guidebook, Sedimentary and Paleolimnological Records of Saline Lakes, Saskatoon 1991
Christiansen EA Gendzwill DJ Meneley WA Howe Lake: a hydrodynamic blowout structure Canadian Journal of Earth Sciences 1982 19 1122 1139
Gendzwill DJ Hajnal Z Seismic investigation of the Crater Lake collapse structure in southeastern Saskatchewan Canadian Journal of Earth Sciences 1971 8 1514 1524
Downey JS Geology of northeastern North Dakota North Dakota Geological Survey Miscellaneous Series 39 1969
Meneley WA Conrol of Brightwater Marsh flowing testhole Saskatchewan Research Council, Saskatoon, Internal Report 1977
Bluemle JP Veblins, hydrodynamic blowouts, and other unusual geologic features in North Dakota NDGS Newsletter 1992 19 7 11
Bluemle JP Aber JS Hydrodynamic blowouts in North Dakota Glaciotectonics and Mapping Glacial Deposits, Proceedings of the INQUA Commission on Formation and Properties of Glacial Deposits, Canadian Plains Research Center, University of Regina, Canadian Plains Proceedings 25-1 1993 259 266
Aber JS Aber JS Geomorphology and structural genesis of the Dirt Hills and Cactus Hills, southern Saskatchewan Glaciotectonics and Mapping Glacial Deposits, Proceedings of the INQUA Commission on Formation and Properties of Glacial Deposits, Canadian Plains Research Center, University of Regina, Canadian Plains Proceedings 25-1 1993 9 35
Christiansen EA Whitaker SH Legget RF Glacial thrusting of drift and bedrock Glacial Till, The Royal Society of Canada Special Publication No 12 1971 121 130
Hind HY Narrative of the Canadian Red River exploring expedition of 1857 and of the Assinniboine and Saskatchewan exploring expedition of 1858 1861 Toronto: MacLear
Waite DT (Ed) Prairie Surface Waters: Problems and Solutions Canadian Plains Proceedings 7 1980
Batt BD Anderson MG Anderson CD Caswell FD van der Valk AG The use of prairie potholes by North American ducks Northern Prairie Wetlands 1989 Iowa State Press 204 227
Richardson JL Arndt JL What use prairie potholes? Journal of Soil and Water Conservation 1989 44 196 198
Adams GD Gentle GC Spatial Changes in waterfoul habitat 1964-1974, on two land types in the Manitoba Newdale Plain. Canadian Wildlife Service, Environment Canada, Occasional Paper Number 38, Saskatoon, Saskatchewan 1978
Cole LH The salt industry of Canada Canadian Department of Mines Publication Number 716 1930
Harper CT (Ed) Geology and Mineral and Petroleum Resources of Saskatchewan Sasktachewan Industry and Reouces Miscellaneous Report 2003–7 2003
Broughton PL Guillet GR, Martin W Sodium sulfate deposits of western Canada The Geology of Industrial Minerals in Canada 1984 29 Canadian Institute of Mining and Metallurgy 195 200
Tremblay JY Industrial Minerals Canadian Mining Journal 1984 105 85 87
Barry GS Sodium sulfate Canadian Mining Journal 1986 107 77
Barry GS Boucher JA Prud'homme M Other industrial minerals Canadian Mining Journal 1985 106 82 83
Eatock WH Gilboy CF, Vigrass LW The Big Quill Lake sulfate of potash project Economic Minerals of Saskatchewan. Saskatchewan Geological Society Special Publication 1987 8 206 210
Last WM Slezak LA Gilboy CF, Vigrass LW Sodium sulfate deposits of western Canada Economic Minerals of Saskatchewan. Saskatchewan Geological Society Special Publication 1987 8 197 205
Marquart RG uPCSM, Mainframe search mathc on an IBM PC Powder Diffraact 1986 1 34 36
Weisman WI Tandy CD Lefond SJ Sodium sulfate deposits Industrial Minerals and Rocks 1975 4 American Institute of Mining and Metallurgy, New York: Petroleum Engineering Inc 1081 1093
Tomkins RV Magnesium in Saskatchewan Saskatchewan Department of Mineral Resources Report No 11 1954
Templeton CW Geological Study and Reserves Report, Metiskow Lakes Salts Deposit, Township 39 Range 6 W 4, Alberta Consultant's Report prepared for Canadian Chemical Company 1968
Alberta Sulfate Limited Evaluation of Remaining Salt Deposits. Company report on Metiskow Lake, Alberta 1981
James F. MacLaren Limited Mineral Aggregate Study of the Southern Interlake Region. Manitoba Department of Energy and Mines Open File Report OF80-2 1980
Groom HD Surficial Geology and Aggregate Resources of the Fisher Branch Area Manitoba Energy and Mines Aggregate Report AR84-2 1985
Mollard JD Mollard DG Gilboy CF, Vigrass LW Location, exploration and production of gravel resources in Saskatchewan Economic Minerals of Saskatchewan. Saskatchewan Geological Society Special Publication 1987 8 211 216
Brown R (Ed) The Collected Poems of Al Purdy 1986 Toronto: McClelland and Stewart
Timms BV Lake Geomorphlogy 1992 Adelaide: Gleneagles Publishing
Håkanson L The influence of wind, fetch, and water depth on the distribution of sediments in Lake Vanern, Sweden Canadian Journal of Earth Sciences 1977 14 397 412
Håkanson L On lake form, lake volume and lake hypsograhic survey Geogr Ann 1977 59 Series A 1 29
Håkanson L Sediments as indicators of contamination – investigations in the four largest Swedish lakes National Swedish Environmental Protection Board, SNV PM 839, Uppsala 1977
Håkanson L A Manual of Lake Morphometry 1981 Berlin: Springer-Verlag
Jones BF van Denburgh AS Geochemical influences on the chemical character of closed lakes Symposium of Garda, Hydrology of Lakes and Reservoirs, Proceedings of the International Association for Scientific Hydrology, Publication 70 1966 438 446
Reeves CC Introduction to Paleolimnology 1968 Amsterdam: Elsevier
Hammer UT Haynes RC The saline lakes of Saskatchewan. II. Locale, hydrology, and other physical aspects Int Rev ges Hydrobiol 1978 63 179 203
Mitchell P Prepas E (Eds) Atlas of Alberta Lakes 1990 Edmonton: University of Alberta Press
Last WM Schweyen TH Sedimentology and geochemistry of saline lakes of the northern Great Plains Hydrobiologia 1983 105 245 263 10.1007/BF00025192
Last WM Rosen MR Paleohydrology of playas in the northern Great Plains: perspectives from Palliser's Triangle Paleoclimate and Basin Evolution of Playa Systems. Geological Society of America, Special Paper 289 1994 69 80
Richardson JL Former lake-level fluctuations – their recognition and interpretation Mitt Internat Verein Limnol 1969 17 78 93
Teller JT Last WM Paleohydrological indicators in playas and salt lakes, with examples from Canada, Australia, and Africa Palaeogeography, Palaeoclimatology, Palaeoecology 1990 76 215 240 10.1016/0031-0182(90)90114-M
Schweyen TH Sedimentology and paleohydrology of Waldsea lake, Saskatchewan, an ectogenic meromictic saline lake MSc thesis 1984 University of Manitoba
Last WM Schweyen TH Late Holocene history or Waldsea Lake, Saskatchewan, Canada Quaternary Research 1985 24 219 234 10.1016/0033-5894(85)90008-0
Last WM Slezak LA Paleohydrology, sedimentology, and geochemistry of two deep saline lakes from southern Saskatchewan Geogr Phys Quat 1986 11 5 15
Reeves CC (Ed) Playa Lake symposium 1972 Lubbock, Texas: International Center for Arid and Semi-arid Land Studies
Rosen MR Rosen MR The importance of groundwater in playas: a review of playa classifications and the sedimentology and hydrology of playas Paleoclimate and Basin Evolution of Playa Systems. Geological Society of America Special Paper 289 1994 1 18
Bell JS Report on the country between Red River and the South Saskatchewan Geological Survey of Canada Report of Progress, 1873–74 1875
Scott MD Scott SA Waite DT Saline lake management for waterfowl resources in western North America: a review of concepts Evaluating Saline Waters in a Plains Environment. Canadian Plains Proceedings 1986 17 23 38
Cameron DR Waite DT Saline waters: agricultural uses Evaluating Saline Waters in a Plains Environment. Canadian Plains Proceedings 1986 17 75 100
Harrington R Johnson B Hunter F (Eds) Responding to Global Climate Change in the Prairies, Volume III of the Canada Country Study: Climate Impacts and Adaptation Environment Canada, Prairie and Northern Region 1997
Moore GE A limnological study of certain lakes of southern Saskatchewan MA Thesis 1939 University of Saskatchewan
Thomas JFJ Industrial water resources of Canada. Nelson River Drainage Basin in Canada, 1953–56 Canadian Water Survey Report No 10 1959
Inland Waters Directorate Water Quality Data – Alberta, 1961–1973 1975 Water Quality Branch, Environment Canada, Ottawa
Inland Waters Directorate Detailed Surface Water Quality Data – Manitoba, 1974–1977 1980 Ottawa: Water Quality Branch, Environment Canada
Inland Waters Directorate Detailed Surface Water Quality Data – Saskatchewan, 1980–1981 1984 Ottawa: Water Quality Branch, Environment Canada
Hammer UT The saline lakes of Saskatchewan, III. Chemical characterization Int Rev Ges Hydrobiol 1978 63 311 335
Last WM Schweyen TH Egan DM Sedimentology, Geochemistry and Stratigraphy of the Saline Lakes of the Canadian Great Plains Final Report to the Max Bell Foundation 1983
Lambert S Hydrogeochemistry of saline lakes in the northern Interior Plains of western Canada and northern United States B.Sc. Honours Thesis 1989 University of Manitoba, Winnipeg
Last WM Sedimentology, Geochemistry, and Evolution of Saline Lakes of the Northern Great Plains Saskatoon, Post-Conference Fieldtrip Guidebook, Sedimentary and Paleolimnological Records of Saline Lakes 1991
Govett GJS Sodium sulfate deposits in Alberta Alberta Research Council Preliminary Report 58-5 1958
Bierhuizen JFH Prepas EE Relationships between nutrients, dominant ions, and phytoplankton standing crop in prairie saline lakes Canadian Journal of Fisheries and Aquatic Sciences 1985 42 1588 1594
Driver EA Pedden DG The chemistry of surface water in the prairie ponds Hydrobiologia 1977 53 33 48 10.1007/BF00021230
Schwartz FW Gallup DN Some factors controlling the major ion chemistry of small lakes; examples from the prairie parkland of Canada Hydrobiologia 1978 58 65 81 10.1007/BF00018896
Lieffers VJ Shay JM Ephemeral saline lakes in the Canadian prairies – their classification and management for emergent macrophyte growth Hydrobiologia 1983 105 85 94 10.1007/BF00025179
Royer LM Investigations of lakes in Moose Mountain Provincial Park 1960 Saskatchewan Fisheries Laboratory, Dept. Of natural Resources
Driver EA Limnological aspects of some saline lakes in west-central Manitoba Journal of Fisheries Research Board of Canada 1965 22 1165 1173
Barica J Geochemistry and nutrient regime of saline eutrophic lakes in the Erickson-Elphinstone district of southwestern Manitoba Canadian Fisheries and Marine Services Research Development Technical Report 511 1975
Williams WD Sherwood JE Definition and measurement of salinity in salt lakes International Journal of Salt Lake Research 1994 3 53 63 10.1007/BF01990642
Desai K Moore EJ Equivalent NaCl determination from ionic concentrations Log Analyst 1969 X
Hem JD Study and Interpretation of the Chemical Characteristics of Natural Waters United States Geological Survey Water-Supply Paper 2254 1985 3 USGS
Drever JI The Geochemistry of Natural Waters 1988 2 Englewood Cliff, New Jersey Prentice Hall
Williams WD Weatherley AH The chemical characteristics of lentic surface waters in Australia Australian Inland Waters and Their Fauna 1967 Canberra: Australian National University Press 18 77
Carpenter AB Origin and chemical evolution of brines in sedimentary basins Oklahoma Geological Survey Circular 79 1978 60 77
Tomkins RV Natural sodium sulfate in Saskatchewan Saskatchewan Department of Mineral Resources Report No 6 1954 2 Government of Saskatchewan
Fraser FJ McLearn FH Russell LS Warren PS Wickenden RTD Geology of southern Saskatchewan Geological Survey of Canada Memoir 176 1935
Winter TC Lerman A, Imboden DM, Gat JR Hydrological processes and the water budget of lakes Physics and Chemistry of Lakes 1995 New York: Springer-Verlag
Gorham E Dean WE Sanger JE The chemical composition of lakes in the north-central United States Limnology and Oceanography 1983 28 287 301
Koch GS Link RF Statistical Analysis of Geological Data 1970 New York: Dover Publications Inc
Braitsch O Salt Deposits, Their Origin and Composition 1971 New York: Springer-Verlag
Hardie LA Smoot JP Eugster HP Matter A, Tucker ME Saline lakes and their deposits: a sedimentological approach Modern and Ancient Lake Sediments. International Association of Sedimentologists, Special Publication No 2 1978 7 42
Rozkowski A The origin of hydrochemical patterns in hummocky morain Canadian Journal of Earth Sciences 1967 4 1065 1092
Wallick EI Chemical evolution of groundwater in a drainage basin of Holocene age, east-central Alberta, Canada Journal of Hydrology 1981 54 245 283 10.1016/0022-1694(81)90163-3
Last WM Modern sedimentology and hydrology of Lake Manitoba, Canada Environmental Geology 1984 5 177 190
Slezak LA Sedimentology of Freefight Lake, Saskatchewan MSc Thesis 1989 University of Manitoba. Winnipeg, Canada
Dean WE Gorham E Major chemical and mineral components of profundal surface sediments n Minnesota Lakes Limnol Ocean 1976 21 259 284
Winter TC Classification of the hydrologic settings of lakes in the North Central United States Water Resources Res 1977 13 753 767
Lieffers VJ Environment and ecology of Scirpus maritimus L. var. apludosus (nels.) Kuk. in saline wetlands of the Canadian Prairies PhD thesis 1981 University of Manitoba
Lockhart EB Origin of gypsum in Lydden Lake, Saskatchewan BSc Honours Thesis 1983 University of Manitoba, Department of Geological Sciences
Lockhart EB Last WM Sedimentology and geochemistry of a small salt playa in southern Saskatchewan Geological Association of Canada Program with Abstracts 1984 9
Hammer UT Parker J Limnology of a perturbed highly saline Canadian lake Arch Hydrobiol 1984 102 31 42
Last WM Rodriguez-Clemente R, Tardy Y Sedimentology, geochemistry, and evolution of a saline playa in the Northern Great Plains of Canada Geochemistry and Mineral Formation in the Earth Surface 1987 Madrid: Consejo Superior de Investigaciones Cientificas, Centre National de la Recherche Scientifique 319 337
Sloan CE Groundwater hydrology of prairie potholes in North Dakota United States Geological Survey Professional Paper 585C 1972
Slezak LA Last WM Glaser JD, Edwards J Geology of sodium sulfate deposits of the northern Great Plains Twentieth Forum on the Geology of Industrial Minerals. Maryland Geological Survey Special Publication No 2 1985 105 115
Van Everdingen RO Turnock AC Surface-water composition in southern Manitoba reflecting discharge of saline subsurface waters and subsurface solution of evaporites Geoscience Studies in Manitoba. Geological Association of Canada Special Paper Number 9 1971 343 352
Grasby SE Saline spring geochemistry, west-central Manitoba Report of Activities 2000, Manitoba 2000 Industry, Trade and Mines, Manitoba Geological Survey 214 216
Wallick EI Krouse HR Sulfur isotope geochemistry of a groundwater-generated Na2SO4/Na2CO3 deposit and the associated drainage basin of Horseshoe Lake, Metiskow, east central Alberta, Canada Strasbourg, France: 2nd International Symposium on Water-rock Interaction 1997 II56 II64
Rozkowska AD Roskowski A Seasonal changes of slough and lake water chemistry in southern Saskatchewan, Canada Journal of Hydrology 1969 7 1 13 10.1016/0022-1694(68)90192-3
Moran SR Groenewold GH Cherry JA Geologic, hydrologic, and geochemical concepts and techniques in characterization for mined land reclamation North Dakota Geological Survey, Report of Investigations, No 63 1978
Timpson ME Richardson JL Ionic composition and distribution ion saline seeps of southwestern North Dakota Geoderma 1986 37 295 305 10.1016/0016-7061(86)90031-5
Arndt JL Richardson JL Robarts RD, Bothwell ML Carbonate and gypsum chemistry in saturated, neutral pH soil environments Aquatic Ecosystems in Semi-Arid Regions, NHRI Symposium No 7 1992 179 187
LaBaugh JW Winter TC Swanson GA Rosenberry DO Nelson RD Euliss NH Changes in atmospheric circulation patterns affect midcontinent wetlands sensitve to climate Limnology and Oceanography 1996 41 864 870
Winter TC Numerical simulation of steady-state three dimensional groundwater flow near lakes Water Resources Research 1978 14 245 254
Hendry MJ Schwartz F Colgan BL Hydrogeology of saline seeps Proceedings of Soil Salinity Conference, Lethbridge 1982 25 53
Doering EJ Colgan BL Water and salt movement in prairie soils Proceedings of Soil Salinity Conference, Lethbridge 1982 55 78
Moorhouse WW The Study of Rocks in Thin Section 1959 New York: Harper & Brothers
Veldman WM Shoreline processes on Lake Winnipeg MSc Thesis 1969 University of Manitoba, Winnipeg, Canada
Sack LA Last WM Lithostratigraphy and recent sedimentation history of Little Manitou Lake, Saskatchewan, Canada Journal of Paleolimnology 1994 10 199 212 10.1007/BF00684033
Kushnir DW Sediments of the South Basin of Lake Winnipeg MSc Thesis 1971 University of Manitoba, Winnipeg
Last WM Todd BJ, Lewis CFM, Thorleifson LH, Nielsen E Bulk composition, texture, and mineralogy of Lake Winnipeg core and surface grab samples Lake Winnipeg Project: Cruise Report and Scientific Results, Geological Survey of Canada Open File Report 3113 1996 209 220
Last WM Todd BJ, Lewis CFM, Thorleifson LH, Nielsen E Geochemistry, texture, and mineralogy of Lake Winnipeg sediments Lake Winnipeg Project: Cruise Report and Scientific Results, Geological Survey of Canada Open File Report 3113 1996 485 584
Henderson P Last WM Holocene sedimentation in Lake Winnipeg, Manitoba, Canada: implications of compositional and textural variations Journal of Paleolimnology 1998 19 265 284 10.1023/A:1007986116660
Dean WE Stuiver M Bradbury JP, Dean WE Stable carbon and oxygen istope studies of the sediments of Elk Lake, Minnesota Elk Lake Minnesota: Evidence for Rapid Climate Change in the North-Central United States. Geological Society of America Special Paper 276 1993 163 180
Last WM Teller JT Paleolimnology of Lake Manitoba, Canada: the lithostratigraphic evidence Geographie physique et Quaternaire 2004 56 135 154
Ehrlich WA Rice WM Post-glacial weathering of Mankato till in Manitoba Journal of Geology 1955 63 527 537
Mills GF Smith RE Soils of the Grahamdale area Manitoba Soil Survey, Soil Report No 16 1971
Teller JT Last WM Postglacial sedimentation and history in Lake Manitoba Manitoba Department of Mines Resources and Environment Report 79-41 1979
Egan DM Mineralogical aspects of hypersaline lakes in southern Saskatchewan M.Sc. Thesis 1984 University of Manitoba, Winnipeg, Canada
Dreimanis A Vagners UJ Goldthwait RP Biomodal distribution of rock and mineral fragments in basal till Till/A Symposium 1971 Ohio State University Press 237 250
Last WM Holocene carbonate sedimentation in Lake Manitoba, Canada Sedimentology 1982 29 691 704
Teller JT Last WM Pedogenic zones in postglacial sediment of lake Manitoba, Canada Earth Sur Proc land 1982 7 367 397
Last WM Teller JT Teller JT, Clayton L Holocene climate and hydrology of the Lake Manitoba basin Glacial Lake Agassiz. Geol Assoc Can Spec Paper 26 1983 333 353
Last WM Geolimnology of Freefight Lake: an unusual hypersaline lake in the northern Great Plains of western Canada Sedimentology 1993 40 431 448
Ghebre-Egziabhier K St Arnaud RJ Carbonate mineralogy of lake sediments and surrounding soils I. Blackstrap Lake Canadian Journal of Soil Science 1983 63 245 257
Ghebre-Egziabhier K St Arnaud RJ Carbonate mineralogy of lake sediments and surrounding soils 2. The Qu'Appelle Lake Canadian Journal of Soil Science 1983 63 259 269
Kelts K Hsü KJ Lerman A Freshwater carbonate sedimentation Lakes – Chemistry, Geology, Physics 1978 New York: Springer-Verlag 295 324
Dean WE Ethridge FG, Flores RM Carbonate minerals and organic matter in sediments of modern north temperate hard-water lakes Recent and Ancient Nonmarine Depositional Environments: Models for Exploration. SEPM Special Publication No 31 1981 213 231
Möller P Kubanek F Role of magneium in nucleation processes of calcite, aragonite and dolomite Neues Jarb Abh 1976 126 199 220
Chilingar GV Zenger DH Bissell HJ Wolf KH Larsen G, Chilingar GV Dolomites and dolomitization Diagenesis in Sediments and Sedimentary Rocks. Developments in Sedimentology 25A 1979 Amsterdam: Elsevier 425 526
Last WM Lacustrine dolomite – an overview of modern, Holocene, and Pleistocene occurrences Earth-Science Reviews 1990 27 221 263 10.1016/0012-8252(90)90004-F
Ginn FM Last WM Londry K Dolomite, Microbiota, And Salt Lakes Of Western Canada CANQUA Water, Ice, Land, and Life: The Quaternary Interface. Program and Abstracts 2005
Zenger DH Mazzullo SJ (Eds) Dolomitization 1982 Stroudsburg: Hutchinson Ross Publishing Company
Shukla V Baker PA (Eds) Sedimentology and Geochemistry of Dolostones SEPM Special Publication No 43 1988
Purser B Tucker M Zenger D Dolomites International Association of Sedimentologists, Special Publication Number 21 1994
Deelman JC Low-temperature formation of dolomite and magnesite 2005 http://www.jcdeelman.demon.nl/dolomite/bookprospectus.html
Callender E The postglacial sedimentology of Devils Lake, North Dakota PhD Thesis 1968 University of North Dakota, Grand Forks
Owen JB Tubb R Anderson DW Callender E The biogeochemistry of Devils Lake, North Dakota United States OWRR Project Report A-006-NDAK 1973
Folk RL Land LS Mg/Ca ratio and salinity: two controls over crystallization of dolomite American Association of Petroleum Geologists Bulletin 1975 59 60 68
Baker PA Kastner M Constraints on the formation of sedimentary dolomite Science 1981 213 214 216 17782787
Morrow DW Diagenesis, 1. Dolomite – Part 1: The chemistry of dolomitization and dolomite precipitation Geoscience Canada 1982 9 95 107
Schweyen TH Last WM Scott MD Sedimentology and paleohydrology of Waldsea Lake, Saskatchewan Candian Plains Proceedings 1983 11 45 59
Last WM Slezak LA Chivas AR, De Deckker P Geolimnology of Freefight Lake, Saskatchewan SLEADS Workshop 87 1987 Canberra: The Australian National University 9 11
Vance RE A Paleobotanical study of Holocene Drought Frequency in Southern Alberta PhD Dissertation 1991 Burnaby, Canada: Simon Fraser University
Vance RE Clague JJ Mathewes RW Holocene paleohydrology of a hypersaline lake in southeastern Alberta Journal of Paleolimnology 1993 8 103 120 10.1007/BF00119784
Vance RE Mathewes RW Clague JJ 7000 year record of lake-level change on the northern Great Plains: a high-resolution record proxy of past climate Geology 1992 20 879 882 10.1130/0091-7613(1992)020<0879:YROLLC>2.3.CO;2
Birks SJ Remenda VH Hydrogeological investigation of Chappice Lake, southeastern Alberta: groundwater inputs to a saline basin Journal of Paleolimnology 1999 21 235 255 10.1023/A:1008041810022
Last WM Vance RE Bedding characteristics of Holocene sediments from salt lakes of the northern Great Plains, western Canada Journal of Paleolimnology 1997 17 297 318 10.1023/A:1007994526745
Timpson ME Richardson JL Keller LP McCarthy GJ Evaporite mineralogy associated with saline seeps in southwestern North Dakota Soil Science of America Journal 1986 50 490 493
Skarie RL Richardson JL McCarthy GJ Maianu A Evaporite mineralogy and groundwater chemistry associated with saline soils in eastern North Dakota Soil Science Society of America Journal 1987 51 1272 1377
Renaut R Rosen MR Carbonate and evaporite sedimentation at Clinton Lake, British Columbia Paleoclimate and Basin Evolution of Playa Systems. Geological Society of America, Special Paper 289 1994 49 68
Renaut R Stead D The saline lakes of southern British Columbia, Canada. Global Geological Record of Lake Basins, Volume 1 1994 New York: Cambridge University Press 415 418
Renaut R Stead D Last Chance Lake, a natric playa-lake in interior British Coumbia, Canada. Global Geological Record of Lake Basins, Volume 1 1994 New York: Cambridge University Press 415 418
Renaut R Stead D Owen RB The saline lakes of the Fraser Plateau, British Columbia, Canada. Global Geological Record of Lake Basins, Volume 1 1994 New York: Cambridge University Press 419 423
Pueyo Mur JJ Inglís-Urpinell M Rodriguez-Clemente R, Tardy Y Substrate mineralogy, pore brine composition and diagenetic processes in the playa lakes of Los Monegros and Bajo Aragón (Spain) Geochemistry and Mineral Formation in the Earth Surface 1986 Madrid: Consejo Superior de Investigaciones Cientificas, Centre National de la Recherche Scientifique 351 372
Ordóñez S Sánchez Moral S García Del Clura MA Rodríguez Badiola E Renaut R, Last WM Precipitation of salts from Mg-(Na)-SO4-Cl playa-lake brines: the endorheic saline ponds of La Mancha, central Spain Sedimentology and Geochemistry of Modern and Ancient Saline Lakes. SEPM Special Publication No 50 1994 61 72
Goudie AS Cook RU Salt efflorescences and saline lakes, a distributional analysis Geoforum 1984 15 563 582 10.1016/0016-7185(84)90025-3
Goudie AS Pans Progress in Physical Geography 1991 15 221 235
Keller LP McCarthy GJ Richardson JL Laboratory modeling of Northern Great Plains salt efflorescence mineralogy Soil Science Society of America Journal 1986 50 1363 1367
Keller CP McCarthy GS Richardson JL Mineralogy and stability of soil evaporites in North Dakota Soil Science Society of America Journal 1986 50 1069 1071
Lyons WB Hines ME Last WM Renaut R, Last WM Sulfate reduction rates in sediments of salt lakes of differing chemistries: implications for organic carbon preservation SEPM Special Publication Number 50, Sedimentary Records of Modern and Ancient Saline Lakes 1994 13 21
Last WM Roberts RD, Bothwell ML Salt Lake Paleolimnology in the northern Great Plains: the Facts, The Fears, The Future Aquatic Ecosystems in Semi-Arid Regions, NHRI Symposium Series 7 1992 Environment Canada 51 62
Sly PG Lerman A Sedimentary processes in lakes Lakes – Chemistry, Geology, Physics 1978 New York: Springer-Verlag 65 90
Lerman A Imboden DM Gat JR (Eds) Physics and Chemistry of Lakes 1995 New York: Springer-Verlag
Boden SE A chemical and sedimentological study of Deadmoose Lake, Saskatchewan BSc Honours Thesis 1985 Winnipeg, Canada: University of Manitoba
Sack LA Geolimnology and paleolimnology of Little Manitou Lake, Saskatchewan BSc Honours Thesis 1993 Winnipeg, Canada University of Manitoba
Last WM Rates of sediment deposition in a hypersaline lake in the northern Great Plains, western Canada International Journal of Salt Lake Research 1993 2 47 58
Last WM Renaut R, Last WM Deep-water evaporite mineral formation in lakes of western Canada Sedimentology and Geochemistry of Modern and Ancient Saline Lakes. SEPM Special Publication No 50 1994 51 60
Kendall AC Walker RG Evaporites Facies Models. Geoscience Canada Reprint Series 1 1984 259 296
Sonnenfeld P Brines and Evaporites 1984 Orlando, Florida: Academic Press
Warren JK Evaporite Sedimentology: Importance in Hydrocarbon Accumulation 1989 Englewood Cliffs, New Jersey: Prentice-Hall, Inc
Einsele G Sedimentary Basins 1992 Berlin: Springer-Verlag
Warren J Evaporites, Their Evolution and Economics 1999 London: Blackwell Science
Holser WT Burns RG Mineralogy of evaporites Marine Minerals. Min Soc Amer Short Course Notes 1979 6 211 294
Beaty CB Climate change, some doubts Geol Soc Amer Bulletin 82 1971 1395 1397
Last WM Slezak LA The salt lakes of western Canada: a paleolimnological overview Hydrobiologia 1988 158 301 316 10.1007/BF00026286
Nambudiri EMV Teller JT Last WM Pre-Quaternary microfossils – a guide to errors in radiocarbon dating Geology 1980 8 123 127 10.1130/0091-7613(1980)8<123:PMGTEI>2.0.CO;2
Environment Canada (Adaptation and Impacts Section) Herrington R, Johnson B, Hunter F Appendix F, Climate change and its impact on water supply and demand in the Prairie Provinces Responding to Global Climate Change in the Prairies, Volume III of the Canada Country Study: Climate Impacts and Adaptation 1997 Environment Canada, Prairie and Northern Region
Vance RE The Geological Survey of Canada's Palliser Triangle Global Change Project: a multidisciplinary geolimnological approach to predicting potential global change impacts on the Northern Great Plains Journal of Paleolimnology 1997 17 3 8 10.1023/A:1007996507331
Lemmen DS Vance RE (Eds) Holocene Climate and Environmental Changes in the Palliser Triangle, Southern Canadian Prairies: Geological Survey of Canada Bulletin 534 1999
Vance RE Last WM Paleolimnology and global change on the southern Canadian Prairies current research 1994-B, Geological Survey of Canada 1994 49 58
Vance RE Beaudoin AB Luckman BH The paleoecological record of 6 ka BP climate in the Canadian Prairie Provinces Geographie physique et Quaternaire 1995 49 81 98
Last WM Gosselin DC Paleolimnology in the Great Plains of North America, Introduction Journal of Paleolimnology 1997 17 1 2 10.1023/A:1007994705514
Lemmen DS Vance RE Wolfe SA Last WM Impacts of future climate change on the southern Canadian prairies: a paleoenvironmental perspective Geoscience Canada 1997 24 121 133
Evans M (Ed) Salt Lake Paleolimnology Special Issue Journal of Paleolimnology 1993 8
Greengrass K Last WM Deleqiat J Sukhan S Waldsea Lake revisited: another look at the recent history of one of western Canada's best-studied meromictic lakes Prairie Perspectives 1999 2 89 115
Last WM Deleqiat J Greengrass K Sukhan S Late Holocene history of Waldsea Lake, Saskatchewan, Canada Sedimentary Geology 2002 148 147 161 10.1016/S0037-0738(01)00215-9
Last WM Rodriguez-Clemente R, Tardy Y Sedimentology, Geochemistry, and Evolution of a Saline Playa in the Northern Great Plains of Canada Geochemistry and Mineral Formation in the Earth Surface 1987 Madrid : Consejo Superior de Investigaciones Cientificas, Centre National de la Recherche Scientifique 319 337
Last WM Campbell ID, Campbell C, Lemmen D, Vance B Evolution of saline lakes in western Canada Climate, Landscape, and Vegetation Change in the Canadian Prairie Provinces 1995 55 64
Boulton Andrew J Brock Margaret A (Eds) Australian Freshwater Ecology Processes and Management 1999 Glen Osmond SA: Gleneagles Publishing
Herbst David B Gradients of salinity stress, environmental stability and water chemistry as a templet for defining habitat types and physiological strategies in inland salt waters Hydrobiologia 2001 466 209 219 10.1023/A:1014508026349
Fritz SC Juggins S Battarbee RW Engstrom DR Reconstruction of past changes in salinity and climate using a diatom based transfer function Nature 1991 352
Wilson Susan E Cumming Brian F Smol John P Assessing the reliability of salinity inference models from diatom assemblages; an examination of a 219 lake data set from western North America Can J Fish Aquat Sci 1996 53 1580 1594 10.1139/cjfas-53-7-1580
Porter Stephen C Sauchyn David J Delorme Denis L The ostracode record from Harris Lake, southwestern Saskatchewan: 9200 years of local environmental change Journal of Paloelimnology 1999 21 35 44 10.1023/A:1008031105186
Macdonald DE Marl resources of Alberta Alberta Research Council Earth Science Report 82-1 1982
Aldrich JM. Ostlie WR Faust TM The status of biodiversity in the Great Plains: Great Plains landscapes of biological significance 1997 The Nature Conservancy, Arlington, VA 135
Last WM Modern sedimentology and hydrology of Lake Manitoba, Canada Environmental Geology 1984 5 177 190
Last WM Teller JT Forester RM Paleohydrology and paleochemistry of Lake Manitoba, Canada: the isotope and ostracode records Journal of Paleolimnology 1994 12 269 283 10.1007/BF00678025
De Lury JS Cole GE 1st annual report on mines and minerals 1928 1 Manitoba Dept. Mines and Natural Resources
Simpson F Dennison EG Subsurface waste-disposal potential of Saskatchewan. Sask Dept Mineral Resources Report 177 1975
Borneuf DM Springs of Alberta. Alberta Geological Survey. Report 82-3 1983
Petch V The 1986 survey of salt flats and pools along the western shore of Lake Winnipegosis, Manitoba Archaeological Quarterly 1987 11 13 22
McKillop WB Patterson RT Delorme LD Nogrady T The origin, physico-chemistry and biotics of sodium chloride dominated saline waters on the western shore of Lake Winnipegosis, Manitoba Canadian Field-Naturalist 1992 106 454 473
Patterson RT McKillop WB Kroker S Nielsen E Reinhardt EG Evidence for rapid avian-mediated foraminiferal colonization of Lake Winnipegosis, Manitoba, during the Holocene Hypsithermal J Paleolimnology 1996 18 131 143 10.1023/A:1007927622654
Bezys RK Bucharme EB Bamburack JD Medikow MAF A geochemical study of saline brine sediments as a guide to prairie-type microdisseminated mineralization and other precious metals in west-central Manitoba (NTS 63C) Manitoba Energy and Mines, Mineral Division, Report of Activities 1997 118 122
Grasby SE Saline spring geochemistry, west-central Manitoba Report of Activities 2000 2000 Manitoba Industry, Trade, and Mines, Manitoba Geological Survey 214 216
Grasby SE Chen Z Subglacial recharge into the Western Canada Sedimentary Basin – Impact of Pleistocene glaciation on basin hydrodynamics GSA Bulletin 2005 117 500 514
Burchill CA Kenkel NC Vegetation-environment relationships of an inland boreal salt pan Can J of Botany 1991 69 722 732
Londry KL Badiou PH Grasby SE Identification of a marine green alga Percursaria percursa, from hypersaline springs in the middle of the North American continent Canadian Field naturalist in press
Nambudiri EMV Shay CT Late Pleistocene and Holocene pollen stratigraphy of the Lake Manitoba basin, Canada Palaeontographica Abt B Bd 1986 202 155 177
Risberg J Sandgren P Teller JT Last WM Siliceous microfossils and mineral magnetic characteristics in a sediment core from Lake Manitoba, Canada: A remnant of glacial Lake Agassiz Canadian Journal of Earth Sciences 1999 26 1299 1314 10.1139/cjes-36-8-1299
Curry BB Paleochemistry of Lake Agassiz and Manitoba based on ostracodes Canadian Journal of Earth Sciences 1997 34 699 708
Pip E A stratigraphic study of mollusks in Lake Manitoba Sediment Walkerana 1990 4 271 277
Garrett Donald E Sodium Sulfate: Handbook of Deposits, Processing, Properties, and Use 2001 Academic Press: New York
Smoot JP Lowenstein TK Melvin JL Depositional environments of non-marine evaporites Evaporites, Petroleum and Mineral Resources 1991 Elsevier, New York 189 348
|
16297237
|
PMC1315329
|
CC BY
|
2022-09-08 23:19:43
|
no
|
Saline Syst. 2005 Nov 18; 1:10
|
utf-8
|
Saline Syst
| 2,005 |
10.1186/1746-1448-1-10
|
oa_comm
|
==== Front
Chiropr OsteopatChiropractic & Osteopathy1746-1340BioMed Central London 1746-1340-13-231630574610.1186/1746-1340-13-23ResearchInfluence of static lumbar flexion on the trunk muscles' response to sudden arm movements Lehman Gregory J [email protected] Stephen [email protected] Robert [email protected] Department of Graduate Studies, Canadian Memorial Chiropractic College, Toronto, ON, Canada2 Undergraduate Department, CMCC, Toronto, ON, Canada2005 23 11 2005 13 23 23 22 9 2005 23 11 2005 Copyright © 2005 Lehman et al; licensee BioMed Central Ltd.2005Lehman 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
Viscoelastic creep of lumbar ligaments (prolonged forward bend) has been shown to negatively influence the spine's muscular reflexive behaviour and spinal stability. No studies to date have investigated the influence of spinall viscoelastic creep on the feedforward response of the trunk muscles to sudden arm raises.
Methods
Surface myoelectric activity was collected from the transversus abdominis/internal oblique, the lower erector spinae and the deltoid muscle during sudden ballistic arm raising before and after 10 minutes of prolonged forward bend in 11 healthy participants free of low back injury. The timing of trunk muscle activity relative to the deltoid muscle was calculated for 5 trials before and 5 trials after the creep procedure.
Results
Viscoelastic creep had no influence on the feedforward response of the trunk muscles during sudden arm raises. A feedforward response of the trunk muscles was not seen in every study participant and during every trial.
Conclusion
Passive trunk muscle fatigue does not appear to influence the timing of the stabilizing role of the investigated trunk muscles to sudden arm flexion.
EMGspine stabilitytrunk musclescreepfeedforward
==== Body
Background
Maintaining adequate spinal stability is considered necessary in the avoidance of low back injury [1]. Spinal stability is theorized by Panjabi [2] to be maintained by the interaction of three systems: the active system, the passive system and the neural control system. The active system is composed of the muscles and related tendons, the passive system consists of the ligaments, discs and structural anatomy, while the neural system coordinates the interaction between the active and passive system via proprioceptive feedback and feed-forward input in response to challenges to spinal stability. The role of the three systems is currently being delineated via human and animal experiments.
When the muscular elements are removed from the spinal column it will buckle under a compressive load as low as 2 kg. Therefore, trunk muscles are necessary for providing spinal stability through out the spine's range of motion and particularly when the spine is in its neutral zone. In the neutral zone the passive ligaments provide little resistance to movement and therefore provide little stability. Muscles function much like guy wires on a ship's mast. The greater the tension in the guy wires and the greater the number of guy wires, the greater the stability and load supported by the mast. Adequate muscular activation is therefore necessary to achieve joint stability. Cholewicki et al [3] demonstrated that at a neutral spinal posture trunk muscles show a co-activation level of 1.7% of their maximum activity (MVC). When the spine is loaded (32 kg) the stability demands increase and muscle co-activation increases to 2.9% of MVC. Patients with low back injury have been shown to have higher trunk muscle co-activation possibly as a response to stabilize their relatively unstable spines compared with a pain free population [4].
The work of Solomonow and colleagues has shown that ligaments appear to function more as proprioceptive feedback than as structural supporting elements [5,6].
Solomonow et al [5] have demonstrated that loading and stimulation of the supraspinal ligaments in a feline model and humans during surgery results in reflexive muscle activation of the multifidus muscle. This reflexive activation suggests that the multifidus is responding to a change in stability (the strain of the ligament) by increasing its activation and in turn attempting to stabilize the spine. This reflexive muscle activation relationship has also been demonstrated between the discs and the multifidus [7] and the SI joint and the gluteus maximus, multfidus and longissimus [8].
External factors may influence the stability of the spine and the interaction between the three systems. Fatigue, vibration and low back pain have been shown to increase the time between sudden loading of the spine (the spine is either abruptly bent forward or backward or perturbed in some manner) and the trunk muscle's response to this perturbation in any attempt to stabilize the spinal system [9,10]. This increased latency response suggests that the spine may become unstable because of external factors.
Viscoelastic creep of the supraspinal ligament due to repetitive flexion has also been shown to compromise spinal stability. In a feline model repetitive static and dynamic flexion (resulting in creep of the ligaments) has resulted in; an attenuation of the protective reflexive response of the multifidus, muscle spasm and a delayed hyperexcitability of trunk muscles [11-13]. The human equivalent of these feline experiments have occurred with participants holding a forward bend posture for more than 10 minutes to induce viscoelastic creep of the lumbar spinal tissues. Solomonow [14] investigated the influence of static and repetitive forward bending on the flexion-relaxation phenomenon (FRP) during forward bending. The flexion relaxation phenomonen is the occurrence of muscle activity silence during a small range of motion around peak forward bending. It is believed that at peak flexion the flexor Moment created by forward bending is resisted by the passive tissues rather than active muscular contraction [15]. Following static lumbar flexion Solomonow [14] found that the creep developed in the viscoelastic structures of the spine caused the erector spinae muscles to remain active longer during anterior flexion and to become active earlier during extension. Granata et al [16] investigated the influence of viscoelastic creep of the lumbar spine on the trunk muscle's reflexive response to sudden loading. Explained simply, this experimental procedure sees a participant suddenly pulled forward. The trunk muscles respond reflexively to the forward displacement of the trunk and become reflexively active to stabilize the spine and volitionally active to return the spine to its neutral position. Experimenters measured the time that the trunk motion occurred, how much force was applied to cause the perturbation and the resulting onset time and amplitude of the EMG signals of the trunk muscles to calculate the timing of reflexive trunk muscle activation as well as the reflexive gain of the trunk muscles. Granata [16] found that following 15 minutes of lumbar viscoelastic creep there was a tendency toward an increase in the gain of the reflexive response but no change in reflex onset latency.
The transverse abdominis has been shown to be an important muscle in providing stability to the lumbar spine and the sacroiliac joint [17,18]. In addition to providing stability to the sacroiliac joint via a force closure mechanism[18] the transverse abdominis appears to act in an anticipatory feedforward manner to self-induced spinal instability/perturbation (i.e. rapid arm raising) [17]. During rapid upper limb movement the transverse abdominis has been shown to become active before the activation or within 50 milliseconds of the primary mover of the upper arm. This feedforward response (defined as occurring within 100 ms before deltoid onset and 50 ms after deltoid onset) is assumed to provide the spinal stability needed to counteract the instability created by the sudden limb movement. In patients with low back injury this feedforward response to sudden limb movement is delayed- the transverse abdominis fails to respond in a feedforward manner [17]. While the influence of viscoelastic creep on muscle function on spine stability properties has been studied extensively, no work has investigated the influence of viscoelastic creep on the superficial transverse abdominis' feedforward response to spine instability induced by sudden arm movement. The aim of this study was to measure the influence of viscoelastic creep on trunk stability as measured by the transverse abdominis' and erector spinae's activation onset timing to sudden rapid arm movement.
Methods
Subject Characteristics and Inclusion Criteria
Eleven healthy males and females with no back, hip or upper limb pain or a history of pain within 3 months and low levels of adipose tissue were recruited from a convenience sample of college age (24 to 30 years old) students. Participants read and signed an information and consent form detailing the experiment approved by the Research Ethics Board of the investigating institution (Canadian Memorial Chiropractic College).
Experimental Procedure
The surface myoelectric activity – and subsequent muscle onsets – of the right transverse abdominis/internal oblique and erector spinae (at L3) was recorded during ballistic raising of the right arm into flexion of 90°. This experimental task occurred immediately before and immediately after a 10 minute static flexion stretch of the lumbar spine-inducing viscoelastic creep.
EMG Collection and Hardware Characteristics
Disposable bipolar Ag-AgCl disc surface electrodes (Bortec EMG, Calgary, AB) with a diameter of one cm were adhered unilaterally over the muscle groups studied with a fixed centre to centre spacing of 1 cm. EMG electrodes were placed parallel with the muscle fibres, on the skin above the right deltoid, right erector spinae (at the level of L3, 3–4 cm lateral to the spinous process, parallel to the muscle fibres) and right transverse abdominis/internal oblique (approximately 2 cm medial and inferior to the Anterior Superior Iliac Spine, at an angle facing the umbilicus). This site for the transverse abdominis/internal oblique has been shown to be a reliable and valid indicator of measuring the feedforward response of the transverse abdominis with surface EMG electrodes [19]. Raw EMG was amplified between 1000 and 20,000 times depending on the subject. The amplifier had a CMRR of 10,000:1 (Bortec EMG, Calgary AB, Canada). Raw EMG was band pass filtered (10 and 1000 Hz) and A/D converted at 2048 Hz using a NI data acquisition system controlled by Delsys EMGWorks software (Delsys EMG, Boston, MA).
Sudden arm movement task
Myoelectric activity from the ipsilateral muscles was collected while the subject ballistically raised their right arm to a position parallel to the floor. Participants faced a wall and the experimenter randomly demanded the initiation of shoulder flexion in an attempt to prevent preactivation of the muscles. Five repetitions of this movement occurred over the course of 20 seconds. If the experimenter noticed volitional preactivation (abdominal bracing) in any of the muscle groups the trials were repeated.
Viscoelastic Creep Stretching Procedure
Subjects warmed up the spine with moderate trunk flexion and extension movements prior to the start of the forward flexion stretch. Subjects were then placed in a seated posture with their knees flexed (between 90–120 degrees) and the hips flexed and externally rotated to allow the soles of the subject's feet to touch one another. The subject then maximally forward flexed their spine and attempted to "hang" on the passive tissues with no attempt to resist the forward bend with muscle activitation. This position was held for 10 minutes.
EMG Processing & Calculation of Muscle Onsets
The latency time from initiation of deltoid activity to trunk muscle activation was found by determining the time between the onset of deltoid firing and the onset of muscle activation for each of the 2 trunk muscles assessed. The threshold for considering when a muscle was considered "on" occurred when its level of activity was greater than 3 standard deviations of the mean activity found during a 400 millisecond baseline period before the five trials of sudden arm movements occurred. The average latency of the 5 trials from each sudden arm movement session (before and after viscoelastic stretching) was calculated.
After the EMG was collected and stored it was processed using EMG analysis software (EMG Works, Delsys, Boston USA) and a Microsoft Excel spreadsheet program. The raw signal first had its bias removed, was full wave rectified, dual pass filtered at 50 Hz (6th order Butterworth digital filter) and then a moving average (50 ms with an overlap 49 ms) was applied to the data. The moving average ensures that the point in time exceeding the calculated threshold exceeds that threshold, on average, for at least 50 ms in an attempt to avoid false muscle onsets. The data was then exported to an Excel file where the mean and standard deviations of the resting signal were calculated and the point in time that the signal exceeded the threshold value was identified mathematically and confirmed visually. This ensured a visual validation of the muscle onset.
Statistical Analysis
A paired t-test at the 5% level of significance was used to assess if there is a difference between the average pre-creep and post-creep latency time for the erector spinae and transverse abdominis/internal oblique. Additionally, trials were classified as satisfying the feedforward criteria (muscle onset within 50 ms after the onset of the deltoid and within 100 ms before the onset of deltoid firing). Descriptive statistics of the percentage of trials satisfying the feedforward criteria were calculated. Last, group averages for the different muscles and conditions were found from the trials satisfying the feedforward criteria and paired t-tests were used to determine if statistically significant differences existed between the pre and post creep conditions for those trials satisfying the feedforward criteria.
Results
The group average for muscle activation delay (after the onset of the deltoid) for transverse abdominus/internal oblique was 59.2 milliseconds (95% Confidence Interval (CI) of 22.5–96.0) for pre-creep and 62.7 (CI = 3 1.5–94.0) for post-creep (positive values indicate that the muscle activation occurred after the deltoid onset). For the erector spinae, delayed muscle activation was 56.9 ms (CI = 1 4.7–99.0) for pre-creep and 69.5 ms (CI = 34.2–104.7) for post-creep. There was no statistical difference in muscle onset latency pre and post the viscoelastic creep protocol. Figure 1 shows a sample (rectified raw EMG) of one trial during the sudden arm raising. The percentage of trials which responded in a feedforward response was also calculated for each muscle and each condition. The transverse abdominis/internal oblique showed a feedforward response during 49% (pre-creep) and 52.7% (post creep) of all trials. The erector spine showed a feedforward response during 60% (pre-creep) and 45.4% (post-creep) of all trials. For individual trial averages, only 4/11 participants demonstrated an average transverse abdominis latency response that could be categorized as feedforward for both the pre and post creep conditions. For the erector spinae muscle 5/11 participants showed average feedforward activation pre creep and 6/11 demonstrated an average feedforward response post-creep.
Figure 1 Myoelectric activity of one trial of during a sudden arm raise. A bias to the EMG amplitude has been added to the TransverseAbdominis/Internal Oblique (TrAb) and the Erector Spinae (ESp) for ease of viewing.
A secondary analysis was done that only included data from trials that satisfied the feedforward criteria. No significant difference was found between the means for both the transverse abdominis and erector spinae pre and post creep protocol. The transverse abdominis group average for those that satisfied the feedforward criteria (n = 8) was 16.7 ms (CI = -1.4–34.9) pre-creep and 26.7 (CI = 13.2–40.2) post creep. The average for the erector spinae pre creep group (n = 11) was 12.0 ms (CI = -5.5–29.5) and post creep was 10.1 ms (CI = -10.5–30.8).
Discussion
The aim of this study was to determine if prolonged trunk flexion influenced the timing of trunk muscle activation during sudden ballistic arm raises. We found that a forward stretching protocol did not result in a change in the timing of muscle activation onsets for the transverse abdominis/internal oblique and the lower erector spinae as measured from the surface EMG. This finding is in agreement with the work of Granata et al [16] who documented no change in the timing of reflexive muscle activation during random trunk perturbations. The viscoelastic creep stretching procedure used in our study is essentially a means of fatiguing the passive components of the soft tissue posterior elements of the spine. Previous to our study no work had documented the effect of passive fatigue on the trunk muscle response to sudden arm raises. However, other researchers have investigated the influence of active muscular fatigue on the trunk muscles response to sudden arm raises. Allison and Henry [20] investigated the influence of active muscular fatigue on the trunk muscles feedforward response to sudden arm movement and found that following erector spinae fatigue the external oblique muscle onset occurred earlier during trials that exhibited feedforward behaviour. When all trunk muscles (rectus abdominis, external oblique, internal oblique, longissimus and transverse abdominis) were grouped together by side the average onset occurred earlier after active fatigue, while individually there were only trends toward a decrease in the muscle activation onset. In the pilot phase of our study we were unable to consistently find muscle activation in the rectus abdominis and external oblique in response to a sudden arm movement, thus those muscles were excluded from the experiment and the subsequent data collection sessions. This was also seen in the study by Allison and Henry [20] who found inconsistencies in the whether a muscle fired in response to sudden arm raising.
We are uncertain why passive fatigue of spinal tissue did not result in changes in the timing of reflexive or anticipatory trunk muscle onsets yet active fatigue of the trunk muscles appears to result in an earlier onset when muscle firing occurs. One possible explanation for the lack of influence of viscoelastic creep on the feedforward trunk muscle latency may be posture dependent. It is assumed that constant forward bend (the viscoelastic creep stretching protocol) would influence the stiffness of the posterior elements of the spine. In an upright neutral posture ligaments provide very little stiffness in the neutral zone. Therefore a disturbance in ligamentous stiffness in the neutral zone may be unnoticed and have little effect on the neuromuscular response to sudden arm raising because of their limited role in providing stiffness in this neutral upright posture. Testing the anticipatory and reflexive muscle activation to perturbations in non neutral postures may better elucidate the influence of viscoelastic creep on the reflexive and anticipatory behaviour of trunk muscles to perturbations in stability.
A limitation of the study was the large degree of variability across participants and within a single participant across trials. This large variability ensures difficulty in finding statistical significance. While previous research [19] using similar procedures have reported high repeatability across days, within subject variability may be normal as seen in the high between trial variability of this study and the large degree of variability reported in previous studies [20]. This study is also limited by only measuring the surface myoelectric activity. In-dwelling electrode readings of the transverse abdominis and from different sections of the muscle may show different findings.
Conclusion
Spinal viscoelastic creep did not influence the anticipatory trunk muscle onset timing following ballistic arm movement. A great degree of variability was seen in the onset timing of the trunk muscles and close to half of all trials failed to satisfy the feedforward criteria.
Authors' contributions
GJL: Conception, design, data collection, data analysis, manuscript preparation.
SS & RM: data collection, data analysis, manuscript preparation.
==== Refs
McGill SM Low back stability: from formal description to issues for performance and rehabilitation Exerc Sport Sci Rev 2001 29 26 31 11210443 10.1097/00003677-200101000-00006
Panjabi MM The stabilizing system of the spine. Part I. Function, dysfunction, adaptation, and enhancement J Spinal Disord 1992 5 383 9 discussion 397 1490034
Cholewicki J Panjabi MM Khachatryan A Stabilizing function of trunk flexor-extensor muscles around a neutral spine posture Spine 22 2207 12 1997, Oct 1 9346140 10.1097/00007632-199710010-00003
Marras WS Davis KG Ferguson SA Lucas BR Gupta P Spine loading characteristics of patients with low back pain compared with asymptomatic individuals Spine 26 2566 74 2001, Dec 1 11725237 10.1097/00007632-200112010-00009
Solomonow M Zhou BH Baratta RV Lu Y Harris M Biomechanics of increased exposure to lumbar injury caused by cyclic loading: Part 1. Loss of reflexive muscular stabilization Spine 24 2426 34 1999, Dec 1 10626304 10.1097/00007632-199912010-00003
Solomonow M Zhou BH Harris M Lu Y Baratta RV The ligamento-muscular stabilizing system of the spine Spine 23 2552 62 1998, Dec 1 9854754 10.1097/00007632-199812010-00010
Indahl A Kaigle AM Reikeras O Holm SH Interaction between the porcine lumbar intervertebral disc, zygapophysial joints, and paraspinal muscles Spine 22 2834 40 1997, Dec 15 9431619 10.1097/00007632-199712150-00006
Indahl A Kaigle A Reikeras O Holm S Sacroiliac joint involvement in activation of the porcine spinal and gluteal musculature Spinal Disord 1999 12 325 30
Radebold A Cholewicki J Panjabi MM Patel TC Muscle response pattern to sudden trunk loading in healthy individuals and in patients with chronic low back pain Spine 25 947 54 2000, Apr 15 10767807 10.1097/00007632-200004150-00009
Radebold A Cholewicki J Polzhofer GK Greene HS Impaired postural control of the lumbar spine is associated with delayed muscle response times in patients with chronic idiopathic low back pain Spine 26 724 30 2001, Apr 1 11295888 10.1097/00007632-200104010-00004
Jackson M Solomonow M Zhou B Baratta RV Harris M Multifidus EMG and tension-relaxation recovery after prolonged static lumbar flexion Spine 26 715 23 2001, Apr 1 11295887 10.1097/00007632-200104010-00003
Solomonow M Baratta RV Zhou BH Burger E Zieske A Gedalia A Miscular dysfunction elicited by creep of lumbar viscoelastic tissue J Electromyogr Kinesiol 2003 13 381 96 12832168 10.1016/S1050-6411(03)00045-2
Solomonow M Zhou BH Baratta RV Burger E Biomechanics and electromyography of a cumulative lumbar disorder: response to static flexion Clin Biomech 2003 18 890 8 10.1016/S0268-0033(03)00173-6
Solomonow M Baratta RV Banks A Freudenberger C Zhou BH Flexion-relaxation response to static lumbar flexion in males and females Clin Biomech 2003 18 273 9 10.1016/S0268-0033(03)00024-X
McGill SM Kippers V Transfer of loads between lumbar tissues during the flexion-relaxation phenomenon Spine 19 2190 6 1994, Oct 1 7809753
Granata KP Rogers E Moorhouse K Effects of static flexion-relaxation on paraspinal reflex behavior Clin Biomech 2005 20 16 24 10.1016/j.clinbiomech.2004.09.001
Hodges PW Richardson CA Altered trunk muscle recruitment in people with low back pain with upper limb movement at different speeds Arch Phys Med Rehabil 1999 80 1005 12 10489000 10.1016/S0003-9993(99)90052-7
Richardson CA Snijders CJ Hides JA Damen L Pas MS Storm J The relation between the transversus abdominis muscles, sacroiliac joint mechanics, and low back pain Spine 27 399 405 2002, Feb 15 11840107 10.1097/00007632-200202150-00015
Marshall PW Murphy B The validity and reliability of surface EMG to assess the neuromuscular response of the abdominal muscles to rapid limb movement Journal of Electromyography and Kinesiology 2003 13 477 489 12932422 10.1016/S1050-6411(03)00027-0
Allison GT Henry SM The influence of fatigue on trunk muscle responses to sudden arm movements, a pilot study Clinical Biomechanics 2002 17 414 417 12084547 10.1016/S0268-0033(02)00029-3
|
16305746
|
PMC1315330
|
CC BY
|
2021-01-04 16:38:22
|
no
|
Chiropr Osteopat. 2005 Nov 23; 13:23
|
utf-8
|
Chiropr Osteopat
| 2,005 |
10.1186/1746-1340-13-23
|
oa_comm
|
==== Front
BMC MicrobiolBMC Microbiology1471-2180BioMed Central London 1471-2180-5-661628751210.1186/1471-2180-5-66Research ArticleEvaluation and selection of tandem repeat loci for Streptococcus pneumoniae MLVA strain typing Koeck Jean-Louis [email protected] Berthe-Marie [email protected] Sonia [email protected] Emmanuelle [email protected] Lassana [email protected] Samina [email protected] Gilles [email protected] Christine [email protected] Laboratoire de biologie clinique, HIA Robert Picqué, 351, route de Toulouse, 33 140 Villenave d'Ornon, France2 Association pour l'Aide à la Médecine Préventive, 25-28 rue du Dr Roux, 75 015 Paris, France3 Centre National de Référence des Pneumocoques, Laboratoire de Microbiologie, Hôpital Européen Georges Pompidou, 20-40 rue Leblanc, 75908 Paris cedex 15, France4 Laboratoire de Virologie-Bactériologie; Centre Hospitalier Universitaire Yalgado Ouédraogo Ouagadougou, Burkina Faso5 Génome, Polymorphisme et Minisatellites (GPMS), Institut de Génétique et Microbiologie, Bat. 400, Université Paris XI, 91405 Orsay cedex, France6 Division of Analytical Microbiology, Centre d'Etude du Bouchet, BP3, 91710 Vert le Petit, France2005 16 11 2005 5 66 66 20 7 2005 16 11 2005 Copyright © 2005 Koeck 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
Precise identification of bacterial pathogens at the strain level is essential for epidemiological purposes. In Streptococcus pneumoniae, the existence of 90 different serotypes makes the typing particularly difficult and requires the use of highly informative tools. Available methods are relatively expensive and cannot be used for large-scale or routine typing of any new isolate. We explore here the potential of MLVA (Multiple Loci VNTR Analysis; VNTR, Variable Number of Tandem Repeats), a method of growing importance in the field of molecular epidemiology, for genotyping of Streptococcus pneumoniae.
Results
Available genome sequences were searched for polymorphic tandem repeats. The loci identified were typed across a collection of 56 diverse isolates and including a group of serotype 1 isolates from Africa. Eventually a set of 16 VNTRs was proposed for MLVA-typing of S. pneumoniae. These robust markers were sufficient to discriminate 49 genotypes and to aggregate strains on the basis of the serotype and geographical origin, although some exceptions were found. Such exceptions may reflect serotype switching or horizontal transfer of genetic material.
Conclusion
We describe a simple PCR-based MLVA genotyping scheme for S. pneumoniae which may prove to be a powerful complement to existing tools for epidemiological studies. Using this technique we uncovered a clonal population of strains, responsible for infections in Burkina Faso. We believe that the proposed MLVA typing scheme can become a standard for epidemiological studies of S. pneumoniae.
==== Body
Background
S. pneumoniae infections remain the major cause of pneumonia, meningitis and otitis in many countries, and a growing number of isolates appear to be resistant to penicillin. Purulent meningitis due to S. pneumoniae was recognized more than twenty years ago to be a serious problem in African countries [1] and is known to occur in a seasonal pattern in sub-Saharan Africa [2]. Serotype 1 is one of the most common pneumococcal serotypes associated with disease, although its prevalence varies among countries [3].
Apart from serotyping based on the variations of the coat exopolysaccharides, different DNA-based methods utilize genetic polymorphism. Macrorestriction and Pulsed-Field Gel Electrophoresis analysis (PFGE) [4], and Multiple Loci Sequence Typing (MLST) [5] are the most frequently used genotyping techniques. An MLST typing system was described by Enright et al. [5] together with an online identification page. The rep-PCR or BOX PCR assay was described in 1996 by van Belkum [6]. The different techniques have been compared in several studies [7,8]. Other methods use the sequencing of PCR product such as the gal U gene [9], or the PCR restriction profile of the cpsA-cpsB genes [10].
Although some of these techniques have proven their capacity to discriminate efficiently among the multiple serotypes, the data are not always reproducible between different laboratories, some may not be amenable to the making of international databases, or they are time consuming and expensive. Polymorphic tandem repeat sequences also called Variable Number of Tandem Repeats (VNTR) are an interesting class of genetic markers. Multiple alleles may be present at a single locus, and size differences are easily resolved by electrophoresis of PCR products. Tandem repeat typing has proved to be highly appropriate for the typing of pathogenic bacterial species [11,12], including species with a very high genetic homogeneity such as the Mycobacterium tuberculosis complex, Bacillus anthracis, and Yersinia pestis [13-15].
The availability of genome sequence data from different S. pneumoniae strains greatly facilitates the search for polymorphic DNA sequences [16]. In this report, we have evaluated the polymorphism of selected tandem repeats, and measured their discrimination power, across a diverse collection of strains.
Results
Selection of VNTRs for MLVA typing
At the onset of this study the genome sequences of two strains, R6 and TIGR4 were available. Comparison of these genome sequences using the approach described by Denoeud et al. [16,17], identified 33 tandem repeats with a repeat unit equal or larger than 12 bp and predicted to display size polymorphism. Preliminary sequence for two additional genomes (Sanger Spanish 23F-1 and TIGR 670-6B) was subsequently made available and was used to select primers for PCR amplification that would match with all four strains. To confirm that the selected markers were indeed polymorphic, a first set of eight isolates including the reference strain R6 and RP28 to RP34 (Table 1 and Table 2) were analyzed (Fig. 1 and data not shown). Eighteen VNTRs were retained to investigate a larger collection, 12 of which with a 45 bp repeat belong to the BOX family of repeated elements [18] (Table 3). Spneu19, a 60 bp repeats, encodes the choline-binding domain of pneumococcal protein A encoded by PcpA [19]. Spneu36, a 45 bp boxB repeat is fused in strain R6 to gene trzA encoding the N-ethylammeline chlorohydrolase a Atz/Trz family protein. The sequence diversity within the repeat units of the 18 VNTRs was calculated using the Tandem Repeat Finder software [20], and is indicated as percent matches (Table 3).
Typing of the reference strain collection
A larger collection of 53 isolates comprising isolates from different origins and with a variety of serotypes was then genotyped (Table 1 and Table 2). As a control, strain R6 was systematically analysed with each set of 5 isolates [15]. The primers listed in Table 3 were used essentially as previously described [14]. The VNTRs were amplified very efficiently in most of the isolates. For a few isolates, no amplification was obtained with Spneu19 and Spneu36. The size variations of the amplicons were as expected for an exact multiple of repeats except in a few cases. With marker Spneu38, an amplicon of intermediate size was observed, marked "1.5", with strain R6 (Figure 1), and for Spneu27 RP36 had a "0.5" intermediate size allele. With marker Spneu25, amplification of RP44 and RP32 (serotype 23F) produced a 1.5 kb amplicon (allele coded "19"). When examining this locus in the different sequenced genomes we found that Sanger strain 23F had an Insertion Sequence (IS) inserted in front of the Spneu25 tandem repeat between the PCR primers. Similarly, for Spneu33, PCR amplification of two isolates, RP43 (serotype 19F) and RP18 (serotype 23F), produced a 2.6 kb amplicon (allele arbitrarily coded "20"), suggesting the presence of an IS element in the repeat. For Spneu 38 and Spneu 42, alleles coded "0.1" correspond to the absence of a VNTR unit although a PCR product is observed.
Spneu26 has a peculiar configuration in strain R6. In this strain 2 repeated elements are observed, a 49 bp repeat with low internal homogeneity inserted inside the usual 51 bp repeat.
The putative MLVA profiles of 4 fully or partially sequenced genomes were determined and used in the clustering analysis shown in Figure 2. The data for Spneu19 and Spneu36 were not used for this clustering because of the existence of null alleles in some isolates as mentioned. Combining the 52 isolates of this study (not including RP45 alias R6, represented by the sequenced genome) plus the 4 sequenced strains, 49 genotypes are observed. The global Hunter Gaston diversity index (HGDI) for the described MLVA assay is 0.995. The HGDI for each VNTR marker is shown in Table 3.
The remarkable similarity between independent isolates with the same serotype strongly suggests that these markers, with the exception of Spneu 39, do not vary at a high frequency. This is confirmed by the fact that the size of the 18 VNTR alleles was the same in two independent R6 isolates, and corresponded exactly to those of the sequenced genome.
To assess the reproducibility of the assay, a series of 10 isolates, RP46 to RP55, were genotyped in duplicate, in Orsay and Bordeaux, giving the same fingerprint. The allele size assignment was performed by eye in one site and using the BioNumerics tools in the other site.
Analysis of two small epidemic groups
Part of the isolates in this study correspond to an epidemic situation. They were isolated in the same geographic area in Burkina Faso during outbreaks of meningitis in year 2002–2003 and 2004. In 2002–2003, 8 out of 9 isolates were of serotype 1 and one was of serotype 25F. The eight serotype 1 isolates from 2002–2003 have identical alleles at all markers except Spneu39. Interestingly, for this marker 6 different alleles are observed, all corresponding to a large number of repeats (8 to 16). In 2004, isolates of 4 different serotypes were obtained, three serotype 6, three serotype 1, three serotype 5 and two serotype 12 isolates. The serotype of RP04 and RP05 could not be determined. All these isolates cluster according to their serotype. The three serotype 1 isolates are identical except for marker Spneu37. Although clearly distinct, they are grouped with the serotype 1 strains isolated in 2002–2003 (shown in a box on Fig. 2).
Discussion
A collection of 18 VNTR markers which can be used to genotype S. pneumoniae strains by simple PCR and agarose gel electrophoresis has been identified. Two of these markers, Spneu19 and Spneu36, belonging respectively to pcpA and trzA, were not kept in the clustering analysis because they were absent from some isolates, although they might be useful in specific situations. Spneu19 is not amplified in isolates of serotype 3 and 6A suggesting that they lack pcpA. It was proposed that the protein encoded by pcpA could be a surface protein involved in cell adhesion with specific proteins of the human extracellular matrix. PCPA is not essential for bacterial growth at least under laboratory conditions as the gene can be knocked-out with no noticeable change in the pneumococcal phenotype [19]. However the polymorphism observed at the choline binding domain might play a role during infection. The effect of Spneu36 polymorphism on trzA is not clear as the 45 bp repeat is apparently not fused to this gene in the TIGR4 genome.
Among the 18 selected markers, 16 consist in 45 to 60 bp repeats with a regular variation and which amplify very efficiently. The size polymorphism can be scored by eye. Two additional markers of 12 bp and 14 bp repeats are also described. In the present study, VNTRs with smaller repeat units, of the microsatellite category (1 to 8 bp long) were not investigated. Due to their sometimes relatively high mutation rate, they may improve the MLVA resolution to investigate local outbreaks.
Most of the VNTRs correspond to the intergenic 45 bp boxB repeat. They belong to a family of elements, present in multiple loci in the S. pneumoniae genome, and composed of three subunits boxA, boxB and boxC. Subunit boxB, 45 bp long, can be tandemly repeated with a high internal sequence similarity [18]. These sequences have been suggested to be regulatory elements shared by coordinately regulated genes. The subunit boxB is the only one that can be tandemly repeated.
The polymorphism of these elements has been used to genotype strains in the BOX-PCR assay. However this assay produces an image (a multi-band pattern) of the added polymorphism of multiple BOX elements whereas the VNTR assay analyses each locus separately. As a result, the MLVA assay is more informative and reproducible, data interpretation is much easier, and genotyping databases can be easily produced [21].
We performed an MLVA analysis on a collection of isolates originating from Africa and France. Clustering was observed both on the basis of serotype and geographical origin although there are some exceptions. Analysis of the sequenced genomes have shown the existence of many gene transfer events which could explain why strains with the same serotype do not always cluster. Conversely, MLVA aggregates a number of strains of serotypes which are known to be close variants, such as serotype 19F and 23F [22], and serotype 14 and 9V [23].
All the African serotype 1 isolates are grouped (Figure 2). Interestingly a cluster of 3 isolates RP07, RP11 and RP02 isolated in 2004 and differing by only one marker, are linked to a second cluster of 9 isolates recovered in 2002–2003 from which they differ at 9 out of 16 markers. This is a large distance suggesting that the existing population of strains able to cause outbreaks is very diverse. A second lineage contains serotype 12 and serotype 5 strains differing at 7 out of 16 markers.
Serotype 1 strains seem to have a propensity to cause meningitis in Burkina Faso, as two related clones of this serotype were found in 50% of the analyzed cases. A similar observation was made in Northern Ghana [24]. In contrast to the majority of other serotypes, nasopharyngeal carriage of serotype 1 is exceptionally observed. This could be linked to 1) a lesser genetic diversity as compared to other serotypes, due to limited exchanges with other streptococci during colonization, 2) a high invasive potential or 3) high attack rates, as supported by the observations of Leimkugel et al. [24]. For serotypes other than serotype 1, distribution across the dendrogram may be associated with the nasopharynx carriage (e.g. 23F).
Conclusion
This preliminary investigation validates a first set of markers for MLVA investigation. The lethality linked to pneumococcal meningitis is high and an appropriate vaccination is necessary, requiring the identification of virulent lineages. The usefulness of the MLVA typing scheme proposed here can now be further determined by investigating a larger population of isolates from Africa which are currently being collected.
However in a species with 90 different genotypes, additional studies will clearly be needed. In particular, it will be useful to see how MLVA compares with MLST [25]. In contrast with MLST, the relatively low cost and moderate expertise required for MLVA typing would allow the systematic typing of any new isolate directly by clinical laboratories within hospitals. All markers proposed here are easy to type with no sophisticated equipment and software, so that it should in principle be feasible to organize networks of clinical laboratories, each one taking in charge the typing of local isolates. To facilitate such projects, shared internet resources enabling the import and analysis of results could be also set-up [21]. Eventually comparison of isolates on such a large scale will provide a precise measure of the stability of each marker, necessary for the optimized interpretation of MLVA typing data.
Methods
Bacterial strains
A total of 53 isolates were analyzed (Table 1 and Table 2). RP14 to RP24 and RP25 to RP35 were generous gifts from respectively Hubert Chardon (CHPA, Aix en Provence, France) and Christine Grandpré (Hôpital d'Instruction des Armées, HIA Percy, France). These isolates were obtained from blood or respiratory specimens (i.e. sputums or bronchoalveolar fluids), from patients with severe pneumonia or septicemia or both. RP36 to RP44 are reference strains for serotypes 1, 6B, 9V, 12F, 14, 18C, 19A, 19F and 23F from the Satens Serum Institute (SSI Denmark) and provided to us by the "Centre National de Référence des Pneumocoques" (CNRP France). RP1 to RP13 and RP46 to RP55 were isolated from cerebro-spinal fluid of patients with meningitis in Bobo-Dioulasso (Burkina Faso) by the "Association pour l'Aide à la Médecine Préventive" (AMP IPP) [2]. Strain R6 (ATCC BAA 255) was used as control under the name RP45. The strain was obtained from two sources, one from the CNRP originating from the SSI in Denmark, and the second from the Institut Pasteur collection (CIP 105880). DNA was purified using the InstaGene kit (Biorad, Marnes la Coquette, France). The minimum inhibitory concentrations of antibiotics (MICs) were determined by the E-test (AB Biodisk). Interpretive criteria for susceptibility or resistance were as recommended by the CASFM (Comité de l'Antibiogramme de la Société Française de Microbiologie) [26] (Last release, January 2005).
Identification of variable number tandem repeats by genomic sequence comparison
The methods previously described [13,16,17,27] were used to identify tandem repeats with a predicted size which differs between the published genome of S. pneumoniae strain R6 [28], strain TIGR4 [29] and the preliminary genome 23F (Sanger Institute) and 670-6B (TIGR) sequence data (obtained respectively from [30] and from [31]).
The different tandem repeat loci are designated by using the nomenclature described previously [14]. For instance Spneu1579_45bp_507bp_7u (Spneu15) is a tandem repeat locus at position 1579 Kb in the R6 genome. It has a 45 bp motif, a total PCR product length of 506 bp in the R6 strain when using the primer set indicated in Table 3. This allele size corresponds to 7 units. Its common laboratory name is Spneu15 (Table 3).
PCR (Polymerase Chain Reaction) amplification and genotyping
PCR amplifications were performed in a total volume of 15 microliters containing 10 ng of DNA, 1× PCR Reaction Buffer, 1U of Taq DNA polymerase (Qbiogen, Illkirch, France), 200 microM of each deoxynucleotide triphosphate, and 0.3 microM of each flanking primer. The primers are listed in Table 3. Amplifications were performed in a MJ Research PTC200 thermocycler. Initial denaturation step at 94°C for 5 min. was followed by 30 cycles of denaturation at 94°C for 30 s, primer annealing at 60°C for 30 s, and elongation at 72°C for 45 s. The final extension step was at 72°C for 7 min. Three microliters of amplification product were loaded on a 2% standard agarose gel and run until the bromophenol blue had reached the 20 cm position. Gels were stained with ethidium bromide, visualized under UV light, and photographed (Vilber-Lourmat, Marnes-la-Vallée, France). The size markers used were a 100-bp ladder (EZ Load 100 pb PCR Molecular Ruler, Biorad, Marnes la Coquette, France) or 20-bp ladder (EZ Load 20 pb Molecular Ruler, Biorad, Marnes la Coquette, France) according to the tandem repeat unit length. Gel images were managed using the BioNumerics software package (version 4.0, Applied-Maths, Sint-Martens-Latem, Belgium).
Data analysis
Band size estimates were converted to number of units within a character dataset. The VNTR data deduced from the sequenced strains R6, TIGR4, 670-6B and Sanger Spanish 23F-1 were added to the MLVA analysis. Clustering analyses used the categorical coefficient and UPGMA (Unweighted Pair Group Method using Arithmetic averages). The use of the categorical parameter implies that the character states are considered unordered. The same weight is given to a large or a small number of differences in the number of repeats at each locus. The polymorphism indexes of individual or grouped VNTRs was calculated using the Hunter-Gaston discriminatory index (HGDI) [32].
List of abbreviations
VNTR; Variable Number of Tandem Repeats
MLVA; Multiple locus VNTR
HGDI; Hunter Gaston discriminatory index
Authors' contributions
SC and SV did most of the typing work and CP and GV did the error checking analysis. JLK was in charge of the definition of the study collection and preparation of the DNA samples. BMN, EV and KS collected and provided bacterial isolates. CP initiated and managed the project. GV was in charge of the BioNumerics database and clustering analyses. CP, JLK, and GV wrote the report. All authors read and approved the final manuscript.
Acknowledgements
We thank Hubert Chardon (CHPA, Aix en Provence, France) and Christine Grandpré (Hôpital d'Instruction des Armées, HIA Percy, France) for providing isolates. Work on the typing and molecular epidemiology of dangerous pathogens is supported by the French ministry of defence. We thank Sanofi Pasteur and the French military health services (Service de Santé des Armées) for their support. Preliminary sequence data for S. pneumoniae strain 670 and strain Spanish 23F-1 was obtained respectively from The Institute for Genomic Research [31] and the Sanger Institute [30]. Sequencing of S. pneumoniae strain 670 was accomplished with support from NIAID (National Institute of Allergy and Infectious Diseases).
Figures and Tables
Figure 1 PCR amplification of 6 VNTRs using DNA of 8 isolates. From left to right R6, RP31, RP28, RP32, RP29, RP33, RP30 and RP34. A 100-bp ladder (100 to 1000 bp) is run on both side of each group of samples. The number of repeat units within each allele is indicated. The rare alleles of intermediate size are coded "#.5" (for instance Strain R6 is coded 1.5 for locus Spneu38).
Figure 2 Dendrogram showing the clustering of 56 isolates by MLVA using 16 markers. The data are expressed as number of repeats for each of the 16 VNTRs. The genotype number is indicated in the first column "Geno". Some strains possess a rare serogroup (RSG) or were not typable (NT). The serotype 1 isolates from Burkina Faso are boxed.
Table 1 S. pneumoniae strains origin and characteristics
Strain id Reference Origin Serotype Year Sitea Country Peni Gb Ery
RP01 100277 AMP (IPP) 12A 2004 CSF Burkina Faso S S
RP02 200801 AMP (IPP) 1 2004 CSF Burkina Faso S S
RP03 300141 AMP (IPP) 12B 2004 CSF Burkina Faso S S
RP04 400202 AMP (IPP) RSGa 2004 CSF Burkina Faso S S
RP05 500205 AMP (IPP) RSG 2004 CSF Burkina Faso S S
RP06 600003 AMP (IPP) 5 2004 CSF Burkina Faso S S
RP07 700025 AMP (IPP) 1 2004 CSF Burkina Faso S S
RP08 800830 AMP (IPP) 6A 2004 CSF Burkina Faso S S
RP09 900518 AMP (IPP) 6A 2004 CSF Burkina Faso S S
RP10 1000244 AMP (IPP) 6A 2004 CSF Burkina Faso S S
RP11 1100236 AMP (IPP) 1 2004 CSF Burkina Faso S S
RP12 1200327 AMP (IPP) 5 2004 CSF Burkina Faso S S
RP13 1300373 AMP (IPP) 5 2004 CSF Burkina Faso S S
RP14 20020876 CHPA 14 2002 Blood France (sud-est) R R
RP15 20020953 CHPA 14 2002 Blood France (sud-est) I S
RP16 20020978 CHPA 18C 2002 Blood France (sud-est) S S
RP17 20021029 CHPA 14 2002 Blood France (sud-est) R R
RP18 20021131 CHPA 23 2002 Blood France (sud-est) I R
RP19 20022003 CHPA 3 2002 Blood France (south-east) S S
RP20 20022042 CHPA 3 2002 Blood France (south-east) S S
RP21 20022215 CHPA 19A 2002 Blood France (south-east) I R
RP22 20022337 CHPA 10A 2002 Blood France (south-east) S S
RP23 20022338 CHPA 23A 2002 Blood France (south-east) I S
RP24 20022339 CHPA 23A 2002 Blood France (south-east) S S
RP25 4030143494 HIA Percy 3 2004 Pulmonary France (Paris area) S S
RP26 488015780 HIA Percy RSGc 2004 Pulmonary France (Paris area) S S
RP27 488019294 HIA Percy NTd 2004 Pulmonary France (Paris area) I R
RP28 4040145233 HIA Percy 9V 2004 Pulmonary France (Paris area) I R
RP29 4050149284 HIA Percy 9V 2004 Pulmonary France (Paris area) R R
RP30 4060150356 HIA Percy RSG 2004 Pulmonary France (Paris area) S S
RP31 4060150868 HIA Percy 9V 2004 Pulmonary France (Paris area) I R
RP32 4060151258 HIA Percy 23 2004 Pulmonary France (Paris area) I R
RP33 488034644 HIA Percy 6A 2004 Pulmonary France (Paris area) S S
RP34 488037640 HIA Percy 19A 2004 Pulmonary France (Paris area) I S
RP35 4070153283 HIA Percy 11A 2004 Blood France (Paris area) S S
RP46 PED012 AMP (IPP) 1 2002 CSF Burkina Faso I S
RP47 MED042 AMP (IPP) 1 2002 CSF Burkina Faso I S
RP48 PED081 AMP (IPP) 1 2003 CSF Burkina Faso S UNe
RP49 MED142 AMP (IPP) 1 2003 CSF Burkina Faso S S
RP50 PED159 AMP (IPP) 1 2003 CSF Burkina Faso S S
RP51 Q06003 AMP (IPP) 25 2003 CSF Burkina Faso S S
RP52 MED126 AMP (IPP) 1 2003 CSF Burkina Faso S S
RP53 Q14004 AMP (IPP) 1 2003 CSF Burkina Faso S S
RP54 PED124 AMP (IPP) 1 2003 CSF Burkina Faso S S
RP55 H01368 AMP (IPP) 1 2003 CSF Burkina Faso S S
(a) Origin of isolate; CSF: Cerebrospinal fluid
(b) Resistance to benzylpenicillin (Pen) or erythromycin (Ery) (S: susceptible; I: intermediate; R: resistant).
(c) RSG rare serogroup
(d) NT non typable
(e) UN unknown
Table 2 S. pneumoniae reference strains
strain id Serotype SSI reference CNRP Reference
RP36 1 134/95 1172
RP37 6B 105/95 1168
RP38 9V 162/95 1158
RP39 12 F 133/95 1171
RP40 14 109/95 1167
RP41 18C Holland 1821 1170
RP42 19A Verbiest 91 1149
RP43 19 F Verbiest 38 1153
RP44 23 F Verbiest 51 1156
RP45 NT R6 105880
Table 3 Characteristics of VNTR markers and sequence of the PCR primers
Name R6 positiona Locus Motif bp Percent matches R6 amplicon PCR primers No of alleles Allele size range (units) HGDId (56 strains)
size bp repeats
Spneu15 1579 BOX 45 94 507 7 TCCAACACGACCTTTATCCA 10 1–12 0.827
AGCCTTGCTCCTCATCTTGA
Spneu17 1783 BOX 45 86 167 3 TCGAAAATCTCTGCAAACCA 12 1–13 0.883
CGGACTAGGCGGCTGATTA
Spneu19 1925 pcpA 60 98 663 10 TCGGGTGTAGTCGTGTTTACT 5 7–10 0.749
AACTGATGTAGCTAAACCTAAAAAGAA
Spneu25 101 BOX 45 96 423 4 TCGCCTTTGCTAATCAAACA 7 1–6c 0.744
GACTGGGTAACAATTCCATT
Spneu26 185 intergenic 51 65 492 6 ATGGAACAGAAGGCGAATTG 7 1–9 0.688
AACAAGGCCAGGATTTTCGT
Spneu27 257 BOX 45 95 347 3 TCAGGAACAGCTATTATCCC 5 0.5–4 0.575
CCAACCTCCTTTTCGTTTCA
Spneu31 557 BOX 45 98 594 9 CTGGAATAGTCCATCGAGCTT 9 1–9 0.763
CGAAGGTAAATGTGAACAAAC
Spneu32 571 BOX 45 86 280 2 AAAGCAAAATAATGCGCTCCT 4 1–4 0.629
ACTACCAAAGCCCAAGTGGT
Spneu33 698 BOX 45 95 407 2 CAGCTGAACATGATGGCAAA 10 1–10c 0.858
CATCACTTCTCCCTTCTAATC
Spneu34 1082 BOX 45 97 239 1 CTCGGTAAAGACGAGGTTCAA 6 1–11 0.598
CATCGTGCAGGGATTTTCTAA
Spneu35 1166 intergenic 49 100 349 4 ACAATCTCAGCTACGCCCTA 5 1–5 0.561
ACGGGATGACATTAAAAACC
Spneu36 1209 trzA 45 86 320 2 GAAATCTTGATCACAAGTCAC 10 1–10 0.866
AAAAAGTTGCCTTCGAGTGAC
Spneu37 1350 BOX 45 91 501 7 ATGCGCAAATCGATTAAGGA 11 2–12 0.876
CGATGTGCTTCACTCACTCC
Spneu38 1911 BOX 45 84 297 1.5b TCAGGAGTAGGTTCCTGACCA 6 0.1–4 0.616
AACCCGAGAAGAAGTGCAAG
Spneu39 1915 BOX 45 83 275 2 CCTTGGACTACCACCTCGTT 14 2–17 0.862
GCCGTGACAGACTTCTGGAA
Spneu40 1611 BOX 45 88 376 3 AGTAGTCTGCAATCGCAGCT 7 2–8 0.832
GGGCCTTAAGTCGCTTTGAT
Spneu41 394 intergenic 14 100 166 2 ACCGTAATGGGACTTCATCT 5 1–4 0.484
ATCTGCACCTAAGACAATCG
Spneu42 1036 intergenic 12 91 120 3 TGCTCCCTCTGAAAAGTCAT 6 0.1–4 0.739
CTTTATATTATATCATGCCTTC
a Position on the chromosome expressed in kbp
b Expected size for a truly 2U amplicon is 312 bp long
c Additional large alleles probably contain an IS element
d Hunter Gaston Diversity index
==== Refs
Mar ID Denis F Cadoz M [Epidemiologic features of pneumococcal meningitis in Africa. Clinical and serotypical aspects (author's transl)] Pathol Biol (Paris) 1979 27 543 548 44737
Parent du Chatelet I Traore Y Gessner BD Antignac A Naccro B Njanpop-Lafourcade BM Ouedraogo MS Tiendrebeogo SR Varon E Taha MK Bacterial meningitis in Burkina Faso: surveillance using field-based polymerase chain reaction testing Clin Infect Dis 2005 40 17 25 15614687 10.1086/426436
Brueggemann AB Spratt BG Geographic distribution and clonal diversity of Streptococcus pneumoniae serotype 1 isolates J Clin Microbiol 2003 41 4966 4970 14605125 10.1128/JCM.41.11.4966-4970.2003
Hall LM Whiley RA Duke B George RC Efstratiou A Genetic relatedness within and between serotypes of Streptococcus pneumoniae from the United Kingdom: analysis of multilocus enzyme electrophoresis, pulsed-field gel electrophoresis, and antimicrobial resistance patterns J Clin Microbiol 1996 34 853 859 8815096
Enright MC Spratt BG A multilocus sequence typing scheme for Streptococcus pneumoniae: identification of clones associated with serious invasive disease Microbiology 1998 144 3049 3060 9846740
van Belkum A Sluijuter M de Groot R Verbrugh H Hermans PW Novel BOX repeat PCR assay for high-resolution typing of Streptococcus pneumoniae strains J Clin Microbiol 1996 34 1176 1179 8727898
Hermans PW Sluijter M Hoogenboezem T Heersma H van Belkum A de Groot R Comparative study of five different DNA fingerprint techniques for molecular typing of Streptococcus pneumoniae strains J Clin Microbiol 1995 33 1606 1612 7650196
Dunne WMJ Kehl KS Holland-Staley CA Brueggeman AB Pfaller MA Doern GV Comparison of results generated by serotyping, pulse-field restriction analysis, ribotyping, and repetitive-sequence PCR used to characterize penicillin-resistant pneumococci from the United States J Clin Microbiol 2001 39 1791 1795 11325992 10.1128/JCM.39.5.1791-1795.2001
Mollerach M Garcia E The galU gene of Streptococcus pneumoniae that codes for a UDP-glucose pyrophosphorylase is highly polymorphic and suitable for molecular typing and phylogenetic studies Gene 2000 260 77 86 11137293 10.1016/S0378-1119(00)00468-6
Lawrence ER Arias CA Duke B Beste D Broughton K Efstratiou A George RC Hall LMC Evaluation of serotype prediction by cpsA-cpsB gene polymorphism in Streptococcus pneumoniae J Clin Microbiol 2000 38 1319 1323 10747101
Lindstedt BA Multiple-locus variable number tandem repeats analysis for genetic fingerprinting of pathogenic bacteria Electrophoresis 2005 26 2567 2582 15937984 10.1002/elps.200500096
Vergnaud G Pourcel C Stackebrandt E Multiple Locus VNTR (Variable Number of Tandem Repeat) Analysis Molecular Identification, Systematics, and Population Structure of Prokaryotes 2006 Berlin Heidelberg: Springer-Verlag 83 104
Le Flèche P Hauck Y Onteniente L Prieur A Denoeud F Ramisse V Sylvestre P Benson G Ramisse F Vergnaud G A tandem repeats database for bacterial genomes: application to the genotyping of Yersinia pestis and Bacillus anthracis BMC Microbiol 2001 1 2 11299044 10.1186/1471-2180-1-2
Le Flèche P Fabre M Denoeud F Koeck JL Vergnaud G High resolution, on-line identification of strains from the Mycobacterium tuberculosis complex based on tandem repeat typing BMC Microbiol 2002 2 37 12456266 10.1186/1471-2180-2-37
Pourcel C Andre-Mazeaud F Neubauer H Ramisse F Vergnaud G Tandem repeats analysis for the high resolution phylogenetic analysis of Yersinia pestis BMC Microbiol 2004 4 22 15186506 10.1186/1471-2180-4-22
Denoeud F Vergnaud G Identification of polymorphic tandem repeats by direct comparison of genome sequence from different bacterial strains: a Web-based resource BMC Bioinformatics 2004 5 4 14715089 10.1186/1471-2105-5-4
The Microorganisms Tandem Repeats Database
Martin B Humbert O Camara M Guenzi E Walker J Mitchell T Andrew P Prudhomme M Alloing G Hakenbeck R A highly conserved repeated DNA element located in the chromosome of Streptococcus pneumoniae Nucleic Acids Res 1992 20 3479 3483 1630918
Sanchez-Beato AR Lopez R Garcia JL Molecular characterization of PcpA: a novel choline-binding protein of Streptococcus pneumoniae FEMS Microbiol Lett 1998 164 207 214 9675866 10.1016/S0378-1097(98)00206-7
Benson G Tandem repeats finder: a program to analyze DNA sequences Nucleic Acids Res 1999 27 573 580 9862982 10.1093/nar/27.2.573
The MLVA web service
Coffey TJ Enright MC Daniels M Morona JK Morona R Hryniewicz W Paton JC Spratt BG Recombinational exchanges at the capsular polysaccharide biosynthetic locus lead to frequent serotype changes among natural isolates of Streptococcus pneumoniae Mol Microbiol 1998 27 73 83 9466257 10.1046/j.1365-2958.1998.00658.x
McEllistrem MC Noller AC Visweswaran S Adams JM Harrison LH Serotype 14 variants of the France 9V(-3) clone from Baltimore, Maryland, can be differentiated by the cpsB gene J Clin Microbiol 2004 42 250 256 14715761 10.1128/JCM.42.1.250-256.2004
Leimkugel J Adams Forgor A Gagneux S Pfluger V Flierl C Awine E Naegeli M Dangy JP Smith T Hodgson A Pluschke G An Outbreak of Serotype 1 Streptococcus pneumoniae Meningitis in Northern Ghana with Features That Are Characteristic of Neisseria meningitidis Meningitis Epidemics J Infect Dis 2005 192 192 199 15962213 10.1086/431151
PubMLST – Multilocus sequence typing (MLST) databases and software
Société Française de Microbiologie
Vergnaud G Denoeud F Minisatellites: mutability and genome architecture Genome Res 2000 10 899 907 10899139 10.1101/gr.10.7.899
Hoskins J Alborn WE JrArnold J Blaszczak LC Burgett S DeHoff BS Estrem ST Fritz L Fu DJ Fuller W Genome of the bacterium Streptococcus pneumoniae strain R6 J Bacteriol 2001 183 5709 5717 11544234 10.1128/JB.183.19.5709-5717.2001
Tettelin H Nelson KE Paulsen IT Eisen JA Read TD Peterson S Heidelberg J DeBoy RT Haft DH Dodson RJ Complete genome sequence of a virulent isolate of Streptococcus pneumoniae Science 2001 293 498 506 11463916 10.1126/science.1061217
The Sanger Institute
The Institute for Genomic Research
Hunter PR Gaston MA Numerical index of the discriminatory ability of typing systems: an application of Simpson's index of diversity J Clin Microbiol 1988 26 2465 2466 3069867
|
16287512
|
PMC1315331
|
CC BY
|
2021-01-04 16:03:40
|
no
|
BMC Microbiol. 2005 Nov 16; 5:66
|
utf-8
|
BMC Microbiol
| 2,005 |
10.1186/1471-2180-5-66
|
oa_comm
|
==== Front
Malar JMalaria Journal1475-2875BioMed Central London 1475-2875-4-551629723410.1186/1475-2875-4-55ResearchEfficacy of sulfadoxine-pyrimethamine in Tanzania after two years as first-line drug for uncomplicated malaria: assessment protocol and implication for treatment policy strategies Mugittu Kefas [email protected] Salim [email protected] Nicole [email protected] Honorati [email protected] Ingrid [email protected] Hassan [email protected] Hans-Peter [email protected] Blaise [email protected] Ifakara Health Research and Development Centre (IHRDC), P. O. Box 53, Ifakara, Tanzania2 Swiss Tropical Institute, Socinstrasse 57, 4002 Basel, Switzerland2005 18 11 2005 4 55 55 23 8 2005 18 11 2005 Copyright © 2005 Mugittu et al; licensee BioMed Central Ltd.2005Mugittu 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
Systematic surveillance for resistant malaria shows high level of resistance of Plasmodium falciparum to sulfadoxine-pyrimethamine (SP) across eastern and southern parts of Africa. This study assessed in vivo SP efficacy after two years of use as an interim first-line drug in Tanzania, and determined the rates of treatment failures obtained after 14 and 28 days of follow-up.
Methods
The study was conducted in the Ipinda, Mlimba and Mkuranga health facilities in Tanzania. Children aged 6–59 months presenting with raised temperature associated exclusively with P. falciparum (1,000–100,000 parasites per μl) were treated with standard dose of SP. Treatment responses were classified according to the World Health Organization (WHO) definition as Adequate Clinical and Parasitological Response (ACPR), Early Treatment Failure (ETF), Late Clinical Failure (LCF) and Late Parasitological Failure (LPF) on day 14 and day 28.
Results
Overall 196 (85.2%) of 230 patients had ACPR on day 14 but only 116 (50.9%) on day 28 (57.7% after excluding new infections by parasite genotyping). Altogether 21 (9.1%) and 13 (5.7%) of the 230 patients assessed up to day 14 and 39 (17.1%) and 55 (24.1%) of the 228 followed up to day 28 had clinical and parasitological failure, respectively.
Conclusion
These findings indicate that SP has low therapeutic value in Tanzania. The recommendation of changing first line treatment to artemether + lumefantrine combination therapy from early next year is, therefore, highly justified. These findings further stress that, for long half-life drugs such as SP, establishment of cut-off points for policy change in high transmission areas should consider both clinical and parasitological responses beyond day 14.
==== Body
Background
There is controversy over the therapeutic life of sulfadoxine-pyrimethamine (SP) when used alone for the treatment of uncomplicated malaria in Africa. Experts do not all agree on which drug efficacy measurements more accurately predict usefulness of a drug in a community. Some consider clearance of symptoms alone [1] or plus parasites by day 14, as advised by the World Health Organization [Regional Office for Africa (WHO/AFRO) Consultative Meeting on Antimalarial Policy in the Africa Region, 14th–15th August 2003, Harare, Zimbabwe]. Others regard clearance of both symptoms and parasites over a much longer period as the most accurate measure of drug effectiveness [2-4].
The assessment methodology has profound implications in terms of treatment policy strategies. Attempts have been made to define the cut-off points for changing first-line malaria treatment. Using the old treatment response classification criteria, the action period was due when a combined Early Treatment Failure (ETF) and Late Treatment Failure (LTF) were between 16 – 24% [4,5]. A couple of years ago, WHO/AFRO recommended 15% clinical and 25% parasitological treatment failure rates at day 14 as cut-off points for implementation of policy change in intense transmission areas (WHO/AFRO consultative meeting on antimalarial policy in the Africa region, 14th–15th August 2003, Harare, Zimbabwe).
Systematic surveillance for efficacy of antimalarial drugs shows increasing levels of Plasmodium falciparum resistance to SP across eastern and southern parts of Africa [6,7]. In 2001, Tanzania adopted SP as interim first-line treatment for uncomplicated malaria while awaiting for the results of different combination therapies trials. As part of the National Malaria Control Programme (NMCP), this study assessed in vivo SP efficacy after two years of widespread use in Tanzania.
Methodology
The study was conducted from July to November 2003 in the Ipinda (south-west), Mlimba (south-east) and Mkuranga (east) health facilities in Tanzania. Malaria transmission in these areas is perennial with peaks between May and July. A slightly modified WHO antimalarial drug efficacy testing protocol [8] was used, so as to conform with another study that was being conducted at the same time under the same project framework in Papua New Guinea, in areas with lower levels of endemicity. Children aged 6–59 months presenting with raised temperature (37.5°C–39.5°C) associated with P. falciparum parasitaemia between 1,000–100,000 parasites per μl were recruited. Exclusion criteria and other procedures were as detailed in the protocol [8]. Patients were treated (under observation) with a standard dose of SP (Fansidar® Roche), i.e 1.25 mg/kg of pyrimethamine and 25 mg/kg of sulfadoxine. The responses were classified according to the new WHO definition as ACPR, ETF, LCF and LPF at day 14 and day 28 [8].
Treatment failures rates were corrected after genotyping the msp2 locus to detect new infections. Extensive diversity in this locus has been observed with over 84 allelic variants in south-eastern Tanzania [9,10] and other investigators observed high genotype complexity with an average of 4.9 genotypes per asymptomatic individual in eastern Tanzania [11]. These observations are suggestive that msp2 alone may sufficiently discriminate recrudescence from reinfection in Tanzania. It has previously been shown that analysis of msp2 locus alone can effectively distinguish recrudescence from reinfection in Uganda [12]. The clinical and molecular data were combined and analysed using Stata version 8.0 (Stata Corporation Inc, Texas, USA).
Results
A total of 241 patients were recruited, of which 13 were lost to follow-up. Table 1 summarizes patient age and clinical parameters recorded on admission day by site. Table 2 provides details of treatment outcome by site. On day 28, only 116 (50.9%) of the 228 patients showed ACPR. Molecular genotyping showed that 27/112 (24.5%) recurrent infections were due to re-infections, therefore were excluded from the analysis and recorded as withdrawn. Hence, PCR-corrected ACPR was 116/201 (57.7%). 196 (85.2%) of the 230 patients had ACPR by day 14.
Table 1 Mean age, temperature, haemoglobin and parasite density on admission day
Site Means
Weight in kg(SD)1 Age in years (SD)1 Temperature in °C (SD)1 Hb1 in g/dl (SD)1 parasites/μl (SD)
Ipinda (n = 73) 10.7(2.9) 1.7(1.3) 38.4(1.0) 9.4(1.6) 31'499 (30,260)
Mlimba (n = 75) 11.0(2.9) 1.7(1.4) 38.3(0.9) 9.0(1.6) 53,206(30,782)
Mkuranga (n = 93) 10.9(3.7) 1.3(1.0) 38.7(0.8) 8.6(1.9) 44,877(38,572)
Hb = haemoglobin; SD = standard deviation
Table 2 Sulfadoxine-pyrimethamine treatment outcomes
Results no. of included patients LF and corr. Evaluable patients ACPR
n (%) ETF
n (%) LCF
n (%) Total CF
n (%) LPF
n (%) Overall TF
n (%)
At D14
Ipinda 73 1 72 62(86.1) 7(9.7) 2(2.8) 9(12.5) 1(1.4) 10(13.9)
Mlimba 75 3 72 63(87.5) 3(4.3) 4(5.7) 7(9.1) 2(2.9) 9(12.9)
Mkuranga 93 7 86 71(82.6) 3(3.5) 2(2.3) 5(6.9) 10(11.6) 15(17.4)
Total 241 11 230 196(85.2) 13(5.7) 8(3.5) 21(9.1) 13(5.7) 34(14.9)
At D28
Ipinda 72 0 72 44(61.1) 7(9.7) 5(6.9) 12(16.7) 16(22.2) 28(38.9)
Mlimba 72 2 70 34(48.6) 3(4.3) 9(12.9) 12(17.1) 24(34.3) 36(51.4)
Mkuranga 86 0 86 38(44.2) 3(3.5) 12(13.9) 15(17.4) 33(38.4) 48(55.8)
Total 230 2 228 116(50.9) 13(5.7) 26(11.4) 39(17.1) 55(24.1) 112(49.1)
At D28 PCR-corrected
Ipinda 72 6 66 44(66.7) 7(9.7) 5(6.9) 12(16.7) 10(13.9) 22(33.3)
Mlimba 70 10 60 34(56.7) 3(4.3) 5(7.1) 8(11.4) 18(25.7) 26(43.3)
Mkuranga 86 11 75 38(50.7) 3(3.5) 8(9.3) 11(12.8) 26(30.2) 37(49.3)
Total 228 27 201 116(57.7) 13(5.7) 18(7.9) 31(13.6) 54(23.7) 85(42.3)
ACPR = Adequate Clinical and Parasitological Response; TF = Treatment failure, ETF = Early Treatment Failure, LCF = Late Clinical Failure; LPF = Late Parasitological Failure; CF = Clinical failure, n = Sample Size; TF = Treatment failure, FP = Lost to Follow-up, corr. = corrected
The total clinical failure by day 14 and 28 was observed in 21 (9.1%) out of 230 and 39 (17.1%) out of 228 patients, respectively. 13 (5.7%) and 55 (24.1%) of the patients had LPF by day 14 and 28, respectively. Thus 34 (14.9%) out of 230 and 112 (49.1%) out of 228 patients had overall treatment failure by day 14 and 28, respectively. After genotyping recurrent infections day-28 treatment failures decreased to 85 (42.3%). In this study Mkuranga recorded the highest rates of both PCR-adjusted and unadjusted treatment failures followed by Mlimba and Ipinda.
Discussion
In 2001 Tanzania replaced chloroquine with SP as interim first-line antimalarial drug. Prior to this change, baseline clinical trials with SP had been conducted throughout the country using the 14 day follow-up protocol, and indicated an average efficacy of 86% on day 14. These findings paved the way for the malaria treatment policy change [13] and SP is still used as first line antimalarial drug in Tanzania. The present study assessed SP efficacy (using a 28 day follow-up) in three sites in Tanzania after two years of use as first-line antimalarial drug. With this extended period of follow-up, half of the patients (49.1%) failed treatment. Even when new infections were taken into account by genotyping, the overall treatment failure rate (42.3%) was still high. This level of resistance is close to that observed in Muheza (45%), an area of high SP resistance in Tanzania [14].
Restricting our analysis to outcomes at day 14 would have led to misleadingly low clinical (9.1%) and parasitological (5.7%) treatment failure rates with the overall treatment failure (14.9%) being equal to that recorded at baseline prior to policy change. Using a shorter follow-up period, another study in parts of Tanzania also recorded an overall SP treatment failure of only 9.2% [15]. According to WHO/AFRO proposed thresholds for policy change (i.e 15% and 25%, respectively), these failure levels would still be considered acceptable. Retention of SP clinical efficacy at day 14 after 10 years of its use as first-line drug has been demonstrated in Malawi [1]. However, extending the follow-up to day 28 the total failure was as high as 66%. Even at day 14 the clinical and total failure rates were far above 15% and 25%, respectively. As in Malawi, the majority of the recurrent infections in our study were LPFs observed between day 14 and 28. The new WHO efficacy testing protocol [8] recommends follow up for 28 days for drug with long half-life such as SP, if genotyping can be done to distinguish recrudescence from re-infections. When efficacy assessment is based only on clearance of symptoms in the first 14 days, the level of parasite resistance can be grossly underestimated. It is widely accepted that clearance of both parasitaemia and symptoms is the most accurate measure of the intrinsic resistance of the parasite to a drug [2-4]. The resistant parasite that is apparently causing asymptomatic infection in LPF is likely to lead in the short-term to a new clinical episode [14] and/or to anaemia, depending on the immunity of the subject.
Our observations show that SP efficacy in Tanzania is compromised and fully justify the recent decision to review the current malaria treatment policy from early next year in favour of artemether + lumefantrine combination therapy. This recommendation should be implemented at a large scale as soon as possible. Such a change would be welcome to protect the use of SP in its indication for the intermittent preventive treatment of pregnant women (IPTp). Indeed, it is at present the only drug that can be used for IPTp purpose because of its good safety profile. As far as methodology is concerned, the findings stress that cut-off points for malaria treatment policy change in high transmission areas should consider both clinical and parasitological responses beyond day 14, coupled with distinction of recrudescence from re-infection using molecular genotyping.
Authors' contributions
B. Genton, H-P. Beck and K. Mugittu designed the study. S. Abdulla and K. Mugittu organised the clinical work. H. Mshinda was the supervisor in Tanzania. N. Falk, I. Felger, H-P. Beck performed molecular genotyping. H. Masanja carried out the statistical analysis. K. Mugittu and B. Genton wrote the article and all others contributed to it.
Ethical approval
The study was approved by the Tanzanian national and institutional ethics bodies
Acknowledgements
We are grateful to the Tanzania Ministry of Health (MoH), National Malaria Control Program (NMCP). We are thankful to the on-site clinical and field officers, nurses and parents/guardians who consented to participate and to Prof. M. Tanner for his critical comments on the manuscript. This study was funded jointly by the European Union (Grant no. QLK2-CT-2002-01503, BBW 03.0001) and the Swiss National Foundation for Science (Grant no. 3100-067260). IHRDC receives core financial support from Swiss Agency for Development and Co-operation. Kefas Mugittu's PhD training programme is supported by Research Training Grants from the UNICEF/UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases (TDR).
==== Refs
Plowe CV Kublin JG Dzinjalamala FK Kamwendo DS Mukadam RA Chimpeni P Molyneux ME Taylor TE Sustained clinical efficacy of sulfadoxine-pyrimethamine for uncomplicated malaria in Malawi after 10 years as first-line treatment: five year prospective studies Brit Med J 2004 328 545 548 Responses from White N, Ringwald P and authors' reply. Sulfadoxine-pyrimethamine for uncomplicated falciparum malaria. BMJ 328:1259–1260. 14757706
Sibley CH Hunt SY Drug resistance in parasites: can we stay ahead of the evolutionary curve? Trends Parasitol 2003 19 532 537 14580966 10.1016/j.pt.2003.09.009
Stepniewska K Taylor WR Mayxay M Price R Smithuis F Guthmann JP Barnes K Myint HY Adjuik M Olliaro P Pukrittayakamee S Looareesuwan S Hien TT Farrar J Nosten F Day NP White NJ In vivo assessment of drug efficacy against Plasmodium falciparum malaria: duration of follow-up Antimicrob Agents Chemother 2004 48 4271 4280 15504852 10.1128/AAC.48.11.4271-4280.2004
World Health Organization The use of antimalarial drugs Report of a WHO informal Consultation, 13–17 November 2000 WHO/CDS/RBM/200133 WHO Geneva
World Health Organization Antimalarial drug combination therapy: Report of a WHO informal Consultation, 4–5 April 2001 WHO/CDS/RBM/200135 2001 WHO Geneva
East Africa Network for Monitoring Antimalarial Treatment Monitoring antimalarial drug resistance within national malaria control programmes: the EANMAT experience Trop Med lnt Health 2001 6 891 898 10.1046/j.1365-3156.2001.00799.x
Wongsrichanalai C Pickard AL Wernsdorfer WH Meshnick SR Epidemiology of drug-resistant malaria Lancet Infect Dis 2002 2 209 218 11937421 10.1016/S1473-3099(02)00239-6
World Health Organization (WHO) Monitoring antimalarial drug resistance In report of a WHO informal consultation, 3-5 December 2001: WHO/CDS/RBM/200239 Geneva
Felger I Smith T Edoh D Kitua A Alonso P Tanner M Beck HP Multiple Plasmodium falciparum infections in Tanzanian infants Trans R Soc Trop Med Hyg 1999 93 29 34 10450423 10.1016/S0035-9203(99)90324-3
Felger I Irion A Steiger S Beck HP Genotypes of merozoite surface protein 2 of Plasmodium falciparum in Tanzania Trans R Soc Trop Med Hyg 1999 93 3 9 10450419 10.1016/S0035-9203(99)90320-6
Magesa SM Mdira KY Babiker HA Alifrangis M Farnert A Simonsen PE Bygbjerg IC Walliker D Jakobsen PH Diversity of Plasmodium falciparum clones infecting children in a holoendemic area in north-eastern Tanzania Acta Trop 2002 84 83 92 12429425 10.1016/S0001-706X(02)00179-1
Cattamanchi A Kyabayinze D Hubbard A Rosenthal PJ Dorsey G Distinguishing recrudescence from re-infection in a longitudinal antimalarial drug efficacy study: comparison of results based on genotyping of MSP-1, MSP-2, and GLURP Am J Trop Med Hyg 2003 68 133 139 12641400
Mugittu K Ndejembi M Malisa A Lemnge M Premji Z Mwita A Nkya W Kataraihya J Abdulla S Beck HP Mshinda H Therapeutic efficacy of sulfadoxine-pyrimethamine and prevalence of resistance markers in Tanzania prior to revision of malaria treatment policy: Plasmodium falciparum dihydrofolate reductase and dihydropteroate synthase mutations in monitoring in vivo resistance Am J Trop Med Hyg 2004 71 696 702 15642957
Mutabingwa T Nzila A Mberu E Nduati E Winstanley P Hills E Watkins W Chlorproguanil-dapsone for the treatment of drug resistant falciparum malaria Lancet 2001 358 1218 1223 11675058 10.1016/S0140-6736(01)06344-9
Lemnge MM Ali AS Malecela EK Sambu E Abdulla R Juma MS Fakih K Abdulla KH Njau RJ Therapeutic efficacy of sulfadoxine-pyrimethamine and amodiaquine among children with uncomplicated Plasmodium falciparum malaria in Zanzibar, Tanzania Am J Trop Med Hyg 2005 73 681 685 16222009
|
16297234
|
PMC1315332
|
CC BY
|
2021-01-04 16:37:31
|
no
|
Malar J. 2005 Nov 18; 4:55
|
utf-8
|
Malar J
| 2,005 |
10.1186/1475-2875-4-55
|
oa_comm
|
==== Front
Biomed Digit LibrBiomedical Digital Libraries1742-5581BioMed Central London 1742-5581-2-71632422210.1186/1742-5581-2-7ReviewThe "impact factor" revisited Dong Peng [email protected] Marie [email protected] Adrian [email protected] Medical Statistics and Epidemiology Group, Bioinformatics Institute, BMRC, A*STAR, Singapore2005 5 12 2005 2 7 7 26 7 2005 5 12 2005 Copyright © 2005 Dong et al; licensee BioMed Central Ltd.2005Dong 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.
The number of scientific journals has become so large that individuals, institutions and institutional libraries cannot completely store their physical content. In order to prioritize the choice of quality information sources, librarians and scientists are in need of reliable decision aids. The "impact factor" (IF) is the most commonly used assessment aid for deciding which journals should receive a scholarly submission or attention from research readership. It is also an often misunderstood tool. This narrative review explains how the IF is calculated, how bias is introduced into the calculation, which questions the IF can or cannot answer, and how different professional groups can benefit from IF use.
==== Body
Background
The number of periodical peer-reviewed scientific publications is conservatively estimated to exceed 16,000 worldwide; nearly 1.4 million articles are published every year [1,2]. Even though electronic formats theoretically allow access to most current publications, the sum of subscription fees charged by most periodicals exceeds the means of academic institutions, not to mention individuals. Accordingly, librarians must limit the quantity of periodical subscriptions. Researchers have a vast number of journals to choose from when considering where to find information, and where to publish their work. Potential employers of scientists who try to evaluate a candidate's bibliography are aware that not all publications are of equal quality. All three parties need objective, preferably quantitative, information to assist publication and subscription decisions, in effect which publications to count as important. A simple descriptive quantitative measurement of a journal's performance is the "impact factor" (IF), the average number of times articles from the journal published in the past two years have been cited in the current year.
Eugene Garfield, the founder of the Institute for Scientific Information (ISI), proposed a bibliographic system for scientific literature – "Citation Indexes for Science" in 1955 [3]. ISI's database was initially developed for cross reference literature searches and identification of individual scientists working on particular topics [4]. The citation index compiled information that was far more useful and convenient than the usual subject indexing and helped to span the gap between authors and researchers. It mainly consisted of a complete alphabetic listing of all periodicals covered and their representative codes. These codes described the bibliographic category (e.g. editorial, original research, review), while a different set of data was assigned to articles referring to an article in question.
The IF was originally conceived as a quantitative assessment of referenced publications in a given journal found in the scientific literature. By processing the data from the citation index, it became possible to calculate a ratio of cites to a journal. Garfield himself explained the meaning of impact, pointing out that a citation indicates an article has influenced someone and therefore, the more often an article is cited, the greater its influence on the scientific community [5]. This ratio was then used to select the journals for inclusion in the Science Citation Index (SCI) [6]. The journal IF is currently calculated by Thomson ISI for all journals contained in the SCI database, then reported in the Journal Citation Reports (JCR) database. Currently 3700 "world-leading" scientific journals are included in the SCI database [7]. Even more journals are "tracked", which means they are monitored but have not been assigned an IF yet. Inclusion criteria into the SCI database and for assignment of an IF were described in detail by Garfield [8].
It is generally understood that the higher the IF, the "better" the journal. As a result, journals with high IF are often preferentially acquired in institutions where subscription funds are limited; researchers are keen to submit their work to journals with high IF to further their career; the editors of journals with high IF are swamped with manuscripts by researchers who want to publish only with the best; some funding agencies expect their scientists to publish in journals above a certain IF; and recruitment officers tend to look for candidates with publications in high-IF journals.
Unfortunately, the IF alone cannot provide the knowledge needed for informed decision-making. Even more unfortunately so, it is often used simply because it is readily available, while alternative measurements are unknown or unavailable to many decision makers [4].
The present narrative review is an introduction to a field of study within scientometrics, and a basis on which librarians, researchers and funding agencies can discuss the usefulness of the IF in planning their publication and funding strategies. Topics covered include how the IF is calculated, some sources of bias in the calculation, an introduction to some alternative assessment scores that complement the IF, and questions the IF can or cannot answer in the informed decision making process.
Calculation of the impact factor
The journal IF defined by the ISI is a ratio of two elements. The denominator is the total number of "citable" articles published in a particular journal within a given timeframe. The numerator is the total number of citations in the current year to any article published in this journal during that given timeframe. The ISI has defined this time frame as two years. The IF of a journal A in a particular year Y is computed following the formula:
By ISI definition, only research articles, technical notes and reviews are "citable" items. Editorials, letters, news items, and meeting abstracts are "non-citable items" for the purpose of calculating the denominator. All items, however, may be counted in the numerator during the calculation
For example, the New England Journal of Medicine published 366 "citable" articles in 2003 and 378 "citable" articles in 2002. Citations in 2004 to any articles published in 2003 and 2002 are 14147 and 14549, respectively. Following the above formula, the IF for this journal in 2004 is:
Factors that bias the calculation of the impact factor
The ready accessibility of the IF and the lack of other well-known quality indicators have rapidly contributed to the attribution of IF as an indicator of journal quality. However, it is important to remember that the calculation of the IF is biased by many factors. These include:
• Coverage and language preference of the SCI database
• Procedures used to collect citations at the ISI
• Algorithm used to calculate the IF
• Citation distribution of journals
• Online availability of publications
• Citations to invalid articles
• Negative citations
• Preference of journal publishers for articles of a certain type
• Publication lag
• Citing behavior across subjects
• Possibility of exertion of influence from journal editors.
Journal coverage by ISI
The coverage of journals and the language preference in the SCI database are important contributors to the limitation of the IF. The SCI covers less than one fourth of peer-reviewed journals worldwide, and exhibits a preference for English language journals [9]. Non-English journals have relatively low IFs due to the limited coverage of such journals by the SCI database. Calculation of the IF for non-English journals in their native countries or regions may be a useful way to complement the data in the SCI database [8,10,11]. At the same time, it must be remembered that at present, English is the lingua franca of science, just as German was in the 19th and early 20th centuries, and Latin and Greek before that [11]. Further bias has been created by a tendency towards self-citation among American scientists [12].
Differences across research fields and subject areas
Different citing behavior across subject field imposes a bias on the IF. Articles in rapidly growing areas tend to cite much more recent references than more traditional research fields, in particular theoretical and mathematical areas [13]. This diversity leads to the wide variance of IFs across subject categories. The IF of underrepresented fields is affected negatively [13].
Collecting citations over only two years post publication has an important effect on the IF. Journals in rapidly growing research fields, such as systems biology and bioinformatics, tend to publish papers with a short time interval from submission to acceptance. A large percentage of papers are cited within two years of their publication. This, in result, leads to a high IF. However, there are many journals with longer citation half-lives. Many papers from such journals are still cited frequently much longer than two years after their publication. ISI defines "citation half-life" as the median age of the articles that were cited in the year for which the half-life is reported. Fields with more "durable" literature have a small percentage of short term citations and thus lower journal IF [13]. This field property together with the low number of references per article gives mathematics, for example, a recorded average citation impact that is only a quarter that of biochemistry [14]. Whitehouse [15] has analyzed this for the British Journal of Radiology as one example of a journal with long citation half-life. Only 12% of the cites to this journal in 1999 quote the previous two years' publications, but more than 50% of the cites in 1999 to the BJR quote papers published in the previous nine years. The scientific impact of the BJR is thus underestimated if the calculation is based only on cites to the previous two years' publications. While this affects most journals to some extent, it seems that the highest ranking journals remain quite stable, regardless of the timeframe used for the calculation of the IF [8,16].
A given research field is often also cited by related fields [13]. For example, clinical medicine draws heavily on basic science. As a result, basic research in medicine is cited three to five times more than clinical medicine. The IF is affected accordingly [17,18].
Differences between journals that have nothing to do with journal quality
A distinct weakness of the IF's algorithm lies in the inclusion of articles into the numerator count that are considered as "non-citable" in the denominator count. Citations to "non-citable" items may dramatically increase a journal's IF [19,20]. Journals publishing large proportion of "non-citable items" can thus achieve higher IFs than journals that predominantly publish "citable" items.
Similarly, the ISI algorithm does not take into account a journals' respective composition of research articles, technical notes and reviews [20]. Reviews are more likely to be cited than original research papers [13,21]. Journals publishing a high proportion of review papers consequently attract more citations and thus are likely to achieve a higher IF.
Editorial preference for longer articles seems to increase a journal's IF. Seglen [21] has shown that the citation rate is proportional to the article length, i.e. longer articles are cited more often.
Given the rapid growth of electronic publications, the online availability of articles has recently become an important factor to influence the IF. Murali et al. [22] determined how the IF of medical journals is affected by their online availability. In that study, a document set obtained from MEDLINE was classified into three groups, namely FUTON (full text on the Net), abstracts only and NAA (no abstract available). Online availability clearly increased the IF. In the FUTON subcategory, there was an IF gradient favoring journals with freely available articles. This is exemplified by the success of several "open access" journals published by BioMed Central (BMC) and the Public Library of Science (PLoS). Open access journals publish full-text online papers free of subscription fees [23]. BioMed Central (BMC) is an "open access" publisher in business since 2000. BMC hosts over 100 biomedical journals ranging from general interest to specialized research. More than twenty journals published by BMC are currently tracked by the ISI and over half of these have IFs available for the recent years. BMC Bioinformatics was assigned its first IF for 2004. At 5.4, it places the journal second in the field, only marginally below the traditional competitor Bioinformatics (IF = 5.7), which has a 20-years' publishing history and is connected to a major learned society within this field of research (International Society for Computational Biology).
PLoS (Public Library of Science) is another example of a successful "open access" publishing strategy. It started publishing two open access journals in biology and medical research in 2003 and 2004 respectively [24]. PLoS Biology was assigned its first IF of 13.9 for 2004. In the ISI subject category "biology", it is thus placed at the number 1 position of 64 in its first year of reporting an IF. FASEB journal at position 2 has an IF of 6.8, but has been in circulation since 1987. Similarly, in the other SCI subject category ("biochemistry and molecular biology")in which PLOS Biology is listed, it ranks at position 8 out of 261. Monitoring the development of such journals' IF will inform the determination of the online-availability bias in the future. This effect will increase in the future with the availability of new search engines with deep penetration such as Google Scholar [25,26], allowing researchers to find relevant articles in an instant, and then choose those with immediately and freely available content over those with barriers, economic and otherwise.
Accuracy of data capture by ISI
Investigations by Nature suggested a significant undercount of "citable" items in Nature Genetics in 1996 and an erroneous inclusion of "citable" items other than those defined by ISI itself for Nature in 2000 [4]. A more recent issue is undercounted citations to articles authored by consortia, rather than by a list of individual authors [27]. The article reporting the draft human genome sequence from the International Human Genome Sequence Consortium [28] is considered as a landmark paper published in Nature in 2001, but was surprisingly absent from the list of "hot papers" in biology, which are published regularly by ISI Science Watch [29]. The examination of the ISI's data showed that the ISI only considered citations to the full list of authors, led by Eric Lander of the Whitehead Institute for Biomedical Research at Cambridge, Massachusetts, and hence led to the grossly undercounted representation. The same applied to other prominent papers authored by consortia.
The accuracy of how citations are collected at ISI significantly influences the final published IF statistics. At ISI, data capture of journal papers is completed by optical character recognition software, important fields are highlighted manually, and the final tagging of every individual article is computerized. Algorithms have been designed to count valid citations. The simple involvement of two highlighted fields, such as the journal title and the year, makes the citation counting for the IF calculation easier. However, this raises systematic bias as citations cannot be matched to individual articles in real time. At this point an incorrect citation leaves its mark. For the field of environmental and occupational medicine, a recent study reported a prevalence of 3.35% incorrect citations; the respective articles receive an incorrect cite count, thus potentially reducing their journals' IF [30].
IF is calculated for a whole journal whereas citations are to individual articles
The IF would reflect a journal's interest to the research community if citations were indeed distributed equally over all articles in the journal. However, this is not the case. Only a small percentage of articles are highly cited. Based on the analysis of three biochemical journals, Seglen [13] found that the most cited 15% of articles account for 50% of the citations and the most cited 50% of articles account for almost all citations (90%). These numbers were confirmed by a later study based on two cardiovascular journals [31]. The most recent study on articles published in Nature showed a similar high skew of citations: 89% of 2004's citations were generated by just 25% of Nature's papers [32]. Apparently, researchers cannot solely depend on the IF to judge the quality of the journal.
Highly cited articles are found mostly in a small subset of journals, regardless of how parameters of the algorithm (e.g. average time-frame) are changed. In Garfield's view, these two combined effects strengthen the ISI's position as a means to point authors and readers to journals with true scientific impact [8]. The argument is that this effect justifies the fact that JCI is not all-inclusive, but rather selective. According to Garfield, JCI could still be considered comprehensive if it covered only the 500 most cited journals.
Invalid articles may pose a considerable bias on the journal IF. Retracted articles may continue to be cited by others as valid work. Pfeifer and Snodgrass [33] identified 82 completely retracted articles, analyzed their subsequent use in the scientific literature, and found that these retractions were still cited hundreds of times to support scientific concepts. Kochan and Budd [34] showed that retracted papers by John Darsee based on fabricated data were still positively cited in the cardiology literature although years had passed since retraction. Budd et al. [35] obtained all retractions from MEDLINE between 1966 and August 1997 and found that many papers still cited retracted papers as valid research long after the retraction notice.
Interesting papers, based on fraudulent data, may attract the scientific community's attention and be cited frequently, thus distorting the true impact of the journal that featured the sensational article. In a notable 2002 case of scientific fraud, Jan Hendrik Schön, a former researcher at Bell Laboratory, published "remarkable" findings on superconductivity, molecular electronics, and molecular crystals in several scientific journals, including Science, Nature and Applied Physics Letters. He was later found out to have falsified or fabricated data in 16 of 24 alleged cases of misconduct [36]. The data of 25 publications were implicated in the perpetuation of dubious claims. The findings of the investigation dismissed research results from "high impact" papers that had been promoted as major breakthroughs in the field.
Active manipulation of IF
Owing to the preference authors and researchers give to high IF journals, editors may be tempted to artificially raise a journal IF. One very crude way to do so is by requesting author self-citation. In 1997, the journal Leukemia was accused of trying to manipulate its IF [37]. This first accusation came from Terry Hamblin, editor of Leukemia Research, a competitor to Leukemia. The evidence he was holding showed that Leukemia had asked authors who had submitted a paper to the journal to cite more articles from Leukemia. Later in 2002, Neuberger and Counsell [38] reported another similar case: they described how one journal editor suggested the inclusion of more references to that journal. In 2004, Sevinc [39] reported yet another incident. The influence of authors' choice of references distorts the perception of the journal within the scholarly community and is considered as highly unethical behavior.
Alternative journal impact measures
The wide use of the IF, combined with obvious flaws, has motivated researchers in scientometrics to try to improve the algorithm for the calculation of the IF or to develop alternative journal citation measures altogether.
Van Leeuwen and Moed [20] have critically analyzed the use and validity of the ISI IF. They focused on four aspects: "non-citable" items included in the numerator of the IF calculation; the relative distribution of research articles; technical notes and reviews, different citing behavior across subject fields; and the fixed two-year citation window. They developed an alternative journal impact measure, the Journal to Field Impact Score (JFIS), to provide solutions to biases incurred from these four aspects. The JFIS includes research articles, technical notes, letters and reviews both in the numerator and the denominator. The JFIS also is field-normalized by comparing the journal's impact with the citation average in the fields it covers. The JFIS takes into account the relative distribution among the four types of distribution. Finally, the JFIS is computed based on a flexible and variable citation and publication window, and the selected publication window can in principle be set to any length. Despite the improvements that the JFIS has over the IF, van Leeuwen and Moed still suggested that more than one indicator should be used in bibliometric journal impact measurements.
Other researchers have focused on refining the ISI IF's limitations, such as the fixed two-year chronologic window. Asai [40] found that more accurate statistics could be calculated if the period count is based on months rather than a year. Accordingly, he proposed an Adjusted Impact Factor to count a weighted sum of citations per month over a time period of four years. Glänzel and Schoepflin [41] conducted a bibliometric study to analyze the time behavior of citations to articles published in seven journals in different subject fields including social sciences, chemistry, medicine and mathematics. The results suggested a three-year citation window to be a good compromise between fast growing disciplines and slowly aging theories.
Sombatsompop et al. [42] introduced the cited half-life into the IF calculation as an alternative to setting the citation window at an absolute number. The proposed indicator, the Cited Half-Life Impact Factor (CHAL-IF), is calculated by replacing the two-year citation window with the journal's cited half-life in the IF computation formula. This study was based on 34 journals in the Polymer Science Category from the ISI subject heading categories. The journal ranking based on the CHAL-IF was different from that based on the ISI IF. The average IF by the CHAL method achieved a better stability than that calculated by the standard ISI method. Rousseau [43] renamed the CHAL-IF to Median Impact Factor (MIF). He further generalized the MIF to create a Percentile IF (pIF). The MIF is a special case of the pIF with p set at 50%. These modified IFs are not meant to replace the ISI IF, but should rather be understood as a complementary assessment tool.
When ranking a list of journals within a subject discipline, it is inadequate to only compare the IF without consideration of subject bias. Hirst [44] introduced what he called the Disciplinary Impact Factor (DIF) to overcome this subject bias. It is based on the average number of times a journal was cited in a sub-field rather than the entire SCI database. A similar approach was chosen by Pudovkin and Garfield [45], who suggested a rank normalized impact factor to be calculated within each subject category. For any journal j, its rnIF is designated as rnIF(j) and equals (K - R-j + 1)/K, where R-j is the descending rank of journal j in its JCR category and K is the number of journals in the category. Ramírez et al. [46] proposed a renormalized IF which was calculated based on the maximum IF and median IF of each category. This quantitative parameter allows the direct comparison among different research areas without introducing other considerations. Sombatsompop [47,48] introduced a new mathematical index, the "Impact Factor Point Average" with the specific aim to allow across-field comparison of IF.
The above-mentioned variants of the IF may improve journal citation methodological aspects. As of now, no database makes use of these derivative algorithms. They are neither widely known nor accessible to the scientific community. There are some commercial alternative databases available that claim to overcome the intrinsic flaws of the SCI database.
The Euro-Factor (EF) database is a moderately successful example of citation analysis innovation. Targeting the language bias and perceived USA-centricity of the SCI database, the Euro-Factor™ (EF) [49] was proposed as an alternative to the ISI IF to meet the citation measurement demand of the European scientific community. The publishing company VICER [50] created the "Euro-Factor" database, in order to collect bibliometric data from biomedical journals in European countries. More than 500 journals were included by means of a peer-reviewed quality selection process. A new algorithm was designed to analyze the biometric relationship between European journals:
Unfortunately, VICER does not provide detailed explanation of the algorithm outside of the simple formula, which arbitrarily sets the EF-Coefficient at a value of 10. The formula does not further the understanding of how a Europe-specific ranking is achieved. The EFs of all European journals covered are calculated every year, and the list of EFs is available from VICER every January. According to VICER, the EF for Lancet and Nature in 2002 is 106.1 and 55 [49], whereas the ISI gives them IF of 15.4 and 30.4 respectively. In these two prominent examples, it seems somewhat naïve to speak of European journals, as both have editorial offices in the United States.
The Prestige Factor (PF) database possessed a dubious and short-lived existence. In an effort to challenge the ISI IF, in 2001 the "Prestige Factor" (PF) was launched at "PrestigeFactor.com". The PF was heralded as a superior assessment tool. It promised to measure the true value of academic journals by including original articles only and hosting a "superior" database compared to SCI. With only minor differences, such as the inclusion of original articles only and a three year citation count window, the underlying premise of both the IF and PF was identical [51]. One detailed analysis of the PF's social sciences subset found essential misrepresentations and misleading data on the company's website [52]. Concerns about the source of citations in the PF database were raised and led to doubts and competitive accusations. In 2002, the company was forced out of business in the wake of a threat from ISI to sue for intellectual property infringements.
What question does the impact factor answer?
Strictly speaking, the journal IF only measures the average citation rate of all the "citable" articles (research articles, technical notes and reviews) in a journal. As such, IF is not a perfect tool to measure the journal quality. However, in most cases, it performs what it promises when various flaws are taken into active consideration. Ready accessibility and regular updates of the ISI IF provides the best available indicator for journal quality, accepted widely within the scientific community. Journals with the highest IF in each discipline are usually the most prestigious ones [8]. It can be considered as a general guideline that helps librarians determine which journals to purchase, helps authors to decide which journal to submit their work to, helps editors and publishers to assess their journals, and helps the funding agencies to shortlist applicants. Garfield [11] points out the IF's surrogate function as a measure of potential future impact of very recent publications, and as a safeguard against hiding ineffective research where funding may have been obtained through political connections rather than research quality. In Garfield's words: "impact simply reflects the ability of journals and editors to attract the best papers available" [53].
What questions does the impact factor not answer?
The IF cannot assess the quality of individual articles, due to the qualitative variety of citations distributed in a journal [13,31,32]. A small proportion of articles count for a large percentage of citations. This means that a typical article in a high IF journal may not be cited more frequently than an average article in a relative low IF journal. As a result, IF alone is not able to judge the individual article's or author's performance.
Even under the assumption that citations were equally distributed among all articles, the IF would only measure the interests of other researchers in an article, but not the article's importance and usefulness. The Guide to Clinical Preventive Services by the US Preventive Services Task Forces (USPSTF) [54] is generally thought to be an example of top-level scientific evidence, the best available knowledge source. Nakayama et al. [55] showed that articles from "low impact factor" journals were also cited frequently in this guide, demonstrating the usefulness of those articles in providing clinical evidence.
In order to determine the relationship between citation factors and a trial's methodological quality, Berghmans et al. [56] analyzed citation indexes including the IF by assessing 181 eligible articles included in nine systematic reviews performed by the European Lung Cancer Working Party (ELCWP). The results showed that journals with higher citation factors did not publish higher quality clinical trials. Furthermore, several studies showed invalid articles continue to be cited after their retraction [33-35]. The high number of citations to these articles may raise the IF of the respective journals, yet this high IF cannot guarantee the significance and performance of an article published in such periodicals.
How may different professional groups take into account the limitations of the impact factor?
Different professional groups need to take into consideration the inherent limitations of the IF. Librarians can use the IF to identify multidisciplinary journals, as a higher IF hints at wider acceptance of the journal. New and very specialized journals, however, must be assessed separately, as an IF might not yet be available or not reflect the importance of the journal within a given field of high specialization. The easiest way to assess the relative position of a particular journal within its field is to browse through the SCI's Subject Category and sort all journals by their IFs in a particular category. It should be noted that some journals may be classified as members of more than one category, and ranked differently across categories. When assessing new journals or journals from highly specialized disciplines, librarians should actively look for guidance from researchers at their institution that might be involved in that particular field of science.
Scientists hope to publish in "prestigious" and widely-read journals primarily to communicate their findings and achieve visibility with peers, enhancing their career prospects. While considering a journal submission target, the most important factors influencing authors' decision are the perceived reputation of the journal (often equated with the IF), closely followed by the international reach and inclusion in abstracting and indexing services [57-59]. As an indirect measure of these qualities, the IF has a place in the process of decision making [11], but should not be paramount. Thought should be given to how well the manuscript's topic fits the journal, the actual circulation numbers, and potential readership. As readers, scientists may customize IF analysis by including only citations from individually chosen trusted journals to other journals in the field and thus identify relatively unknown journals of interest to this research topic. This technique was suggested (and used) by Garfield in 1972 [60], but may be somewhat obsolete in an era when digital library readers can quickly access and scan the abstracts of interesting articles.
Editors and publishers must have a strong determination to publish valid articles without regard to the possibility of a potentially high citation count. Publishers must also analyze how articles are cited (Do citing authors agree or disagree? Do they cite a technique, or a conclusion? Are citations to valid articles, or was a retraction overlooked?) if they want to improve the quality of the research they publish. Editors might put additional effort in identifying the "best" quality papers, rewarding the successful author(s) a distinction of merit, useful when preparing an academic promotion portfolio [51]. BMC has recently introduced such a feature by labeling articles as "Highly accessed" if they are accessed more frequently than would be expected in the subject category. BMC does not, however, disclose the exact usage benchmark. In the end, nothing replaces innovative, and even good controversial research [11] as a promotional too for a scientific journal.
ISI cautions against the use of IF for the evaluation of individual researchers [4], yet funding agencies continue to track the IF record of applicants as an individual's investigator's assessment. Finland demonstrates an extreme example of IF canonized into law. Finnish government funding for university hospitals there partially depends on "publication points", which are derived from the IF of journals wherein the researchers publish their work [4]. Due to the IF's inability to compare individual articles, funding agencies should develop a detailed assessment of how an individual's work impacts on the scientific community, including how a submission decision was made. The IF is an indirect, affiliated measure of a researcher's work at best. A recent publication in a high-impact journal with high editorial standards and strict peer-review leads to the assumption of quality for the individual article [11]. Individual researcher assessments by funding agencies or potential employers would be best advised to make use of subject category-specific derivates of the IF, such as the rank-normalized IF [45].
Conclusion
The present narrative review gives an introduction to the scientometrics of the ISI IF to non-specialist librarians, researchers and administrators. It describes the IF as a bibliometric tool with limited explanatory power. The IF must be used with full knowledge of its limitations and can then serve an indirect affiliated indicator of research quality. More precise information can be gained if some of the described alternative measures are appropriately used.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
Author 1 (PD) and author 3 (AM) participated in literature review, while author 1 (PD), author 2 (ML) and author 3 (AM) drafted the manuscript.
==== Refs
Mabe M Amin M Growth dynamics of scholarly and scientific journals Scientometrics 2001 51 147 162 10.1023/A:1010520913124
Mabe M The Growth and Number of Journals Serials 2003 16 191 197
Garfield E Citation indexes for science; a new dimension in documentation through association of ideas Science 1955 122 108 111 14385826
Adam D The counting house Nature 2002 415 726 729 11845174
Garfield E Which medical journals have the greatest impact? Ann Intern Med 1986 105 313 320 3524345
Garfield E Journal impact factor: a brief review Cmaj 1999 161 979 980 10551195
Science Citation Index [http://www.isinet.com/products/citation/sci/]
Garfield E How ISI selects Journals for Coverage: Quantitative and Qualitative Considerations Current Contents 1990 22 5 13
Moed HF Burger WJM Frankfort JG Van Raan AFJ On the measurement of research performance: the use of bibliometric indicators 1987 Leiden, Science Studies Unit, LISBON-Institute, Universith of Leiden
Bordons M Fernández MT Gómez I Advantages and limitations in the use of impact factor measures for the assessment of research performance in a peripheral country Scientometrics 2002 53 195 206 10.1023/A:1014800407876
Garfield E The meaning of the Impact Factor International Journal of Clinical and Health Psychology 2003 3 363 369
Moller AP National citations Nature 1990 348 480
Seglen PO Why the impact factor of journals should not be used for evaluating research Bmj 1997 314 498 502 9056804
Moed HF Burger WJM Frankfort JG Van Raan AFJ The application of bibliometric indicators: important field- and time-dependent factors to be considered. Scientometrics 1985 8 177 203 10.1007/BF02016935
Whitehouse GH Citation rates and impact factors: should they matter? Br J Radiol 2001 74 1 3 11227769
Garfield E Long-term vs. short-term journal impact: does it matter? Physiologist 1998 41 113 115 9652169
Seglen PO Bruk av siteringsanalyse og andre bibliometriske metoder i evaluering av forskningsaktivitet. Tidsskr Nor Laegeforen 1989 104 331 335
Narin F Pinski G H GH Structure of the biomedical literature J Am Soc Inf Sci 1976 27 25 45
Jacsó P A deficiency in the algorithm for calculating the Impact Factor of scholarly journals: The Journal Impact Factor Cortex 2001 37 590 594 11721873
van Leeuwen TN Moed HF Development and application of journal impact measures in the Dutch science system Scientometrics 2002 53 249 266 10.1023/A:1014808709694
Seglen PO Weingart P, Sehringer R and Winterhager M Evaluation of scientists by journal impact Representations of science and technology 1992 Leiden, DSWO Press 240 252
Murali NS Murali HR Auethavekiat P Erwin PJ Mandrekar JN Manek NJ Ghosh AK Impact of FUTON and NAA bias on visibility of research Mayo Clinic Porceedings 2004 79 1001 1006
BioMed Central - the Open Access Publisher [http://www.biomedcentral.com/]
PLoS - Public Library of Sicence [http://www.plos.org/]
Dong P Loh M Mondry A Relevance similarity: an alternative means to monitor information retrieval systems Biomed Digit Libr 2005 2 6 16029513 10.1186/1742-5581-2-6
Banks MA The excitement of Google Scholar, the worry of Google Print Biomed Digit Libr 2005 2 2 15784147 10.1186/1742-5581-2-2
Errors in citation statistics Nature 2002 415 101 10.1038/415101a
Initial sequencing and analysis of the human genome Nature 2001 409 860 921 11237011 10.1038/35057062
Science Watch [http://www.sciencewatch.com/]
Gehanno JF Darmoni SJ Caillard JF Major inaccuracies in articles citing occupational or environmental medicine papers and their implications J Med Libr Assoc 2005 93 118 121 15685284
Opthof T Coronel R Piper HM Impact factors: no totum pro parte by skewness of citation Cardiovasc Res 2004 61 201 203 14736534 10.1016/j.cardiores.2003.11.023
Not-so-deep impact Nature 2005 435 1003 1004 10.1038/4351003a
Pfeifer MP Snodgrass GL The continued use of retracted, invalid scientific literature JAMA 1990 263 1420 1423 2406475 10.1001/jama.263.10.1420
Kochan CA Budd JM The persistence of fraud in the literature: the Darsee case J Am Soc Inf Sci 1992 43 488 493 11653988 10.1002/(SICI)1097-4571(199208)43:7<488::AID-ASI3>3.0.CO;2-7
Budd JM Sievert ME Schultz TR Phenomena of Retraction: Reasons for Retraction and Citations to the Publications JAMA 1998 280 296 297 9676689 10.1001/jama.280.3.296
Brumfiel G Misconduct findings at Bell Labs shakes physics community Nature 2002 419 419 421
Smith R Journal accused of manipulating impact factor Bmj 1997 314 463
Neuberger J Counsell C Impact factors: uses and abuses Eur J Gastroenterol Hepatol 2002 14 209 211 11953682 10.1097/00042737-200203000-00001
Sevinc A Manipulating impact factor: an unethical issue or an Editor's choice? Swiss Med Wkly 2004 134 410 15389359
Asai I Adjusted age distribution and its application to impact factor and immediacy index Journal of the American Society for Information Science 1981 32 172 174
Glänzel W Schoepflin U A bibliometric study on aging and reception processes of scientific literature Journal of Information Science 1995 21 37 53
Sombatsompop N Markpin T Premkamolnetr N A modified method for calculating the Impact Factors of journals in ISI Journal Citation Reports: Polymer Science Category in 1997-2001 Scientometrics 2004 60 217 235 10.1023/B:SCIE.0000027794.98854.f6
Rousseau R Median and percentile impact factors: A set of new indicators Scientometrics 2005 63 431 441 10.1007/s11192-005-0223-1
Hirst G Discipline impact factor: a method for determining core journal lists Journal of the American Society for Information Science 1978 29 171 172
Pudovkin AI Garfield E Rank-normalized impact factor: A way to compare journal performance across subject categories MANAGING AND ENHANCING INFORMATION: CULTURES AND CONFLICTS Bryans JB 2004 41 507 515
Ramírez AM García EO Río JAD Renormalized impact factor Scientometrics 2000 47 3 9 10.1023/A:1005600807292
Sombatsompop N Markpin T Yochai W Saechiew M An evaluation of research performance for different subject categories using Impact Factor Point Average (IFPA) index: Thailand case study Scientometrics 2005 65 293 305 10.1007/s11192-005-0275-2
Sombatsompop N Markpin T Making an equality of ISI impact factors for different subject fields Journal of the American Society for Information Science and Technology 2005 56 676 683 10.1002/asi.20150
Hofbauer R Frass M Gmeiner B Kaye AD Euro-Factor - The new european scientific currency 2002 Vienna, VICER Publishing
VICER [http://www.vicer.org/]
Walter G Bloch S Hunt G Fisher K Counting on citations: a flawed way to measure quality Med J Aust 2003 178 280 281 12633486
Jacsó P The mirage of prestige [http://www2.hawaii.edu/~jacso/extra/infotoday/prestige/mirage prestige.htm]
Garfield E How can impact factors be improved? Bmj 1996 313 411 413 8761234
US Preventive Services Task Force [http://odphp.osophs.dhhs.gov/pubs/guidecps/]
Nakayama T Fukui T Comparison between impact factors and citations in Evidence-Based Practice Guidelines 2005
Berghmans T Meert AP Mascaux C Paesmans M Lafitte JJ Sculier JP Citation indexes do not reflect methodological quality in lung cancer randomised trials Ann Oncol 2003 14 715 721 12702525 10.1093/annonc/mdg203
Rowland F Two large-scale surveys of electronic publication in the United Kingdom 1999 131 136
Swan A Brown S 'What authors want' Report of the ALPSP research study on the motivations and concerns of contributors to learned journals Learned Publishing 1999 12 74 10.1087/09531519950145878
Mabe MA An overview of existing user behaviour research 2004 in www.alpsp.org/events/previous/mab040703.ppt
Garfield E Citation analysis as a tool in journal evaluation Science 1972 178 471 479 5079701
|
16324222
|
PMC1315333
|
CC BY
|
2021-01-04 16:38:25
|
no
|
Biomed Digit Libr. 2005 Dec 5; 2:7
|
utf-8
|
Biomed Digit Libr
| 2,005 |
10.1186/1742-5581-2-7
|
oa_comm
|
==== Front
BMC CancerBMC Cancer1471-2407BioMed Central London 1471-2407-5-1511631645910.1186/1471-2407-5-151Research ArticleFirst-line therapy with gemcitabine and paclitaxel in locally, recurrent or metastatic breast cancer: A phase II study Allouache Djelila [email protected] Sulochana R [email protected] Michele [email protected] Nicole [email protected] Sophie [email protected] Fawzia [email protected] Laurence [email protected] Jean-Yves [email protected] Centre F. Baclesse, Lion sur mer, 14076 CAEN Cedex 05, France2 Eli Lilly and Company, Lilly Corporate, Indianapolis, IN 46285, USA3 Ctr René Huguenin, 35 rue Dailly, 92210, Saint-Cloud, France4 Hopital Dupuytren, 2 avenue Martin Luther King, 87042, Limoges, France5 Hospital Curie, 26 rue Ulm, 75005, Paris, France6 Lilly France, 13 rue Pagès 92158 Suresnes Cedex, France2005 29 11 2005 5 151 151 1 12 2004 29 11 2005 Copyright © 2005 Allouache 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
This phase II study evaluated the efficacy and safety of gemcitabine (G) plus paclitaxel (T) as first-line therapy in recurrent or metastatic breast cancer.
Methods
Patients with locally, recurrent or metastatic breast cancer and no prior chemotherapy for metastatic disease received G 1200 mg/m2 on days 1 and 8, and T 175 mg/m2 on day 1 (before G) every 21 days for a maximum of 10 cycles.
Results
Forty patients, 39 metastatic breast cancer and 1 locally-advanced disease, were enrolled. Their median age was 61.5 years, and 85% had a World Health Organization performance status (PS) of 0 or 1. Poor prognostic factors at baseline included visceral involvement (87.5%) and ≥2 metastatic sites (70%). Also, 27 (67.5%) patients had prior adjuvant chemotherapy, 25 of which had prior anthracyclines. A total of 220 cycles (median 6; range, 1–10) were administered. Of the 40 enrolled patients, 2 had complete response and 12 partial response, for an overall response rate of 35.0% for intent-to-treat population. Among 35 patients evaluable for efficacy the response rate was 40%. Additional 14 patients had stable disease, and 7 had progressive disease. The median duration of response was 12 months; median time to progression, 7.2 months; median survival, 25.7 months. Common grade 3/4 toxicities were neutropenia in 17 (42.5%) patients each, grade 3 leukopenia in 19 (47.5%), and grade 3 alopecia in 30 (75.0%) patients; 1 (2.5%) patient had grade 4 thrombocytopenia.
Conclusion
GT exhibited encouraging activity and tolerable toxicity as first-line therapy in metastatic breast cancer. Phase III trials for further evaluation are ongoing.
==== Body
Background
In most developed countries, breast cancer is second only to lung cancer as the most common cause of cancer-related death in women, [1] and thus represents a serious health-care problem. Systemic therapy for patients with metastatic breast cancer (MBC) consists of hormonal therapy and cytotoxic chemotherapy. Anthracyclines such as doxorubicin and epirubicin can yield response rates of around 20% to 40% in MBC patients when used as single agents, and up to 60% when given as part of combination regimens [2]. However, the efficacy achieved with anthracyclines comes at the cost of high toxicity. New cytotoxic drugs with high activity, such as taxanes (paclitaxel and docetaxel), vinorelbine, gemcitabine, and capecitabine (all of which were introduced in the 1990s), have raised the hopes of patients with MBC to experience higher efficacy with tolerable toxicity. Recent studies suggest that combination chemotherapy may be more effective than single-agent therapy [3,4].
Paclitaxel is a mitotic spindle poison that promotes microtubular aggregation and interferes with essential cellular functions such as mitosis, cell transport, and cell motility [5,6]. It has shown remarkable activity in both chemonaive and anthracycline-resistant patients with MBC. Single-agent paclitaxel produced response rates of 32% to 62% in MBC patients not pretreated with chemotherapy, and from 6% to 48% in those who relapsed after treatment with anthracyclines [7-12].
Gemcitabine (difluorodeoxycytidine), an analog of cytosine arabinoside (ara-C), is a pyrimidine antimetabolite [13] that acts as a competitive substrate for incorporation into DNA where it brings about chain termination. It has undergone considerable testing for various malignancies and has exhibited activity in many solid tumors, including advanced or MBC. In MBC, single-agent gemcitabine has yielded response rates of up to 37% in chemonaive patients, [14-16] and 26% in those pretreated with anthracyclines [16-18]. Median progression-free survival with gemcitabine monotherapy was in the range of 2 to 6 months [14-20].
Preclinical studies of gemcitabine and paclitaxel have suggested that mechanisms of resistance affecting one drug may not have a significant effect on the other [21,22]. In pharmacokinetic studies, paclitaxel increased accumulation of dFdCTP, which might enhance the antitumor activity of gemcitabine in clinical studies [23]. The administration of paclitaxel prior to gemcitabine indicated an additive effect and was considered a better choice than a reverse sequence [24]. Study of this combination in treatment of ovarian cancer has also indicated that paclitaxel before gemcitabine is a less toxic sequence[25].
In summary, gemcitabine and paclitaxel are 2 agents with unique mechanisms of action, noncross-resistance, and the potential for synergistic antitumor activity. They are both active as single agents in the treatment of patients with MBC, and when used together, the treatment was well tolerated with promising activity [26-28]. A 3-week schedule of gemcitabine/paclitaxel (with gemcitabine administered on days 1 and 8, and paclitaxel on day 1) was tolerated better than a 4-week schedule [29,30].
Based on this information, we initiated a phase II, open-label, multicenter, nonrandomized study of gemcitabine in combination with paclitaxel given as first-line therapy in patients with recurrent or MBC. The primary objectives were to evaluate the response rate and duration of response. The secondary objectives were to assess safety, and to determine the time to disease progression and overall survival of this combination.
Methods
Eligibility criteria
Patients with unresectable, locally recurrent, or metastatic breast cancer that was not amenable to surgery or radiation of curative intent were enrolled. The qualifying patients were required to have histologically or cytologically confirmed breast carcinoma, with bidimensionally measurable lesions at least 1 cm × 1 cm (or 2 cm × 2 cm by physical examination). Prior chemotherapy was not allowed unless patients had local or metastatic relapse more than 12 months after the end of prior adjuvant or neoadjuvant chemotherapy. Patients could not have received previous therapy with gemcitabine or a taxane. Patients aged 18 to 75 years were further required to have a World Health Organization (WHO) performance status of 0 to 2, and an estimated life expectancy of at least 12 weeks. It was also necessary for patients to have adequate bone marrow function (absolute granulocyte count [AGC] ≥1.5 × 109/L, platelet count ≥100 × 109/L, and hemoglobin ≥90 g/L), adequate liver function (bilirubin ≤1.5 times the upper limit of normal [ULN], alanine transaminase [ALT] and aspartate transaminase [AST] ≤3 times ULN, or up to 5 times the ULN in patients with known metastatic liver disease), and adequate renal function (creatinine ≤2.5 times above the ULN). Prior radiation was permitted only if measurable disease was outside a previously irradiated area, if radiotherapy was not given to more than 50% of bone marrow volume, and if it was terminated at least 4 weeks prior to enrollment. Prior bone marrow transplantation, as adjuvant therapy, was allowed, and antitumoral hormonal treatment had to be terminated before enrollment. Patients with inflammatory breast cancer without evidence of metastatic disease, as well as patients who were pregnant, who had a neurological disorder WHO grade ≥2, or serious concomitant systemic disorders incompatible with the study could not participate. Patients were also excluded for active cardiac disease not controlled by therapy and/or myocardial infarction within the previous 6 months. Additional exclusion criteria included active infection, presence of severe psychiatric disease, or second malignancy (except in situ carcinoma of the cervix or adequately treated basal cell carcinoma of the skin).
All the patients who presented and met eligibility criteria were entered into the study. Physicians obtained signed informed consent from all patients prior to administering treatment. The study was conducted per the guidelines of good clinical practice and the Declaration of Helsinki.
Treatment plan
Paclitaxel 175 mg/m2 was administered intravenously (iv), before gemcitabine, on day 1 over a period of 3 hours. Gemcitabine 1200 mg/m2 was given iv on days 1 and 8 over a period of 30 to 60 minutes (ideally 30 minutes), followed by a 1-week rest period. Each 21-day (3-week) period defined a cycle of therapy. Multiple injections of both drugs were administered for a total of at least 2 cycles unless it was clearly not in the patient's best interest to continue. Treatment was stopped in case of intolerable toxicity or disease progression. Additional cycles, up to a maximum of 10, could be administered in patients exhibiting a complete response (CR) or partial response (PR). All patients were premedicated with dexamethasone and chlorpheniramine, or clemastine or ranitidine or cimetidine prior to paclitaxel administration to prevent severe hypersensitivity reactions.
Dose adjustments during treatment (day 8) were made based on weekly AGC and platelet counts performed within 24 hours prior to the start of therapy and clinical assessment of nonhematologic toxicities. For an AGC <1.0 (×109/L) and/or a platelet count <75 (×109/L) or a WHO nonhematologic toxicity grade 3 (except nausea/vomiting and alopecia) or grade 4, doses of gemcitabine were held. A dose missed for any reason was not given at a later time.
The day-1 dose of each subsequent cycle depended on the toxicity seen in the previous cycle. The treatment was delayed until the AGC returned to 1.5 and the platelet count to 100. Otherwise, full doses of both drugs were given, except in patients with WHO grade 4 granulocytopenia lasting for more than 1 week, or grade 4 neutropenia associated with fever ≥38.5°C, or grade 4 thrombocytopenia. In these circumstances, after recovery, the day 1 and 8 doses of both drugs were given at 75% of the dose given on day 1 of the last cycle. The observed nonhematologic toxicities (except alopecia and vomiting) had to return to WHO grade 0 to 1, or baseline conditions, before resuming injections of both drugs. Doses in subsequent cycles were reduced to 75% or held for any grade 3 nonhematologic toxicity (except nausea/vomiting and alopecia), and were reduced to 50% or held for any grade 4 nonhematologic toxicity. Patients were withdrawn from the study after 3 weeks of treatment delay due to any toxicity.
Baseline and treatment assessments
Assessments performed at baseline and throughout the study included history and physical examination (including weight and height), WHO performance status, and tumor measurement of palpable or visual lesions. Radiological tests of computed tomography (CT) scan, magnetic resonance imaging (MRI), or nuclear medicine scan were used, if necessary, for tumor measurement of lesions not evaluable by other imaging modalities. Chest x-rays were used in patients with chest metastasis. Full blood count (with differential and platelet counts), blood chemistries, electrocardiogram, and vital signs were done for all patients before and at regular intervals during the study. Additionally, coagulation studies were assessed as appropriate and the number of units required for transfusions every 4 weeks.
All patients who received at least 2 cycles of therapy and 1 radiologic evaluation were eligible for the efficacy analysis which was done based on WHO criteria. Complete response (CR) was defined as the disappearance of all known disease, and PR was defined as at least a 50% decrease in the total tumor size of the lesions, both determined by two observations not less than 4 weeks apart, and without appearance of new lesions. Stable disease (SD) was documented if a 50% decrease in total tumor size could not be established, nor a 25% increase in the size of one or more measurable lesions demonstrated. Progressive disease (PD) was defined as an increase ≥25% in measurable or evaluable tumor size and/or the appearance of new tumor sites. The safety analysis was performed on data from all patients who received at least 1 dose of the study drugs. The same assessment method used to determine the disease status at baseline was used consistently throughout the study for efficacy evaluation. Patients were clinically evaluated for response at the start of each cycle. Every 3 cycles during therapy, then every 3 months until disease progression, patients were assessed by radiologic imaging studies and chest radiography. Survival was measured from time of first-dose administration until the date of death. Time to disease progression was calculated from the administration of the first dose until the date of progression. Overall duration of response was measured from the first day of treatment to the date of the first observation of progressive disease. Toxicity ratings, based on WHO criteria, were assessed before the beginning of each next cycle.
Data analysis
The response rate for this study was anticipated to be in the region of 40%. If 35 patients qualified for the efficacy analysis and 14 responses (40%) were observed, then the 95% confidence interval (CI) would be 26%–54%. The evaluability criteria for efficacy analysis were predefined in the protocol as; any patient who received at least 2 cycles of therapy and one radiological evaluation. It was anticipated that approximately 10% of the patients recruited might not qualify for the efficacy analysis. Consequently, this study intended to recruit 39 patients.
For the primary analysis, a 95% CI for the true response rate was calculated using the normal approximation to the binomial distribution. Estimates of duration of response and time to disease progression were calculated using the Kaplan-Meier method, with medians and quartiles derived from these estimates.
Results
Patient characteristics
From October 1999 to December 2001, a total of 40 female patients were entered, and all were enrolled (received study treatment). These patients had a median age of 61.5 years (range, 30–76 years), and the majority (85%) had a WHO performance status of 0 or 1. The baseline patient and disease characteristics are summarized in Table 1. This patient population presented with multiple poor prognostic factors. Overall, 39 (97.5%) patients presented with metastatic disease, and 35 (87.5%) had visceral involvement. Eight (20.0%) patients reported both liver and lung metastases at baseline, and 28 (70.0%) patients had at least 2 sites of metastatic disease at baseline. Three patients had 4 or more tumor burden sites, including 1 patient with 6 sites of metastatic disease. Of the 35 patients with known hormone receptor status, 12 (34.3%) were negative for estrogen/progesterone receptors (ER/PR). In addition, 27 (67.5%) patients had received previous adjuvant chemotherapy, 25 of whom had prior anthracyclines. A total of 24 (60.0%) patients had received both adjuvant chemotherapy and hormonal therapy.
Response and time-to-event measures
Among the 40 patients enrolled, there were 2 CR and 12 PR. Based on an Intent-to-treat analysis, the overall response rate was 35.0%. An additional 14 (35.0%) patients had stable disease, and 7 patients (17.5%) had a best response of progressive disease. Of the remaining 5 patients, 4 did not undergo further evaluation of their lesions after baseline, and 1 patient did not have histologically or cytologically proven breast carcinoma. As per the predefined criteria in the protocol these 5 patients were not eligible for efficacy analysis.
According to per-protocol efficacy analysis, 2 CR and 12 PR, gave an overall response rate of 40.0% (14/35) (95% CI, 23.8%–56.2%). The response rate was lower among 22 patients who had prior anthracycline treatment compared to the remaining 13 with no anthracycline exposure (27.3% vs 61.5%) (Table 2).
At the time of this analysis, the median follow-up time (the period from patient enrollment to last visit) was 13.7 months (range, 1.6–28.3 months). A total of 14.3% (n = 5) of patients were censored for the estimate of duration of response because they were alive and progression-free, or lost to follow-up at the time of the analysis. The estimate of median duration of response was 12.0 months (95% CI, 10.0–15.0 months). The probability of duration of response lasting at least 9 and 12 months was estimated as 78.6% (95% CI, 57.1%–100%) and 47.1% (95% CI, 20.0%–74.3%), respectively. The estimate for median time to progressive disease was 7.2 months (95% CI, 4.6–10.0 months). The progression-free probability at 6 and 9 months was estimated as 62.9% (95% CI, 46.8%–78.9%) and 40.0% (95% CI, 23.8%–56.2%), respectively. A total of 11.4% (n = 4) of patients who were alive and had not progressed, or were lost to follow-up, were censored for this analysis. The most common sites of disease progression were liver and lung, which were the most commonly involved sites at baseline. Fourteen patients had died, and the remaining 60% (n = 21) of patients who were alive or lost to follow-up were censored for the survival analysis. The estimate of median survival was 25.7 months (95% CI, 14.7-xx months). The probability of surviving beyond 12 and 18 months was estimated as 74.0% (95% CI, 59.3%–88.6%) and 62.0% (95% CI, 44.5%–79.5%), respectively.
Dose administration
All 40 patients received at least 1 cycle of therapy, for a total of 220 cycles. The median number of cycles given was 6 (range, 1–10 cycles). Three (7.5%) patients completed the study and received the maximum number of 10 treatment cycles. The most frequently cited reason for discontinuation from the study was progressive disease (14 patients; 35.0%), as indicated by radiologic and/or physical assessment. Five (12.5%) patients discontinued the study because of adverse events, and 1 patient died in cycle 2 due to progression.
Patients received mean doses of 681.0 mg/m2 gemcitabine and 56.7 mg/m2 paclitaxel per week, which were 85.1% and 97.2% of the planned weekly mean doses, respectively. There were 48 (10.9% of doses administered) dose omissions and 1 (0.2%) dose reduction of gemcitabine, and 4 (1.8%) omissions and 4 (1.8%) reductions of paclitaxel. Neutropenia was the most frequent cause of gemcitabine dose omission (35 of 48 doses omitted; 72.9%). All gemcitabine dose omissions due to neutropenia occurred on day 8, and the majority of dose omissions (22 of the 35 doses) occurred during the first 3 cycles. Paresthesia was the cause of 2 paclitaxel dose reductions and all 4 paclitaxel dose omissions. Overall, there were 34 cycle delays of administration of gemcitabine plus paclitaxel. The clinically relevant reasons for cycle delays included hepatitis in 1 patient (delay at cycles 2 and 3), pyrexia in 2 patients, neutropenia in 3 patients, and asthenia, edema, paresthesia, and pleural disorder in 1 patient each.
Toxicity
All 40 patients entered into the study were evaluable for toxicity. The WHO grade 3 and 4 toxicities observed during this study were primarily hematologic (Table 3). These included grade 3 and 4 neutropenia in 17 (42.5%) patients each, grade 3 leukopenia in 19 (47.5%) patients, grade 3 anemia in 2 (5.0%) patients, and grade 3 and 4 thrombocytopenia in 1 (2.5%) patient each. Although 2 patients with grade 4 neutropenia had concurrent fever, neither patient was hospitalized for febrile neutropenia. The patient who reported grade 4 thrombocytopenia required a platelet transfusion. A total of 8 units of red blood cells were given; the recipients included both of the patients with grade 3 anemia.
No grade 4 nonhematologic toxicities were reported during the study. The most common grade 3 toxicity was alopecia in 30 (75.0%) patients. Three (7.5%) patients reported grade 3 peripheral neurotoxicity, which occurred during later cycles (cycles 4 through 8). Grade 2 neurotoxicity was experienced by 12 (30.0%) patients. There was only 1 report of a grade 2 pulmonary toxicity that was described as dyspnea. One patient, who was previously treated in the adjuvant setting with epirubicin, 5-FU, and cyclophosphamide combination therapy and radiotherapy, reported cardiac toxicity. The patient was hospitalized for severe dyspnea and was diagnosed with arrhythmia and tachycardia. The patient also had grade 3 infection described as severe sepsis. The patient discontinued the study due to cardiac rhythmic events.
There were 2 deaths, 1 of which occurred on-study and 1 during the 30-day follow-up period after administration of the last dose of study drug. Both of the deaths were considered related to disease progression. There were no deaths due to study drug toxicity.
Discussion
The results of this phase II trial demonstrated that the combination of gemcitabine 1200 mg/m2 on days 1 and 8, plus paclitaxel 175 mg/m2 on day 1, administered in a 21-day cycle was effective in patients with unresectable, locally recurrent or MBC. In a first-line setting, this regimen resulted in a response rate of 40.0% (35.0% for Intent-to-treat population), with a complete response rate of 5.7%. The response rate in the patients previously exposed to anthracyclines was lower than in those not exposed to anthracyclines. The median duration of response was 12 months. In addition, the median time to progressive disease was 7.2 months, which was higher than that obtained with either single agent. Although the overall survival data are not fully mature, the median survival time of 25.7 months is also quite promising. This is a phase II non-randomized trial, and hence there are limitations in interpreting these results, however the efficacy results compare favorably with those seen with single-agent paclitaxel or single-agent gemcitabine in MBC,[11,14,31,32] and are also comparable with the results of this combination seen in other trials (see Table 4). It is important to note that these efficacy results were achieved in a patient population with multiple poor prognostic factors such as: visceral involvement, at least 2 sites of metastatic disease at baseline, negative hormone receptor status, and prior exposure to chemotherapy, including anthracyclines, in an adjuvant setting.
Since the initiation of this trial, the combination of gemcitabine and paclitaxel has been tested in different doses and schedules, as well as in various patient populations with different levels of exposure to prior chemotherapy. In a phase I dose-finding study,[33] fixed doses of gemcitabine (1000 mg/m2) on days 1 and 8 were administered with escalating doses of paclitaxel (range, 90–270 mg/m2) on day 1 of a 21-day cycle in patients with pretreated MBC or ovarian cancer. At the paclitaxel dose level of 270 mg/m2, the dose-limiting toxicities (DLT) of grade 4 neutropenia and thrombocytopenia were noted, but there were no unexpected toxicities. Among 30 evaluable patients with MBC, 4 CRs (13%) and 12 PRs (40%) were observed, for an overall response rate of 53%. The median duration of response was 7.2 months.
In another study, heavily pretreated MBC patients were treated with 2500 mg/m2 gemcitabine and 135 mg/m2 paclitaxel, both of which were given on days 1 and 15 [29]. The overall response rate achieved was 45%. The median time to progression was 7 months, and overall survival time was 11 months. However, 34% of the patients needed growth factors. In a similar trial, Vici et al [26] used 1500 mg/m2 gemcitabine and 135 mg/m2 paclitaxel on days 1 and 15 every 4 weeks in heavily pretreated advanced breast cancer patients who had received 1 to 4 cycles of prior chemotherapy. Although the doses were lower than those used in the Sanchez-Rovira trial, patients were supported with G-CSF injections. In the preliminary results, the overall response rate in 20 evaluable patients was 45% with a median time to progression of 8 months.
Murad et al [30] evaluated gemcitabine plus paclitaxel in heavily pretreated MBC patients with history of 2 or 3 relapses following treatment with anthracycline-containing regimens. The initial schedule was gemcitabine (1000 mg/m2) on days 1, 8, and 15 plus paclitaxel (175 mg/m2) on day 1, given every 4 weeks. However, due to occurrence of unacceptable toxicity (thrombocytopenia) in the first 5 patients, the schedule was modified to every 3 weeks with gemcitabine given on days 1 and 8. The modified regimen was well tolerated, with a significantly lower incidence of grade 3 or 4 thrombocytopenia (18.5% in the day-28 schedule vs 5.4% in the day-21 schedule), and resulted in an overall response rate of 55%, with 17% complete responses. The median response duration was 8 months, and the median overall survival was 12 months.
The combination of gemcitabine and paclitaxel was also evaluated as first-line treatment of advanced or MBC. Forty-three chemonaive patients with histologically confirmed metastatic breast carcinoma received paclitaxel 150 mg/m2 followed by gemcitabine 2500 mg/m2, both on day 1 of a 2-week cyclem [27]. Among the 38 evaluable patients, the overall response rate was 68%, with moderate neutropenia seen in 32% of patients.
Delfino et al [28] assessed the efficacy and toxicity of the gemcitabine/paclitaxel combination in a first-line setting using a 3-week schedule. Chemonaive patients with advanced or metastatic breast cancer were given paclitaxel (175 mg/m2) on day 1 and gemcitabine (1200 mg/m2) on days 1 and 8 every 3 weeks. The overall response rate was 67%, with 22% complete responders. The median time to tumor progression was 11 months, and the median duration of response was 18 months.
The results of these studies (Table 4), along with those of our current study, clearly establish the efficacy of the gemcitabine and paclitaxel combination in chemonaive and pretreated MBC patients. At present, anthracycline-based combinations are the mainstay of chemotherapy in the early treatment of breast cancer, but their effectiveness decreases in later treatments in the metastatic setting. The cardiotoxicity associated with anthracyclines also limits the total amount of the drug that can be used in a patient. Thus, there is a strong need to develop newer treatment regimens that are not cross-resistant with anthracyclines yet have antitumor activity in MBC. Gemcitabine and paclitaxel are known to possess considerable cytotoxic activity individually with minimally overlapping toxicity profiles. Both drugs act on different cellular targets with indications of noncross-resistance to each other. In pharmacokinetic studies, gemcitabine and paclitaxel did not interfere with each other, and in vitro studies did not demonstrate any synergism between these 2 drugs [23,34]. However, increased accumulation of dFdCTP by paclitaxel might augment the antitumor activity of gemcitabine in clinical studies[23]. Based on this information, it is not unreasonable to expect that the addition of paclitaxel to gemcitabine may produce additional efficacy that is superior to either drug given alone. To explore this possibility further, a phase randomized III trial comparing gemcitabine plus paclitaxel versus paclitaxel alone is currently under way. The efficacy and toxicity results of this study will offer significant insight into the clinical implications of combining these agents in patients with advanced or metastatic breast cancer.
Conclusion
In conclusion, gemcitabine in combination with paclitaxel has demonstrated notable activity along with an acceptable and tolerable safety profile. Further evaluation of this regimen is warranted in the treatment of metastatic breast cancer.
Competing interests
Laurence Bozec was employed by Eli Lilly France as a clinical research physician at the time of the writing of this manuscript. Sulochana Gawande is currently employed by Eli Lilly and Company.
The remaining authors, Djelila Allouache, Michele Tubiana-Hulin, Nicole Tubiana-Mathieu, Sophie Piperno-Neumann, Fawzia Mefti, and Jean-Yves Genot, have no competing interests.
Authors' contributions
Djelila Allouache conducted patient trial, participated in patient enrollment and manuscript review.
Sulochana R. Gawande helped in data analysis, data interpretation and writing the manuscript.
Michele Tubiana-Hulin conducted patient trial, participated in patient enrollment and manuscript review.
Nicole Tubiana-Mathieu conducted patient trial, participated in patient enrollment and manuscript review.
Sophie Piperno-Neumann conducted patient trial, participated in patient enrollment and manuscript review.
Fawzia Mefti conducted patient trial, participated in patient enrollment and manuscript review.
Laurence Bozec helped in conducting the trial, data collection and writing the manuscript.
Jean-Yves Genot conducted patient trial, participated in patient enrollment and manuscript review.
Pre-publication history
The pre-publication history for this paper can be accessed here:
Acknowledgements
This study was funded by Eli Lilly and Company (Study Code B9E-MC-S024). Eli Lilly and Company helped the investigators with study design, analysis and interpretation of data; writing of the manuscript, and in the decision to submit the manuscript for publication. The investigators in this study were responsible for data collection.
Figures and Tables
Table 1 Patient characteristics (N = 40)
Patients entered and enrolled 40
Median age, years (range) 61.5 (30–76)
WHO performance status, n (%)
0 18 (45.0)
1 16 (40.0)
2 6 (15.0)
Histology, n (%)
Breast 1 (2.5)
Ductal breast 31 (77.5)
Lobular breast 8 (20.0)
Number of metastatic sites, n (%)
1 12 (30.0)
2 21 (52.5)
3 4 (10.0)
≥4 3 (7.5)
Metastatic sites of diseasea, n (%)
Visceral 35 (87.5)
Liver 26 (65.0)
Lung (including pleura) 17 (42.5)
Nonvisceral only 5 (12.5)
Menopausal status, n (%)
Pre-menopausal 20 (50.0)
Menopausal 8 (20.0)
Post-menopausal 10 (25.0)
Unknown 2 (5.0)
Estrogen/progesterone receptors, n (%)
Positive 23 (57.5)
Negative 12 (30.0)
Unknown 5 (12.5)
aPatients may be counted in more than one category.
Table 2 Summary of best tumor response according to prior anthracycline exposure
No. of Pts No. of Responders Percent of Pts 95% CI
Evaluable patients 35 14 40.0% (23.8%–56.2%)
Prior anthracycline exposure
Yes 22 6 27.3% (8.7%–45.9%)
No 13 8 61.5% (35.1%–88.0%)
CI = confidence interval.
Table 3 Summary of maximum WHO toxicity grades (N = 40)
Grade 3 Grade 4
Toxicity n (%) n (%)
Hematologic
Anemia 2 5.0 0 0
Leukopenia 19 47.5 0 0
Neutropeniaa 17 42.5 17 42.5
Thrombocytopenia 1 2.5 1 2.5
Non-hematologic
Cardiac function 1 2.5 0 0
Alopecia 30 75.0 0 0
Infection 1 2.5 0 0
Nausea/vomiting 1 2.5 0 0
Peripheral neurotoxicity 3 7.5 0 0
aSegmented neutrophils have been converted to WHO scores using granulocyte count criteria.
Table 4 Phase II studies of gemcitabine/paclitaxel in patients with advanced or metastatic breast cancer
Author Regimen No. of Pts ORR % (CR %) Median TTP (mos) Comments
Vici et al26 G 1500 mg/m2 d1, 15 + T150 mg/m2 d1, 15 q4wks 20 45% (10%) 8 Heavily pretreated pts; supported with G-CSF
Colomer et al27 G 2500 mg/m2 d1 + T150 mg/m2 d1 q2wks 43 68% (21%) 9 Previously treated pts; hematologic toxicity
Delfino et al28 G 1200 mg/m2 d1, 8 + T 175 mg/m2 d1 q3wks 45 67% (22%) 11 Chemonaïve pts; mild hematologic toxicity; manageable toxicity
Sanchez-Rovira et al29 G 2500 mg/m2 d1, 15 + T 135 mg/m2 d1, 15 q4wks 44 45% (7%) 7 Previously treated pts; hematologic and neurologic toxicities
Murad et al30 G 1000 mg/m2 d1, 8 + T 175 mg/m2 d1 q3wks 29 55% (17%) 8 Heavily pretreated pts; 28-day reduced to a 21-day schedule due to hematologic toxicity
Genot et al (current study) G 1200 mg/m2 d1, 8 T 175 mg/m2 d1 q3wks 40 40% (5.7%) 7.2 No prior treatment for metastatic disease, neutropenia prominent, other toxicities mild
ORR = overall response rate; CR = complete response; TTP = time to progression; mos = months; G = gemcitabine; T = paclitaxel.
==== Refs
Jemal A Thomas A Murray T Thun M Cancer statistics, 2002 CA Cancer J Clin 2002 52 23 47 11814064
Mouridsen HT Systemic therapy of advanced breast cancer Drugs 1992 17 28 1283847
O'Shaughnessy J Miles D Vukelja S Moiseyenko V Ayoub JP Cervantes G Fumoleau P Jones S Lui WY Mauriac L Twelves C Van Hazel G Verma S Leonard R Superior survival with capecitabine plus docetaxel combination therapy in anthracycline-pretreated patients with advanced breast cancer: Phase III trial results J Clin Oncol 2002 20 2812 2823 12065558 10.1200/JCO.2002.09.002
Fossati R Confalonieri C Torri V Ghislandi E Penna A Pistotti V Tinazzi A Liberati A Cytotoxic and hormonal treatment for metastatic breast cancer: A systematic review of published randomized trials involving 31,510 women J Clin Oncol 1998 16 3439 3460 9779724
Rowinsky EK Cazenave LA Donehower RC Taxol: A novel investigational antimicrotubule agent J Natl Cancer Inst 1990 82 1247 1259 1973737
Kuhn JG Pharmacology and pharmacokinetics of paclitaxel Ann Pharmacother 1994 28 S15 S17 7915153
Holmes FA Walters RS Theriault RL Forman AD Newton LK Raber MN Buzdar AU Frye DK Hortobagyi GN Phase II trial of taxol, an active drug in the treatment of metastatic breast cancer J Natl Cancer Inst 1991 83 1797 1805 1683908
Holmes FA Valero V Walters RS Theriault RL Booser DJ Fraschini G Buzdar AU Frye D Gibbs HR Hortobagyi GN The M.D. Anderson Cancer Center experience with Taxol in metastatic breast cancer J Natl Cancer Inst Monogr 1993 15 161 169 7912522
Seidman AD Reichman BS Crown JP Yao TJ Currie V Hakes TB Hudis CA Gilewski TA Baselga J Forsythe P Paclitaxel as second and subsequent therapy for metastatic breast cancer: Activity independent of prior anthracycline response J Clin Oncol 1995 13 1152 1159 7537798
Abrams JS Vena DA Baltz J Adams J Montello M Christian M Onetto N Desmond-Hellmann S Canetta R Friedman MA Paclitaxel activity in heavily pretreated breast cancer: A National Cancer Institute Treatment Referral Center trial J Clin Oncol 1995 13 2056 2065 7543562
Nabholtz JM Gelmon K Bontenbal M Spielmann M Catimel G Conte P Klaassen U Namer M Bonneterre J Fumoleau P Winograd B Multicenter, randomized comparative study of two doses of paclitaxel in patients with metastatic breast cancer J Clin Oncol 1996 14 1858 1867 8656254
Perez EA Vogel CL Irwin DH Kirshner JJ Patel R Multicenter phase II trial of weekly paclitaxel in women with metastatic breast cancer J Clin Oncol 2001 19 4216 4223 11709565
Hertel LW Kroin JS Misner JW Tustin JM Synthesis of 2-Deoxy-2,2-difluoro-D-ribose and 2-Deoxy-2,2-difluoro-D-ribofuranosyl Nucleosides J Org Chem 1988 53 2406 2409 10.1021/jo00246a002
Blackstein M Vogel CL Ambinder R Cowan J Iglesias J Melemed A Gemcitabine as first-line therapy in patients with metastatic breast cancer: A phase II trial Oncology 2002 62 2 8 11810037 10.1159/000048240
Possinger K Kaufmann M Coleman R Grosse Y Sezer O Mergenthaler HG Possinger K Phase II study of gemcitabine as first-line chemotherapy in patients with advanced or metastatic breast cancer Anticancer Drugs 1999 10 155 162 10211545
Carmichael J Possinger K Phillip P Beykirch M Kerr H Walling J Harris AL Advanced breast cancer: A phase II trial with gemcitabine J Clin Oncol 1995 13 2731 2736 7595731
Spielmann M Llombart-Cussac A Kalla S Espie M Namer M Ferrero JM Dieras V Fumoleau P Cuvier C Perrocheau G Ponzio A Kayitalire L Pouillart P Single-agent gemcitabine is active in previously treated metastatic breast cancer Oncology 2001 60 303 307 11408796 10.1159/000058524
Brodowicz T Kostler WJ Moslinger R Tomek S Vaclavik I Herscovici V Wiltschke C Steger GG Wein W Seifert M Kubista E Zielinski CC Single-agent gemcitabine as second- and third-line treatment in metastatic breast cancer The Breast 2000 9 338 342 14965758 10.1054/brst.2000.0170
Schmid P Akrivakis K Flath B Grosse Y Sezer O Mergenthaler HG Possinger K Phase II trial of gemcitabine as prolonged infusion in metastatic breast cancer Anticancer Drugs 1999 10 625 631 10507311
Smorenburg CH Bontenbal M Seynaeve C van Zuylen C de Heus G Verweij J de Wit R Phase II study of weekly gemcitabine in patients with metastatic breast cancer relapsing or failing both an anthracycline and a taxane Breast Cancer Res Treat 2001 66 83 87 11368414 10.1023/A:1010679127390
Gazitt Y Rothenberg ML Hilsenbeck SG Fey V Thomas C Montegomrey W Bcl-2 overexpression is associated with resistance to paclitaxel, but not gemcitabine, in multiple myeloma cells Int J Oncol 1998 13 839 848 9735415
Jensen PB Holm B Sorensen M Christensen IJ Sehested M In vitro cross-resistance and collateral sensitivity in seven resistant small-cell lung cancer cell lines: Preclinical identification of suitable drug partners to taxotere, topotecan and gemcitabine[sic] Br J Cancer 1997 75 869 877 9062409
Kroep JR Giaccone G Voorn DA Smit EF Beijnen JH Rosing H van Moorsel CJ van Groeningen CJ Postmus PE Pinedo HM Peters GJ Gemcitabine and paclitaxel: Pharmacokinetic and pharmacodynamic interactions in patients with non-small-cell lung cancer J Clin Oncol 1999 17 2190 2197 10561275
Kroep JR Giaccone G Tolis C Voorn DA Loves WJ Groeningen CJ Pinedo HM Peters GJ Sequence dependent effect of paclitaxel on gemcitabine metabolism in relation to cell cycle and cytotoxicity in non-small-cell lung cancer cell lines Br J Cancer 2000 83 1069 1076 10993656 10.1054/bjoc.2000.1399
Jungnelius JU Ridderheim M Perren T Gawande S Roychowdhury D Poole C Optimal sequence leads to improved dose delivery for gemcitabine (G)/paclitaxel (P) combination in recurrent ovarian cancer: a phase I study [abstract] Proc Am Soc Clin Oncol 2002 21 223a
Vici P Foggi P Capomolla E Santiccioli S Cauchi C X Giacinti L Lopez M Biweekly paclitaxel/gemcitabine (P/G) as salvage treatment in breast cancer patients (pts): Preliminary results [abstract] Proc Am Soc Clin Oncol 2002 21 61b
Colomer R Llombart A Lluch A Ojeda B Barnadas A Caranana V Fernandez Y De Paz L Guillem V Alonso S Paclitaxel/gemcitabine administered every two weeks in advanced breast cancer: preliminary results of a phase II trial Semin Oncol 2000 20 24 10697032
Delfino C Caccia G Gonzales LR Mickiewicz E Rodger J Balbiani L Morales DF Comba AZ Brosio C Gemcitabine plus paclitaxel as first-line chemotherapy for patients with advanced breast cancer Oncology 2004 66 18 23 15031594 10.1159/000076330
Sanchez-Rovira P Medina MB Mohedano N Jaen A Porras I González E Fernández M Lozano A Results from a phase II study of Gemcitabine in combination with paclitaxel in metastatic breast cancer [abstract] Ann Oncol 1998 16
Murad AM Guimaraes RC Aragao BC Scalabrini-Neto AO Rodrigues VH Garcia R Phase II trial of the use of paclitaxel and gemcitabine as a salvage treatment in metastatic breast cancer Am J Clin Oncol 2001 24 264 268 11404498 10.1097/00000421-200106000-00011
Winer EP Berry DA Woolf S Duggan D Kornblith A Harris LN Michaelson RA Kirshner JA Fleming GF Perry MC Graham ML Sharp SA Keresztes R Henderson JC Hudis C Muss H Norton L Failure of higher-dose paclitaxel to improve outcome in patients with metastatic breast cancer: Cancer and Leukemia Group B trial 9342 J Clin Oncol 2004 22 2061 2068 15169793 10.1200/JCO.2004.08.048
Spielmann M Llombart-Cussac A Kalla S Espie M Namer M Ferrero JM Dieras V Fumoleau P Cuvier C Perrocheau G Ponzio A Kayitalire L Pouillart P Single-agent gemcitabine is active in previously treated metastatic breast cancer Oncology 2001 60 303 307 11408796 10.1159/000058524
Iaffaioli RV Tortoriello A Santangelo M Turitto G Libutti M Benassai G Frattolillo A Ciccarelli PD De Rosa P Crovella F Carbone I Barbarisi A Phase I dose escalation study of gemcitabine and paclitaxel plus colony-stimulating factors in previously treated patients with advanced breast and ovarian cancer Clin Oncol 2000 12 251 255
De Pas T de Braud F Danesi R Sessa C Catania C Curigliano G Fogli S del Tacca M Zampino G Sbanotto A Rocca A Cinieri S Marrocco E Milani A Goldhirsch A Phase I and pharmacologic study of weekly gemcitabine and paclitaxel in chemo-naïve patients with advanced non-small-cell lung cancer Ann Oncol 2000 11 821 827 10997809 10.1023/A:1008319923516
|
16316459
|
PMC1315334
|
CC BY
|
2021-01-04 16:03:08
|
no
|
BMC Cancer. 2005 Nov 29; 5:151
|
utf-8
|
BMC Cancer
| 2,005 |
10.1186/1471-2407-5-151
|
oa_comm
|
==== Front
Comp HepatolComparative Hepatology1476-5926BioMed Central London 1476-5926-4-71633664910.1186/1476-5926-4-7ResearchRegenerative and fibrotic pathways in canine hepatic portosystemic shunt and portal vein hypoplasia, new models for clinical hepatocyte growth factor treatment Spee Bart [email protected] Louis C [email protected] den Ingh Ted SGAM [email protected] Brigitte [email protected] Jooske [email protected] Sluijs Frederik J [email protected] Jan [email protected] Department of Clinical Sciences of Companion Animals, Faculty of Veterinary Medicine, Utrecht University, Utrecht, The Netherlands2 Department of Pathobiology, Faculty of Veterinary Medicine, Utrecht University, Utrecht, The Netherlands2005 7 12 2005 4 7 7 10 2 2005 7 12 2005 Copyright © 2005 Spee et al; licensee BioMed Central Ltd.2005Spee 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
We analyzed two spontaneous dog diseases characterized by subnormal portal perfusion and reduced liver growth: (i) congenital portosystemic shunts (CPSS) without fibrosis and (ii) primary portal vein hypoplasia (PPVH), a disease associated with fibrosis. These pathologies, that lack inflammation or cholestasis, may represent simplified models to study liver growth and fibrosis. To investigate the possible use of those models for hepatocyte growth factor (HGF) treatment, we studied the functionality of HGF signaling in CPSS and PPVH dogs and compared this to aged-matched healthy controls.
Results
We used quantitative real-time polymerase chain reaction (Q-PCR) to analyze the mRNA expression of HGF, transforming growth factor β1 (TGF-β1), and relevant mediators in liver biopsies from cases with CPSS or PPVH, in comparison with healthy control dogs. CPSS and PPVH were associated with a decrease in mRNA expression of HGF and of MET proto-oncogene (c-MET). Western blot analysis confirmed the Q-PCR results and showed that intracellular signaling components (protein kinase B/Akt, ERK1/2, and STAT3) were functional. The TGF-β1 mRNA levels were unchanged in CPSS whereas there was a 2-fold increase in PPVH indicating an active TGF-β1 pathway, consistent with the observation of fibrosis seen in PPVH. Western blots on TGF-β1 and phosphorylated Smad2 confirmed an activated pro-fibrotic pathway in PPVH. Furthermore, Q-PCR showed an increase in the amount of collagen I present in PPVH compared to CPSS and control, which was confirmed by Western blot analysis.
Conclusion
The pathophysiological differences between CPSS and PPVH can adequately be explained by the Q-PCR measurements and Western blots. Although c-MET levels were reduced, downstream signaling seemed to be functional and provides a rational for HGF-supplementation in controlled studies with CPSS and PPVH. Furthermore both diseases may serve as simplified models for comparison with more complex chronic inflammatory diseases and cirrhosis.
==== Body
Background
Chronic liver disease is characterized by decreased regeneration of hepatocytes and increased formation of fibrous tissue. These characteristics may be the sequel of various chronic processes such as cholestasis, viral infections, toxin exposure, and metabolic disorders. Dogs have complex liver diseases such as hepatitis and cirrhosis which are highly comparable with the human counterparts. Moreover, coding sequences of dogs proved highly homologous to the human sequences [1], especially compared to the rodent genome. Thus, dogs may fulfill a role as a spontaneous animal model in between toxin-induced or surgical models in rodents, and spontaneous diseases in man. The complex interplay of many factors active in chronic liver disease makes it difficult to unravel the roles of different individual pathogenetic pathways. Dogs display liver diseases, which are potentially valuable models to compare complex with simple pathologic entities.
We have chosen these two congenital dog diseases for comparative analysis of liver growth/regeneration, fibrosis, and hepatic homeostasis: congenital portosystemic shunt (CPSS) and primary portal vein hypoplasia (PPVH). CPSS is characterized by an abnormal single large communication between the portal vein and a major systemic vein (cava or azygos). This results in the virtual absence of portal vein perfusion to the liver from birth onwards. Liver growth remains nearly absent but there is essentially no liver pathology [2,3]. PPVH is a developmental abnormality in which the terminal vein branches are not or only partially present and, in most cases, in combination with congenital portal fibrosis, but without inflammation [4]. PPVH is associated with portal hypertension and reduced liver growth. Thus, these two congenital diseases represent relatively simple models for reduced liver growth associated with fibrosis (PPVH) or without fibrosis (CPSS). Both diseases have a decrease in liver growth due to differences in portal perfusion which results in a massive reduction of liver size.
Because hepatocyte growth factor (HGF) is one of the most important genes involved in liver growth/regeneration [5-7], abnormal expression of HGF could play a major role in the decreased liver size in CPSS or PPVH. Therefore, treatment of dogs with HGF could be a possible therapeutic approach. A pre-requisite for treatment is that HGF signaling components are unaffected in those dogs. Consequently, we focused on measuring gene products involved in signaling of HGF and counteracting transforming growth factor β1 (TGF-β1). All biological responses induced by HGF are elicited by binding to its receptor, a transmembrane tyrosine kinase encoded by the MET proto-oncogene (c-MET). The signaling cascade triggered by HGF begins with phosphorylation of the receptor and is mediated by concomitant activation of different cytoplasmic effectors that bind to the same multifunctional binding site. The c-MET mediated response includes two key pathways involved in cell survival and mitogenesis [8]. The first; protein kinase B (PKB/Akt) is activated by phosphoinositide 3-kinase (PI3K) and elicits cell survival [9,10]. The second; ERK1/2 (also known as p42/44 MAPK), a member of the mitogen-activated protein (MAP) kinase family, is activated by the RAS-RAF-MEK pathway and is responsible for mitogenesis [11]. A third response of HGF is the branching morphogenesis which next to the PKB and ERK pathways requires involvement of the signal transducer and activator of transcription (STAT) 3 pathway [12].
It is well established that an increase of TGF-β1 in liver promotes the formation of extracellular matrix (ECM) components and suppresses hepatocyte proliferation [13,14]. Prolonged overexpression of TGF-β1 in non-parenchymal cells causes hepatic fibrosis in humans and experimental animals. In several fibrosis models, fibrotic lesions are associated with an increase in collagens and TGF-β1 mRNAs [15]. The intracellular pathway that is activated by TGF-β1 receptors is mediated by Smads. Smad2 is activated via carboxy-terminal phosphorylation by TGF-β1 type I receptor kinases. When bound with co-Smads, they act as TGF-β1-induced transcriptional activators of target genes [16].
Cell homeostasis is the result of balance between cell death, cell proliferation, and growth-arrest. Therefore we investigated expression levels of pro-apoptotic Fas ligand and caspase-3, anti-apoptotic Bcl-2 [17], cell-cycle stimulating TGFα, and cell-cycle inhibitor p27kip. All of these gene-products are regulated directly or indirectly by PKB [9].
The present study was designed to describe the differential gene-expression of the above indicated crucial pathways involved in growth/regeneration, fibrosis, and cellular homeostasis in liver tissues of dogs with CPSS (reduced growth/regeneration without fibrosis) and PPVH (reduced growth/regeneration and fibrosis) in comparison with healthy animals. These simple congenital dog models may be used to unravel the roles of different gene products in those pathways. These well-defined large animal models are intended to serve as the first spontaneous liver diseases to investigate novel regenerative/anti-fibrotic therapies, such as HGF treatment. This study may also serve as a basis for future comparison with more complex diseases like chronic hepatitis and cirrhosis.
Results
Histological grading of fibrosis
No fibrosis was seen in liver biopsies of CPSS dogs. In the PPVH dogs histological examination revealed slight portal fibrosis in one dog, slight to moderate portal fibrosis associated with slight to moderate centrolobular fibrosis in four dogs, and marked portal fibrosis with biliary proliferation in three dogs. The control dogs showed a normal liver without fibrosis. Examples of histological examination of CPSS and PPVH are included as Figures 1A and 1B, respectively.
Figure 1 Histological grading of fibrosis. (A) CPSS, Portal area without recognizable portal vein and arteriolar proliferation. Van Gieson stain. (B) PPVH, Markedly enlarged portal area with fibrosis and extensive arteriolar and ductular proliferation. Van Gieson stain.
HGF/c-MET signaling pathway involved in regeneration and growth
One of the main in vivo events during regeneration and growth is the signaling via phosphorylation of the HGF receptor c-MET. Q-PCR analysis revealed that HGF mRNA levels in both CPSS and PPVH were decreased three-fold in comparison with healthy dogs (Figure 2). Moreover, the c-MET levels in CPSS and PPVH were significantly decreased (two- and three-fold, respectively). The levels of the mRNAs for TGFα (proliferation) were decreased six-fold in both CPSS and PPVH. The serine-protease HGF activator mRNA was doubled in dogs with CPSS. In contrast, it was halved in dogs with PPVH. Although not significantly in dogs with CPSS, the cell-cycle inhibitor p27kip mRNA was decreased in both conditions.
Figure 2 Quantitative real-time PCR of genes involved in regeneration and growth. Representative data of mRNA levels of congenital portosystemic shunt (CPSS, n = 11 dogs) is shown in (A). Representative data of mRNA levels of primary portal vein hypoplasia (PPVH, n = 8 dogs) is shown in (B). Data represent mean ± 2SD.
TGF-β1 cascade signaling pathway involved in fibrosis
The fibrosis signaling pathway is activated through bindings of the active TGF-β1 dimer to the heteromeric type-I and type-II serine/threonine receptor kinases. As shown in Figure 3, TGF-β1 mRNA levels were increased two-fold in dogs with PPVH, whereas the levels in dogs with CPSS were not changed significantly. The receptor type-I, was induced in both liver diseases but only significantly in PPVH. Receptor type-II was increased in both CPSS and PPVH (4- and 5-fold, respectively), indicating an increased binding capacity. One of the proteolytic enzymes involved in activation of TGF-β1 is urokinase plasminogen activator (uPA). The uPA mRNA level was decreased two-fold in dogs with CPSS and, in contrast, doubled in dogs with PPVH.
Figure 3 Quantitative real-time PCR of genes involved in fibrosis. Representative data of mRNA levels of congenital portosystemic shunt (CPSS, n = 11 dogs) is shown in (A). Representative data of mRNA levels of primary portal vein hypoplasia (PPVH, n = 8 dogs) is shown in (B). Data represent mean ± 2SD.
Gene-expression of apoptosis-related signaling proteins and hypoxia induced factor
We measured three well-known basic apoptotic components of which two are pro-apoptotic (caspase-3 and Fas ligand) and one is anti-apoptotic (Bcl-2). Figure 4 shows that pro-apoptotic mediator Fas ligand was severely inhibited in both dogs with CPSS and in dogs with PPVH (14- and 8-fold, respectively). Moreover, caspase-3 was halved in both CPSS and PPVH. On the other hand, no induction of the anti-apoptotic Bcl-2 was seen in dogs with CPSS, whereas Bcl-2 in dogs with PPVH was doubled. The mechanisms underlying progressive fibrosis are unknown, but fibrosis and hypoxia could have been a fibrogenic stimulus. Hypoxia coordinately up-regulates matrix production and hypoxia induced factor 1 alpha (HIF1α) [18]. These direct hypoxic effects on the expression of genes involved in fibrogenesis was shown in our dogs with PPVH which indeed had elevated levels of HIF1α.
Figure 4 Quantitative real-time PCR of apoptosis genes and a hypoxia related gene. Representative data of mRNA levels of congenital portosystemic shunt (CPSS, n = 11 dogs) is shown in (A). Representative data of mRNA levels of primary portal vein hypoplasia (PPVH, n = 8 dogs) is shown in (B). Data represent mean ± 2SD.
Gene-expression of extracellular matrix gene products
The analysis of ECM expression was performed on three collagens (I, III and IV) and one glycoprotein (fibronectin). Interstitial collagens types I and III are the most commonly found collagens, collagen type IV is a basal membrane collagen. In Figure 5, collagen I was shown to be significantly increased in PPVH (two-fold), whereas CPSS was unchanged. Collagen III and IV were not significantly changed in both groups. Fibronectin showed to be halved in the CPSS group where PPVH remained normal.
Figure 5 Quantitative real-time PCR of extracellular matrix gene products. Representative data of mRNA levels of congenital portosystemic shunt (CPSS, n = 11 dogs) is shown in (A). Representative data of mRNA levels of primary portal vein hypoplasia (PPVH, n = 8 dogs) is shown in (B). Data represent mean ± 2SD.
Western blot analysis of HGF, c-MET, PKB, STAT3, ERK, TGF-β1, Smad2, Collagen I, and Caspase-3
PKB plays a pivotal role in liver regeneration and growth upon activation of the c-MET-HGF signaling pathway [10]. Western blot analysis of HGF showed an immunoreactive band at 82 kDa with no apparent quantitative differences (Figure 6A). Non-phosphorylated c-MET was detected in all samples, where it was present as an immunoreactive band of 145 kDa. Results showed a decrease in the amount of c-MET in both diseases. On the other hand, the anti-phosphorylated c-MET antibody showed an immunoreactive band in all samples with no apparent quantitative differences. Non-phosphorylated PKB was detected in all samples, where it was present as a single band of 60 kDa. The anti-phosphorylated PKB antibody showed an immunoreactive band in all samples. Two immunoreactive bands at 42 and 44 kDa representing the MAP kinase ERK1/2 showed to be equally present at the protein level between the diseased groups and healthy controls. Interestingly, this also applied for the phosphorylated form where no apparent quantitative differences were found. The 80 kDa STAT3 protein showed a similar result with no apparent quantitative differences in the non-phosphorylated form; however, the STAT3 protein seemed to be somewhat less phosphorylated at the serine 727 residue in the PPVH group. TGF-β1 exerts its actions through complex intracellular signaling pathways. All downstream signaling routes following binding of an active TGF-β1 to its receptors type-I and II elicit phosphorylation of Smad2. TGF-β1 was seen in all diseases as a single band of 25 kDa under non-denaturing conditions (Figure 6B). Interestingly, the amount of TGF-β1 was induced in PPVH compared to CPSS and controls. Non-phosphorylated Smad2 was detected in all samples, where it was present as a single band of 58 kDa, with no apparent changes in quantity. Also interestingly, the anti-phosphorylated Smad2 antibody showed a slight band in CPSS whereas in PPVH a phosphorylated Smad2 is clearly present. Moreover, anti-collagen I showed an increase in the amount of protein in PPVH compared to CPSS and healthy controls, all together emphasizing the differences in fibrosis between CPSS and PPVH. Although reduced in the CPSS and PPVH group, inactive or uncleaved caspase-3 was detected in all samples (Figure 6C), where it was present as a single band of 34 kDa. Finally, the processed forms of 20 and 13 kDa showed to be increased in CPSS and PPVH towards healthy controls.
Figure 6 Western blot analysis of liver homogenates of controls, CPSS, and PPVH. Detection of HGF, c-MET, PKB, STAT, and ERK shown in (A), detection of the TGF-β1, Smad2, and Collagen I in (B), and detection of the Caspase-3protein, uncleaved/inactive 34 kDa, and cleaved/active products of 20 kDa and 13 kDa in (C). Western blot analysis of liver homogenates (n = 6 dogs per group, randomly chosen from original group). Lane samples: 1 = control; 2 = congenital portosystemic shunt; 3 = primary portal vein hypoplasia.
Discussion
In order to analyze the possibility of growth factor therapy, two congenital canine liver diseases were molecularly dissected. The expression of a total of 17 gene products involved in liver growth/regeneration, fibrosis, ECM, and cellular homeostasis was measured and normalized to the average amount of two reference genes (Q-PCR). Western blot analysis confirmed the quantitative mRNA results and, furthermore, showed activated pathways. These two independent techniques provided insight into the effects of portal venous hypoperfusion in two canine hepatic diseases; congenital portosystemic shunt (CPSS) without fibrosis and primary portal vein hypoplasia (PPVH) with fibrosis. Taken together, the obtained data provided insights in the feasibility for HGF-treatment.
The normalization performed in this study was obtained by averaging the amount of two different reference genes, glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and hypoxanthine phosphoribosyl transferase (HPRT). No samples were more than 5 percent apart from the individual measured reference genes levels (data not shown). This normalization strategy, using the average amount of two reference genes, is taken as a prerequisite for accurate Q-PCR expression profiling which enables us to measure small expression differences and allows the study of their biological relevance [19].
It is well known that HGF plays an essential role in development [20] and regeneration of the liver, and increases hepatocyte viability. The found decrease in gene-expression of both HGF and its receptor agrees with the reduced liver size in these canine disorders. However, and in contrast to the c-MET levels which correlate nicely with the found protein levels, the amount of HGF mRNA does not seem to reflect protein levels. This can be contributed to HGF which can be a paracrine but also an endocrine factor. Extra-hepatic HGF could have been present in the pancreas or intestinal tract [21].
Although HGF and c-MET mRNA levels were decreased, downstream targets of this tyrosine cascade signaling pathway were still active. Downstream targets, such as Fas ligand and p27kip, were chosen as direct or indirect targets of the HGF-cMET-PI3K-PKB axis. Fas ligand transcription is regulated by FOXO (forkhead box, sub-group "O" transcription factors). Therefore, the decrease in Fas ligand can be explained by an active PKB which directly phosphorylates FOXO [22]. A similar result can be seen in the reduced levels of p27kip mRNA, as this is down-regulated at the gene-transcription level by active PKB [23]. Combined, this indicates that PKB is active in both diseases, which was confirmed by Western blot analysis. It remains to be seen whether other receptor tyrosine kinases (e.g., EGF receptor or insulin receptor) activate this pathway in these dogs [24]. Next to the activated PKB pathway, we have analyzed other c-MET mediated responses in CPSS and PPVH. ERK1/2 showed to be activated in both diseases to a similar level as the healthy controls. The significance of the slightly reduced phosphorylated STAT3 in PPVH, which is phosphorylated by HGF on serine 727 [25], needs to be further investigated. Taken together, the pathways which elicit all major biological functions of c-MET showed to be active in CPSS and PPVH.
Prolonged or overexpression of TGF-β1 acts to suppress cell proliferation, and induces a deposition of ECM proteins, resulting in fibrosis in major organs such as liver [26,27]. We showed that in PPVH the TGF-β1 pathway through Smad2 is activated, consistent with the fibrosis seen in PPVH. Measurements on fibrosis related gene products revealed no elevated activity of the TGF-β1 pathway in CPSS. Gene expression levels related to the TGF-β1 pathway, including its receptors, and the proteolytic activator of TGF-β1 (uPA) were elevated in PPVH, thus indicating an active Smad pathway that could subsequently lead to fibrosis. Western blot analysis confirmed found TGF-β1 levels. Measurements on collagen gene-expression, especially collagen I, confirm the current paradigm of TGF-β1 signaling in fibrous tissues like PPVH [28]. Contrary, non-fibrotic CPSS did not show any alterations in collagen expression. The observation of phosphorylated Smad2 in healthy liver tissue showed that the phosphorylation of Smad2 is a dynamic process and has already been described in other publications [29,30].
The expressions of the pro-apoptotic genes Fas ligand and caspase-3 were clearly decreased. Bcl-2 gene-expression was elevated two times in PPVH; but not in CPSS (Figure 4). Western blot analysis showed that the unprocessed form of caspase-3 was present in lesser amount in CPSS and PPVH; however, the amount of processed or active bands compared to healthy control was higher in the diseases compared to healthy controls. This indicates that although the total amount of caspase-3 is lower, there is more cleavage of the caspase-3 to its active forms in the diseases, possibly leading to an increase in apoptosis.
Both HGF and TGF-β1 need extracellular processing to become biologically active. The serine protease HGF activator is responsible for activation of proHGF [31]. Our studies revealed that HGF activator gene-expression was doubled in dogs with CPSS and halved in case of PPVH. This indicated an increased HGF activation in CPSS. Although levels of HGF activator were reduced in PPVH, this does not necessarily indicate a lack of extracellular processing of HGF. Interestingly uPA, the activator of TGF-β1, was expressed at an increased level in dogs with PPVH. This may, via active TGF-β1-receptor interaction, indicate an activation of Smads and thus the formation of collagens.
Differential gene expression measurements on hepatic diseases have been performed in the past; nevertheless, little is known about levels of genes that play an important role in fibrosis. There have been measurements on cirrhosis in man and rat that indicate an up or down-regulated expression of several proteins [32]. Although these results might be significant in severe forms of fibrosis, these data depict an end-point of the disease whereas earlier stages may be more informative.
Regeneration with recombinant HGF has been achieved in rodent models of liver failure [33,34]. Moreover, besides its regenerative capacity, HGF is known to have an antifibrogenic effect [35,36] and thus reduces or prevent fibrosis in PPVH. TGF-β1 intervention to halt the progression of liver fibrosis and positively effect regeneration, has been applied successfully [37] even in cirrhosis [38]. The measured gene products involved in fibrosis in PPVH make it a good spontaneous animal model to investigate new therapeutic strategies to influence the HGF and/or TGF-β1 pathways in vivo. Furthermore, most fibrogenic models are induced by toxins such as dimethylnitrosamine (DMN), CCl4, or thioacetamide [39]. The canine PPVH model is not drug-induced; therefore, may be better to compare with human diseases and thus fill the gap between induced rodent models and human diseases.
This study is the first to measure expression profiles of crucial pathways of liver growth/regeneration, fibrosis, and hepatic homeostasis in spontaneous canine liver diseases. The present findings in two diseases with relatively simple pathogenesis may also serve as basis for evaluation of more complex diseases like hepatitis and cirrhosis. Evaluation of such complex diseases in dogs is highly suitable for comparative studies on the roles of different pathways in the pathogenesis of liver diseases in man. Two further conclusions can be deduced from the data presented here. First, the pathophysiological differences between CPSS and PPVH can nicely be explained by the Q-PCR measurements and Western blots. Second, although c-MET levels were reduced, downstream signaling seemed to be functional and provides a rational background to design controlled studies for HGF-supplementation in CPSS and PPVH.
Methods
Animals
All samples are obtained from different dog breeds appearing in the clinic with spontaneous diseases. Samples were randomly chosen and aimed to encompass different dog-breeds and both sexes in each group. The procedures were approved by the Ethical Committee as required under Dutch legislation.
Groups
The congenital portosystemic shunt (n = 11 dogs) and primary portal vein hypoplasia group (n = 8 dogs) were compared with a group of healthy dogs (n = 11 dogs). The inclusion criteria for CPSS were increased fasting plasma ammonia concentration, abnormal ammonia tolerance test (peak ammonia ≥ 150 μmol/l plasma) and ultrasonographic visualization of a small liver and a congenital portosystemic shunt with a diameter as wide as the portal vein trunk. The presence of the shunt was further confirmed with surgery, during which a wedge liver biopsy was taken and immediately put in liquid nitrogen and stored at -70°C, until analysis. In CPSS there is no portal hypertension. The inclusion criteria for PPVH were the visualization of a small liver with ultrasonography, presence of multiple small acquired portosystemic collaterals due to portal hypertension, and an abnormal ammonia tolerance test (peak ammonia ≥ 150 μmol/l plasma). Liver tissue of dogs with PPVH was obtained under local anaesthesia by ultrasound-guided biopsy with a true cut 16G biopsy needle. Two biopsies were immediately immersed in liquid nitrogen, and stored at -70°C, until analysis. The healthy control dogs were age-matched, and had AP, ALT, and fasting bile acids in plasma within the reference range. Ultrasonographically the control dog livers had a normal size, shape, and structure, and there were no histological abnormalities in stained histological sections.
Histological grading of fibrosis
Liver samples were fixed in 10% buffered formalin and routinely embedded in paraffin. Sections (4 μm) were stained with haematoxylin-eosin, the Van Gieson stain, and the reticulin stain according to Gordon and Sweet. Histologically, the presence of fibrosis was evaluated semi-quantitatively (absent, slight, moderate, or marked) as well as with respect to its localization. Fibrosis scoring was performed according to Scheuer, a defined scoring method for fibrosis in hepatitis. The slides were independently examined by one certified veterinary pathologist.
RNA isolation and reverse-transcription polymerase chain reaction
Total cellular RNA was isolated from each frozen canine liver tissue in duplicate, using the RNeasy Mini Kit (Qiagen, Leusden, The Netherlands) according to the manufacturer's instructions. The RNA samples were treated with Dnase-I (Qiagen Rnase-free DNase kit). In total 3 μg of RNA was incubated with poly(dT) primers at 42°C for 45 min, in a 60 μl reaction volume, using the Reverse Transcription System from Promega (Promega Benelux, Leiden, The Netherlands).
Quantitative measurements of the mRNA levels of HGF, TGF-β1, and other related signaling molecules
Q-PCR based on the high affinity double-stranded DNA-binding dye SYBR® green I (BMA, Rockland, ME) was performed in triplicate in a spectrofluorimetric thermal iCycler® (BioRad, Veenendaal, The Netherlands). Data were collected and analyzed with the provided application software. For each Q-PCR, 2 μl (of the 2 times diluted stock) of cDNA was used in a reaction volume of 50 μl containing 1× manufacturer's buffer, 2 mM MgCl2, 0.5 × SYBR® green I, 200 μM dNTP's, 20 pmol of both primers, 1.25 units of AmpliTaq Gold (Applied Biosystems, Nieuwerkerk a/d IJssel, The Netherlands), on 96-well iCycler iQ plates (BioRad). Primer pairs, depicted in Table 1, were designed using PrimerSelect software (DNASTAR Inc., Madison, WI). All PCR protocols included a 5-minute polymerase activation step and continued for 40 cycles at 95°C denaturation for 20 sec, annealing for 30 sec and elongation at 72°C for 30 sec with a final extension for 5 min at 72°C. Annealing temperatures were optimized at various levels ranging from 56°C till 67°C (Table 1). Melt curves (iCycler, BioRad), agarose gel electrophoresis, and standard sequencing procedures were used to examine each sample for purity and specificity (ABI PRISM 3100 Genetic Analyser, Applied Biosystems). Standard curves constructed by plotting the relative starting amount versus threshold cycles were generated using serial 4-fold dilutions of pooled cDNA fractions from both healthy and diseased liver tissues. The amplification efficiency, E (%) = (10(1/-s)-1) * 100 (s = slope), of each standard curve was determined and appeared to be > 95%, and < 105%, over a wide dynamic range. For each experimental sample, the amount of the gene of interest, and of the endogenous references glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and hypoxanthine phosphoribosyl transferase (HPRT) were determined from the appropriate standard curve in autonomous experiments. If relative amounts of GAPDH and HPRT were constant for a sample, data were considered valid and the average amount was included in the study (data not shown). Results were normalized according to the average amount of the endogenous references. The normalized values were divided by the normalized values of the calibrator (healthy group) to generate relative expression levels [40].
Table 1 Nucleotide Sequences of Dog-Specific Primers for Real-Time Q-PCR.
Gene Primer Sequence (5'-3') °C Product size (bp) Accession number
GAPDH Forward TGT CCC CAC CCC CAA TGT ATC 58 100 AB038240
Reversed CTC CGA TGC CTG CTT CAC TAC CTT
HPRT Forward AGC TTG CTG GTG AAA AGG AC 56 100 L77488/L77489
Reversed TTA TAG TCA AGG GCA TAT CC
HGF Forward AAA GGA GAT GAG AAA CGC AAA CAG 58 92 BD105535
Reversed GGC CTA GCA AGC TTC AGT AAT ACC
c-MET Forward TGT GCT GTG AAA TCC CTG AAT AGA AATC 59 112 AB118945
Reversed CCA AGA GTG AGA GTA CGT TTG GAT GAC
TGFα Forward CCG CCT TGG TGG TGG TCT CC 63 136 AY458143
Reversed AGG GCG CTG GGC TTC TCG T
HGF activator Forward ACA CAG ACG TTT GGC ATC GAG AAG TAT 60 128 AY458142
Reversed AAA CTG GAG CGG ATG GCA CAG
p27kip Forward CGG AGG GAC GCC AAA CAG G 60 90 AY455798
Reversed GTC CCG GGT CAA CTC TTC GTG
TGF-β1 Forward CAA GGA TCT GGG CTG GAA GTG GA 66 113 L34956
Reversed CCA GGA CCT TGC TGT ACT GCG TGT
TGF-β1 R I Forward CAG TCA CCG AGA CCA CAG ACA AAG T 59 101 AY455799
Reversed TGA AGA TGG TGC ACA AAC AAA TGG
TGF-β1 R II Forward GAC CTG CTG CCT GTG TGA CTT TG 61 116 AY455800
Reversed GGA CTT CGG GAG CCA TGT ATC TTG
UPA Forward CTG GGG AGA TGA AGT TTG AGG TGG 64.5 105 AY455801
Reversed TGG AAC GGA TCT TCA GCA AGG C
Bcl-2 Forward TGG AGA GCG TCA ACC GGG AGA TGT 61 87 AB116145
Reversed AGG TGT GCA GAT GCC GGT TCA GGT
Fas Ligand Forward GGG GTC AGT CCT GCA ACA ACA A 54 94 AY603042
Reversed ATC TTC CCC TCC ATC AGC ATC AG
Caspase-3 Forward ATC ACT GAA GAT GGA TGG GTT GGT 58 140 AB085580
Reversed GAA AGG AGC ATG TTC TGA AGT AGC ACT
HIF1α Forward TTA CGT TCC TTC GAT CAG TTG TCA 61 106 AY455802
Reversed GAG GAG GTT CTT GCA TTG GAG TC
Collagen I Forward GTG TGT ACA GAA CGG CCT CA 61 111 AF056303
Reversed TCG CAA ATC ACG TCA TCG
Collagen III Forward ATA GAG GCT TTG ATG GAC GAA 65 134 AB042266
Reversed CCT CGC TCA CCA GGA GC
Collagen IV Forward CAC AGC CAG ACA ACA GAT GC 67 151 U07888
Reversed GCA TGG TAC TGA AGC GAC G
Fibronectin Forward AGG TTG TTA CCA TGG GCA 61 91 U52106
Reversed GCA TAA TGG GAA ACC GTG TAG
Statistical analysis
A Kolmogorov-Smirnov test was performed to establish a normal distribution and a Levene's test for the homogeneity of variances. All samples included in this study were normally distributed and homogeneous in variance. The statistical significance of differences between diseased and control animals was determined by using the Student's t-test. A p-value < 0.05 was considered statistically significant. Analysis was performed using SPSS software (SPSS Benelux BV, Gorinchem, The Netherlands).
Western blot analysis
Used antibodies are described in Table 2. For Western blot analysis 30 mg of liver tissue from at least six samples of each group (n = 6 dogs per group, randomly chosen from original group) were pooled and analyzed. Liver tissues were homogenized in RIPA buffer containing 1% Igepal, 0.6 mM phenylmethylsulfonyl-fluoride, 17 μg/ml aprotinine, and 1 mM sodium-orthovanadate (Sigma chemical Co., Zwijndrecht, The Netherlands). Protein concentrations were obtained using a Lowry-based assay (DC Protein Assay, BioRad). Twenty μg of protein of the supernatant was denatured for 3 min at 95°C and electroforesed on 7.5% Tris-HCl polyacrylamide gels (BioRad) and the proteins were transferred onto Hybond-C Extra Nitrocellulose membranes (Amersham Biosciences Europe, Roosendaal, The Netherlands) using a Mini Trans-Blot® Cell blot-apparatus (BioRad). Immunodetection was based on an ECL Western blot analysis system, performed according to the manufacturer's instructions (Amersham Biosciences Europe). The membranes were incubated with 4% ECL blocking solution in TBS for 1 hour under gentle shaking. The incubation of the primary antibody was performed at 4°C over-night for all antibodies (see Table 2) in TBS with 0.1% Tween-20 (Boom B.V., Meppel, The Netherlands). After washing, the membranes were incubated with their respective horseradish peroxidase-conjugated secondary antibody (R&D systems, Europe Ltd., Abingdon, UK) at room temperature for 1 h and exposed to Kodak BioMax Light-1 films (Sigma chemical Co.). Densitometric analysis of immunoreactive bands was performed with a Gel Doc 2000 system coupled to the Quantity One 4.3.0 Software (BioRad).
Table 2 Used antibodies in Western blot experiments.
Antigen Product Size (kDa) Dilution Manufacturer Secondary antibody Dilution
HGF 82 1:100 Neomarkers Anti-mouse HRP 1:20,000
p-c-MET (Tyr 1230/1234/1235) 145 1:750 Abcam Anti-rabbit HRP 1:20,000
c-MET 145 1:750 Sigma Anti-goat HRP 1:20,000
p-PKB (Thr 308) 60 1:1,000 Cell-Signaling Anti-mouse HRP 1:20,000
PKB 60 1:250 BD Biosciences Anti-mouse HRP 1:20,000
p-STAT3 (Ser 727) 86 1:1,000 Cell Signalling Anti-rabbit HRP 1:20,000
STAT3 86 1:2,500 BD Biosciences Anti-rabbit HRP 1:20,000
p-Erk1/2 (Thr 202/Tyr 204) 42/44 1:1,500 Cell Signalling Anti-rabbit HRP 1:20,000
ERK1/2 42/44 1:1,000 Cell Signalling Anti-rabbit HRP 1:20,000
TGF-β1 25 1:1,000 Abcam Anti-rabbit HRP 1:20,000
p-Smad2 (Ser 465/467) 58 1:2,000 Cell-Signaling Anti-rabbit HRP 1:20,000
Smad2 58 1:500 BD Biosciences Anti-mouse HRP 1:20,000
Collagen I 95/210 1:500 Calbiochem Anti-mouse HRP 1:20,000
Caspase-3 34/20/18 1:1,000 Calbiochem Anti-rabbit HRP 1:20,000
Beta-actin (pan Ab-5) 42 1:2,000 Neomarkers Anti-mouse HRP 1:20,000
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
BS performed most Q-PCR measurements and wrote the manuscript. LP participated in the setup of Q-PCR measurements and helped to draft the manuscript. TI histochemically examined samples described in this manuscript. BA helped perform the0 Western blot experiments. JIJ histochemically examined samples described in this manuscript. FS helped collect all samples. JR participated in the study design and helped to draft the initial manuscript. All authors read and approved the final manuscript.
Acknowledgements
The authors are indebted to Dr. Alexandra Pietersen, Dr. Bernard Roelen, and Dr. Peter ten Dijke for their invaluable advice.
==== Refs
Guyon R Lorentzen TD Hitte C Kim L Cadieu E Parker HG Quignon P Lowe JK Renier C Gelfenbeyn B Vignaux F DeFrance HB Gloux S Mahairas GG Andre C Galibert F Ostrander EA A 1-Mb resolution radiation hybrid map of the canine genome Proc Natl Acad Sci U S A 2003 100 5296 5301 12700351 10.1073/pnas.0831002100
Murray CP Yoo SJ Babyn PS Congenital extrahepatic portosystemic shunts Pediatr Radio 2003 33 614 620 10.1007/s00247-003-1002-x
van den Ingh TS Rothuizen J Meyer HP Circulatory disorder of the liver in dogs and cats Vet Q 1995 17 70 76 7571284
van den Ingh TS Rothuizen J Meyer HP Portal Hypertension associated with primary hypoplasia of the hepatic portal vein in dogs Vet Rec 1995 137 424 427 8560700
Michalopoulos KM DeFrances MC Liver regeneration Science 1997 276 60 66 9082986 10.1126/science.276.5309.60
Trusolino L Comoglio PM Scatter-factor and semaphorin receptors: cell signalling for invasive growth Nat Rev Cancer 2002 2 289 300 12001990 10.1038/nrc779
Fausto N Liver regeneration J Hepatol 2000 19 31 10728791 10.1016/S0168-8278(00)80412-2
Funakoshi H Nakamura T Hepatocyte growth factor: from diagnosis to clinical applications Clin Chim Acta 2003 327 1 23 12482615 10.1016/S0009-8981(02)00302-9
Scheid MP Woodgett JR Unravelling the activation mechanisms of protein kinase B/Akt FEBS Lett 2003 546 108 112 12829245 10.1016/S0014-5793(03)00562-3
Coffer PJ Jin J Woodgett JR Protein kinase B (c-Akt): a multifunctional mediator of phosphatidylinositol 3-kinase activation Biochem J 1998 335 1 13 9742206
Paumelle R Tulasne D Kherrouche Z Plaza S Leroy C Reveneau S Vandenbunder B Fafeur V Tulashe D Reveneau S Hepatocyte growth factor/scatter factor activates the ETS1 transcription factor by a RAS-RAF-MEK-ERK signaling pathway Oncogene 2002 21 2309 2319 11948414 10.1038/sj.onc.1205297
Boccaccio C Ando M Tamagnone L Bardelli A Michieli P Battistini C Comoglio PM Induction of epithelial tubules by growth factor HGF depends on the STAT pathway Nature 1998 391 285 288 9440692 10.1038/34657
Derynck R Akhurst RJ Balmain A TGF-β signaling in tumor suppression and cancer progression Nature Genet 2001 29 117 129 11586292 10.1038/ng1001-117
Piek E Heldin CA ten Dijke P Specificity, diversity, and regulation in TGF-β superfamily signalling FASEB J 1999 13 2105 2124 10593858
Aihara Y Kasuya H Onda H Hori T Takeda J Quantitative analysis of gene expressions related to inflammation in canine spastic artery after subarachnoid hemorrhage Stroke 2001 32 212 217 11136939
Massague J How cells read TGF-beta signals Nat Rev Mol Cell Biol 2000 1 169 178 11252892 10.1038/35043051
Hengartner MO The biochemistry of apoptosis Nature 2000 407 770 776 11048727 10.1038/35037710
Norman JT Clark IM Garcia PL Hypoxia promotes fibrogenesis in human renal fibroblasts Kidney Int 2000 58 2351 2366 11115069 10.1046/j.1523-1755.2000.00419.x
Vandesompele J De Preter K Pattyn F Poppe B Van Roy N De Paepe A Speleman F Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes Genome Biol 2002 3 research0034.1 research0034.11 12184808 10.1186/gb-2002-3-7-research0034
Schmidt C Bladt F Goedecke S Brinkmann V Zschiesche W Sharpe M Gherardi E Birchmeier C Scatter factor/hepatocyte growth factor is essential for liver development Nature 1995 373 699 702 7854452 10.1038/373699a0
La Rosa S Uccella S Capella C Erba S Sessa F Localization of Hepatocyte Growth Factor and Its Receptor met in Endocrine Cells and Related Tumors of the Gut and Pancreas: An Immunohistochemical Study Endocr Pathol 2000 11 315 329 12114756 10.1385/EP:11:4:315
Barthelemy C Henderson CE Pettmann B Foxo3a induces motoneuron death through the Fas pathway in cooperation with JNK BMC Neurosci 2004 5 48 15569384 10.1186/1471-2202-5-48
Chandramohan V Jeay S Pianetti S Sonenshein GE Reciprocal control of Forkhead box O 3a and c-Myc via the phosphatidylinositol 3-kinase pathway coordinately regulates p27Kip1 levels J Immunol 2004 172 5522 5527 15100294
Hanada M Feng J Hemmings BA Structure, regulation and function of PKB/AKT – a major therapeutic target Biochim Biophys Acta 2004 1697 3 16 15023346
Nakagami H Morishita R Yamamoto K Taniyama Y Aoki M Matsumoto K Nakamura T Kaneda Y Horiuchi M Ogihara T Mitogenic and antiapoptotic actions of hepatocyte growth factor through ERK, STAT3, and AKT in endothelial cells Hypertension 2001 37 581 586 11230338
Diez-Marques L Ortega-Velazquez R Langa C Rodriguez-Barbero A Lopez-Novoa JM Lamas S Bernabeu C Expression of endoglin in human mesangial cells: modulation of extracellular matrix synthesis Biochim Biophys Acta 2002 1587 36 44 12009422
Zhu HJ Burgess AW Regulation of transforming growth factor-beta signaling Mol Cell Biol Res Commun 2001 4 321 330 11703090 10.1006/mcbr.2001.0301
Tsukada S Parsons CJ Rippe RA Mechanisms of liver fibrosis Clin Chim Act
Liu C Gaca MD Swenson ES Vellucci VF Reiss M Wells RG Smads 2 and 3 are differentially activated by transforming growth factor-beta (TGF-beta) in quiescent and activated hepatic stellate cells. Constitutive nuclear localization of Smads in activated cells is TGF-beta-independent J Biol Chem 2003 278 11721 11728 12547835 10.1074/jbc.M207728200
Kondou H Mushiake S Etani Y Miyoshi Y Michigami T Ozono K A blocking peptide for transforming growth factor-beta1 activation prevents hepatic fibrosis in vivo J Hepatol 2003 39 742 748 14568256 10.1016/S0168-8278(03)00377-5
Kataoka H Miyata S Uchinokura S Itoh H Roles of hepatocyte growth factor (HGF) activator and HGF activator inhibitor in the pericellular activation of HGF/scatter factor Cancer Metastasis Rev 2003 22 223 236 12784998 10.1023/A:1023051500010
Wang HT Chen S Wang J Ou QJ Liu C Zheng SS Deng MH Liu XP Expression of growth hormone receptor and its mRNA in hepatic cirrhosis World J Gastroenterol 2003 9 765 770 12679928
Kosai K Matsumoto K Funakoshi H Nakamura T Hepatocyte growth factor prevents endotoxin-induced lethal hepatic failure in mice Hepatology 1999 30 151 159 10385651 10.1002/hep.510300102
Yasuda H Imai E Shiota A Fujise N Morinaga T Higashio K Antifibrogenic effect of a deletion variant of hepatocyte growth factor on liver fibrosis in rats Hepatology 1996 24 636 642 8781336
Ueki T Kaneda Y Tsutsui H Nakanishi K Sawa Y Morishita R Matsumoto K Nakamura T Takahashi H Okamoto E Fujimoto J Hepatocyte growth factor gene therapy of liver cirrhosis in rats Nat Med 1999 5 226 230 9930873 10.1038/5593
Matsuda Y Matsumoto K Yamada A Ichida T Asakura H Komoriya Y Nishiyama E Nakamura T Preventive and therapeutic effects in rats of hepatocyte growth factor infusion on liver fibrosis/cirrhosis Hepatology 1997 26 81 89 9214455
Ohara K Kusano M Anti-transforming growth factor-beta1 antibody improves survival rate following partial hepatectomy in cirrhotic rats Hepatol Res 2002 24 174 12270747 10.1016/S1386-6346(02)00031-1
Nakamura T Sakata R Ueno T Sata M Ueno H Inhibition of transforming growth factor β prevents progression of liver fibrosis and enhanced hepatocyte regeneration in dimethylnitrosamine-treated rats Hepatology 2000 32 247 255 10915731 10.1053/jhep.2000.9109
Sato M Kakubari M Kawamura M Sugimoto J Matsumoto K Ishii T The decrease in total collagen fibers in the liver by hepatocyte growth factor after formation of cirrhosis induced by thioacetamide Biochem Pharmacol 2000 59 681 690 10677585 10.1016/S0006-2952(99)00359-7
Kimura Y Leung PS Kenny TP van de Water J Nishioka M Giraud AS Neuberger J Benson G Kaul R Ansari AA Coppel RL Gershwin ME Differential expression of intestinal trefoil factor in biliary epithelial cells of primary biliary cirrhosis Hepatology 2002 36 1227 1235 12395334 10.1053/jhep.2002.36157
|
16336649
|
PMC1315335
|
CC BY
|
2021-01-04 16:38:24
|
no
|
Comp Hepatol. 2005 Dec 7; 4:7
|
utf-8
|
Comp Hepatol
| 2,005 |
10.1186/1476-5926-4-7
|
oa_comm
|
==== Front
RetrovirologyRetrovirology1742-4690BioMed Central London 1742-4690-2-721629319410.1186/1742-4690-2-72ReviewReview of the twelfth West Coast retrovirus meeting Barry Sheila M [email protected] Marta [email protected] Philippe [email protected] Thomas J [email protected] Department of Cell and Molecular Biology, College of Medicine, Northwestern University, Chicago, IL 60611, USA2 Department of Microbiology and Immunology, College of Medicine, University of Illinois at Chicago, Chicago, IL 60612, USA3 The Scripps Research Institute, La Jolla, CA 92037, USA2005 17 11 2005 2 72 72 16 11 2005 17 11 2005 Copyright © 2005 Barry et al; licensee BioMed Central Ltd.2005Barry 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.
Every year the Cancer Research Institute from University of California at Irvine organizes the West Coast Retrovirus Meeting where participants have a chance to discuss the latest progress in understanding the pathology of retroviruses. The 12th meeting was held at the Hyatt Regency Suites in Palm Springs, California from October 6th to October 9th 2005, with the major focus on human immunodeficiency virus (HIV) pathogenesis. Philippe Gallay from The Scripps Research Institute and Thomas J. Hope from Northwestern University organized the meeting, which covered all the steps involved in the lifecycle of retroviruses with an emphasis on virus:host interactions. The trend in research appeared to be on the restriction of viral infection, both by the endogenous, cellular restriction factors, as well as by the potential antimicrobial compounds of known or unknown mechanisms. Additionally, new stories on the inevitable feedback from the host immune system were presented as well. HIV still represents a challenge that an army of motivated people has been working on for over 20 years. And yet, the field has not reached the plateau in knowledge nor enthusiasm, which was proven again in October 2005 in Palm Springs.
==== Body
Review
Viral Entry
John Young of the Salk Institute began this session by describing work his lab has recently completed in understanding cellular requirements for replication of Murine Leukemia Virus (MLV) [1]. Through use of chemically mutagenized CHO cells, they identified five clones that became resistant to MLV infection. Additional studies revealed this restriction was specific to the MLV core. After confirming the virus was blocked prior to integration, the clones were separated into two phenotypes, those which blocked reverse transcription early and those which allowed reverse transcription and nuclear entry, but prevented viral integration. Young and colleagues are currently identifying cellular factors involved in the latter phenotype. While the exact identities of these cellular factors were not revealed, Young shared that they believe one is an enzyme and the other a putative transcription factor.
Pankaj Kumar from Lorraine Albritton's lab at the University of Tennessee continued this theme by examining cellular factors involved in Moloney MLV entry. Previous work found that the exposure of MLV to proteases enhanced the viral infectivity and certain cell lines, including XC cells, innately possessed proteases that could facilitate MLV infection. The group decided to focus on cathepsins, since expression of these cellular proteases is induced under these conditions. They found a broad spectrum cathepsin inhibitor as well as a cathepsin B-specific inhibitor reduced Moloney MLV infectivity. Additionally, treatment of viral particles with cathepsin B resulted in cleavage of the surface glycoprotein (SU). They postulate Moloney MLV encounters cathepsin B within early lysosomes and the ensuing cleavage of SU facilitates fusion and entry steps.
Two talks turned attention to the involvement of HIV envelope glycoprotein gp41 in early steps of viral infection. In work previously published by his lab, John Day of the University of California San Diego and others determined the membrane proximal tyrosine based sorting signal of gp41, Y712xxL, was important in viral entry and infectivity and was involved in virion incorporation of the envelope glycoprotein (Env) only in some cell lines [2]. They hypothesized this enhancement of viral infectivity resulted from the virus using adaptor protein complexes to traffic Env to specific cellular membranes. Gp41 has few motifs that are known to interact with adaptor proteins (AP); Y712xxL interacts with AP-2, while the C-terminal double leucine motif (LL855/856) binds to AP-1. Thus, both signals were evaluated for their ability to affect intracellular localization and viral infectivity. In studies using CXCR4 tropic HIV-1, LL855/856 was found to have no effect on viral infectivity or entry, a sharp contrast from the observed viral dependence on Y712xxL. However, no difference was observed in intracellular localization of either mutant compared to wildtype. This suggests the Env sorting signals may not be involved in targeting viral morphogenesis to specific cellular membranes. Interestingly, when these signals were evaluated with CCR5 tropic HIV-1, neither the LL855/86 nor the Y712xxxL sorting signal had any effect on viral infectivity. This observation indicated the significance of the tyrosine-sorting signal in viral infectivity is dependent on the tropism of the HIV Env ectodomain.
Michael Kay from the University of Utah presented his lab's efforts in understanding the ineffectiveness of vaccine development against the N trimer of gp41. Following gp120 binding to coreceptor, gp41 undergoes conformational changes, from a pre-hairpin state where both N and C peptides are exposed, to the formation of a six-helix bundle, where a trimer of N peptides (N trimer) is surrounded by three C peptides. Within this N-trimer is a highly conserved pocket which has become the target of most vaccine development. Unfortunately, little progress has been made in creating an effective anti N trimer vaccine. Kay and collaborators considered a potential obstacle to vaccine development was the accessibility of the region to bulky inhibiting proteins. To evaluate this possibility, this group used a C-peptide inhibitor that was attached via a flexible linker to several cargo proteins of various sizes. They found the potency of this inhibitor decreased with increasing cargo protein size. Increasing the length of the flexible linker region could restore potency [3]. This suggests a severe steric block in gp41 to neutralizing antibodies.
The session ended with a talk by Marta Melar from Thomas Hope's lab at Northwestern University on coreceptor dependent signaling during HIV entry. By measuring changes in intracellular calcium (Ca2+) levels as a marker for signaling through coreceptor, Melar observed that signaling was coreceptor specific, responsive to both monomeric and virion bound gp120, and dependent on CD4. The fluorescent microscopic technique employed in these studies allowed Melar to quantate the number of virions bound to Ca2+- fluxing cells. An average of four virions was determined to be sufficient for Ca2+ mobilization in primary unstimulated CD4+ T cells.
Vif, Vpr and Nef
Several interesting talks emphasized the ability of these accessory HIV proteins to evade the host immune system in order to make a perfect niche for viral replication in the hostile target cells. The stories on Vif protein focused on its ability to protect the virions from incorporation of the cellular apolipoprotein B mRNA-editing enzyme-catalytic polypeptide-like-3G (APOBEC-3G or A3G) [4]. A3G has cytidine deamination activity and can use newly reverse transcribed viral genome as a substrate, leading to the loss of viral fitness through introduction of G-to-A hypermutations in the plus strand of the cDNA
Jason Kreisberg, a graduate student from Warner Greene's lab from University of California at San Francisco, presented the ongoing work in the lab regarding the mechanism of A3G dependant HIV restriction in secondary lymphoid organs. This work is the continuation of already published data [5] on two existing forms of A3G, high and low molecular weight A3G, where only a low molecular weight form exhibits enzymatic activity. RNase treatment was shown to facilitate the switch from the high into low molecular weight form. Kreisberg emphasized the correlation between the presence of the A3G molecular weight form and permissiveness of the cell type to infection. They found resting peripheral CD4+ T cells that are not permissive for infection express the enzymatically active version of A3G. However, when isolated from tonsils and cultured in conditioned media, this cell type becomes permissive to HIV infection. Cytokines, specifically IL-2 and IL-15, may have a role in this in vivo switch from low to high molecular weight A3G. These data could shed some light on the role of the target cell A3G opposing the well-established mutagenizing role of A3G on the HIV genome in the producer cells. Since only high molecular weight A3G is incorporated into ΔVif virions, it was not known how A3G gains its activity in the target cells. The virally encoded enzyme RNaseH may be doing the virus a contra-favor, by functioning as the facilitator of A3G cytidine deaminase activation.
The session continued with another keynote lecture given by Nathaniel Landau from the Salk Institute for Biological Studies from San Diego. This talk was focused on the species-specificity of the Vif:A3G interaction. The ability of Vif to block the antiviral activity of A3G is species-specific [6], where the positive charge of single Asp in the position 128 within human A3G is responsible for recognition of HIV Vif and its interaction. From mutational analyses, Landau and his collaborators found that out of two active sites within APOBEC family of enzymes, the first active site (AS1) plays a role in encapsidation of the enzyme into the ΔVif virions, where AS2 is responsible for deamination of the substrate, the negative DNA single strands in a newly synthesized viral genome. As well, the group found a graded deamination frequency, from low at the 5'-end to higher towards the 3'-end, most likely a phenomenon affected by the mechanism of the reverse transcription reaction and the availability of negative strand cDNA to the A3G-induced mutation.
The following talk from Michael Emerman’s group at the Fred Hutchinson Cancer Research Center continued the discussion on different aspects of antiviral properties of APOBEC enzymes. Shari Kaiser addressed the question if the uracil DNA glycosylase 2 (UNG2) is involved in the antiviral effects of A3G. Previously, this enzyme was postulated to work one step downstream of A3G, enabling G-to-A hypermutations to occur. However, Kaiser found that virus replication in either target or producer cells was not affected as compared to the positive control in either ung-/- cell line or after the UNG2 inhibitor treatment in the producer cells. This implied UNG2 was dispensable for the fitness of the virus contrasting with a recent publication [7].
Another focus on host:virus interaction came from Morehouse School of Medicine in Atlanta, where Michael Powell's group works on HIV infectivity enhancement through the direct Nef and CypA interaction. This work was based on the hypothesis that CypA acts as a linker between HIV Nef and the viral core, interacting with Nef at its N-terminus and the core through its C-terminus. They speculate this interaction between Nef and CypA can facilitate the uncoating process in the target cells, since induction of natural endogenous reverse transcription (NERT) in intact virions could overcome the lack of either protein. They also showed a Nef:CypA fusion protein, which efficiently got incorporated into virions, restored infectivity of ΔNef virions. Interestingly, the group also suggested that the ability of SIV Nef to bind core directly might mask the restriction effect of cellular restriction factor TRIM5α that is known to interact with viral core, since HIV virions expressing SIV Nef were able to bypass the restriction point of simian TRIM5α and replicate in simian MAGI cells. That was also the case with NERT induced wild type HIV in simian MAGI cells.
The mechanism of MHC class II invariant chain (Ii) up-regulation was another Nef function discussed during this session. Richard Mitchell from University of California at San Diego presented work on the importance of the di-leucine sorting motif E160xxxLL found at the C-terminus of HIV Nef and its potential role in providing a sorting endocytic signal for down-regulation of the surface expression of CD4, coreceptors CXCR4 and CCR5, MHC I and II and up-regulation of MHC II-Ii complexes at the cell surface. By using yeast three-hybrid system and GST-pulldown assays, the group found that residues E160 and LL were important for up-regulation of the surface Ii expression. This is another report explaining the role of this accessory HIV protein in enhancing the infectivity of the virus, by altering the immune response of the host.
Uncoating and budding
The next panel began with two keynote lectures, both focusing on the issue of viral restriction in different hosts. Jaquelin Dudley from University of Austin, Texas, introduced us to the world of mouse resistance to multiple pathogens. Her group observed that certain strains of inbred mice carry an endogenous mouse mammary tumor virus (MMTV) that is replication deficient but does express the virally encoded superantigens (Sags). Sags expression results in a depletion of specific T cell subsets. These mice, when infected with exogenous MMTV, are prone to the development of mammary gland tumors. The group created MMTV-negative mice, which were found to be protected from a replication-competent, exogenous MMTV, type B leukemogenic virus and Vibrio cholerae. Subsequently infected with MMTV, MMTV-null mice lacked an immune response to the virus and lacked the tumor development. Genetic analysis revealed that the susceptibility to MMTV infection of endogenously infected mice was a recessive feature and that a single MMTV gene product was rendering these animals susceptible to infection, implying a novel mechanism of resistance to both viral and bacterial pathogens.
Another story on resistance to viral infection dealt with HIV restriction in Old World monkeys by a cellular restriction factor named TRIM5α. This molecule is a big hit in HIV research, ever since the Sodroski group from Harvard University published data from a primary rhesus monkey lung fibroblasts cDNA library screen for the resistance to HIV-1 infection [8]. Matt Stremlau gave us an insight on how this restriction factor might work in order to block the incoming virus at the post-entry step but pre-integration. Previously defined interaction of TRIM5α with the viral core served as a starting point to speculate that TRIM5α could either stabilize the capsid core, cause rapid disassembly of the core or target the capsid (CA) for proteasomal degradation. All three outcomes could have an impact on the very time-sensitive process of the reverse transcription. From their work on in vitro assembled HIV cores, representing highly ordered tubular structure of p24 CA hexamers [9], the group found that TRIM5α in its functional trimeric form binds only to the core composed out of CA hexamers, but does not bind to p24 CA monomers. Since their data indicate that the proteasomal inhibitors did not recover the loss of the oligomeric into the monomeric CA form, the group speculated that TRIM5α most probably acts to rapidly disassemble the core and that would impair the reverse transcription process, also implying the species-specific blocking mechanism on the conformational level.
On the other hand, Philippe Gallay from The Scripps Research Institute showed recent data arguing that HIV CA but not the matrix protein was being targeted for degradation, although other than through proteasomal pathway, since proteasomal inhibitors did not fully rescue the RhMTRIM5α mediated degradation of the HIV CA. This group argued that TRIM5α restriction occurs at the level of accelerated degradation of the core, possibly also affecting the nuclear import of the preintegration complex.
Microscopy based approach to study the cellular localization of TRIM5α in living cells came from Thomas Hope group at Northwestern University. The audience had a chance to see that both exogenous and endogenous TRIM5α formed cytoplasmic bodies, but the proteins were also found in the nuclei. The cytoplasmic bodies are highly dynamic hollow structures and their formation is speculated to be relevant in the TRIM5α function as a restriction factor. The morphology of the bodies could be altered with the proteasome inhibitor MG132, where the smaller bodies merged to form bigger structures. The group is currently investigating the effect of MG132 on the TRIM5α restriction potency.
An interesting study came from Bruce Torbett's group, where Christina Swan presented work on the design of HIV based vectors for gene therapy in human stem and T cells based on the HIV tropism. However, since monkeys would be the animal model for the vector design trials, the problem of the intrinsic cellular restriction of incoming HIV virions by the RhTRIM5α arose. In order to overcome this restriction problem, the group decided to test numerous HIV CA mutants and found that incorporation of the naturally occurring four amino acid substitutions in the CypA binding site of HIV Gag/Pol allowed for the restriction escape and therefore higher transduction efficiencies in primary human and monkey cell lines. These mutations allowed independence of CypA in human cells and loss of TRIM5α recognition because of the lack of CypA incorporation into the virions in monkey cells.
Another way to block HIV infection besides engaging the endogenous restriction machinery is to test the potential antimicrobial compounds. Christopher Aiken from Vanderbilt University introduced us to the HIV-1 maturation inhibitor 3-O-{3',3'- dimethylsuccinyl}-betulinic acid (DSB). DSB specifically inhibits HIV replication by delaying the last step in the Gag maturation: the release of the spacer peptide SP1 from the C-terminus of CA. However, the inhibitory effect was not due to the protease (PR) inhibition, since PR inactivation stabilized the DSB:CA complex. The escape mutants in CA-SP1 junction were not incorporating DSB and were now rendered resistant to it. Moreover, Aiken showed data supporting the hypothesis that DSB binds to a pocket formed by Gag oligomerization, an interaction that sterically inhibits PR from binding [10]. The compound had to be present at the time of the viral assembly in order to inhibit the viral replication in a dose-dependant manner and was also shown to be a weak fusion inhibitor.
The session on viral uncoating and budding was concluded by the talk from Wesley Sundquist's group from University of Utah. Their research focuses on structural proteomics to understand the process of ubiquitinated Gag recognition by the cellular sorting machinery through endosomal sorting complexes required for transport (ESCRT I-III), utilization of multivesicular bodies formation and the energy of ATP hydrolysis in the viral protein sorting, assembly and budding. Melissa Stuchell-Brereton presented recently published data on the latest structural analysis of one of the players in this cellular machinery that mediates recycling of the sorting apparatus from the cargo, namely VPS4A AAA ATPase [11]. Stuchell-Brereton described the novel three-dimensional structure of VPS4A C-terminal helix and N-terminal fragment: a microtubule interacting and transport domain (MIT). Data suggested that the VPS4A MIT domain directly binds the C-terminus of one of the ESCRT-III proteins, allowing the formation of the ring structure, where VPS4 proteins may serve to unfold, translocate and therefore recycle the members of ESCRT-III family through the ring pore, indirectly facilitating HIV budding.
HIV Inhibition and Activation
David Margolis of the University of North Carolina at Chapel Hill gave the keynote lecture of this session, recapping work his lab has completed in depleting latent HIV infection from resting CD4+ cells [12]. In the twenty years since the discovery of HIV, several anti-retroviral therapies have been attempted, many of which have terrible side effects and are not well tolerated by patients. In addition, while viremia may be reduced during treatment, viral load increases significantly once therapy is stopped. A major obstacle to eradication of HIV infection is the persistence of a latent viral reservoir within resting CD4+ cells. Therefore, stimulating HIV expression from these resting CD4+ T cells would allow the immune system to recognize infected cells and target the infection more efficiently. Histone deacetylase 1 (HDAC1) is instrumental in maintaining latency of integrated HIV, thus inhibitors of HDAC1, such as the FDA-approved valproic acid (VPA), may assist in expression of HIV from resting CD4+ cells. To examine this hypothesis, Margolis' group supplemented the treatment of four patients with therapeutic doses of VPA. Infection of CD4+ cells decreased in all patients, with three exceeding expectations. While considerable work still remains to be completed, these results suggest VPA may be a promising addition to HIV treatment.
The subsequent two talks examined the participation of certain transcription factors in HIV expression. Jonathan Karn from Case School of Medicine and his lab have recently completed research studying the molecular mechanisms of NF-κB and other transcription factors in expression of integrated HIV. To conduct these studies, they created a population of T cells that possessed stably integrated proviral HIV genomes that encoded GFP. The group used these cells to evaluate the activation of HIV transcription, as they turn green following treatment with TNF-α. Additionally, they were able to evaluate the distribution of RNA polymerase (RNA pol) II along HIV LTR as well as the kinetics of proviral activation following recruitment of TFIIH and NF-κB to the promoter and provirus by using chromatin immunoprecipitation (ChIP). These studies revealed recruitment of NF-κB coincided with an accumulation of RNA pol and TFIIH within the nucleus. Interestingly, induction of transcription was found to be transient, with levels of RNA pol, TFIIH, and NF-κB returning to pretreatment levels within 90 minutes following activation, only to increase in a second cycle of induction 3 to 5 hours later. Although the mechanism is more complicated, T cells stimulated though the T cell receptor CD3 experienced a similar trend. Initially, NFAT was observed to be selectively mobilized, only to be replaced by NF-κB within 30 minutes. These observations suggest the induction of HIV transcription is a multifactorial process that is cyclical in nature, not the sustained event as previously supposed.
Andrew Rice from Baylor College of Medicine presented his lab's investigation of the role of 7SK small nuclear RNA (7SK) in P-TEFb function and, in doing so, challenged the previously described model for these proteins in HIV expression [13]. P-TEFb is a RNA pol II transcription factor that is composed of Cdk9 and cyclin T1, T2 or K. The HIV Tat protein targets the Cdk9/cyclin T1 P-TEFb to activate transcription of the viral genome. Much of this P-TEFb is complexed to 7SK and HEXIM proteins, however, and this complex has been demonstrated to have decreased kinase activity in vitro. Rice and colleagues examined 7SK and HEXIM in primary cells and found expression of these proteins positively correlated with the activation state of the cells. Additionally, there was no observed difference in expression of endogenous genes or integrated HIV provirus when siRNA was used to deplete 7SK, although expression of reporter plasmids increased. Another interesting observation was that apoptosis was induced within 72 hours in 7SK depleted cells. This group postulates these findings indicate 7SK plays a significant role in P-TEFb function, one that merits further investigation.
Wendong Yu from Baylor College of Medicine at Houston, Texas discussed the function of cyclin T1 in Mono-Mac-6 (MM6) cells as a model for primary monocytes-to-macrophages differentiation. The work was based on the observation that the differentiation of monocytes into macrophages (MΦs) is followed by the increasing levels of CycT1, which together with CDK9 constitutes for P-TEFb, a factor necessary for Tat-induced transcriptional activation. In the early MΦs, both CycT1 and Tat levels were elevated, but there was a significant loss of CycT1 expression in late MΦs that could be restored with PMA, IFNγ or LPS induced signaling. Indeed, when CycT1 was knocked out in MM6 cells using a shRNA approach, microarray analysis revealed downregulation of ~13% genes, where ~11% genes were PMA-inducible ones. This data emphasized the role of the CycT1 induction in MΦ differentiation and upregulation of ~11% genes.
Mary Lewinski from Bushman's group gave us an insight into the integration target specificity of HIV and MLV [14]. After extensive integration site cloning, mapping to the genome and considerable statistical analyses, the group concluded that the chromosomal environment influences the expression of integrated sequences and that different retroviruses show disparate preferences for integration of their genome into the host chromosomes. To understand which viral proteins orchestrate the choice for the integration location within the host genome, numerous chimeric viruses between MLV and HIV were tested for the preferential sites for integration. The interesting conclusion was that not one, but a pair of genes, Gag and integrase, worked synergistically to determine the integration site specificity.
The last two talks in this session were reserved for potential antiviral agents. The talk from Vanderbilt University by Derya Unutmaz focused on VacA toxin, produced by bacterium Helicobacter pylori. The group observed that the infectivity levels in primary activated T cells, normally susceptible to HIV infection, dropped almost 100% when pre-treated with VacA. The block was determined to be post-reverse transcription, but pre-integration, possibly at the level of nuclear membrane. VacA was not affecting TCR signaling, but was shown to downregulate IL-2 production and secretion, leading to abrogation of proliferation, an effect similar to rapamycin. However, the group is still investigating the host target(s) of this toxin.
Roland Wolkowicz from Stanford University explained the method for screening of relatively large number of random peptide libraries for resistance to HIV infection. The rationale behind random screening for antiviral compounds was found in a possible steric block between the viral and host proteins involved in HIV lifecycle, that could lead to the gain of resistance of the cells transduced with retroviral vector carrying the peptide library formed in silico. Using this approach, Wolkowicz and his collaborators confirmed the positive role of the signalosome and Casein Kinase II in HIV lifecycle, as a randomly chosen peptide could interact with these proteins and block the HIV replication.
VPU and HIV Transinfection
The first two talks in this session explored the function of HIV accessory protein Vpu. While Vpu has been well characterized to enhance virus release, the mechanism by which Vpu accomplishes this has remained unknown. Edward Stephens from the University of Kansas and his lab investigated the role of the transmembrane (TM) domain of Vpu as well as its ion channel properties by exchanging this domain for the M2 protein domain from influenza A [15]. While this exchange had little effect on replication, viral maturation or pathogenicity, the mutant virus now became susceptible to antiviral drugs that specifically targeted the M2 ion channel, namely amantadine and rimantadine. Studies of the M2 protein mapped the ion channel's function to its HxxxW motif. By replacing a single alanine residue with histidine, this group was able to construct a Vpu protein, which possessed this HxxxW motif within its TM domain. This alteration was sufficient to render HIV susceptible to rimantadine. These studies suggest the Vpu ion channel may be an effective target for anti HIV therapeutics.
Beth Noble from Paula Cannon's lab at Childrens Hospital Los Angeles presented work on the involvement of the cytoplasmic tail of Vpu in to enhancing viral release. Microscopic analysis revealed Vpu in a mutant HeLa cell line (HeLa-T17) was aberrantly concentrated in the perinuclear region; a phenotype which the group hypothesized was the result of improper trafficking with adaptor protein 3 (AP-3). AP-3 depletion by siRNA and alteration of a specific motif within the cytoplasmic tail of Vpu seem to support this hypothesis. Together, these studies have identified two specific regions of Vpu that affect viral release, namely the transmembrane ion channel and the AP-3 interacting cytoplasmic tail.
Sheila Barry from Thomas Hope's lab at Northwestern University began the discussion of HIV transinfection, by describing recent studies investigating the role of Langerhans cells (LCs) in mediating transinfection. Previously, considerable efforts have been invested in studying the effect of DC-SIGN-expressing dendritic cells (DCs) on HIV infection. Such DCs are confined to deep tissue layers, where they may not readily encounter HIV. In contrast, Langerhans cells reside in surface epithelial tissue, and can send dendritic processes across intact tight junctions to sample pathogens prior to host infection. Using a luciferase reporter assay, this group demonstrated that LCs exposed to X4-tropic virus could enhance viral infectivity in a manner similar to mature DCs. In addition, fluorescent microscopy revealed GFP-labeled HIV was found in CD1a+ compartments within activated LCs and this overlap continued in recipient T cells. These results suggest LCs can enhance HIV infectivity without becoming infected themselves, and viral delivery potentially takes place through an infectious synapse resulting in delivery of both virus and LC specific proteins to target cells.
In his second talk of the conference, Derya Unutmaz presented evidence that antimicrobial peptides derived from amphibian skin (A-AMPs) inhibit both HIV infection and viral transfer between DCs and T cells [16]. Using GFP labeled virus, they observed three A-AMPs, caerin 1.1, caerin 1.9 and maculatin 1.1, could inhibit HIV infection of target cells within minutes of exposure at concentrations that did not affect cell viability. Further, caerin 1.9 could inhibit both HIV and MLV regardless of Env, while it had no effect on the non-enveloped reovirus. As DCs have previously been shown to mediate HIV transinfection by internalizing the virus and protecting viral particles from intracellular degradation, Unutmaz and colleagues considered A-AMPs might affect viral transfer from DCs to T cells. Addition of A-AMPs to HIV-pulsed DCs up to 8 hours post virus exposure was found to inhibit DC-mediated transinfection of T cells, however pretreating DCs with peptides prior to virus exposure had no effect on viral infectivity. A-AMPs were either neutralizing virus at the cell surface or trafficking to the same intracellular compartment as HIV and inactivating virus there. Through fluorescent microscopy, the group observed A-AMPs neutralized GFP labeled HIV and were confined to the surface of DCs. This suggests internalized virus may be continually cycling to the surface.
Retrovirus Pathogenesis
Maribeth Eiden from the NIH discussed her lab's efforts in tracking the evolution of gammaretroviruses in gibbon apes and koalas. Gibbon ape leukemia virus (GALV) was originally identified in captive gibbon apes in the 1970s. Recently, koala retrovirus (KoRV) was isolated from captive koalas. Interestingly, KoRV shares a 78% nucleotide identity with GALV, despite the fact that GALV is an exogenous retrovirus affecting gibbon apes while KoRV is endogenously found in koalas. This suggests GALV and KoRV probably originated from a common ancestor, with KoRV diverging at an earlier time point than GALV. One potential source could be an infectious murine gammaretrovirus, as elements related to the envelope genes of GALV and KoRV were found in the genomes of several Asian feral mice species. In an attempt to identify potential vectors for transmission between koalas in Australia and gibbon apes in Thailand, Eiden et al found both GALV and KoRV were able to infect mosquito cells, thus establishing the possibility that insects could have acted as an infectious intermediate.
Stacey Hull from Hung Fan's lab at the University of California-Irvine closed the conference by presenting her work on the cytoplasmic tail of Jaagsiekte sheep retrovirus (JSRV) Env mediating transformation. They found JSRV-mediated transformation transpired by signaling through either the PI-3K-Akt-mTOR or the Ras-MEK-MAPK pathways, transformation was negatively regulated by p38 signaling, and phosphatidylinositol 3-kinase (PI3-K) binding site (YxxM) found in the cytoplasmic tail was necessary for transformation [17]. By conducting an alanine scan across the full length of the cytoplasmic tail, they created mutant Env proteins that affected transformation efficiency. Interestingly, one mutant increased JSRV transformation efficiency, although it was unaffected by inhibitors of the mTOR or Ras pathways, implying signaling in this mutant may be taking place through another unknown pathway. To further investigate the importance of the PI3-K binding motif, Hull exchanged the methionine residue either to aspartic acid, lysine, serine, or isoleucine. Interestingly, the isoleucine mutant, which has essentially been transformed into the binding motif for Src, had a greater transformation efficiency as compared to wildtype, thus suggesting Src may play some role in JSRV transformation.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
Every author meets the criteria of author as defined by the Retrovirology journal. SMB and MM contributed equally to the drafting and revising of the manuscript. PG and TJH also made considerable intellectual contributions to this review. All authors approved of this version for publication.
Acknowledgements
We thank the wonderful speakers for their enthusiastic participation in the meeting. We thank the Cancer Research Institute of the University of California at Irvine, Debiopharm S.A., and Debioinnovation for their organizational and financial support of this meeting. TJH is an Elizabeth Glaser scientist. We also acknowledge those who provided assistance in the development of this review.
==== Refs
Bruce JW Bradley KA Ahlquist P Young JA Isolation of cell lines that show novel, murine leukemia virus-specific blocks to early steps of retroviral replication J Virol 2005 79 12969 12978 16188999 10.1128/JVI.79.20.12969-12978.2005
Day JR Munk C Guatelli JC The membrane-proximal tyrosine-based sorting signal of human immunodeficiency virus type 1 gp41 is required for optimal viral infectivity J Virol 2004 78 1069 1079 14722262 10.1128/JVI.78.3.1069-1079.2004
Hamburger AE Kim S Welch BD Kay MS Steric accessibility of the HIV-1 gp41 N-trimer region J Biol Chem 2005 280 12567 12572 15657041 10.1074/jbc.M412770200
Sheehy AM Gaddis NC Choi JD Malim MH Isolation of a human gene that inhibits HIV-1 infection and is suppressed by the viral Vif protein Nature 2002 418 646 650 12167863 10.1038/nature00939
Chiu YL Soros VB Kreisberg JF Stopak K Yonemoto W Greene WC Cellular APOBEC3G restricts HIV-1 infection in resting CD4+ T cells Nature 2005 435 108 114 15829920 10.1038/nature03493
Mariani R Chen D Schrofelbauer B Navarro F Konig R Bollman B Munk C Nymark-McMahon H Landau NR Species-specific exclusion of APOBEC3G from HIV-1 virions by Vif Cell 2003 114 21 31 12859895 10.1016/S0092-8674(03)00515-4
Priet S Gros N Navarro JM Boretto J Canard B Querat G Sire J HIV-1-associated uracil DNA glycosylase activity controls dUTP misincorporation in viral DNA and is essential to the HIV-1 life cycle Mol Cell 2005 17 479 490 15721252 10.1016/j.molcel.2005.01.016
Stremlau M Owens CM Perron MJ Kiessling M Autissier P Sodroski J The cytoplasmic body component TRIM5alpha restricts HIV-1 infection in Old World monkeys Nature 2004 427 848 853 14985764 10.1038/nature02343
Ganser-Pornillos BK von Schwedler UK Stray KM Aiken C Sundquist WI Assembly properties of the human immunodeficiency virus type 1 CA protein J Virol 2004 78 2545 2552 14963157 10.1128/JVI.78.5.2545-2552.2004
Zhou J Huang L Hachey DL Chen CH Aiken C Inhibition of HIV-1 maturation via drug association with the viral Gag protein in immature HIV-1 particles J Biol Chem 2005
Scott A Chung HY Gonciarz-Swiatek M Hill GC Whitby FG Gaspar J Holton JM Viswanathan R Ghaffarian S Hill CP Sundquist WI Structural and mechanistic studies of VPS4 proteins Embo J 2005 24 3658 3669 16193069 10.1038/sj.emboj.7600818
Lehrman G Hogue IB Palmer S Jennings C Spina CA Wiegand A Landay AL Coombs RW Richman DD Mellors JW Coffin JM Bosch RJ Margolis DM Depletion of latent HIV-1 infection in vivo: a proof-of-concept study Lancet 2005 366 549 555 16099290 10.1016/S0140-6736(05)67098-5
Haaland RE Herrmann CH Rice AP siRNA depletion of 7SK snRNA induces apoptosis but does not affect expression of the HIV-1 LTR or P-TEFb-dependent cellular genes J Cell Physiol 2005 205 463 470 16152622 10.1002/jcp.20528
Bushman F Lewinski M Ciuffi A Barr S Leipzig J Hannenhalli S Hoffmann C Genome-wide analysis of retroviral DNA integration Nat Rev Microbiol 2005 3 848 858 16175173 10.1038/nrmicro1263
Hout DR Gomez ML Pacyniak E Gomez LM Fegley B Mulcahy ER Hill MS Culley N Pinson DM Nothnick W Powers MF Wong SW Stephens EB Substitution of the transmembrane domain of Vpu in simian-human immunodeficiency virus (SHIV(KU1bMC33)) with that of M2 of influenza A results in a virus that is sensitive to inhibitors of the M2 ion channel and is pathogenic for pig-tailed macaques Virology 2005
VanCompernolle SE Taylor RJ Oswald-Richter K Jiang J Youree BE Bowie JH Tyler MJ Conlon JM Wade D Aiken C Dermody TS KewalRamani VN Rollins-Smith LA Unutmaz D Antimicrobial peptides from amphibian skin potently inhibit human immunodeficiency virus infection and transfer of virus from dendritic cells to T cells J Virol 2005 79 11598 11606 16140737 10.1128/JVI.79.18.11598-11606.2005
Maeda N Fu W Ortin A de las Heras M Fan H Roles of the Ras-MEK-mitogen-activated protein kinase and phosphatidylinositol 3-kinase-Akt-mTOR pathways in Jaagsiekte sheep retrovirus-induced transformation of rodent fibroblast and epithelial cell lines J Virol 2005 79 4440 4450 15767444 10.1128/JVI.79.7.4440-4450.2005
|
16293194
|
PMC1315336
|
CC BY
|
2021-01-04 16:36:41
|
no
|
Retrovirology. 2005 Nov 17; 2:72
|
utf-8
|
Retrovirology
| 2,005 |
10.1186/1742-4690-2-72
|
oa_comm
|
==== Front
Cancer Cell IntCancer Cell International1475-2867BioMed Central London 1475-2867-5-331631862810.1186/1475-2867-5-33EditorialCell biology as the basis of a better understanding of cancer Wheatley Denys N Editor-in-Chief, Cancer Cell [email protected] Director of BioMedES, Leggat House, Keithhall, Inverurie, Aberdeen AB51 0LX, UK2005 30 11 2005 5 33 33 26 8 2005 30 11 2005 Copyright © 2005 Wheatley; licensee BioMed Central Ltd.2005Wheatley; 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.
Clinicians will argue that cancer can only really receive the treatment that is needed through thorough understanding of medicine. However, even empirical approaches to therapy result in experimental analysis of the agencies involved on test cells, usually in culture. From the obverse perspective, cell biologists will argue that until we fully understand cell cycle regulation, tumour management will be too imprecise to make the best advances. A forum is needed whereby the fundamental studies on cells prior to, during and after transformation in vitro can be freely reported (open access) and discussed. The action of anticancer agents and cancer preventative substances can more easily be studied in vitro before the often excessive complexity of making similar studies in experimental and human cancers is tackled. Cancer Cell International is committed to providing such a forum. Ironically within a few months of launching this open access journal, Elsevier had much the same idea, and there one has to pay for the privilege of downloading vital papers in this biomedical field.
==== Body
Cancer journals have an interesting history. Many of them are the official journals from established institutions or societies that have a fine track record in research and medical practice, with an avid and faithful readership. There are probably more journals devoted to cancer than any other field of medical science. The reason is obvious in that cancer is a scourge, a disease that manifests itself in a vast array of different forms and affects young and old human beings, animals, and plants. Perhaps one of the most remarkable things in the whole of biology is the development of an adult organism from an egg, but during this process or at some stage after it is complete, some cells lose their co-ordinates and start to grow anomalously in a relatively unregulated manner. They will continue to grow in circumstances where normal cells would be constrained. So the problem being addressed is one that comes down to the very heart of cellular biology, to the regulation of the cell cycle, the process of differentiation and the control at the next level of organisation in the development of tissues, organs and bodies. Why is it that cancer cells carry on dividing under circumstances where normal cells become constrained? It is a keen academic problem because we seek to find out from the pathology what has gone wrong.
In many ways, advances in cancer research have been rapid largely because we have begun to understand fundamental cell cycle control, genetic alterations consistent with transformation, and the ways in which drugs interact with tumour as well as normal cells. Controversy continues; some believe it is not so much that a series of lesions occur within individual cells themselves as a case of inappropriate communication that leads to "a society of cells" breaking away from the "normal" constraints when certain stresses are placed upon them. Looked at this way, it is perhaps surprising that cancer is not even more rife than we find, especially among longer lived species.
Whatever way we wish to view cancer, it continues to attract good research funding because, although cancer is not the major killer amongst mankind, it is a frightening (malignant) disorder that has for many years not seen novel and better therapies emerging An exception is seen with cervical cancer, but the cost of regular screening and reading of exfoliative cytology samples is both a time-consuming and expensive business, which was supported from early days and has traditionally been maintained.
Much work can be done not only on human and animals cells that have a malignant phenotype, but on quite remote organisms such as yeast and fungi, which give information about control mechanisms and their derangement. The use of animals to test carcinogens and cancer drugs has been drastically reduced, bringing cell models including spheroids (3-D cultures) much more into vogue. The number of articles being produced on cancer each year around the world has reached quite phenomenal proportions, with many journals being devoted to it. But some of those that in recent times used to "serve" the grass-root work on fundamental cancer research have become increasingly selective, in that average to good studies are seen as not sufficiently high profile for the prestigious pages of the best known journals. The result is that many researchers will inevitably have to seek other journals that do not indulge so heavily in "cherry-picking". We also need to have fast publication with open access to the readers without the need to pay (the Varmos principle).
Cancer Cell International was set up with the specific intention of addressing some of the above problems, and was one of the very first niche or specialist – and now referred to as "independent" journals – at BioMed Central . My aim has been to have a journal immediately available to everyone worldwide through the internet that is devoted to the cellular aspects of cancer. Some four months after launch, Elsevier had the same idea and launched Cancer Cell. Cell Press, now part of Elsevier, had developed a strong track record with its publication of Cell and Neuron. Our position turned almost immediately into a David and Goliath encounter, and the saving grace of David persisting is that you have the continued opportunity of an online, immediately accessible, free of charge journal, unlike Cancer Cell, which is both exclusive and expensive. Compare the costs of getting a paper into Cancer Cell – if you are so lucky – with that of getting it into Cancer Cell International. [Forget for a moment the impact factor problem – almost all articles wherever published can be accessed one way or another these days, so it's the article that counts, not the journal.] I call upon those of you who do good scientific work in the field of cancer cell biology to submit manuscripts online to us, and have the type of journal that meets modern demands fairly across the board and for countries less advantaged than those privileged ones of North American and Western Europe. It will not exclude the average and good in order to concentrate on the best. It will review all papers fairly, wherever they come from.
We have no page limit on the number of papers we can publish per month or year. Colour is not an issue, and you retain the copyright of your article. So, why not make the most of this opportunity. We would greatly appreciate seeing your reports, and if you need presubmission appraisal, we can advise you on this matter before you submit online.
|
16318628
|
PMC1315337
|
CC BY
|
2021-01-04 16:40:10
|
no
|
Cancer Cell Int. 2005 Nov 30; 5:33
|
utf-8
|
Cancer Cell Int
| 2,005 |
10.1186/1475-2867-5-33
|
oa_comm
|
==== Front
BMC BiolBMC Biology1741-7007BioMed Central London 1741-7007-3-251628394310.1186/1741-7007-3-25Research ArticleDynamic regulation of integrin activation by intracellular and extracellular signals controls oligodendrocyte morphology Olsen Inger Marie [email protected] Charles [email protected] Departments of Pathology and Medical Genetics, University of Cambridge, Tennis Court Road, Cambridge CB2 1QP, UK2 Centre for Basic Psychiatric Research, Aarhus University Hospital, Skovagervej 2, DK-8240 Risskov, Denmark2005 12 11 2005 3 25 25 17 6 2005 12 11 2005 Copyright © 2005 Olsen and ffrench-Constant; 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
Myelination requires precise control of oligodendrocyte morphology and myelin generation at each of the axons contacted by an individual cell. This control must involve the integration of extracellular cues, such as those on the axon surface, with intrinsic developmental programmes. We asked whether integrins represent one class of oligodendrocyte cell-surface receptors able to provide this integration.
Results
Integrins signal via a process of activation, a conformational change that can be induced either by "outside-in" signals comprising physiological extracellular matrix ligands (mimicked by the pharmacological use of the divalent cation manganese) or "inside-out" signalling molecules such as R-Ras. Increasing levels of outside-in signalling via the laminin receptor α6β1 integrin were found to promote oligodendrocyte processing and myelin sheet formation in culture. Similar results were obtained when inside-out signalling was increased by the expression of a constitutively-active R-Ras. Inhibiting inside-out signalling by using dominant-negative R-Ras reduces processes and myelin sheets; importantly, this can be partially rescued by the co-stimulation of outside-in signalling using manganese.
Conclusion
The balance of the equilibrium between active and inactive integrins regulates oligodendrocyte morphology, which is itself regulated by extrinsic and intrinsic cues so providing a mechanism of signal integration. As laminins capable of providing outside-in signals are present on axons at the time of myelination, a mechanism exists by which morphology and myelin generation might be regulated independently in each oligodendrocyte process.
==== Body
Background
The process of myelination in the CNS requires a remarkable morphological transformation by newly-formed oligodendrocytes, with processes contacting and extending along each axon before elaborating a myelin membrane to enwrap the axon multiple times to create a sheath. This differentiation step is tightly controlled, as indicated by the formation of processes each with sufficient membrane for a sheath thickness that has a precise relationship with final axon diameter [1]. In order to ensure that the precise amount of myelin is formed at the right developmental stage and in the correct place, a key component of oligodendrocyte behaviour during myelin formation must be the integration of multiple extrinsic signals at the axon surface along with intrinsic programmes, such as autonomous developmental timers of differentiation. These points of integration are therefore important for our understanding of myelination and may facilitate the development of strategies to promote remyelination.
One important group of candidate integrative molecules are the integrins, the cell surface receptors of extracellular matrix proteins. Integrins comprise two transmembrane chains, termed α and β, with a ligand-binding site formed by the head domain of the two chains [2]. Recent work has established that integrins exist in at least three different confirmations on the cell surface, each in a dynamic equilibrium with one another (Fig 1A) [3-7]. Inactive integrins are folded over, have a low binding affinity for ligand and do not signal. Primed integrins are straightened, and bind ligand with higher affinity as a result of shape changes within the head domain. Activated integrins have bound ligand leading to receptor clustering, and have undergone a further shape change in the β chain leading to separation of the two cytoplasmic domains, thereby allowing formation of the signalling complex (termed "outside-in" signalling). Since the change of shape can be transmitted across the membrane in either direction, activation can also be achieved by so called "inside-out" signals. These separate cytoplasmic domains and induce changes in the extracellular ligand-binding site that increase receptor affinity, leading to ligand binding, integrin clustering and signalling. As a result, integrin activation and formation of the signalling complex is regulated by the integration of both extrinsic ligand concentrations and the activity of (intrinsic) 'inside out' signalling pathways.
Oligodendrocytes express 5 integrins; αVβ1, αVβ3, αVβ5 and αVβ8 as well as α6β1 [8,9]. In cell culture, the αV integrins promote proliferation and migration [10,11], while the laminin receptor, α6β1, promotes differentiation (as measured by myelin sheet formation) and survival [12,13]. In vivo, mice lacking α6 show increased apoptosis of newly-formed oligodendrocytes [14]. Transplantation of cells expressing dominant-negative β1 integrins into focal demyelinated lesions in the adult rat reduced the extent of remyelination [15]. We therefore focused on α6β1 to examine the hypothesis that integrin activation provides a mechanism for the integration of extrinsic and intrinsic signals in oligodendrocyte differentiation. Our goal was to test two key predictions of the hypothesis (Fig 1B and 1C): (i) that either extracellular or intracellular signals regulating integrin activation can control oligodendrocyte morphology, and (ii), that the effects of inhibition of one class of signals could be partially overcome by stimulation of the other, so demonstrating integration of signals by the integrin.
Results
We have shown previously that laminin-2 substrates promote the formation of a differentiated morphological phenotype in oligodendrocytes, as evidenced by increased elaboration of processes and myelin sheets, and that α6 and β1 integrin subunits are found throughout the process network of oligodendrocytes [12]. To confirm that the change in phenotype was α6β1 integrin dependent (and can therefore be used as an assay of integration of inside-out and outside-in signalling), oligodendrocyte morphology was analyzed in the presence of the monoclonal antibody Ha2/5 (that blocks β1 integrins) and RGD peptides (that mimic the peptide recognition sequence within the extracellular matrix molecules recognised by αV integrins and so block all the oligodendrocyte αV integrins [16]). While confirming previous results with polyclonal anti-β1 antibody, we also found that the monoclonal Ha2/5 antibody blocked morphological differentiation on laminin-2 (Fig 2A). In contrast, this antibody had little effect on fibronectin substrates, which promote process outgrowth but not myelin sheet formation and on which the RGD peptide was an effective inhibitor (Fig 2A). Since αVβ1 and α6β1 are the only β1 integrins expressed on oligodendrocytes, and because RGD peptides would be expected to block αVβ1, we conclude that morphological differentiation leading to myelin sheet formation by oligodendrocytes on laminin-2 is α6β1-dependent.
To confirm that extracellular signals regulate oligodendrocyte morphology by activation of α6β1, oligodendrocyte precursor cells were plated on increasing concentrations of laminin-2 and allowed to differentiate into oligodendrocytes. Oligodendrocyte morphology was scored as having either simple (primary) processes only, complex (secondary and tertiary) processes or myelin sheets, reflecting the different changes associated with axon contact and sheath formation. Increasing the laminin-2 coating concentration from 1 to 10 μg/ml resulted in a significantly greater degree of morphological differentiation (Fig 2B). To exclude the possibility that this represented signalling by a second non-integrin receptor rather than α6β1 integrin, we also examined the effect of adding Mn2+ to the tissue culture media. This divalent cation replaces the Ca2+ normally present in the integrin ligand-binding domain and activates the integrin as a result of the conformational changes associated with the substitution. At low coating concentrations of laminin-2 (0.1 μg/ml), Mn2+ significantly increased morphological differentiation, but had no effect at higher concentrations (Fig 2B). This result would be expected if either high ECM concentrations (1 μg/ml and above) or Mn2+ were sufficient to activate the oligodendrocyte α6β1 integrins effectively, while lower ECM concentrations only elicited partial activation. Consequently, Mn2+ promotes further differentiation at low ECM concentrations (by activating those integrins not already activated by the ECM) but has no significant additive effect at the high ECM concentrations.
The effect of intracellular signalling pathways that regulate integrin activation was explored. Both RT-PCR and western blotting studies confirmed the presence of R-Ras in oligodendroglial cells (Fig 3A,B), shown to regulate integrin activation in other cell types [17,18]. To manipulate R-Ras signalling, transient transfections of primary oligodendrocyte precursor cell populations were used to express either wild-type (control) or mutant forms of R-Ras [19]. Expression of a constitutively active mutant form of R-Ras (38V) resulted in a dramatic increase in sheet formation by the oligodendrocytes (Fig 3C). Quantification revealed that this increase was significantly greater than that seen in cells expressing wild-type R-Ras at the highest laminin-2 concentrations tested, with >80% of the cells expressing sheets at 10 μg/ml (Fig 3F).
Conversely, a dominant negative R-Ras construct (43N) reduced the degree of morphological differentiation seen in response to laminin-2 (Fig 3D). When compared with cells expressing wild-type R-Ras, the percentage of oligodendrocytes with a complex morphology was significantly reduced at laminin-2 concentrations above 0.1 μg/ml, with the percentage exhibiting a simple morphology correspondingly increased (Fig 3G). Together with the data using the constitutively-active R-Ras38V, these results point to the presence on the surface of the oligodendrocyte of a dynamic equilibrium between inactive and activated α6β1 that can be displaced in either direction by manipulation of R-Ras signalling. Moreover, when considered with the experiments using high laminin-2 coating concentrations and/or Mn2+, they confirm our first prediction that either intracellular or extracellular cues regulating activation can control oligodendrocyte morphology.
To test the second prediction of the hypothesis that this equilibrium provides a mechanism for integrating intracellular and extracellular cues, we determined whether the effects of inhibiting activation by dominant negative R-Ras (an "inside-out" signal) could be overcome by stimulating activation using Mn2+ (an "outside-in" signal), as illustrated in Fig 1C. Mn2+ increased the extent of sheet formation, and decreased the percentage of cells with a simple morphology, in cells expressing the dominant negative R-Ras construct at the two lower laminin concentrations tested (Fig 4). Because Mn2+ acts directly on the integrin ligand binding domain, we can conclude that the increase in sheet formation reflects a direct effect on the integrin and therefore on the equilibrium between inactive and active integrins, rather than via modulation of separate downstream signalling pathways.
Discussion
These results show that the equilibrium between active and inactive integrins regulates oligodendrocyte morphology and myelin sheet formation. The balance of this equilibrium is determined by a combination of intracellular signals and extracellular ligand concentrations, with the two classes of signal integrated by individual integrin heterodimers. Manipulation of the equilibrium provides one mechanism by which both the number of myelin sheaths formed by an individual oligodendrocyte, as well as their thickness, could be regulated during CNS development.
To be relevant to myelination in vivo, our results require the presence of signals in the developing brain that can promote both outside-in and inside-out integrin signalling. We have shown previously that laminin α2 (the α chain present in the laminin trimers 2,4 and 12) is present on the surface of axons in embryonic brainstem at the time of myelination [14], extending a previous report of laminin α2 expression by Purkinje cells in the myelinating postnatal cerebellum [20]. Binding of this laminin would provide an outside-in signal, as shown in the experiments using increasing laminin concentrations. A number of different extracellular signals might trigger the intracellular pathways that promote inside-out signalling. For example, our previous work has defined a role for growth factors and their receptors in integrin activation via "inside-out" signalling. In studies on cell proliferation of oligodendrocyte precursors, αvβ3 integrin was activated by PDGF [21]. Activation of αvβ3 had been previously described in response to VEGF in endothelial cells [22], suggesting this might represent a general mechanism for integrin/growth factor interactions. Two other extracellular signalling systems, eph/ephrins and semaphorins, have also been shown to alter integrin activation [23-25]. Equally, the key intracellular components of the inside-out signalling pathway are present in oligodendrocytes. We have reported here that R-Ras is expressed in oligodendroglia. Recent work has shown that the interaction of the cytoskeletal protein talin with the β integrin cytoplasmic domain represents a final common pathway for the signals that promote activation [26], and we have also found that talin is expressed in oligodendroglial cells (Uli Forster and C ff-C, unpublished observations) consistent with such a role in these cells.
These observations show that there are multiple potential signalling pathways that might regulate the equilibrium of active and inactive integrins. Together with our present results, they support a model by which this equilibrium provides a dynamic means of integrating multiple signals to control the timing of myelination and the extent of myelin membrane. The model is based on the assumption that, under physiological conditions during normal development where growth factors are present in limiting concentrations, two classes of signals are required to generate sufficient numbers of activated integrins. So, for example, the laminin on the axon surface would be necessary for integrin signalling and myelination, but may not be sufficient without a further contribution to integrin activation from intracellular ("inside-out") signals. These "inside-out" signals could be generated by growth factor signalling, or alternatively from intracellular pathways driven by autonomous developmental clocks [27], and would then effectively act as timers of myelination. Equally, while these intracellular signals may initiate the myelination process, the amount of myelin formed (and hence the number of wraps) would then be determined by the concentration of extracellular ligand on the axon surface, with higher concentrations increasing integrin activation and so the amount of myelin produced. A precedent for wrap-number being determined by the concentration of axon-surface molecules is provided by the PNS, where transmembrane isoforms of neuregulin regulate Schwann cell myelination [28,29]. There is no evidence that laminins provide quantitative signals to the myelinating oligodendrocytes, but the observation that laminin-2 deficient dy/dy mutant mice show an increased G ratio (reflecting thinner myelin) in the CNS [30] would be consistent with such signals, and further studies are required.
Conclusion
Oligodendrocyte morphology is regulated at least in part by the balance of active and inactive integrins (Fig 1A). Such a model provides important insights into two aspects of oligodendrocyte biology. First, a single oligodendrocyte can myelinate multiple axons, and the thickness of each sheath (and therefore the amount of myelin membrane produced at the end of each process) is independently regulated by the axon it ensheaths. Clearly, this remarkable example of local control of cell-cell interactions cannot be regulated entirely by changes in transcription or translation initiated in the cell body. However, the control of integrin signalling by alterations of the equilibrium between active and inactive integrins provides a plausible mechanism by which integrin ligands on the axon could regulate membrane formation and morphological changes independently within each process. Second, one hypothesis for the failure of remyelination in MS (the dysregulation hypothesis) is that the loss of the coordination between promyelinogenic signals, rather than the loss of any one specific factor, prevents repair [31]. The possibility that integrative signalling molecules such as integrins might provide targets for drug therapies designed to promote remyelination by activation and promotion of signalling in the absence of the usual upstream cues therefore requires further investigation.
Methods
Oligodendrocyte cultures
Purified oligodendrocyte precursor cells were obtained from primary cultures of newborn Sprague-Dawley rat forebrains and were isolated by mechanical dissociation and differential adhesion as described previously [8]. The precursor cells were then plated on 8-well Permanox chamber slides (Nunc) pre-coated with fibronectin (10 μg/mL; Sigma) or laminin-2 (0.01, 0.1, 1 and 10 μg/mL; Sigma human placental laminin) at 20,000 cells/well. For both the blocking experiments and the Mn2+ experiments, the cells were allowed to attach to the substrate for 2 h at 37°C before adding either blocking reagents or Mn2+. Ha2/5 antibody (PharMingen) was used at 10 μg/mL, blocking peptides (RGD and RAD; Calbiochem) were used at 0.1 μg/mL and Mn2+ (MnCl2·4H2O; Sigma) was freshly prepared and used at 50 μg/mL. After 3 days of differentiation in SATO medium [8] at 37°C, 7.5% CO2, cells were fixed with 4% paraformaldehyde for 15 min at room temperature. The cells were permeabilised and blocked in a one-step procedure with 10% normal goat serum (NGS, Sigma) and 0.1% TritonX-100 (NBS Biologicals) in PBS for one hour at room temperature prior to immunostaining.
Transfection
After 10–12 days in vitro, the forebrain cultures were rinsed twice with DMEM (Sigma) supplemented with 10% fetal bovine serum (Sigma) and subsequently transfected overnight with the following constructs; pEGFP-C1/R-Ras-wt, pEGFP-C1/R-Ras-38V and pEGFP-C1/R-Ras-43N using FuGene 6.0 (Roche). The R-Ras constructs have been previously described [19], and were a kind gift of Prof Alan Hall (UCL, London, UK). They were recloned into the pEGFP-C1 vector (Clontech) following digestion with BamHI and XbaI. The cells were then washed three times with DMEM (10% fetal bovine serum) and 24 h later they were plated onto laminin-2 substrates as described above at 40,000 cells/well. The cells were differentiated for 3 days before fixation, permeabilisation and blocking in preparation for the immunostaining.
Immunostaining
Cells were washed three times with PBS and were incubated with appropriate dilutions of the primary antibodies in a buffer of PBS, 1% NGS and 0.1% Triton X-100 for 1–2 h at room temperature or overnight at 4°C. Myelin basic protein (MBP) staining was carried out using a 1:50 dilution of rat anti-MBP (82–87) (Serotec). A 1:100 dilution of mouse anti-2',3'-cyclic nucleotide 3'-phosphohydrolase (CNPase clone 11-5B, Sigma) was used for CNPase staining. For enhanced green fluorescence protein (EGFP) staining rabbit anti-GFP IgG (1 μg/mL corresponding to 1:1000, Molecular Probes) was used. After washing with PBS, the cells were incubated with secondary antibodies (Jackson ImmunoResearch Laboratories INC.) diluted 1:100 in PBS with 1% NGS for 1 h at room temperature.
RT-PCR and Western blotting
Total RNA was isolated from oligodendrocyte precursor cells and cDNA was made using a first strand cDNA synthesis kit (Amersham) according to the manufacturer's instructions. For the RT-PCR the primer sequences were: ADAM9 (control) GCCTGACGAAGCCTACA and GTCCACGCTTCCTCCTAT; R-Ras GGTTGGGAACAAGGCAGATC and GCAGGACACAGGGACATCC. For western blotting, differentiated oligodendrocytes were lysed and scraped off in 250 μL lysis buffer (50 mM Tris-HCl, 5 mM EDTA, 150 mM NaCl, 1% Triton X-100, pH7.4), with the protease inhibitors phenylmethylsulfonyl fluoride (2 mM), pepstatin (1 μg/mL), leupeptin (5 μg/mL) and aprotinin (2 μg/mL)) per 10 cm tissue culture dish. The cell lysates were incubated for 30 min and centrifuged (13,000 rpm) for 10 min at room temperature. Protein concentrations were determined using a detergent compatible Lowry protein assay (BioRad DcProtein assay). Denatured, boiled proteins were separated by SDS-PAGE (12% pre-cast Tris-HCl minigels; BioRad) and transferred onto a 0.45 μm nitrocellulose membrane (HybondC, Amersham). The membrane was blocked in 4% bovine serum albumin (Sigma) in TBST (TBS with 0.1% Tween 20) for 1 h at room temperature, and further incubated with rabbit anti-R-Ras (BD PharMingen; 1:1000) in blocking buffer overnight at 4°C. After washing with TBST, the membrane was incubated with a horseradish peroxidase-conjucated secondary antibody (Amersham) in 2% blocking buffer for 1 h at room temperature. The membrane was washed and the immunoreactive proteins were detected using enhanced chemiluminescence (Amersham).
Scoring
MBP+/EGFP+ cells were assigned to one of three morphological categories. Cells that had only primary processes were assigned to the first category, which was denoted 'simple processes'. Cells with secondary and tertiary processes were assigned to a second category denoted 'complex processes'. Cells with myelin sheets (as evidenced by continuous staining filling in between the processes) were assigned to a third category denoted 'myelin sheets'. In each assay, 50 MBP+/EGFP+ oligodendrocytes were scored, with the results being the M ± SEM of 3 independent experiments.
Authors' contributions
IMO carried out all experimental analyses and drafted the manuscript. C ff-C conceived of the study, participated in its design and coordination and helped to draft the manuscript. Both authors read and approved the final manuscript.
Acknowledgements
We are grateful to Alan Hall (UCL) for providing the R-Ras constructs from which the expression constructs used in this study was made, to Claire Acquaviva (Cambridge) for the pEGFP-C1 expression vector, to John Ivins (Houston) for much helpful advice on integrin activation and to the ffrench-Constant lab for helpful comments on the manuscript. The study was funded by the Danish Research Agency and the Wellcome Trust.
Figures and Tables
Figure 1 Strategies for manipulating integrin activation in oligodendrocytes. Panel A shows the equilibrium between 3 different conformations of integrin; inactive, primed and activated. Only the latter assembles a signalling complex and promotes morphological differentiation of oligodendrocytes, as manifested by complex processes and the formation of myelin sheets. Panel B shows the 2 methods used in this study to promote activation; outside-in signalling using high extracellular matrix (ECM) ligand concentrations or the divalent cation Mn2+ (results in Fig 2), and inside-out signalling using active R-Ras (Fig 3). Panel C illustrates the logic of the experiment shown in Fig 4 to confirm that integrins integrate these extracellular and intracellular cues, showing how outside-in signalling would be predicted to overcome the effects of R-Ras inhibition on oligodendrocyte morphology and myelin sheet formation.
Figure 2 Outside-in integrin signalling in oligodendrocytes. Panel A shows that morphological differentiation of oligodendrocytes on laminin-2 requires integrins. A monoclonal anti-β1 integrin-blocking antibody (Ha2/5) inhibits oligodendrocyte process formation on laminin-2 (Ln) but not fibronectin (Fn). In contrast, RGD peptides that block αV integrins inhibit the morphological differentiation on Fn. No change is seen with control RAD peptides that do not perturb integrin function. Panel B shows that integrin activation by outside-in signalling promotes differentiation. Increasing laminin-2 concentrations are associated with greater process complexity and myelin sheet formation, with significant increases seen between 1 and 10 μg/ml. Mn2+ (Mn) also promotes differentiation, with significant effects seen at the lowest laminin-2 concentration (0.1 μg/ml). Note that the addition of Mn2+ has no significant effect on morphology at the higher laminin concentrations. Single and double asterix indicate significance levels of p < 0.05 and p < 0.01 respectively, calculated using Student's paired t-test. Scale bar in panel A is 20 μm.
Figure 3 Inside-out integrin signalling in oligodendrocytes. Panel A shows an RT-PCR experiment on cDNA obtained from cultured oligodendrocytes, grown as described in the Methods. Lane 1 – size markers (nt); lane 2 – ADAM9 (control); lane 3 – R-Ras. Panel B shows a western blot of lysates from 3 different sets of cultured oligodendrocytes (lanes 1–3) using anti-R-Ras antibodies. All 3 lanes show a band at the predicted size (23 kD) of R-Ras (arrow). Size markers (kDa) shown on the left. These 2 experiments therefore confirm that oligodendrocytes express R-Ras. Panel C shows the effect of expressing constitutively active R-Ras (R-Ras38V) in oligodendrocytes, while Panel D shows the effect of expressing the dominant-negative R-Ras43N; all 4 micrographs in each panel show representative examples of transfected cells visualised by antibodies against EGFP (green) and CNPase (red). Note the different morphologies observed, with elaborate processes and large sheets seen in some cells expressing R-Ras38V and stunted processes in some expressing R-Ras43N. The effects of R-Ras38V and R-Ras43N at different laminin-2 concentrations are quantified in Panels F and G, with control cells (expressing wild-type R-Ras) in these experiments shown in panel E. These show the increased complexity of R-Ras38V cells at all laminin-2 concentrations when compared with wild-type cells, in contrast to the decreased complexity of the R-Ras43N cells. Single and double asterix indicate significance levels of p < 0.05 and p < 0.01 respectively, calculated using Student's paired t-test. Scale bars in panels C and D are 20 μm.
Figure 4 Outside-in integrin signalling rescues the effect of dominant-negative R-Ras43N. In the same set of experiments illustrated in Fig 3, Panel G, addition of Mn2+ significantly increases oligodendrocyte complexity in the cells expressing dominant-negative R-Ras. *p < 0.05.
==== Refs
ffrench-Constant C Colognato H Franklin RJ Neuroscience. The mysteries of myelin unwrapped Science 2004 304 688 689 15118149 10.1126/science.1097851
Hynes RO Integrins: versatility, modulation, and signaling in cell adhesion Cell 1992 69 11 25 1555235 10.1016/0092-8674(92)90115-S
Hynes RO Integrins: bidirectional, allosteric signaling machines Cell 2002 110 673 687 12297042 10.1016/S0092-8674(02)00971-6
Liddington RC Ginsberg MH Integrin activation takes shape J Cell Biol 2002 158 833 839 12213832 10.1083/jcb.200206011
Giancotti FG A structural view of integrin activation and signaling Dev Cell 2003 4 149 151 12586058 10.1016/S1534-5807(03)00034-0
Xiao T Takagi J Coller BS Wang JH Springer TA Structural basis for allostery in integrins and binding to fibrinogen-mimetic therapeutics Nature 2004 432 59 67 15378069 10.1038/nature02976
Mould AP Humphries MJ Cell biology: adhesion articulated Nature 2004 432 27 28 15525967 10.1038/432027a
Milner R ffrench-Constant C A developmental analysis of oligodendroglial integrins in primary cells: changes in alpha v-associated beta subunits during differentiation Development 1994 120 3497 3506 7821217
Milner R Frost E Nishimura S Delcommenne M Streuli C Pytela R ffrench-Constant C Expression of alpha vbeta3 and alpha vbeta8 integrins during oligodendrocyte precursor differentiation in the presence and absence of axons Glia 1997 21 350 360 9419010 10.1002/(SICI)1098-1136(199712)21:4<350::AID-GLIA2>3.0.CO;2-7
Milner R Edwards G Streuli C ffrench-Constant C A role in migration for the alpha V beta 1 integrin expressed on oligodendrocyte precursors J Neurosci 1996 16 7240 7252 8929432
Blaschuk KL Frost EE ffrench-Constant C The regulation of proliferation and differentiation in oligodendrocyte progenitor cells by alphaV integrins Development 2000 127 1961 1969 10751184
Buttery PC ffrench-Constant C Laminin-2/integrin interactions enhance myelin membrane formation by oligodendrocytes Mol Cell Neurosci 1999 14 199 212 10576890 10.1006/mcne.1999.0781
Frost EE Buttery PC Milner R ffrench-Constant C Integrins mediate a neuronal survival signal for oligodendrocytes Curr Biol 1999 9 1251 1254 10556090 10.1016/S0960-9822(99)80506-5
Colognato H Baron W Avellana-Adalid V Relvas JB Evercooren AB Georges-Labouesse E ffrench-Constant C CNS integrins switch growth factor signalling to promote target-dependent survival Nat Cell Biol 2002 14 14
Relvas JB Setzu A Baron W Buttery PC LaFlamme SE Franklin RJ ffrench-Constant C Expression of dominant-negative and chimeric subunits reveals an essential role for beta1 integrin during myelination Curr Biol 2001 11 1039 1043 11470408 10.1016/S0960-9822(01)00292-5
Ruoslahti E RGD and other recognition sequences for integrins Annu Rev Cell Dev Biol 1996 12 697 715 8970741 10.1146/annurev.cellbio.12.1.697
Zhang Z Vuori K Wang H Reed JC Ruoslahti E Integrin activation by R-ras Cell 1996 85 61 69 8620538 10.1016/S0092-8674(00)81082-X
Ivins JK Yurchenco PD Lander AD Regulation of neurite outgrowth by integrin activation J Neurosci 2000 20 6551 6560 10964960
Self AJ Caron E Paterson HF Hall A Analysis of R-Ras signalling pathways J Cell Sci 2001 114 1357 1366 11257001
Powell SK Williams CC Nomizu M Yamada Y Kleinman HK Laminin-like proteins are differentially regulated during cerebellar development and stimulate granule cell neurite outgrowth in vitro J Neurosci Res 1998 54 233 247 9788282 10.1002/(SICI)1097-4547(19981015)54:2<233::AID-JNR11>3.0.CO;2-5
Baron W Shattil SJ ffrench-Constant C The oligodendrocyte precursor mitogen PDGF stimulates proliferation by activation of alpha(v)beta3 integrins Embo J 2002 21 1957 1966 11953315 10.1093/emboj/21.8.1957
Byzova TV Goldman CK Pampori N Thomas KA Bett A Shattil SJ Plow EF A mechanism for modulation of cellular responses to VEGF: activation of the integrins Mol Cell 2000 6 851 860 11090623
Huai J Drescher U An ephrin-A-dependent signaling pathway controls integrin function and is linked to the tyrosine phosphorylation of a 120-kDa protein J Biol Chem 2001 276 6689 6694 11053419 10.1074/jbc.M008127200
Davy A Robbins SM Ephrin-A5 modulates cell adhesion and morphology in an integrin-dependent manner Embo J 2000 19 5396 5405 11032807 10.1093/emboj/19.20.5396
Serini G Valdembri D Zanivan S Morterra G Burkhardt C Caccavari F Zammataro L Primo L Tamagnone L Logan M Class 3 semaphorins control vascular morphogenesis by inhibiting integrin function Nature 2003 424 391 397 12879061 10.1038/nature01784
Tadokoro S Shattil SJ Eto K Tai V Liddington RC de Pereda JM Ginsberg MH Calderwood DA Talin binding to integrin beta tails: a final common step in integrin activation Science 2003 302 103 106 14526080 10.1126/science.1086652
Durand B Raff M A cell-intrinsic timer that operates during oligodendrocyte development Bioessays 2000 22 64 71 10649292 10.1002/(SICI)1521-1878(200001)22:1<64::AID-BIES11>3.0.CO;2-Q
Garratt AN Voiculescu O Topilko P Charnay P Birchmeier C A dual role of erbB2 in myelination and in expansion of the schwann cell precursor pool J Cell Biol 2000 148 1035 1046 10704452 10.1083/jcb.148.5.1035
Michailov GV Sereda MW Brinkmann BG Fischer TM Haug B Birchmeier C Role L Lai C Schwab MH Nave KA Axonal neuregulin-1 regulates myelin sheath thickness Science 2004 304 700 703 Epub 2004 Mar 2025. 15044753 10.1126/science.1095862
Chun SJ Rasband MN Sidman RL Habib AA Vartanian T Integrin-linked kinase is required for laminin-2-induced oligodendrocyte cell spreading and CNS myelination J Cell Biol 2003 163 397 408 14581460 10.1083/jcb.200304154
Franklin RJ Why does remyelination fail in multiple sclerosis? Nat Rev Neurosci 2002 3 705 714 12209119 10.1038/nrn917
|
16283943
|
PMC1315338
|
CC BY
|
2021-01-04 16:02:56
|
no
|
BMC Biol. 2005 Nov 12; 3:25
|
utf-8
|
BMC Biol
| 2,005 |
10.1186/1741-7007-3-25
|
oa_comm
|
==== Front
BMC Ear Nose Throat DisordBMC Ear, Nose and Throat Disorders1472-6815BioMed Central London 1472-6815-5-101631646910.1186/1472-6815-5-10Research ArticlePharmacological reversal of endothelin-1 mediated constriction of the spiral modiolar artery: a potential new treatment for sudden sensorineural hearing loss Scherer Elias Q [email protected] Wolfgang [email protected] Philine [email protected] Cell Physiology Laboratory, Dept. Anatomy & Physiology, Kansas State University, Manhattan, KS 66506, USA2 Department of Otorhinolaryngology, Head and Neck Surgery, Technical University of Munich, Klinikum rechts der Isar, Ismaninger Str. 22, 81675 Munich, Germany2005 29 11 2005 5 10 10 16 6 2005 29 11 2005 Copyright © 2005 Scherer 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
Vasospasm of the spiral modiolar artery (SMA) may cause ischemic stroke of the inner ear. Endothelin-1 (ET-1) induces a strong, long-lasting constriction of the SMA by increasing contractile apparatus Ca2+ sensitivity via Rho-kinase. We therefore tested several Rho-kinase inhibitors and a cell-permeable analogue of cAMP (dbcAMP) for their ability to reverse ET-1-induced constriction and Ca2+-sensitization.
Methods
The present study employed SMA isolated from gerbil temporal bones. Ca2+sensitivity was evaluated by correlating vascular diameter and smooth muscle cell [Ca2+]i, measured by fluo-4-microfluorometry and videomicroscopy.
Results
The Rho-kinase inhibitors Y-27632, fasudil, and hydroxy-fasudil reversed ET-1-induced vasoconstriction with an IC50 of 3, 15, and 111 μmol/L, respectively. DbcAMP stimulated a dose-dependent vasodilation (Ec50 = 1 mmol/L) and a reduction of [Ca2+]i (EC50 = 0.3 μmol/L) of ET-1-preconstricted vessels (1 nmol/L). Fasudil and dbcAMP both reversed the ET-1-induced increase in Ca2+ sensitivity.
Conclusion
Rho-kinase inhibition and dbcAMP reversed ET-1-induced vasoconstriction and Ca2+-sensitization. Therefore, Rho-kinase inhibitors or cAMP modulators could possess promise as pharmacological tools for the treatment of ET-1-induced constriction, ischemic stroke and sudden hearing loss.
==== Body
Background
The inner ear's blood supply depends solely on the spiral modiolar artery (SMA), a functional end artery. Vasospasm/constriction of the SMA can cause an ischemic stroke of the inner ear, leading to sudden sensorineural hearing loss (SSHL). Thus, investigating the mechanisms controlling the inner ear microcirculation is a prerequisite for the development of new strategies to treat SSHL.
Capillary blood flow is primarily regulated by the resistance of precapillary arteries. The vascular resistance is a function of the contractile status of the vascular smooth muscle cells (VSMCs). Constriction of VSMCs results from an increase in intracellular Ca2+ ([Ca2+]i) and/or by an increase in the Ca2+ sensitivity of the contractile apparatus [1,2]. One key mechanism enhancing Ca2+ sensitivity and thus vascular tone is Rho-kinase signalling, which results in inhibition of myosin light chain phosphatase [2,3]. Rho-kinase activation has been shown to cause vasospasm of coronary, cerebral and spiral modiolar arteries [4-9].
One of the strongest Rho-kinase activators described so far is the vasoconstrictor endothelin-1 (ET-1). The synthesis of ET-1 by endothelial cells is activated by physiological stimuli such as shear stress, insulin, thrombin and other vascular factors [10]. ET-1 and ETA receptors play a fundamental role in the maintenance of basal vasomotor tone in resistance arteries [11]. The synthesis of ET-1 can be increased by hypoxia and elevated oxidized low-density lipoproteins [12,13] and has been implicated in the pathogenesis of a number of cerebrovascular disorders, including stroke, ischemia, and, in particular, cerebral vasospasm [14,15]. Thus, ET-1 possesses pathological potential in addition to its physiological functions. ET-1 is present in the SMA and induces strong, long-lasting constriction via ETA-receptor-mediated Rho-kinase activation [9,16,17]. Taken together ET-1 is likely an endogenous regulator of inner ear microvascular tone.
We have previously shown that CGRP is able to reverse ET-1-induced constrictions in the SMA via an increase in vascular smooth muscle cAMP [18]. CGRP is present in perivascular nerves of the SMA and therefore is a potential endogenous vasodilator of the SMA. We propose, therefore, that reversal of ET-1-induced constriction is not necessarily limited to inhibition of ET-1-related mechanisms (e.g., Rho-kinase signalling).
These findings provide a clinical perspective for a new treatment of SSHL, because both Rho-kinase signalling and cAMP can be targeted via pharmacological agents. Therefore, we assessed the potency of clinically relevant Rho-kinase inhibitors and a cell-permeable analogue cAMP (dbcAMP) in terms of reversing ET-1-induced constriction and Ca2+-sensitization in the SMA.
Methods
Drugs and solutions
The physiologic salt solution (PSS) contained (in mmol/L) 150 NaCl, 3.6 KC1, 1.0 MgCl2, 1.0 CaCl2, 5.0 HEPES, and 5.0 glucose, pH 7.4. Extracellular Ca2+ concentration ([Ca2+]ex) was raised to 3 and 10 mmol/L by addition of CaCl2. A maximal vasodilation was induced by the removal of extracellular Ca2+. The nominally Ca2+-free solution contained (in mmol/L) 150 NaCl, 3.6 KC1, 1.0 MgCl2, 1.0 EGTA, 5.0 HEPES, and 5.0 glucose, pH = 7.4. Fluo-4-AM (Molecular Probes) was dissolved in anhydrous DMSO and stored in 1 mmol/L aliquots. Y-27632 was kindly provided by Welfide. Fasudil was obtained from Calbiochem. Fasudil (obtained from Tocris Cookson) was modified to hydroxyfasudil by Dr. Duy Hua, Dept of Chemistry, Kansas State University. All other chemicals were obtained from Sigma.
Preparation of the spiral modiolar artery (SMA)
Experiments were conducted on tissues isolated from gerbils under a protocol that was approved by the Institutional Animal Care and Use Committee at Kansas State University. Gerbils were anesthetized with sodium pentobarbital (100 mg/kg i.p.) and decapitated. Temporal bones were removed, opened and placed into a micro-dissection chamber containing PSS at 4°C. The SMA was isolated from the cochlea by micro-dissection as described previously [19]. Briefly, the cochlea was opened. The bone surrounding the modiolus was carefully removed and the SMA, which is only loosely attached to the eighth cranial nerve, was isolated. Care was taken to not stretch the artery.
Simultaneous measurement of vascular diameter and [Ca2+]i
The simultaneous measurement of vascular diameter and [Ca2+]i has been described previously [17]. Briefly, the smooth muscle cells of vessel segments were loaded with the Ca2+ indicator dye fluo-4 by incubation in PSS containing 5 μmol/L fluo-4-AM for 35 min at 37°C. After loading, vessel segments were washed with PSS and maintained at 4°C for 20 minutes prior to experimentation at 37°C. Vessel segments were transferred into a bath chamber mounted on the stage of an inverted microscope (Nikon). Fluorescence emitted by fluo-4 (518–542 nm) in response to excitation at 488 nm (Photon Technology International) was detected by a photon counter (Photon Technology International). For measurements of the vascular diameter, the vessel was illuminated at 605–615 nm and the transmission image was recorded with a chilled CCD camera (Hamamatsu). The outer vascular diameter was measured by two video edge detectors (Crescent). Fluorescence and calibrated diameter signals were digitized and recorded simultaneously (Photon Technology International).
Experimental protocols
Experiments were started 20 min after loading with fluo-4. Vessel segments were superfused at a rate of 9 ml/min with PSS. This flow rate corresponds to an exchange rate of 2 bath chambers volumes/sec, given the bath chamber volume of 75 μl. Upon start of the superfusion, the unpressurized artery develops a spontaneous vascular tone that is sensitive to removal of extracellular Ca2+ and inhibition of L-type Ca2+ channels with nanomolar concentrations of nifedipine [19]. The viability of each vessel was assessed by its constrictor response to 10 mmol/L [Ca2+]ex. The [Ca2+]i was monitored as fluorescence intensity and was normalized to the basal fluorescent emission prior to the beginning of each experiment. The fluorescence and diameter values taken for statistical analysis represent averages of the [Ca2+]i-fluorescence and the vascular diameter over 30 sec beginning 30 sec after the onset of stimulation. We carefully validated the [Ca2+]i -measurements with the non-ratiometric Ca2+-dye fluo-4 by (i) excluding artefacts due to vessel diameter changes (evaluated by [Ca2+]ex-versus ET-1 dose-response curves), (ii) excluding significant differences in the magnitude of [Ca2+]i -changes between preparations, and (iii) assessing the same optimal dye-loading conditions in each experiment [9,17].
Affinity constants (KDB) and concentrations that cause a half-maximal inhibition (IC50) were determined in cumulative experiments and averaged after logarithmic transformation (pKDB and pIC50) as previously described [17].
The Ca2+ sensitivity of the contractile apparatus was determined by a correlation of [Ca2+]i and the vascular diameter as described previously [9]. Changes in [Ca2+]i were induced by changes in [Ca2+]ex. Stepwise increases in [Ca2+]ex from 0 to 1, 3 and 10 mmol/L caused increases in [Ca2+]i and decreases in the vascular diameter. Correlations were found to be linear (r > 0.95) within the measured range (the relation between calcium concentration and tension gets sigmoidal if the [Ca2+]ex and thus the [Ca2+]i is further increased [9]). Slopes were quantified in the arbitrary unit μm/Ca2+, where μm represent the change in the vascular diameter and Ca2+ represents the normalized change in the cytosolic Ca2+ concentration. Linear slopes were obtained to compare the Ca2+ sensitivity within paired experiments. In each vessel segment, the Ca2+ sensitivity was assessed under control and under experimental conditions.
Experiments of dbcAMP induced dilations were bracketed by Ca2+ free manoeuvres that were performed to induce a maximal vasodilation and reduce the [Ca2+]i to a minimal level. Vascular diameter or fluorescence intensity in the absence of Ca2+ was considered as baseline. The magnitude of the vascular tone or fluorescence intensity was determined as the difference between the recorded value and the baseline value. Measurements were averaged over a period of 1 min to average vasomotion. DbcAMP-induced effects were normalized to the magnitude of the vascular tone or fluorescence intensity during 1 min immediately prior to the application of dbcAMP.
Statistical analysis
All results are expressed as average ± SEM of n experiments with n representing the number of vessel segments. The significance of changes in the vascular diameter and of changes in the Ca2+ sensitivity were determined using Student's paired t-test. Differences were considered to be significant at error probabilities less than 0.05 (P < 0.05).
Results
This report is based on recordings of 50 vessels from 34 animals. The average vascular diameter was 65 ± 1 μm.
ET-1-induced constriction is reversed via Rho-kinase inhibitors
ET-1 (10 nmol/L) induced a transient increase in [Ca2+]i, a strong and long-lasting vasoconstriction and a robust increase in the vasomotion of the gerbil spiral modiolar artery (Fig. 1A). The [Ca2+]i returned to almost resting levels after the transient increase, while the constriction was maintained. The ET-1-induced vasoconstriction was not readily reversible upon removal of ET-1 from the perfusate. The constriction and the increased vasomotion were observed without a significant change for at least 20 minutes after removal of ET-1 from the superfusate (data not shown). Note that ET-1-induced a transient [Ca2+]i increase and a sustained vasoconstriction while exposure to 10 mmol/l Ca2+ induced an increase in [Ca2+]i and a parallel vasoconstriction.
We tested the potency of different Rho-kinase inhibitors reversing ET-1-induced constriction. Figure 1B shows an original recording of an ET-1-induced constriction which is antagonized by increasing concentrations of fasudil. Fasudil mediated vasodilation were induced without significantly altering [Ca2+]i-levels. The Rho-kinase inhibitors Y-27632, fasudil and hydroxy-fasudil reversed ET-1-induced constriction (10 nmol/L) in a dose-dependent manner (Figure 1C). The 1C50 for Y-27632-, fasudil- and hydroxy-fasudil- mediated reversion of constriction was 3 μmol/L (pIC50 = 5.50 ± 0.31; n = 6), 15 μmol/L (pIC50 = 4.71 ± 0.13; n = 7) and 111 μmol/L (pIC50 = 3.95 ± 0.24; n = 6), respectively. The Ca2+ sensitivity of the contractile apparatus was assessed as linear slopes obtained from correlations of [Ca2+]i and vascular diameter. ET-1 (100 pmol/L) increased the Ca2+ sensitivity (-36 ± 9 versus -62 ± 13 μm/Ca2+, n = 8; Figure 1D), fasudil (3 μmol/L) prevented the ET-1-induced increase in the Ca2+ sensitivity (-17 ± 3 versus -16 ± 2 μm/Ca2+, n = 8, Figure 1E). Taken together, these observations demonstrate that ET-1-induced constriction in the SMA is maintained by a Rho-kinase-mediated increase of the Ca2+ sensitivity of the contractile apparatus, which can be effectively reversed by Rho-kinase inhibition.
Exogenous, cell-permeable cAMP (dbcAMP) reverses ET-1-induced constriction and Ca2+sensitization
DbcAMP induced dose-dependant decreases in [Ca2+]i and reversal of constriction induced by 1 nmol/L ET-1 with an EC50 of 1 mmol/L and 0.3 μmol/L (pEC50 = 2.97 ± 0.09 and 6.49 ± 0.07, n = 8, Fig. 2A and 2B), respectively. Note that dbcAMP-induced decreases in [Ca2+]i were less pronounced than those induced by removal of Ca2+ from the extracellular solution although dilations were comparable (Fig. 2A). This observation suggests that the dilatory effect of cAMP is at least in part due to a decrease of the Ca2+ sensitivity of the contractile apparatus. Consistent with this interpretation is the apparent rightward shift of the dose-response curve of dbcAMP for change in [Ca2+]i (Fig. 2B). If cAMP and ET-1 have opposing effects on the Ca2+ sensitivity, it should be possible to prevent the ET-1-induced increase in the Ca2+ sensitivity with dbcAMP. Indeed, 10 μmol/L dbcAMP prevented the increase in the Ca2+ sensitivity induced by 100 pmol/L ET-1 (-29 ± 4 versus -27 ± 3 μm/Ca2+, n = 7; Figure 2C). These observations support the hypothesis that cAMP and Rho-kinase can interact as functional antagonists at the level of the Ca2+ sensitivity of the contractile apparatus.
Discussion
The primary observation of the present study is that the ET-1-induced SMA vasoconstriction is reversed by the Rho-kinase inhibitors Y-27632, fasudil and hydroxy-fasudil and by the cAMP analogue dbcAMP. All four agents decreased VSMC contractile apparatus Ca2+ sensitivity.
In general, ET-1-induced constriction has been found to be elicited by different Ca2+ mobilizing mechanisms, including Ca2+ release from intracellular Ca2+ stores via a phospholipase C mediated activation of IP3-receptors and activation of L-type and non-selective Ca2+ channels [20-22]. Ca2+ mobilization has been considered to be the main mechanism of ET-1-induced constriction. In contrast, according to our previous results, ET-1-induced Ca2+ mobilization in the SMA appears to play a minor role [17]. The major mechanism of ET-1-induced constriction is an increase in the Ca2+ sensitivity of the contractile apparatus [9].
The increase in the Ca2+ sensitivity appears to be mediated by a Rho-kinase dependent inactivation of MLCP. This hypothesis is supported by two observations. First, inhibition of Rho-kinase with the selective Rho-kinase inhibitor Y-27632 abolished the ET-1-induced increase in the Ca2+ sensitivity. Y-27632 at concentrations of up to 10 μmol/L has been shown to be a selective Rho-kinase inhibitor [23-25]. Second, inhibition of Rho-kinase with the selective Rho-kinase inhibitors fasudil and its functional metabolite hydroxy-fasudil also reversed ET-1-induced constriction and Ca2+ sensitization of the contractile apparatus. These inhibitors have also been shown to be selective Rho-kinase inhibitors up to 20 μmol/L [24,26,27]. Since Y-27632 and the fasudil derivatives are structurally unique, but were observed to have similar functional effects, the concern of non-specific effects of the inhibitors can be minimized.
It has been shown that the Rho-kinase-dependent inhibition of the MLCP results from phosphorylation of the myosin-binding subunit (MBS) of the enzyme [28]. All Rho-kinase inhibitors employed in the present study reversed ET-1-induced constriction, with a clinically relevant EC50 range below the level reported to cause systemic side effects (especially hypotension). This possibly indicates that basal Rho-kinase activity is rather low under normal conditions [23,29]. Thus, an up-regulation of Rho-kinase expression/activity under pathophysiological conditions (hypertension, cerebral and coronary vasospasm) would impart Rho-kinase inhibitors with pharmacological relevance [29].
We show that cAMP is a potent vasodilating second messenger in the SMA, acting by mechanisms which decrease [Ca2+]i and Ca2+ sensitivity. Although cAMP reduces intracellular Ca2+ levels, the present data demonstrate that the main pathway targets the Ca2+ sensitivity of the contractile apparatus. This interpretation is consistent with findings obtained in other preparations such as permeabilized intestinal and bronchial smooth muscle. In these studies the authors also could show that cAMP mediates a Ca2+ desensitization of the contractile apparatus [30,31]. cAMP seems to be the main second messenger of CGRP-induced Ca2+ desensitization and vasodilation in the SMA [18]. The data supports our hypothesis that the reversal of ET-1-induced constriction is not limited to inhibition of ET-1-related mechanisms (e.g., Rho-kinase signalling) and that ET-1-independent mechanisms can be targeted pharmacologically to reverse ET-1-mediated constriction. Clinical relevance arises from the possibility of modulating pharmacologically VSMC cAMP concentrations.
Conclusion
Although the sudden loss of hearing causes substantial distress and pronounced long-term effects in affected individuals, adequate strategies to clinically treat the disorder are lacking. One subgroup of SSHL is believed to arise from SMA vasospasm(s) which ultimately lead to ischemic stroke of the inner ear [32]. The targeting of two distinct signalling mechanisms, Rho-kinase and cAMP, were both effective in reversing SMA constriction. Of note, two of the agents employed in the present study (fasudil and hydroxy-fasudil) are currently used in the clinical setting and were found to be effective at concentrations below the threshold of systemic side effects [8,33]. Thus, this study presents two novel, clinically relevant approaches for the treatment of SSHL. We therefore propose that clinical investigation into the use of these agents for SSHL treatment is warranted.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
EQS conducted all of the experiments and was involved in all aspects of data analysis and manuscript preparation. PW assisted with all facets of this study, assisting with data analysis and manuscript preparation. PW also provided laboratory space, equipment and financial support. WA provided financial support and made substantial contributions to conception and design of the present study.
All authors have been involved in drafting the manuscript and revising it critically for important intellectual content and given final approval of the version to be published.
Pre-publication history
The pre-publication history for this paper can be accessed here:
Acknowledgements
The support by Research Grant RO1-DC04280 from the National Institute on Deafness and Other Communication Disorders, National Institutes of Health is gratefully acknowledged. The authors thank Dr. Steffen-Sebastian Bolz and Dr. Darcy Lidington for critically reviewing this manuscript and for giving important suggestions for improvement.
Figures and Tables
Figure 1 Rho-kinase inhibitors reverse ET-1-induced constriction. A) Effect of 10 nmol/L endothelin-1 (ET-1) on the cytosolic Ca2+ concentration ([Ca2+]i) and the vascular diameter of the spiral modiolar artery. Note that ET-1 caused a transient increase in [Ca2+]i and a sustained vasoconstriction and increase in vasomotion. This constriction outlasted the stimulation with ET-1. Increasing extracellular Ca2+ concentration ([Ca2+]ex) to 10 mmol/L ("Ca") induced an increase in [Ca2+]i with parallel vasoconstriction. This increase of the [Ca2+]ex from 1 to 10 mmol/L ("Ca") served as a control experiment. Measurements of [Ca2+]i were normalized to the value obtained prior to the admission of ET-1 (value at time 'x' was set to 1). B) In the presence of 10 nmol/L ET-1, increasing concentrations of fasudil (1–100 μmol/L) induce dose-dependent relaxation. [Ca2+]i values were normalized, with the baseline value prior to application of ET-1 designated as 1.0 (denoted as "x"). C) Dose-response curves for Y-27632-, fasudil- and hydroxy-fasudil-induced reversal of ET-1 (10 nmol/L) -mediated vasoconstriction. C) ET-1 stimulated a significant increase in VSMC contractile apparatus Ca2+ sensitivity. D) Fasudil completely reversed the ET-1-induced Ca2+-sensitization.
Figure 2 dbcAMP reverses ET-1 (1 nmol/L) -induced vasoconstriction in a dose-dependent manner. A) A representative recording of [Ca2+]i-diameter measurements. The [Ca2+]i was normalized to the value prior to application of ET-1 (value at timepoint "x" was set to 1.0). Maximal vasodilation was determined by exposure of the SMA to Ca-free conditions ("0Ca"). An increase of extracellular Ca2+ to 10 mmol/L ("Ca") induced a [Ca2+]i increase and parallel vasoconstriction. B) Dose-response curves of dbcAMP-induced vasodilation and [Ca2+]i decrease of ET-1 (1 nmol/L) preconstricted SMA. C) dbcAMP prevented ET-1-induced Ca2+-sensitization.
==== Refs
Hartshorne DJ Johnson LR Biochemistry of the Contractile Process in Smooth Muscle Physiology of the Gastrointestinal Tract 1987 New York: Raven Press 423 482
Somlyo AP Somlyo AV Ca2+ Sensitivity of Smooth Muscle and Nonmuscle Myosin II: Modulated by G Proteins, Kinases, and Myosin Phosphatase Physiol Rev 2003 83 1325 1358 14506307
Kimura K Ito M Amano M Chihara K Fukata Y Nakafuku M Yamamori B Feng J Nakano T Okawa K Iwamatsu A Kaibuchi K Regulation of myosin phosphatase by Rho and Rho-associated kinase (Rho- kinase) [see comments] Science 1996 273 245 248 8662509
Sato M Tani E Fujikawa H Kaibuchi K Involvement of Rho-kinase-mediated phosphorylation of myosin light chain in enhancement of cerebral vasospasm Circ Res 2000 87 195 200 10926869
Fassbender K Hodapp B Rossol S Bertsch T Schmeck J Schutt S Fritzinger M Horn P Vajkoczy P Wendel-Wellner M Ragoschke A Kuehl S Brunner J Schurer L Schmiedeck P Hennerici M Endothelin-1 in subarachnoid hemorrhage: An acute-phase reactant produced by cerebrospinal fluid leukocytes Stroke 2000 31 2971 2975 11108758
Kandabashi T Shimokawa H Miyata K Kunihiro I Kawano Y Fukata Y Higo T Egashira K Takahashi S Kaibuchi K Takeshita A Inhibition of Myosin Phosphatase by Upregulated Rho-Kinase Plays a Key Role for Coronary Artery Spasm in a Porcine Model With Interleukin-1 {beta} Circulation 2000 101 1319 1323 10725293
Shimokawa H Rho-kinase as a Novel Therapeutic Target in Treatment of Cardiovascular Diseases J Cardiovasc Pharmacol 2002 39 319 327 11862109 10.1097/00005344-200203000-00001
Masumoto A Mohri M Shimokawa H Urakami L Usui M Takeshita A Suppression of Coronary Artery Spasm by the Rho-Kinase Inhibitor Fasudil in Patients With Vasospastic Angina Circulation 2002 105 1545 1547 11927519 10.1161/hc1002.105938
Scherer EQ Herzog M Wangemann P Endothelin-1-Induced Vasospasms of Spiral Modiolar Artery Are Mediated by Rho-Kinase-Induced Ca2+ Sensitization of Contractile Apparatus and Reversed by Calcitonin Gene-Related Peptide Stroke 2002 33 2965 2971 12468798 10.1161/01.STR.0000043673.22993.FD
Levin ER Endothelins N Engl J Med 1995 333 356 363 7609754 10.1056/NEJM199508103330607
Haynes WG Endothelins as regulators of vascular tone in man Clin Sci (Colch) 1995 88 509 517
Rakugi H Tabuchi Y Nakamaru M Nagano M Higashimori K Mikami H Ogihara T Suzuki N Evidence for endothelin-1 release from resistance vessels of rats in response to hypoxia Biochem Biophys Res Commun 1990 169 973 977 2194458 10.1016/0006-291X(90)91989-6
Xie H Bevan JA Oxidized low-density lipoprotein enhances myogenic tone in the rabbit posterior cerebral artery through the release of endothelin-1 Stroke 1999 30 2423 2429 10548680
Chow M Dumont AS Kassell NF Endothelin receptor antagonists and cerebral vasospasm: an update Neurosurgery 2002 51 1333 1341 12445337 10.1097/00006123-200212000-00002
Lan C Das D Wloskowicz A Vollrath B Endothelin-1 modulates hemoglobin-mediated signaling in cerebrovascular smooth muscle via RhoA/Rho kinase and protein kinase C Am J Physiol Heart Circ Physiol 2004 286 H165 H173 14500131 10.1152/ajpheart.00664.2003
Jinnouchi K Mechanism of endothelin 1 production in the cochlea of rats ORL J Otorhinolaryngol Relat Spec 2001 63 6 11 11174056
Scherer EQ Wonneberger K Wangemann P Differential desensitization of Ca2+ mobilization and vasoconstriction by ETA receptors in the gerbil spiral modiolar artery J MembrBiol 2001 182 183 191
Herzog M Scherer EQ Albrecht B Rorabaugh B Scofield MA Wangemann P CGRP Receptors in the Gerbil Spiral Modiolar Artery Mediate a Sustained Vasodilation via a Transient cAMP-mediated Ca2+-decrease J Membr Biol 2002 189 225 236 12395287 10.1007/s00232-002-1017-5
Wangemann P Cohn ES Gruber DD Gratton MA Ca2+-dependence and nifedipine-sensitivity of vascular tone and contractility in the isolated superfused spiral modiolar artery in vitro Hear Res 1998 118 90 100 9606064 10.1016/S0378-5955(98)00017-3
Goto K Kasuya Y Matsuki N Takuwa Y Kurihara H Ishikawa T Kimura S Yanagisawa M Masaki T Endothelin activates the dihydropyridine-sensitive, voltage-dependent Ca2+ channel in vascular smooth muscle Proc Natl Acad Sci USA 1989 86 3915 3918 2542956
Enoki T Miwa S Sakamoto A Minowa T Komuro T Kobayashi S Ninomiya H Masaki T Functional coupling of ETA receptor with Ca(2+)-permeable nonselective cation channel in mouse fibroblasts and rabbit aortic smooth-muscle cells J Cardiovasc Pharmacol 1995 26 S258 S261 8587381
Guibert C Beech DJ Positive and negative coupling of the endothelin ETA receptor to Ca2+-permeable channels in rabbit cerebral cortex arterioles J Physiol (Lond) 1999 514 843 856 9882755 10.1111/j.1469-7793.1999.843ad.x
Uehata M Ishizaki T Satoh H Ono T Kawahara T Morishita T Tamakawa H Yamagami K Inui J Maekawa M Narumiya S Calcium sensitization of smooth muscle mediated by a Rho-associated protein kinase in hypertension Nature 1997 389 990 994 9353125 10.1038/40187
Davies SP Reddy H Caivano M Cohen P Specificity and mechanism of action of some commonly used protein kinase inhibitors Biochem J 2000 351 95 105 10998351 10.1042/0264-6021:3510095
Ishizaki T Uehata M Tamechika I Keel J Nonomura K Maekawa M Narumiya S Pharmacological properties of Y-27632, a specific inhibitor of rho- associated kinases Mol Pharmacol 2000 57 976 983 10779382
Shimokawa H Seto M Katsumata N Amano M Kozai T Yamawaki T Kuwata K Kandabashi T Egashira K Ikegaki I Asano T Kaibuchi K Takeshita A Rho-kinase-mediated pathway induces enhanced myosin light chain phosphorylations in a swine model of coronary artery spasm Cardiovasc Res 1999 43 1029 1039 10615430 10.1016/S0008-6363(99)00144-3
Nakamura K Nishimura J Hirano K Ibayashi S Fujishima M Kanaide H Hydroxyfasudil, an active metabolite of fasudil hydrochloride, relaxes the rabbit basilar artery by disinhibition of myosin light chain phosphatase J Cereb Blood Flow Metab 2001 21 876 885 11435800 10.1097/00004647-200107000-00013
Kimura K Ito M Amano M Chihara K Fukata Y Nakafuku M Yamamori B Feng J Nakano T Okawa K Iwamatsu A Kaibuchi K Regulation of myosin phosphatase by Rho and Rho-associated kinase (Rho- kinase) Science 1996 273 245 248 8662509
Mukai Y Shimokawa H Matoba T Kandabashi T Satoh S Hiroki J Kaibuchi K Takeshita A Involvement of Rho-kinase in hypertensive vascular disease: a novel therapeutic target in hypertension FASEB J 2001 15 1062 1064 11292668
Bonnevier J Fassler R Somlyo AP Somlyo AV Arner A Modulation of Ca2+ Sensitivity by Cyclic Nucleotides in Smooth Muscle from Protein Kinase G-deficient Mice J Biol Chem 2004 279 5146 5151 14610087 10.1074/jbc.M306532200
Ise S Nishimura J Hirano K Kara N Kanaide H Theophylline attenuates Ca2+ sensitivity and modulates BK channels in porcine tracheal smooth muscle Br J Pharmacol 2003 140 939 947 14517178 10.1038/sj.bjp.0705508
Lee H Sohn SI Jung DK Cho YW Lim JG Yi SD Lee SR Sohn CH Baloh RW Sudden Deafness and Anterior Inferior Cerebellar Artery Infarction Stroke 2002 33 2807 2812 12468774 10.1161/01.STR.0000038692.17290.24
Shibuya M Suzuki Y Sugita K Saito I Sasaki T Takakura K Nagata I Kikuchi H Takemae T Hidaka H Effect of AT877 on cerebral vasospasm after aneurysmal subarachnoid hemorrhage. Results of a prospective placebo-controlled double-blind trial J Neurosurg 1992 76 571 577 1545249
|
16316469
|
PMC1315339
|
CC BY
|
2021-01-04 16:03:26
|
no
|
BMC Ear Nose Throat Disord. 2005 Nov 29; 5:10
|
utf-8
|
BMC Ear Nose Throat Disord
| 2,005 |
10.1186/1472-6815-5-10
|
oa_comm
|
==== Front
BMC GenetBMC Genetics1471-2156BioMed Central London 1471-2156-6-541628750810.1186/1471-2156-6-54Research ArticleTargeted oligonucleotide-mediated microsatellite identification (TOMMI) from large-insert library clones Chen Kefei [email protected] Christoph [email protected] Kirsten [email protected] Jun [email protected] Lusheng [email protected] Gary A [email protected] Bertram [email protected] Institute of Veterinary Medicine, University of Göttingen, Burckhardtweg 2, 37077, Göttingen, Germany2 Key Laboratory for Animal Biotechnology of Jiangxi Province and the Ministry of Agriculture, Jiangxi Agricultural University, Nanchang 330045, P. R. China3 US Meat Animal Research Center, Agricultural Research Service, US Department of Agriculture, Clay Center, NE 68933-0166, USA2005 15 11 2005 6 54 54 24 8 2005 15 11 2005 Copyright © 2005 Chen et al; licensee BioMed Central Ltd.2005Chen 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
In the last few years, microsatellites have become the most popular molecular marker system and have intensively been applied in genome mapping, biodiversity and phylogeny studies of livestock. Compared to single nucleotide polymorphism (SNP) as another popular marker system, microsatellites reveal obvious advantages. They are multi-allelic, possibly more polymorphic and cheaper to genotype. Calculations showed that a multi-allelic marker system always has more power to detect Linkage Disequilibrium (LD) than does a di-allelic marker system [1]. Traditional isolation methods using partial genomic libraries are time-consuming and cost-intensive. In order to directly generate microsatellites from large-insert libraries a sequencing approach with repeat-containing oligonucleotides is introduced.
Results
Seventeen porcine microsatellite markers were isolated from eleven PAC clones by targeted oligonucleotide-mediated microsatellite identification (TOMMI), an improved efficient and rapid flanking sequence-based approach for the isolation of STS-markers. With the application of TOMMI, an average of 1.55 (CA/GT) microsatellites per PAC clone was identified. The number of alleles, allele size distribution, polymorphism information content (PIC), average heterozygosity (HT), and effective allele number (NE) for the STS-markers were calculated using a sampling of 336 unrelated animals representing fifteen pig breeds (nine European and six Chinese breeds). Sixteen of the microsatellite markers proved to be polymorphic (2 to 22 alleles) in this heterogeneous sampling. Most of the publicly available (porcine) microsatellite amplicons range from approximately 80 bp to 200 bp. Here, we attempted to utilize as much sequence information as possible to develop STS-markers with larger amplicons. Indeed, fourteen of the seventeen STS-marker amplicons have minimal allele sizes of at least 200 bp. Thus, most of the generated STS-markers can easily be integrated into multilocus assays covering a broader separation spectrum. Linkage mapping results of the markers indicate their potential immediate use in QTL studies to further dissect trait associated chromosomal regions.
Conclusion
The sequencing strategy described in this study provides a targeted, inexpensive and fast method to develop microsatellites from large-insert libraries. It is well suited to generate polymorphic markers for selected chromosomal regions, contigs of overlapping clones and yields sufficient high quality sequence data to develop amplicons greater than 250 bases.
==== Body
Background
Almost all of the applied protocols to isolate microsatellites de novo include construction of partial genomic libraries (selected for small insert size) followed by cumbersome screening steps with hybridization probes [2]. Here, we introduce an improved approach called TOMMI (Targeted Oligonucleotide-Mediated Microsatellite Identification) to develop microsatellites by straightforward sequencing of clones isolated from large-insert libraries like PAC (P1-derived Artificial Chromosome) and BAC (Bacterial Artificial Chromosome) with repeat-containing oligonucleotides. The need to specifically identify and isolate STS-markers from these types of libraries is unquestionable. First, large-insert libraries are predominantly used in animal genetics, e.g. [3,4], as tools to identify candidate genes or to generate overlapping contigs of chromosomal regions that are associated with quantitative or economic trait loci (QTL or ETL). Secondly, the overall number of microsatellites present in a genome depends mainly on their complexity and size. Assuming a total size of 3 × 109 bp and an estimated frequency of a dinucleotide repeat every 30–50 kb in mammals (as reviewed by [5]), a genome-wide figure of 100,000 microsatellite markers of that kind can be assumed [6]. However, only approximately 1,200 porcine microsatellites have been reported so far [7]. Furthermore, both the total number and the distribution of the loci are still not sufficient to have well-distributed microsatellite coverage throughout the genome or for several chromosomes, e.g. SSC18 [8]. The objective of the present study was the selective generation of microsatellites from PAC-clones, which were prior to STS development isolated from the porcine PAC library TAIGP714 [3] by a three-dimensional PCR screening strategy [9]. Eight of the eleven clones harbored functional or positional candidate genes involved in health, reproduction, production, and regulation, whereas the other three clones have been used in the attempt to construct a PAC contig covering SSC16q11-13 (Table 1).
Table 1 Primers used for selecting the PAC clones from TAIGP714 large-insert library
PAC Clone Gene/Marker Forward Primer (5'-3')
Reverse Primer (5'-3') Size (bp) References
TAIGP714M09100Q GUSB GTCTGTGTCTGACTTCTACACTCTC
GCGGTCACAGGCTGCATCACCT 504 [24]
TAIGP714N07113Q HEXB GGAAGCTATTCTTTGTCTCATGT
CTTTTCCCCAAGACCGTGAAT 132 [25]
TAIGP714I04060Q CALCA CCCTCACTCTTACCTCTAACC
AGCTAAGCGGTGCAGTAATC 398 [26]
TAIGP714P05202Q HoxA10 CAGCCAACTGGCTCACGGCA
AGTTGGCTGTGAGCTCCCGG 239 [27]
TAIGP714O11196Q HYAL3 GATTGGGAGGAGTGGTGTC
GAGGTAGATGCTGGGGAAG 363 [28]
TAIGP714I22103Q SPRMTK GCGAGATGACATGACTGTCT
CTGTACCCATGGCACGCACA 254 [29]
TAIGP714F10061Q SW813 TCAGTTATTTCTGGCTATCATCTC
TTGATGTAGACCACCCAGCTAGTG 98–114 [12]
TAIGP714L02061Q S0111 AGTTGATTTAAAATGTTGTGCCA
AATATTTCAAAAAAAGGAATGCG 150–178 [10]
TAIGP714I23038Q SW742 AATTCTACTTCTGGGGAGAGGG
CTTTTGGGAACATTTCTGCC 193–227 [11]
TAIGP714N18001Q LAMR1P1 GTCGTAACTTAAAGGGAG
ATTTGGAAGTCAAGGTTGG 128 unpublished
TAIGP714H02175Q PGK1 CTTCCATCCCAAGCATC
TTCCCTTCTTCCTCCAC 384 [30]
Results and discussion
Fifteen of the seventeen microsatellites (Table 2) were developed with sequencing primers containing one selective nucleotide at the 3'-end: (CA)8T (S0701, S0703, and S0767), (CA)8A (S0702, S0704, and S0710), (CA)8G (S0705, S0706, S0712, and S0766), (AC)8C (S0709), (AC)8G (S0707 and S0715), (AC)8T (S0708 and S0711). Characterization of microsatellites S0713 and S0714 was only accomplished by an improved discrimination of the PAC clone sequences with sequencing primers further extended at the 3'-end with a second nucleotide [(CA)8AT for S0713 and (CA)8GC for S0714]. The second nucleotide became necessary because the respective clones TAIGP714L02061Q (for S0713) and TAIGP714I23038Q (for S0714) contained additional (CA)8A or (CA)8G primer binding regions or motifs. Contrary, a further extension with three nucleotides at the 3'-ends of the primers did not result in additional microsatellites in any of the PAC clones or was not required. Therefore, we conclude that repeat primers with two 3'-nucleotides next to the repeat motif are sufficient to detect and sequence all repeats potentially present on a large-insert library clone. The results of our isolation strategy also indicate that two sequencing reactions (the reverse sequencing primer was designed based on the obtained sequences) seem to be sufficient in most cases to gain sequence information of high quality to amplify microsatellites (Table 2). Usage of sequencing primers degenerated at the 3'-end proved, however, to be inadequate as no sequence information at all was achieved. Also, to avoid overlapping primary sequences, oligonucleotides that basically extend the dinucleotide repeat at the 3'-end – such as (CA)8C and (AC)8A – are not recommended. TOMMI proved to be an efficient and reliable isolation strategy. Besides new STS-markers, six previously described microsatellites were also detected. Three of these loci, microsatellites S0111 [10], SW742 [11], and SW813 [12], were initially used as probes for the isolation of clones TAIGP714L02061Q, TAIGP714I23038Q, and TAIGP714F10061Q. The other three already described microsatellite sequences reside on TAIGP714C09004Q [GenBank: AJ440949 (repeat location: 3172–3231) and GenBank: AJ440950 (repeat location: 15831–15860 and 16007–16038)]. They were not further considered in this study as they were not regarded as novel. Independently of our effort, two other groups [13,14] introduced similar sequencing approaches to generate microsatellites from large-insert libraries. There are, however, several differences between our approach and the ones of the other groups in terms of sequence generation and selective amplification of microsatellites. Here, contrary to Waldbieser and colleagues [14] – who used trinucleotide repeat containing primers for sequencing – both gene-specific primers are not 5'-tailed with extra nucleotide stretches to enable either product labeling or to promote alleged non-template adenylation. Fujishima-Kanaya's group [13] used larger repeat compounds contributing to the primer [(CA/GT)(10) instead of (CA/GT)(8)]. Secondly, the sequencing primers consisted generally of three selective nucleotides at the 3'-end adjacent to the repeat motif (e.g. CNA/GVG). There, the first of the three terminal nucleotides was always identical with the starting nucleotide of the dinucleotide repeat primer used. In addition, primers contained a degenerated base according to the International Union of Biochemistry (IUB) codes at the second position from or directly at the 3'-end. Thirdly, determination of the double-stranded primary DNA sequence stretch was achieved by four sequencing reactions using both a CA-repeat containing primer plus a GT-repeat containing primer heading in the opposite direction and two reverse primers were developed based on the obtained sequence. Finally, they always designed an additional primer pair for the specific amplification of the microsatellite. In contrast, we used the single reverse sequencing primer in combination with a newly developed sequence specific primer (S0766 and S0767) or designed a new primer pair to amplify the microsatellite (S0701 to S0715).
Table 2 Forward and reverse sequencing primer
Single reverse sequencing primer
Locus Initial sequencing primer (5'-3') Primer sequence (5'-3') Primer location (GenBank)
S0701 (CA)8T CCCAGGAGATTGAATATAG 223–241 (AY253989)
S0702 (CA)8A AAAAGCACCCAAAAAAGCC 156–174 (AY253990)
S0703 (CA)8T CTTATGGAGGTTCTCAGG 55–72 (AY253991)
S0704 (CA)8A GAGTGTGGGATAGACTG 119–135 (AY253992)
S0705 (CA)8G GAAGGGTAGGTTAAAGGG 252–269 (AY253993)
S0706 (CA)8G GGGAAACAGAAAATGGGG 156–173 (AY253994)
S0707 (AC)8G GTGAGCAAATAATTCAGTG 116–134 (AY253995)
S0708 (AC)8T CACAATTACTGCTTCTCTC 77–95 (AY253996)
S0709 (AC)8C GAGTGAGCACCATTCTAAG 117–135 (AY253997)
S0710 (CA)8A GCTTCATCACCCTGTTC 61–77 (AY253998)
S0711 (AC)8T CATTTTTCAGAGGGAAGAG 86–104 (AY253999)
S0712 (CA)8G GACCCTGGCATAGTATC 254–270 (AY254000)
S0713 (CA)8AT GAAGATACTGTTCTATGGATAG 1–22 (AY254001)
S0714 (CA)8GC CGTGTAGGTTGAAGACAAG 123–141 (AY254002)
S0715 (AC)8G GAGTTGTGTTTTATGGAGTTG 43–63 (AY254003)
S0766 (CA)8G AGACCTCCTATTAGAGGTGGA 519–539 (AY731063)
S0767 (CA)8T CTAGAATGGAAAACAATCTGA 367–387 (AY731064)
The observed number of alleles per locus (monomorphic locus S0709 is not included in this calculation) in the heterogeneous sampling was as low as 2 (S0702) and as high as 22 (S0713), leading to an average number of 9.94 alleles, NE ranged from 1.05 to 11.54 and both HT and PIC from 0.05 to 0.91 (Table 3).
Table 3 Characteristics of TOMMI-microsatellites
PAC Locus Chr1) Primer pair sequence (5'-3') Ta Size range Alleles NE PIC HT Repeat motif GenBank
TAIGP714M09100Q S0701 3p16-p14 GCAGAGTGATTCAGTTATAC 60 366–372 3 1.89 0.38 0.47 i(GT)14i(AT)6 AY253989
TCATCTTCCCTACCACC
TAIGP714M09100Q S0702 3p16-p14 TTTGGGGGGTTTGTTTTTG 57 346–348 2 1.18 0.14 0.16 (GT)9 AY253990
AATATAATTGGTGGCTCGG
TAIGP714N07113Q S0703 2p13-p11 AACCCACTGAACAAGGC 58 239–259 7 3.37 0.70 0.70 i(GT)11(AT)9 AY253991
GCAAGACAGATACTACAGG
TAIGP714N07113Q S0704 2p13-p11 AGCTATCATCAGGAAATGC 58 265–285 11 5.34 0.81 0.81 (GT)18 AY253992
GTTCTGTCGATTTTCTACTG
TAIGP714I04060Q S0705 2q21-q22 CAGGGGGTTAAAGATCAG 59 292–326 9 1.99 0.47 0.50 (GT)13 AY253993
GGGGCACATAAAAGGAAG
TAIGP714P05202Q S0706 18q23-q24 CTGGGTTGCTAAAGAGAC 56 211–215 3 1.05 0.05 0.05 (GT)6 AY253994
CACCTGAAGGATGTGAG
TAIGP714O11196Q S0707 13q21 GGTAGGGCTTACTTAACTC 56 163–196 11 6.89 0.85 0.86 i(GT)19i(GC)7(GA)10 AY253995
GAGAGGGATGAGAATCAG
TAIGP714I22103Q S0708 3q11-q12 GTTAGTTTCAGGCGTATAG 56 349–397 16 6.57 0.84 0.85 (GT)15i(AT)25(GT)14(AT)11 AY253996
CTGTGGTATAGGTCGAAG
TAIGP714F10061Q S0709 16q11-13 TTTAAGACACAGACAGCAG 58 151 1 1 0 0 i(GT)9 AY253997
CAGCATCTACATCCAGAC
TAIGP714F10061Q S0710 16q11-13 CTCAGCACCTTACAAACC 58 326–387 14 3.88 0.72 0.74 i(TAAA)7(GT)9 AY253998
TCCCAAACCAATCCACAC
TAIGP714F10061Q S0711 16q11-13 CAGAATCTAGCCTCAGCGTC 58 201–209 8 3.06 0.66 0.67 (GT)6(G)10 AY253999
CACTCCATCCCTCCCAAG
TAIGP714L02061Q S0712 16q11-13 TGGCATTGCTATGGCTG 57 251–310 14 5.57 0.82 0.82 (GT)12 AY254000
CACAACCACCTACATATCATC
TAIGP714L02061Q S0713 16q11-13 CATAATGCCCTCCACATC 54 263–317 22 11.54 0.91 0.91 (GT)17 AY254001
CCATATCATCCAGCAATTC
TAIGP714I23038Q S0714 16q11-13 TCTAGCTGTCGTGTAGG 55 199–207 8 3.75 0.70 0.73 (GT)7 AY254002
GAGGGATTACTCTGAGTTAAG
TAIGP714I23038Q S0715 16q11-13 GCCCTCCAGGACAAAAC 58 208–242 14 7.16 0.86 0.86 (GT)10i(GC)7i(GT)14 AY254003
GCTGTGACGTAGGTTGG
TAIGP714N18001Q S0766 6q27-28 GTGTAGATATGTGTCTGTACA 58 439–471 14 6.98 0.86 0.86 (GAAA)4(CA)6i(CA)16 AY731063
AGACCTCCTATTAGAGGTGGA
TAIGP714H02175Q S0767 Xq12-q13 TGACCATGTCTTGTGGTAA 53 239–247 3 2.04 0.40 0.51 (CA)11 AY731064
CTAGAATGGAAAACAATCTGA
Ta = Annealing temperature; NE = Effective allele number; PIC = Polymorphism Information Content; HT = Heterozygosity; i = Interrupted sequence.
1) Physical assignment: S0701 to S0708, and S0767 [24-30]; S0709 to S0715, and S0766 (physical localisation unpublished, but results are available through the website of the INRA-UMN porcine rodent hybrid IMpRH panel [17]
Due to their isolation from partial genomic libraries selected for small insert sizes most of the publicly available porcine microsatellites lie within DNA-fragments of about 80 to 200 bp. Their potential combination in multiplex assays – also considering different annealing temperatures and technical limitations of the automated sequencers (limited number of available fluorescent dyes) – is therefore hampered. Hence, an enhanced number of genotypes per run can only be achieved by the integration of STS-markers covering a larger allelic spectrum. Thus, we intended and focused on the development of large amplicons for microsatellites by utilizing as much sequence information as possible for primer design. Indeed, fourteen STS-markers had allele sizes of at least 200 bp and for five of the isolated microsatellites, sequence information proved to be good enough to amplify allele sizes of at least 300 bp (Table 3).
By the guided isolation of STS-markers S0709 to S0715 from three SSC16q derived PAC clones (relative position 0 cM to 9.3 cM [7]; 2.33 STS-markers per clone), the marker density in this chromosomal region was improved remarkably. An average of 1.55 new microsatellites was isolated from PAC clones harboring functional candidate genes (S0701-S0708; S0766 and S0767). Considering all used PAC clones and developed STS-markers, 1.55 microsatellites per clone were isolated. As the PAC clones had an average length of 80 kb (as shown by pulsed-field-gel electrophoresis) the frequency of dinucleotide repeats every (30 to) 50 kb [5] was more or less confirmed. TOMMI holds therefore the potential to identify existing STS-markers linked/adjacent to e.g. candidate genes on large-insert library clones. Thus, in combination with a genome scan, respective putative candidate genes could either be transformed to or excluded as positional candidate genes prior to their complete structural characterization including SNP detection. Linkage mapping results for S0701, S0705, S0707, S0711, S0712, S0713, S0715, and S0766 are presented in Table 4. A comparison of their mapping positions with QTL positions (Pig Quantitative Trait Loci (QTL) database [15] reveal that S0705 (64.22 cM), S0707 (43.19 cM), and S0766 (102.50 cM) reside on the respective chromosomes exactly at QTL locations (S0705: backfat between the last 3th and 4th rib; S0707: early growth rate and water holding capacity; S0766: backfat thickness at first rib and intra-muscular fat). The other STS-markers are located in QTL spans of ± 5 cM. This indicates their immediate potential to further dissect these respective QTL regions.
Table 4 MARC marker information and linkage mapping results
Marker Forward Primer (5'-3') Reverse Primer (5'-3') Number of Alleles Allele Size Range (bp) Number of Meioses Linkage Position (Chr: cM)
S0701 TGTTTCAGGTACACAGCAGAGTG AACGCGGTTTTGACCTACAG 3 161–167 48 3:31.4
S0705 TGGTTCAGATTGCTGTGGAG ATACCTGCAAACGCTGACCT 5 182–200 62 2:64.3
S0707 GGTAGGGCTTACTTAACTC GAGAGGGATGAGAATCAG 4 165–195 80 13:42.5
S0711 TCTGTTGCTGGCCATGAGT GTTCTGGCAACCCAGTCCT 2 116–119 63 16:5.9
S0712 TTTGCACTCTGCTTTTGAAGA GACCTGCACAACCACCTACA 6 173–221 134 16:0.6
S0713 AGCATAATGCCCTCCACATC GTGGCACCAACAGATGAATG 8 157–185 162 16:0.0
S0715 CCCTCCAGGACAAAACATTC TTTGAGGGAAAGAGGTGGAG 6 191–213 114 16:9.3
S0766 TAGAAACCTGCCCATTGAGG AGGCAGGGACAGGGTCTATT 7 122–144 122 6:102.6
Conclusion
The sequencing strategy described in this study provides a targeted, inexpensive and fast method to develop microsatellites from large-insert libraries. It is also well suited to generate polymorphic markers for selected chromosomal regions and contigs of overlapping clones and yielded sufficient high quality sequence data to develop marker amplicons greater than 250 bases.
Methods
PAC clone isolation and physical mapping
Prior to STS development, a total of 11 clones were isolated from the porcine PAC library TAIGP714 [3] by a three-dimensional PCR screening strategy. PAC-DNA preparations were done according to the manufacturer's protocol (Qiagen, Hilden, Germany). The physical assignment of the PAC clones was performed by Fluorescence in situ Hybridization (FISH) as described in [16] or alternatively by analysis of the INRA-UMN porcine radiation hybrid (IMpRH) panel [17]. Microsatellite primers (Table 3) were used to RH map S0703, S0704 and S0708 – S0715. Marker assignment of S0701, S0702, S0705 – S0707, S0766 and S0767 was performed with primers from further sequence segments of the PAC clones.
Microsatellite generation and characterization
All sequencing reactions and the separation of microsatellites were performed on an ABI PRISM® 3100 DNA analyzer (ABI, Weiterstadt, Germany). Sequencing reactions were done using the BigDye™ Terminator (v 3.0) Cycle Sequencing Kit (ABI, Weiterstadt, Germany). DNA sequencing was performed using 10 pmol of the respective oligonucleotide, 1 μl BigDye Premix and 50–100 ng of purified plasmid DNA as template in a total volume of 10 μl. Sequencing conditions were 96°C for 30 s followed by 30 cycles of 96°C for 10 s, the respective annealing temperature for 5 s and 60°C for 4 min. The optimal annealing temperature for the repeat containing primer was between 50°C and 52°C, except for the generation of sequences for S0714, which were at 56°C. To generate STS-markers, oligonucleotides containing repeat motifs (CA)8 respectively (AC)8 at the 5'-end and few (one or two) non-repetitive bases at the 3'-end were originally used as sequencing primers. Based on the obtained sequence, specific primers were developed and used as reverse oligonucleotides to determine the composition of the repeat region and its 5'-flanking region (Table 2; Figure 1). BLAST comparison followed sequence determination to verify the novelty and uniqueness of the obtained sequences. Depending on the quality of the sequenced stretch, primers were developed to amplify seventeen STS-markers (S0701 to S0715; S0766 and S0767; Table 3). To confirm the sequence identity of the respective microsatellites [GenBank: AY253989 to AY254003, AY731063, and AY731064] on genomic DNA, the resulting PCR products were subcloned into the polylinker of the pGEM®-T vector (Promega, Mannheim, Germany) and three independent clones each were bi-directionally sequenced using standard sequencing primers SP6 (5'-ATT TAG GTG ACA CTA TAG AA-3') and T7 (5'-TAA TAC GAC TCA CTA TAG GG-3').
Figure 1 Generation of STS-markers by TOMMI.
Evaluation of microsatellites and size determination of alleles were done with appropriate ABI-softwares GENESCAN (3.7) and GENOTYPER (3.6) using GENESCAN™-500ROX™ as internal size standard. Oligonucleotides were designed with the Oligo Selection Program [18] and synthesized by MWG Biotech (Ebersberg, Germany). To characterize size range, number of alleles, polymorphism information content (PIC), average heterozygosity (HT) and effective allele number (NE) of the microsatellites, STS-markers were separately amplified. PCR assays were performed at 54°C for S0706, S0708, S0712, S0713, S0714, and S0767, at 56°C for S0701, S0702, S0703, S0705, S0707 and S0715, and at 58°C for S0704, S0709, S0710, S0711, and S0766 in a RoboCycler Gradient 96® (Stratagene, LaJolla, USA) using PURE Taq Ready-To-Go PCR Beads® (Amersham Biosciences, Freiburg, Germany), along with the respective oligonucleotides (one labeled at the 5'-end alternatively with fluorescent dyes FAM, JOE or NED) and 50 ng of genomic porcine DNA in a volume of 12.5 μl (the concentration of each dNTP is 100 μM in 10 mM Tris-HCl (pH 9.0 at room temperature), 50 mM KCl and 1.5 mM MgCl2). In total, 336 unrelated pigs representing nine European breeds (9 Angeln Saddleback, 18 Bunte Bentheimer, 9 German Edelschwein, 15 German Landrace, 30 Hampshire, 27 Göttingen Minipig, 31 Pietrain, 12 Swabian-Haellian Swine, and 7 European Wild Boar), and six Chinese breeds (30 Chinese Jiangquhai, 28 Chinese Luchuan, 30 Chinese Minpig, 30 Chinese Rongchang, 30 Chinese Tibetan, and 30 Chinese Yushanhei) were investigated. The standard PCR profile was as follows: pre-denaturation at 92°C for 2 min, followed by 35 cycles of 92°C for 30 s, the optimal annealing temperature for 30 s, and 72°C for 30 s. The final cycle had an extension at 72°C for 10 min. PIC, HT and NE were estimated based on algorithms as introduced by Botstein and colleagues [19], Nei [20], and Kimura and Crow [21].
Linkage mapping of STS-markers on the USDA-MARC linkage map
Seven families of the MARC Swine Reference Population were genotyped as described [22]. Amplified DNA was radioactively labeled, separated by denaturing polyacrylamide gel electrophoresis and visualized with autoradiography. To ensure accurate sizing and discrimination of alleles, amplification primers were redesigned to yield smaller products for all markers except S0706, S0707 and S0709. S0767 was not tested in this population. Four markers were not informative in the MARC Swine Reference Population (S0702, S0706, S0709 and S0714) and four primer sets failed to produce reliable products (S0703, S0704, S0708 and S0710). Genotypes were determined and entered into the MARC Genome Database. Each marker was initially assigned to a chromosome based on TWOPOINT results of CRIMAP [23], then multipoint linkage analyses determined the final location of each marker. Genotypic data were evaluated with CHROMPIC and corrections made if necessary. The final position reported is based on the current MARC swine linkage map. Amplification primers for the eight successfully mapped markers are presented in Table 4.
Authors' contributions
KFC conducted the lab work to isolate and characterize S0701 to S0715 and CK to isolate and characterize S0766 and S0767. CK shared manuscript preparation and editing with KFC, supervised KFC's Ph.D. thesis, evaluated microsatellite data, and organized and provided DNA of the European pig breeds. KBK optimized and conducted fragment analysis and was responsible for evaluation of microsatellite data. JR assisted KFC in the beginning of the project. LSH organized DNA of the Chinese pig breeds. GAR conducted linkage mapping of the markers and edited the manuscript. BB proposed the idea, supervised and commented on the project, was responsible for funding and manuscript editing, and acts as head of the research group in Göttingen.
Acknowledgements
The authors would like to thank A. Siebels for expert technical assistance. This research project was supported by a grant of the Erxleben Research & Innovation Council to B. Brenig (ERIC-BR1959-2001-06).
==== Refs
Chapman NH Wijsman EM Genome screens using linkage disequilibrium tests: optimal marker characteristics and feasibility Am J Hum Genet 1998 63 1872 1885 9837839 10.1086/302139
Zane L Bargelloni L Patarnello T Strategies for microsatellite isolation: a review Mol Ecol 2002 11 1 16 11903900 10.1046/j.0962-1083.2001.01418.x
Al-Bayati HK Duscher S Kollers S Rettenberger G Fries R Brenig B Construction and characterization of a porcine P1-derived artificial chromosome (PAC) library covering 3.2 genome equivalents and cytogenetical assignment of six type I and type II loci Mamm Genome 1999 10 569 572 10341087 10.1007/s003359901046
Buitkamp J Kollers S Durstewitz G Welzel K Schafer K Kellermann A Lehrach H Fries R Construction and characterization of a gridded cattle BAC library Anim Genet 2000 31 347 351 11167520 10.1046/j.1365-2052.2000.00675.x
Hearne CM Ghosh S Todd JA Microsatellites for linkage analysis of genetic traits Trends Genet 1992 8 288 294 1509520 10.1016/0168-9525(92)90256-4
Wintero AK Fredholm M Thomsen PD Variable (dG-dT)n.(dC-dA)n sequences in the porcine genome Genomics 1992 12 281 288 1740337 10.1016/0888-7543(92)90375-3
USDA-MARC linkage map
Campbell EM Fahrenkrug SC Vallet JL Smith TP Rohrer GA An updated linkage and comparative map of porcine chromosome 18 Anim Genet 2001 32 375 379 11736809 10.1046/j.1365-2052.2001.00782.x
Cai L Taylor JF Wing RA Gallagher DS Woo SS Davis SK Construction and characterization of a bovine bacterial artificial chromosome library Genomics 1995 29 413 425 8666390 10.1006/geno.1995.9986
Ruyter D Verstege AJ van der Poel JJ Groenen MA Five porcine polymorphic microsatellite markers Anim Genet 1994 25 53 8161024
Rohrer GA Alexander LJ Keele JW Smith TP Beattie CW A microsatellite linkage map of the porcine genome Genetics 1994 136 231 245 8138161
Alexander LJ Rohrer GA Stone RT Beattie CW Porcine SINE-associated microsatellite markers: evidence for new artiodactyl SINEs Mamm Genome 1995 6 464 468 7579888 10.1007/BF00360655
Fujishima-Kanaya N Toki D Suzuki K Sawazaki T Hiraiwa H Iida M Hayashi T Uenishi H Wada Y Ito Y Awata T Development of 50 gene-associated microsatellite markers using BAC clones and the construction of a linkage map of swine chromosome 4 Anim Genet 2003 34 135 141 12648097 10.1046/j.1365-2052.2003.00967.x
Waldbieser GC Quiniou SM Karsi A Rapid development of gene-tagged microsatellite markers from bacterial artificial chromosome clones using anchored TAA repeat primers Biotechniques 2003 35 976 979 14628671
Pig Quantitative trait loci (QTL) database
Chen KF Beck J Huang LS Knorr C Brenig B Assignment of the phosphoglycerate kinase 2 (PGK2) gene to porcine chromosome 7q14-q15 by fluorescence in situ hybridization and by analysis of somatic cell and radiation hybrid panels Anim Genet 2004 35 71 72 14731237 10.1046/j.1365-2052.2003.01066.x
Yerle M Pinton P Robic A Alfonso A Palvadeau Y Delcros C Hawken R Alexander L Beattie C Schook L Milan D Gellin J Construction of a whole-genome radiation hybrid panel for high-resolution gene mapping in pigs Cytogenet Cell Genet 1998 82 182 188 9858812 10.1159/000015095
Hillier L Green P OSP: a computer program for choosing PCR and DNA sequencing primers PCR Methods Appl 1991 1 124 128 1842929
Botstein D White RL Skolnick M Davis RW Construction of a genetic linkage map in man using restriction fragment length polymorphisms Am J Hum Genet 1980 32 314 331 6247908
Nei M Estimation of average heterozygosity and genetic distance from a small number of individuals Genetics 1978 89 583 590 17248844
Kimura M Crow JF The Number of Alleles That Can Be Maintained in a Finite Population Genetics 1964 49 725 738 14156929
Rohrer GA Alexander LJ Hu Z Smith TP Keele JW Beattie CW A comprehensive map of the porcine genome Genome Res 1996 6 371 391 8743988
Green P Falls K Crooks S Documentation for CRI-MAP, version 2.4 1990 Washington University, School of Medicine, St. Louis, MO
Beck J Knorr C Habermann F Fries R Brenig B Assignment of the beta-glucuronidase (GUSB) gene to porcine chromosome SSC3p16-->p14 by FISH and confirmation by hybrid panel analyses Cytogenet Genome Res 2002 97 277G 12866501 10.1159/000066610
Mueller A Knorr C Habermann F Slanchev K Zwilling D Fries R Brenig B Assignment of the beta-N-acetylhexosaminidase gene (HEXB) to porcine chromosome SSC2q21-->q22 by fluorescence in situ hybridization and by analysis of somatic cell and radiation hybrid panels Cytogenet Genome Res 2003 101 178 14619878 10.1159/000074176
Knorr C Kollers S Fries R Brenig B Assignment of the CALC-A/alpha-CGRP gene (CALCA) to porcine chromosome SSC2p13-->p11 by fluorescence in situ hybridization and by analysis of somatic cell and radiation hybrid panels Cytogenet Genome Res 2002 97 140F 12438759 10.1159/000064050
Knorr C Uibeleisen AC Kollers S Fries R Brenig B Assignment of the homeobox A10 gene (HOXA10) to porcine chromosome SSC18q23-->q24 by FISH and confirmation by hybrid panel analyses Cytogenet Cell Genet 2001 93 145 146 11474203 10.1159/000056972
Gatphayak K Knorr C Habermann F Fries R Brenig B Assignment of the porcine hyaluronidase-3 (HYAL3) gene to SSC13-->q21 by FISH and confirmation by hybrid panel analyses Cytogenet Genome Res 2003 101 178 14619892 10.1159/000074181
Bull L Jansen S Habermann F Fries R Knorr C Brenig B Assignment of the sperm protein zona receptor tyrosine kinase gene (SPRMTK) to porcine chromosome SSC3q11-->q12 by fluorescence in situ hybridization and by analysis of somatic cell and radiation hybrid panels Cytogenet Genome Res 2003 101 178 14619881 10.1159/000074177
Chen KF Beck J Huang LS Knorr C Brenig B Assignment of the phosphoglycerate kinase 1 (PGK1) gene to porcine chromosome Xq12-q13 by fluorescence in situ hybridization and hybrid panel analyses Anim Genet 2004 35 143 145 15025578 10.1111/j.1365-2052.2004.01092.x
|
16287508
|
PMC1315340
|
CC BY
|
2021-01-04 16:30:39
|
no
|
BMC Genet. 2005 Nov 15; 6:54
|
utf-8
|
BMC Genet
| 2,005 |
10.1186/1471-2156-6-54
|
oa_comm
|
==== Front
BMC BioinformaticsBMC Bioinformatics1471-2105BioMed Central London 1471-2105-6-2671627447810.1186/1471-2105-6-267Research ArticleIdentification of clustered microRNAs using an ab initio prediction method Sewer Alain [email protected] Nicodème [email protected] Pablo [email protected] Alexei [email protected] Sébastien [email protected] Michael J [email protected] Thomas [email protected] Nimwegen Erik [email protected] Mihaela [email protected] Biozentrum, Universität Basel, Basel, Switzerland2 Laboratory of RNA Molecular Biology, Rockefeller University, New York, USA3 J. Craig Venter Institute, Functional Genomics, Rockville, USA4 IBMP-CNRS, Strasbourg, France2005 7 11 2005 6 267 267 20 5 2005 7 11 2005 Copyright © 2005 Sewer et al; licensee BioMed Central Ltd.2005Sewer 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
MicroRNAs (miRNAs) are endogenous 21 to 23-nucleotide RNA molecules that regulate protein-coding gene expression in plants and animals via the RNA interference pathway. Hundreds of them have been identified in the last five years and very recent works indicate that their total number is still larger. Therefore miRNAs gene discovery remains an important aspect of understanding this new and still widely unknown regulation mechanism. Bioinformatics approaches have proved to be very useful toward this goal by guiding the experimental investigations.
Results
In this work we describe our computational method for miRNA prediction and the results of its application to the discovery of novel mammalian miRNAs. We focus on genomic regions around already known miRNAs, in order to exploit the property that miRNAs are occasionally found in clusters. Starting with the known human, mouse and rat miRNAs we analyze 20 kb of flanking genomic regions for the presence of putative precursor miRNAs (pre-miRNAs). Each genome is analyzed separately, allowing us to study the species-specific identity and genome organization of miRNA loci. We only use cross-species comparisons to make conservative estimates of the number of novel miRNAs. Our ab initio method predicts between fifty and hundred novel pre-miRNAs for each of the considered species. Around 30% of these already have experimental support in a large set of cloned mammalian small RNAs. The validation rate among predicted cases that are conserved in at least one other species is higher, about 60%, and many of them have not been detected by prediction methods that used cross-species comparisons. A large fraction of the experimentally confirmed predictions correspond to an imprinted locus residing on chromosome 14 in human, 12 in mouse and 6 in rat. Our computational tool can be accessed on the world-wide-web.
Conclusion
Our results show that the assumption that many miRNAs occur in clusters is fruitful for the discovery of novel miRNAs. Additionally we show that although the overall miRNA content in the observed clusters is very similar across the three considered species, the internal organization of the clusters changes in evolution.
==== Body
Background
MicroRNAs (miRNAs) form a recently-discovered family of single-stranded RNA molecules of length approximatively 22 nucleotides that are present in all higher eukaryotes [1,2]. As shown by the growing number of specific examples, they regulate gene expression at a post-transcriptional level by binding to specific mRNA targets whose translation is thereby inhibited [3]. Although some details of miRNA biogenesis are still missing, a consensus scenario has now emerged: primary miRNA (pri-miRNAs) are transcribed generally by polymerase II [4], these transcripts are processed in the nucleus by the Drosha endonuclease [5] and exported as individual pre-miRNA stem loops to the cytoplasm by Exportin 5 [6]. In the cytoplasm, the mature forms are produced through the action of the Dicer endonuclease [7]. It appears that a crucial feature throughout these processing steps is a stem loop secondary structure [8].
An upper bound on the number of miRNAs present in the human genome was initially set by Lim et al. to a few hundred [9]. Recently however, this number has been re-evaluated by Berezikov et al. who argued that mammalian genomes encode close to a thousand miRNAs [10]. Thus, the debate about the number and identity of the miRNAs in mammalian genomes is open, especially considering that these estimates concern only miRNAs that are conserved between relatively distant species such as primates and rodents and not miRNAs that are of a more recent evolutionary origin.
The complete miRNA transcription units (pri-miRNA) remain to be defined, although some studies have already associated miRNAs with cDNAs sequences corresponding presumably to pri-miRNAs that can be found in sequence databases [11]. This and other studies (as well as our own unpublished data) show that some miRNAs are transcribed as polycistronic transcripts which are several kb long. Additional support for this hypothesis comes from a recent study that revealed that miRNAs that are found within 50 kb of each other on the same strand display correlated expression in microarray experiments [12]. Therefore the genomic regions around the loci of known miRNAs appear particularly promising for discovering additional miRNAs.
In the past few years several algorithms have been designed for detecting (pre-)miRNAs, and they proved to be extremely efficient in supporting experimental mature miRNAs discovery [10,13-15]. Very generally, these methods identify specific secondary structures corresponding to miRNA precursors in regions of the genome that are conserved between species. Recent experiments have uncovered, however, a number of miRNAs that do not have close homologs in the sequenced genomes available to date, such as for example the miRNAs encoded by the Epstein-Barr virus (EBV) [16]. This finding emphasized that it would be desirable to have a method able to predict miRNAs in a single genome, without an absolute requirement for cross-species conservation. We developed such a prediction method and we used it to discover miRNAs in a number of members of the herpes virus family [17]. By similarity with protein-coding gene prediction methods that only scan genomic regions looking for signals characteristic to protein-coding genes and do not use external transcripts or other genomes, we called our method ab initio.
Here we apply our method to search for novel miRNAs that are in close proximity, and may be co-transcribed, with already known miRNAs. As the set of known miRNAs we take the human, mouse and rat sequences from the April 2005 release of the Rfam miRNA repository [18]. To evaluate the performance of the method, we use a growing set of mammalian sequences that are cloned in the Tuschl laboratory [19]. In the following we first present the general ideas behind our pre-miRNA prediction method, then show the results of two validation tests and finally move on to its application to the discovery of potentially co-transcribed miRNAs in human, mouse and rat.
Results
Overview of the pre-miRNA prediction method
The general idea of our approach to pre-miRNA prediction is to design a computational method that can be used to better understand the constraints that define miRNA precursors in relationship to their processing enzymes. We start with the observation that one of the generic features shared by all miRNA genes is the secondary structure assumed by the transcript region surrounding the mature miRNA. Indeed, mature miRNAs appear to reside inside one arm (5' or 3') of a stem loop with good, though not perfect, base pairing during the various steps of the biogenesis [1,2]. This stem loop structure is important for miRNA precursor recognition by RNAase III enzymes Drosha [5] and Dicer [7] as well for the export of the miRNA precursor from the nucleus [6]. The intermediates in this processing pathway differ in the length of the sequence surrounding the mature miRNA, implying that the stem loop structure of the pre-miRNA persists independently of the precise sequence context that varies from several kb for the pri-miRNA transcripts to 50–70 nucleotides for the relatively short pre-miRNA. We thus design our prediction method to identify such "context-robust" (or shortly, "robust") stem loops and then characterize their compositional and secondary structure properties in relationship to those of known pre-miRNA as well as negative examples. Our approach consists in three steps:
1. From the input sequence we first extract a set of genomic regions that are predicted to exhibit the same stem loop secondary structure, irrespectively of the size of the larger transcript in which they are embedded. We call these "robust" stem loops.
2. Each of the stem loops thus detected is assigned a score that reflects its similarity to known cases of human pre-miRNAs. Since the precise structural features that contribute to miRNA precursor recognition by processing enzymes are not known, we use a device from statistical machine learning called "support vector machine" (SVM), as follows. We describe any given stem loop in terms of sequence and structure features. Then, using two training sets consisting of known human miRNA precursor sequences as positive examples and random subsequences from genomic regions, tRNA, rRNA and mRNA genes as negative examples we build a model which describes the relative contribution (weight) of each of the features to the score assigned to any given stem loop. The score measures the distance (in our feature space) from the candidate stem loop to the hypersurface that best separates the positive from the negative examples. The weights associated with the features that we used to describe the stem loops give us insight into the constraints that appear to be most important for the recognition of miRNA precursors by the processing enzymes.
3. In order to guide experimental investigations, we develop a probabilistic mathematical framework that enables us to estimate the pre-miRNA content of the input genomic sequence from the scores assigned to all the stem loops identified in this sequence. This framework that has the advantage of not being dependent of the somehow arbitrary score cut-off used to define the predicted miRNA precursor stem loops.
Further details about each of these components are presented in the "Methods" section.
Validation of the method
Identification of viral miRNAs
We initially validated our method on a set of eight human pathogenic viruses for which experimental investigations were simultaneously undertaken [17]. We made 32 pre-miRNA predictions, out of which 13 were confirmed by the cloning study, giving a confirmation rate of 41%. As explained in the "Method" section, the first step of our prediction method consists in extracting genomic regions which are able to form robust stem loop structures. The number of such regions grows linearly with the genome size (Figure 1, upper panel). The number of predicted pre-miRNAs, however, is not merely a linearly function of the genome size [17]. This property shows that the classifier that we have developed captures specific features of pre-miRNAs that are not uniformly distributed across the input genomic sequence. We have also developed a method for estimating the expected number of pre-miRNAs in a given genomic sequence (see Equation 4 from the "Methods" section). As shown in the lower panel of Figure 1, the expected number of pre-miRNAs in a genome is, as expected, strongly correlated with the number of predicted pre-miRNAs (i.e. the number of stem loops with a positive prediction score). At the same time this statistics is more robust against small variations in the value of the score threshold defining the predictions. An additional advantage of using this estimation approach is that it enables us to identify genomic regions that are likely to give rise to miRNAs without having to pinpoint their precise location. This is useful for directing experimental studies to promising genomes or genomic regions.
Figure 1 Robust stem loops and pre-miRNA predictions. The upper panel shows a plot of the number of robust stem loops versus the size of the genomic sequence they originate from. The data come from the application of our method to viruses [17] and, for the three at the extreme right-hand side, from the present study (see Subsection "Identification of novel clustered miRNAs"). The linear dependence is very clear and corresponds to an average of 1.2 robust stem loop every kb. The lower panel shows the relation between the number of predicted pre-miRNAs (stem loops with positive SVM score) and the expected number of pre-miRNAs, given by Equation 4. The linear dependence is also clear here, but the slope is smaller than 1, which would have corresponded to a strict equality between the two sets.
Recovering known miRNA in distantly-related species
It is perhaps not surprising that we can predict viral miRNAs, given that they are processed by the miRNA processing machinery of the human host, as are the miRNAs with which we have trained our model. To test the ability of our method to identify pre-miRNAs in distantly-related species, we applied it to the regions known to encode miRNA loci in the invertebrate Caernorhabditis elegans. Of the 116 known miRNA precursors, we recover 50 (43%) as predictions. This results indicates that our prediction method has a reasonable sensitivity for a wide range of genomes from worms, to vertebrates to animal viruses.
Predictions: clustered miRNAs in human, mouse and rat
Identification of novel clustered miRNAs
miRNAs are often found in genomic clusters, some of which are believed to be transcribed as a single transcript (polycistronic pri-miRNA). A nice example is the cluster of hsa-mir-17, whose elements reside within a 1 kb interval on human chromosome 13 and are indeed co-transcribed (cDNA Genbank accession number BC040320). Motivated by the recent microarray study by Baskerville and Bartel showing that correlated expression of miRNAs can be shown up to the order of 50 kb of relative genomic distance [12], we set to discover novel miRNAs in the neighborhood of confirmed miRNAs from the Rfam database [18]. We adopt the following strategy. We first group into clusters the known miRNAs from human/mouse/rat Rfam6.0 that are closer than 10 kb from each other and have the same transcription sense. We find that 105/111/82 of the total 227/232/186 pre-miRNAs (i.e. 46%/48%/44%) belong to a cluster of at least two elements. By adding these to the miRNAs that so far appear to be isolated, we obtain a total of 162/161/138 clusters. We then extend the genomic regions of all of these clusters by 10 kb on each side and submit the resulting sequences to the prediction method. In total, we analyze 3.36/3.35/2.84 million nucleotides and we find that 224/192/208 of the 3829/3537/3034 candidate stems are classified as predicted pre-miRNA by our prediction method. After filtering out the known pre-miRNAs in these clusters, we obtain a total of 89/66/105 predictions, given in the Additional files 1, 2, and 3.
In order to validate these predictions, we have searched a large database of small RNAs from human, mouse and rat that have been cloned in the Tuschl lab [19]. We consider a prediction to be validated if one of the arms of the stem loop matches perfectly a cloned small RNA that is not known to be derived from a rRNA, tRNA, snRNA or snoRNA. Additionally, although our predictions include sequences coming from repeated regions, we discard cases where the cloned small RNA has more than two perfect mappings to the genome of the species it originates from. We then find that 20/17/6 of the predictions have a match from the same species, and these numbers raise to 22/20/26 if matches from small RNAs from all three species are allowed. This corresponds to confirmation rates between 25% and 30%, which are somewhat lower than the one obtained with the earlier application of our method to viruses (40%). The pre-miRNAs predicted in repetitive elements are partially responsible for these lower confirmation rates. The complete list of the confirmed predictions together with the sequences of the cloned miRNAs are given in the Additional files 4, 5, and 6.
The false negative rate of our prediction method, as determined from the cloning data, is 34%, a value which is close to the false negative rate of 29% that we obtained for our SVM training set, using a threshold score of 0. This indicates that the prediction method behaves as expected. For completeness, the false negative predictions are shown in the Additional files 7, 8, and 9.
If we consider our results at the level of genomic clusters, we find 5 novel clusters in human, 5 in mouse and 6 in rat. By "novel clusters" we mean a set of precursor miRNAs that contains, beside the confirmed predictions, known cases that were not considered to be in clusters, i.e. which did not have another known pre-miRNA with same transcription sense at distance smaller than 10 kb from its genomic location. This corresponds to an increase of the total number of clusters from 40/40/34 to 44/45/38 in human/mouse/rat.
Phylogenetic conservation of the clustered miRNAs
Since cross-species conservation was not used in the process of generating our predictions, we can now go back and ask the question of whether the predicted pre-miRNAs are indeed conserved between human, mouse and rat. We define the "conservation" relation between two given species at three progressive levels, corresponding to the three columns labeled "Conservation" in the Additional files 1, 2, and 3. The first level requires that a homologous sequence is found for a predicted miRNA precursor in another genome (sequences alignment E-value given by the WU-Blast program ≤ 10-5). The second level requires that both the predicted precursor as well as the homologous sequence fold into simple (not branched) stem loops. The third level requires that the two homologous sequences are predicted to be miRNA precursors by a method that uses cross-species conservation. For this purpose, we use the web interface to the MiRscan method with the default parameters (threshold at score value 14) [13].
As shown in the Additional files 1, 2, and 3, the predictions that are conserved across species are more likely to be experimentally confirmed than the one that are not. Indeed the confirmation rates are 1%, 22% and 49% for predicted precursors with homologs in none, precisely one and both species, respectively. We thus conclude that the cross-species information strongly improves the specificity of the prediction method. If we now restrict ourselves to predictions that have experimental confirmation, we find that for almost all of them (95%) a strong sequence homology is equivalent to a conserved stem loop structure. Additionally, we learn from the third conservation column that only 68% of the conserved and confirmed miRNAs that we predict are also classified as miRNA precursors by MiRscan.
In some cases we discovered miRNAs that are known in some species, but were not reported in others. This is the case of the rat homolog of mmu-mir-1, which corresponds to our predicted RP-79. The predicted precursor has over 97% identity relative to the mouse precursor, and the region corresponding to the mature miRNA is perfectly conserved. This miRNA has not been cloned in rat. In other cases we discovered miRNAs that are conserved across all three species but that are found in the neighborhood of a miRNA only known to exist in one of the species. This is the case of RP-97, which is close to rno-mir-421. For this miRNA we found cloning and conservation evidence in mouse and human as well. Note, however, that we do not report here candidate pre-miRNAs that are homologous to some of our predictions but that were not part of our predicted set because in their species of origin they are not found in the neighborhood of a known miRNA.
Genomic locations of the miRNA clusters
Most of the confirmed predictions come from a relatively small number of clusters. These are the following:
1. The orthologous loci located on chromosome 14 in human, chromosome 12 in mouse and chromosome 6 in rat, each of which is less than 200 kb in length. The human locus has been shown to be imprinted [20]. Only a few miRNAs from these loci have been deposited in Rfam, although other studies have also published computational predictions matching them [10,20,21]. The fine-grained structure of these loci has some species-specific aspects, as illustrated by Figures 2 and 3. The figures show all the validated miRNAs in these regions, including those with suboptimal prediction scores from the Additional files 7, 8, and 9. We find that some miRNAs that are related in sequence, and have presumably arisen by duplication (such as the mir-368/mir-376-related sequences) have different numbers of copies in rodents and human. We also find rodent- (MP-33/RP-30, MP-34/RN-4, MP-44/RP-49) and human-specific (HP-31) miRNAs, meaning that they do not have very close mature form homologs in the other species. Yet in other cases the human and mouse sequence have diverged slightly, as the predicted miRNAs HP-30/MP-32 and HP-41/MP-41. Because the mature forms of these miRNAs differ by a deletion or a substitution, and because we only considered perfectly matching small RNAs from human as validation, only the human miRNA genes end up being validated (see the Additional files 1 and 2). Finally, there are cases of more complicated species-specific evolutionary pattern. For example, mir-329 appears to have diverged between human and mouse: at the syntenic location in human we find two identical copies of a miRNA distantly-related to mmu-mir-329 (HP-33 and HP-34). Furthermore, this cluster has a complex composition, containing other related sequences {hsa-mir-323, HP-33, HP-34, HP-35, HN-6} whereas the corresponding mouse cluster {mmu-mir-323, mmu-mir-329, MP-35, MN-7, MP-37} additionally contains a rodent-specific sequence, MN-7, which is not related to the other sequences in the cluster. A similar situation can be found in the human cluster {hsa-mir-368, HP-37, HN-7, hsa-mir-376a} which corresponds to the mouse {mmu-mir-376a, mmu-mir-376b, MP-38}.
Figure 2 Mappings between the human and mouse imprinted loci. For human and mouse we take all the sequences of the mature miRNAs belonging to the imprinted loci on chromosomes 14 and 12 that are present in Rfam6.0, in our set of confirmed predictions (Additional files 4 and 5) and in the false negatives set (Additional files 7 and 8). We sort them according to their genomic coordinates and display on the graph the sequence similarity (i.e. the number of matches in a CLUSTALW alignment) for all possible pairs between the two set. Although a clear diagonal signal indicating cross-species conservation is visible, it is also very clear that species-specific features altering it are numerous, as described in the text.
Figure 3 Mappings between the human and mouse imprinted loci. The same as Figure 2, but for mouse and rat. We clearly observe a much better diagonal signal, as expected. Off-diagonal bright spots indicate miRNAs that are related in sequence and that have probably arisen by duplication of a common ancestor.
2. Chromosome X also contains a substantial number of novel miRNAs in all three species. They are spread over the full chromosome and eventually form small clusters with only a few elements. In the mir-17 cluster paralog on chromosome X, whose evolution has been analyzed in detail by Tanzer and Stadler [22], we found two additional miRNAs that are conserved in all three species. In the order of genome location, the cluster then reads: mir-106a, HN-14/MP-56/RP-100, mir-19b-2, mir-92-2, and HP-85/MN-8/RP-99. Consistent with the evolutionary scenario proposed by these authors we find that these novel miRNAs are relatively close in sequence to other miRNAs in the cluster. For instance the mature miRNA sequence of MP-56 has only two mismatches with mmu-mir-17, mmu-mir-20 and mmu-mir-106a. We observe a similar situation for another cluster on human chromosome X: hsa-mir-188, HN-11, HP-77, HN-12, and HN-13 (in transcription sense order). Whereas only HP-77 is a confirmed prediction and possesses a close homolog in mouse (MN-9), this cluster contains further three related miRNAs that have negative scores but that have been confirmed experimentally.
3. Apart from the above mentioned miRNAs, we found a few other cases of clustered, and potentially co-transcribed miRNAs: two in human (chromosomes 16 and 17), four in mouse (chromosomes 3, 10, 11 and Un-random) and three in rat (chromosomes 9, 10 and 18).
Comparison with other prediction methods
To evaluate the performance of our approach relatively to other large-scale miRNA prediction methods, we perform the following test. We take the (pre-)miRNA predictions sets provided by the most extensive predictions studies in the recent past [10,21,23]. Since these results have been published at different times and some of the predictions sets contain also known miRNAs, we set the Rfam6.0 release as the reference set of known miRNAs. The set of miRNAs used to perform sensitivity/specificity tests contains those that have been introduced in the most recent version of the miRNA repository, Rfam7.0 and that only became available while our manuscript was under revision, as well as those (pre-)miRNAs that are predicted by any of the four methods (including ours) and that are confirmed by the cloning data [19]. This set comprises 38 miRNAs. Table 1 shows that, when tested on miRNAs that are conserved between human and mouse or human and rat, our method has comparable performance to methods that make predictions from conserved genomic regions. Overall, we have somewhat higher sensitivity, at the expense of somewhat lower specificity compared to the methods of Berezikov et al. and Xie et al.. Interestingly, each of four methods is able to predict some (pre-)miRNAs that are not predicted by any of the other ones.
Table 1 Sensitivity and specificity. This table shows a comparison of the performances of various methods when applied to the genomic loci considered in this work. The sensitivity was calculated by taking the union of the miRNAs predicted by any method but confirmed experimentally. This set contains 38 elements.
Altuvia et al. [23] Berezikov et al. [10] Xie et al. [21] Our method
Conserved predictions in analyzed loci 87 179 29 36
Predictions not in Rfam 6.0 78 27 23 36
Predictions in Rfam 7.0 or in cloning set [19] 18 21 20 23
Sensitivity 47% 55% 53% 64%
Specificity 23% 77% 87% 64%
Uniquely predicted pre-miRNAs 4 4 1 4
Overlap of our predictions with others 10 14 14 -
Discussion
We have developed a computational method for ab initio prediction of precursor miRNAs that we applied here to the problem of identifying clustered, probably co-transcribed miRNAs. As explained in details in the "pre-miRNA prediction method" subsection and in the "Methods" section, our approach is based on a mechanistic model for the action of enzymes like Drosha or Dicer, and uses only the information contained in the input sequence and secondary structure. In doing so we neglect important aspects such as the fact that any stem loop that we consider a candidate pre-miRNA has to be be transcribed and accessible to all the processing enzymes responsible for ultimately producing a mature miRNA. This aspect is, in part, responsible for a relatively high number of false positives that we would obtain if we were to run the prediction method on an entire mammalian genome. But by applying the method to regions around loci of already known miRNAs, we believe that we circumvent the issue of whether the genomic regions that we analyze are transcribed and are accessible to all the processing enzymes. Our results (68 novel, experimentally confirmed cases for 260 predictions, or a 26% hit rate) show that this assumption was justified. This is also indicated by another recent study that identified 8 of our 68 predicted and confirmed miRNAs [23]. The given percentage is a lower bound for the performance of our method on these genomic regions, since it cannot be excluded that some of the predicted pre-miRNAs are not detected or are expressed in tissues or developmental stages other than the ones that have been used in the experiments. Furthermore, we did not filter out from our predictions those that fall inside repeat elements and may have higher likelihood of being false positives. Although this inclusion lead to a lower hit rate, we have seen above that this allowed us to discover a rich structure in the imprinted clusters.
Virtually all of precursors of the validated miRNAs have some homolog (albeit somewhat different at the level of the mature miRNA) in at least one other species, and a number of factors may contribute to this effect.
1. We have focused our search on regions that are already known to contain conserved miRNAs and, as can be observed from out comparison of human and mouse loci, miRNAs that are close to each other in the genome are frequently related in sequence.
2. We have used for validation data from all three species, and we have considered a predicted pre-miRNA to be confirmed even in cases in which the supporting small RNA cloning data came from another species.
3. Finally, the mouse and rat genomes are quite close to each other and we therefore expect that almost all of the mouse miRNAs have rat homologs, and vice versa.
Note however that 17 out of the 46 confirmed mouse and rat miRNAs appear to be rodent-specific (i.e. 37%), and one confirmed miRNA appears to be human specific (HP-31). These miRNAs would be difficult to discover using other methods either due to lack of cross-species conservation or because the genomes that are sharing the miRNA are too close.
Although our "ab initio" approach to pre-miRNA discovery was initially designed and successfully applied to detect species-specific miRNAs [17], we find that it retains its value when applied to a situation where cross-species conservation plays a important role. In fact, we were able to discover conserved miRNAs that were missed by three methods that use cross-species comparisons to make their predictions [10,15,21]. Concretely, in the large imprinted clusters from human chromosome 14, mouse chromosome 12 and rat chromosome 6, almost half (48%) of the miRNAs that we predicted and were confirmed experimentally are novel (see the Additional files 1, 2, and 3).
It would be, of course, very instructive to understand what factors contribute to the different results obtained by different prediction methods. Although this is not the topic of our current study, we discuss some of these factors below. Generally, all these approaches, including ours, are based on a two-level strategy: first identify a relatively small set of candidates and then examine these candidates in detail to make predictions.
1. A fundamental difference between the first layer of our method and any cross-species-based analysis is the fact that we have tried to take into account mechanistic considerations rather than evolutionary conservation or statistical properties. That is, by first identifying "robust" stem loops (see "Methods") we not only pick up genomic regions which are likely to form suitable RNA secondary structure (which may be gotten using programs such as RNALfold [24]), but we also take into account the fact that the stem loop has to be present in the various stages of the miRNA biogenesis. This implies a dependency of the stem loop secondary structures that we identify on the flanking genomic regions, which should not compete too strongly for pairing with nucleotides within the miRNA precursor and thereby destroying its secondary structure. This mechanism naturally gives a basis to the observation that the functional RNAs, and in particular miRNA precursors, have thermodynamically more stable structures than randomized sequences with the same (di-)nucleotide composition [25] (this property has been used to predict precursor miRNAs [10], see next paragraph). The same property, namely robustness of functional RNA secondary structure elements with respect to varying sequence context, has been described in RNA viruses [26].
2. The second layer of the approach consists in scoring the candidate regions. In our case, these regions are already predicted to form stem loop secondary structures. Our choice of implementing a support vector machine (SVM, see "Methods") to distinguish between "good" and "bad" precursor miRNA candidates has several advantages. First it also includes information about what a miRNA precursor should not be, unlike the MiRscan scoring scheme which is only based on positive examples. Additionally, our procedure enables us to use the positive and negative examples to compute the weights with which the various features in our model should contribute to the score (see Tables 2, 3, 4, and 5). This detailed description of the sequence composition and structural features of the miRNA precursors enables our classifier to perform better than classifiers based simply on the thermodynamic stability of miRNA precursor stem loops [25]: on our training data, the SVM has at least 3-fold lower false positive rate compared to a classifier based on the RANDfold algorithm for any given rate of false negative predictions (data not shown). Additionally, the weights assigned by the SVM for individual features give us a deeper understanding of miRNA processing than procedures that only use statistics of secondary structure stability of pre-miRNAs relative to randomized variants [10].
Table 2 SVM features to describe stem loops, part 1. These quantities are calculated over the entire stem loop structure. The weights are normalized with respect to the first feature, which turns out also to be the one with smallest value.
Index Feature description SVM weight
1 Free energy of folding -1
2 Length of the longest simple stem 0.547
3 Length of the hairpin loop 0.193
4 Length of the longest perfect stem 0.030
5 Number of nucleotides in symmetrical loops -0.006
6 Number of nucleotides in asymmetrical loops 0
7 Average distance between internal loops -0.029
8 Average size of symmetrical loops 0.207
9 Average size of asymmetrical loops -0.171
10/11/12/13 Proportion of A/C/G/U nucleotides in the stem 0.005/-0.003/-0.004/0.001
14/15/16 Proportion of A-U/C-G/G-U base pairs in the stem 0.015/-0.002/-0.006
Table 3 SVM features to describe stem loops, part 2. These quantities are calculated over the longest symmetrical region of the stem loop, i.e. the longest region without any asymmetrical loop.
Index Feature description SVM weight
17 Length 0.353
18 Distance from the hairpin loop 0.126
19 Number of nucleotides involved in internal loops 0.041
20/21/22/23 Proportion of A/C/G/U nucleotides 0.082/0.241/0.078/0.059
24/25/26 Proportion of A-U/C-G/G-U base pairs 0.211/0.254/-0.131
Table 4 SVM features to describe stem loops, part 3. These quantities are calculated over the longest region in which the difference between the 5' and 3' components of asymmetrical loops is not larger than Δl (we will call this "relaxed symmetry region").
Index Feature description SVM weight
27 Length 0.189
28 Distance from the hairpin loop 0.116
29 Number of nucleotides involved in symmetrical internal loops -0.220
30 Number of nucleotides involved in asymmetrical internal loops -0.176
31/32/33/34 Proportion of A/C/G/U nucleotides 0.024/0.077/-0.079/0.149
35/36/37 Proportion of A-U/C-G/G-U base pairs 0.317/0.123/-0.156
Table 5 SVM features to describe stem loops, part 4. These quantities are calculated over all windows of length corresponding to miRNA length lm that we could place on the candidate stem loop.
Index Feature description SVM weight
38 Maximum number of base pairs -0.140
39 Minimum number of nucleotides in asymmetrical loops -0.025
40 Minimum asymmetry over the internal loops in this region 0.026
Adding our confirmed predictions (around 25 per species) to the already known miRNAs from Rfam6.0 (227/232/188 for human/mouse/rat), we have reached the upper bound on the number of miRNAs that was estimated by Lim et al. to be around 255 [13]. It is now important to realize that this estimate was based on an assumption of miRNA conservation over an evolutionary distance up to the pufferfish fugu rubripes. The more recent estimate of Berezikov et al., with less stringent assumptions about conservation lies in the range of about a thousand miRNAs [10]. From the present work we have learned that these methods have missed some conserved miRNAs. Moreover, some miRNAs are only represented in closely-related species (such as mouse and rat) and there are also families of closely related miRNAs that differ in precise composition across species. All these considerations lead us to conclude that the miRNA discovery is still not completed, and moreover, that hundreds of miRNAs are yet awaiting experimental confirmation.
Although the hypothesis that miRNA loci tend to occur in clusters which are probably co-transcribed has been useful in the discovery of novel miRNAs, there are interesting open questions about the expression of the co-transcribed miRNAs. In particular, not all of the known examples of co-transcribed miRNAs show strongly correlated expression patterns [27]. This indicates that yet unknown processing factors lead to differential expression of the clustered miRNAs by either making the processing enzymes having different efficiencies on different templates or by directing transcription from alternative transcription start sites.
Conclusion
We have developed a computational method to estimate the pre-miRNA content and to predict the location of precursor miRNAs in genomic sequences. This method can be used to guide experiments to find both miRNAs that are evolutionarily conserved as well as species-specific miRNAs such as those known now to exist in viruses. Here we applied our method to the discovery of clustered, probably co-transcribed, miRNAs in human, mouse and rat. We have shown that our method successfully identifies evolutionarily conserved miRNAs that have been missed by various other methods that are based on cross-species comparisons. Most of the novel miRNAs that we discovered reside in a conserved imprinted locus from chromosome 14 in human, 12 in mouse and 6 in rat and on chromosome X (in all species). In these regions we found species-specific patterns of miRNA duplication and diversification. The web interface to our prediction method can be accessed on the world-wide-web [28].
Methods
Extraction of genomic regions with robust secondary structures
We determine genomic regions with context-independent stem loop secondary structures (shortly "robust" secondary structures) as follows. We move a window of length L across the entire input RNA sequence in a stepwise manner, shifting by an amount ΔL at each step. For each window position we compute the minimal free-energy secondary structure of the corresponding sequence using the RNAfold program of the Vienna package [29]. We store the nucleotide pairs of this structure into a table with all the pairs that occurred in at least one structure, and at the end, we determine the preservation rate ("robustness") r for every nucleotide pair (i, j) in the table. This is defined as:
r=number of windows containing the nucleotide pair (i,j)number of windows containing both nucleotides i and j. (1)
MathType@MTEF@5@5@+=feaafiart1ev1aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacH8akY=wiFfYdH8Gipec8Eeeu0xXdbba9frFj0=OqFfea0dXdd9vqai=hGuQ8kuc9pgc9s8qqaq=dirpe0xb9q8qiLsFr0=vr0=vr0dc8meaabaqaciGacaGaaeqabaqabeGadaaakeaacqWGYbGCcqGH9aqpdaWcaaqaaiabb6gaUjabbwha1jabb2gaTjabbkgaIjabbwgaLjabbkhaYjabbccaGiabb+gaVjabbAgaMjabbccaGiabbEha3jabbMgaPjabb6gaUjabbsgaKjabb+gaVjabbEha3jabbohaZjabbccaGiabbogaJjabb+gaVjabb6gaUjabbsha0jabbggaHjabbMgaPjabb6gaUjabbMgaPjabb6gaUjabbEgaNjabbccaGiabbsha0jabbIgaOjabbwgaLjabbccaGiabb6gaUjabbwha1jabbogaJjabbYgaSjabbwgaLjabb+gaVjabbsha0jabbMgaPjabbsgaKjabbwgaLjabbccaGiabbchaWjabbggaHjabbMgaPjabbkhaYjabbccaGiabbIcaOiabdMgaPjabcYcaSiabdQgaQjabcMcaPaqaaiabb6gaUjabbwha1jabb2gaTjabbkgaIjabbwgaLjabbkhaYjabbccaGiabb+gaVjabbAgaMjabbccaGiabbEha3jabbMgaPjabb6gaUjabbsgaKjabb+gaVjabbEha3jabbohaZjabbccaGiabbogaJjabb+gaVjabb6gaUjabbsha0jabbggaHjabbMgaPjabb6gaUjabbMgaPjabb6gaUjabbEgaNjabbccaGiabbkgaIjabb+gaVjabbsha0jabbIgaOjabbccaGiabb6gaUjabbwha1jabbogaJjabbYgaSjabbwgaLjabb+gaVjabbsha0jabbMgaPjabbsgaKjabbwgaLjabbohaZjabbccaGiabdMgaPjabbccaGiabbggaHjabb6gaUjabbsgaKjabbccaGiabdQgaQbaacqGGUaGlcaWLjaGaaCzcamaabmaabaGaeGymaedacaGLOaGaayzkaaaaaa@BBCB@
Choosing a minimal robustness value rmin, we reconstruct the secondary structure elements (including stem loops) that occur with rate r ≥ rmin using the following property of the nearest-neighbor energy model used in the the secondary structure calculation [30]. Given two different sequences containing both the subsequence corresponding to the interval [i - 1, j + 1], if the pair (i, j) is present in the minimal free-energy structures of the two sequences, then the secondary structure pairing pattern of the subsequence [i, j] is exactly the same for the two sequences. Therefore, if a pair (i, j) appears with a robustness r in the table, then all pairs in the table belonging to the common minimal free energy structure of the subsequence [i + 1, j - 1] in overlapping windows will appear with a robustness of at least r. We use this property to reconstruct the secondary structure elements preserved with a minimal rate of rmin. In our implementation we take rmin = 0.9, and keep only the genomic regions where a single stem of at least 15 nucleotides is present in the robust secondary structure. Finally, we fix the windowing parameters L and ΔL. L should correspond to the length of known miRNA precursors, which can vary between hundreds and thousands of nucleotides for the primary transcripts and between 50 and 70 nucleotides for pre-miRNAs. The constraint on ΔL is set such as to allow us to collect sufficient data to make a statistically meaningful estimation of structure robustness. To fulfill all these requirements while keeping the calculation time within reasonable limits, we use two combinations of (L,ΔL) values, namely (500,25) and (1000,50), and then take the intersection of the sets of predicted regions with robust secondary structures.
Application of the above procedure to genomic regions of 20 kb around the Rfam6.0 miRNA loci [18] shows that 89%/88%/89% (203/205/166 out of 227/232/186) of the known precursors of human/mouse/rat miRNAs overlap with a robust stem loop. This cross-check shows that the secondary structure "robustness" criteria is very appropriate to make a first selection of candidate miRNA precursor stem loops.
Classification of stem loops using a support vector machine
In order to construct a support vector machine (SVM) classifier for the candidate stem loops, we need a set positive and a set of negative examples of miRNA precursors. As positive examples, we use the complete set of human pre-miRNAs in the Rfam repository. As negative examples we use random subsequences isolated from tRNA, rRNA and mRNA genes. In cases where the input sequence was too short to enable us to perform the robust stem identification as described above (e.g. in the case of the short tRNA genes), we pad the input sequence with random sequence using the mono-nucleotide frequencies of the input sequence. To get a good sampling of the space of sequences that the transcription and miRNA processing machinery may encounter in the cell, we add to the set of negative examples genomic regions isolated from random positions in the human genome, as well as the genomic regions from various viruses. Given that the fraction of genome encoding miRNA genes is quite small, it is highly unlikely that the sequences that we have chosen this way contain miRNA precursors. We have a total of 178 positive examples (i.e. the human precursor miRNAs from Rfam4.0) and 5395 negative examples. Thus the fraction of positives is the training set is of the order of a few percent, similar to what we think is the case in the human genome. Analyzing the known examples [18], we and others find that miRNA precursors generally assume simple hairpin structures (except some member of the let-7 family), longer than about 50 nucleotides. The fraction of paired nucleotides in the stem is relatively high, and the internal loops, if present, tend to be symmetrical. The hairpin loop appears to be relatively short, of at most 20 nucleotides. The nucleotide composition in the stem is generally balanced, as is the number of A-U, G-C and G-U pairs. Based on such considerations we use the RNAfold program to calculate the minimal free energy secondary structure [29] and then we describe each stem loop (positive/negative example or candidate pre-miRNA) in terms of the following sets of features:
1. Statistics computed over the entire hairpin structure: 16 features given in Table 2;
2. Statistics computed over the longest symmetrical region of the stem, i.e. the longest region without any asymmetrical loops: 10 features given in Table 3;
3. Statistics computed over the longest region in which the difference between the 5' and 3' components of asymmetrical loops is not larger than Δl (this region is called "relaxed symmetry region"): 11 features given in Table 4.
4. Statistics computed over all windows of length corresponding to mature miRNA length lm that we could place on the candidate stem loop, in order to zoom onto the region of the mature miRNAs: 3 features given in Table 5.
Note that there are only two parameters in these computations: the maximally allowed asymmetry Δl in a region with "relaxed symmetry", and the assumed length of the mature miRNA lm. We choose these parameters so as to minimize the number of misclassified examples in our training set. This minimum occurs at the values 4 and 20, respectively.
For the SVM, we use the "SVMlight" implementation of Joachims and a quadratic kernel [31]. This implementation allows us to choose an additional cost for penalizing misclassified positive relative to negative examples. We chose this value such as to get a reasonable recovery rate of known pre-miRNAs, while keeping a low false positive rate. The distributions of scores of the training sets are shown in Figure 4. In short, the model that we have constructed recovers 71% of the positive pre-miRNA examples with robust stems, with a false positive rate of 3%. The features to which the SVM has assigned the largest positive weights are the stem length, the length of the longest symmetrical region, number of A-U and number of G-C base pairs in the "relaxed symmetry" region. The features with the largest negative weights are the free energy of folding, the number of nucleotides in symmetrical and asymmetrical loops in the "relaxed symmetry" region, and the average size of asymmetrical loops. These conform to prior knowledge [13].
Figure 4 SVM training set score distributions. Normalized distributions of the SVM scores for the positive and negative stem loop examples used for the SVM training. The good separation between the two sets indicates that the SVM performs well in recognizing the miRNA precursor specific features.
Estimation of the pre-miRNA content of a sequence
The number of pre-miRNA predictions obtained from the SVM classifier depends on the value of the score threshold, which is somewhat arbitrary. We show here that our approach can provide an estimate of the pre-miRNA content of a sequence which is independent of this threshold.
According to the assumptions made previously, the features that we use to decide whether a candidate stem loop is a miRNA precursor are contained in RNA sequence and secondary structures only, and are combined into a score using the SVM model. If this assumption holds, then the probability p that a robust stem loop contains a miRNA is a function of the score s only, and we can compute the overall pre-miRNA content of a sequence based on the distributions of scores for the set of positives (S+), the set of negatives (S-), and the set of candidate stems in a given genomic sequence (SC). We expect that p(s) has a step-like monotonic behavior, being close to 0 for large negative values of s and becoming asymptotically close to 1 for large positive scores. In order to concretely specify the function p(s), we first choose a suitable parametric functional expression such as
p(s)=12(1+tanh (as+b)) (2)
MathType@MTEF@5@5@+=feaafiart1ev1aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacH8akY=wiFfYdH8Gipec8Eeeu0xXdbba9frFj0=OqFfea0dXdd9vqai=hGuQ8kuc9pgc9s8qqaq=dirpe0xb9q8qiLsFr0=vr0=vr0dc8meaabaqaciGacaGaaeqabaqabeGadaaakeaacqWGWbaCcqGGOaakcqWGZbWCcqGGPaqkcqGH9aqpdaWcaaqaaiabigdaXaqaaiabikdaYaaacqGGOaakcqaIXaqmcqGHRaWkcqqG0baDcqqGHbqycqqGUbGBcqqGObaAcqqGGaaicqqGOaakcqWGHbqycqWGZbWCcqGHRaWkcqWGIbGycqGGPaqkcqGGPaqkcaWLjaGaaCzcamaabmaabaGaeGOmaidacaGLOaGaayzkaaaaaa@4846@
and then fix its parameters a and b by maximizing the likelihood of the training set, defined as
L(a,b)=∏i∈S+p(si)∏j∈S−p(sj). (3)
MathType@MTEF@5@5@+=feaafiart1ev1aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacH8akY=wiFfYdH8Gipec8Eeeu0xXdbba9frFj0=OqFfea0dXdd9vqai=hGuQ8kuc9pgc9s8qqaq=dirpe0xb9q8qiLsFr0=vr0=vr0dc8meaabaqaciGacaGaaeqabaqabeGadaaakeaacqWGmbatcqGGOaakcqWGHbqycqGGSaalcqWGIbGycqGGPaqkcqGH9aqpdaqeqbqaaiabdchaWjabcIcaOiabdohaZnaaBaaaleaacqWGPbqAaeqaaOGaeiykaKcaleaacqWGPbqAcqGHiiIZcqWGtbWudaahaaadbeqaaiabgUcaRaaaaSqab0Gaey4dIunakmaarafabaGaemiCaaNaeiikaGIaem4Cam3aaSbaaSqaaiabdQgaQbqabaGccqGGPaqkcqGGUaGlcaWLjaGaaCzcamaabmaabaGaeG4mamdacaGLOaGaayzkaaaaleaacqWGQbGAcqGHiiIZcqWGtbWudaahaaadbeqaaiabgkHiTaaaaSqab0Gaey4dIunaaaa@5369@
If we now make the assumption that all candidates i ∊ SC are independent from another, the number of miRNA precursors E is given by a sum of independent Bernoulli distributions, each of them being characterized by its own probability 0 ≤ p(si) ≤ 1. As a consequence, the expected value of E and its error ΔE are given by
E±ΔE=∑i∈SCp(si)±∑i∈SCp(si)(1−p(si)). (4)
MathType@MTEF@5@5@+=feaafiart1ev1aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacH8akY=wiFfYdH8Gipec8Eeeu0xXdbba9frFj0=OqFfea0dXdd9vqai=hGuQ8kuc9pgc9s8qqaq=dirpe0xb9q8qiLsFr0=vr0=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@5EB8@
We can also calculate the probability that a given genome encodes any given integer number m of miRNAs precursors. Defining a generating function as
G(z)=∏i∈SC(1−p(si)+zp(si)) (5)
MathType@MTEF@5@5@+=feaafiart1ev1aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacH8akY=wiFfYdH8Gipec8Eeeu0xXdbba9frFj0=OqFfea0dXdd9vqai=hGuQ8kuc9pgc9s8qqaq=dirpe0xb9q8qiLsFr0=vr0=vr0dc8meaabaqaciGacaGaaeqabaqabeGadaaakeaacqWGhbWrcqGGOaakcqWG6bGEcqGGPaqkcqGH9aqpdaqeqbqaaiabcIcaOiabigdaXiabgkHiTiabdchaWjabcIcaOiabdohaZnaaBaaaleaacqWGPbqAaeqaaOGaeiykaKIaey4kaSIaemOEaONaemiCaaNaeiikaGIaem4Cam3aaSbaaSqaaiabdMgaPbqabaGccqGGPaqkcqGGPaqkcaWLjaGaaCzcamaabmaabaGaeGynaudacaGLOaGaayzkaaaaleaacqWGPbqAcqGHiiIZcqWGtbWudaWgaaadbaGaem4qameabeaaaSqab0Gaey4dIunaaaa@4F4A@
the expression for the probability of having exactly m miRNAs is found to be
P(m)=1m!∂m∂zmG(z)|z=0 (6)=∑m−uplesj1,…,jm∈SCp(sj1)…p(sjm)∏j∉{j1,…,jm}(1−p(sj))
MathType@MTEF@5@5@+=feaafiart1ev1aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacH8akY=wiFfYdH8Gipec8Eeeu0xXdbba9frFj0=OqFfea0dXdd9vqai=hGuQ8kuc9pgc9s8qqaq=dirpe0xb9q8qiLsFr0=vr0=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@8F83@
The quantiles of the distribution P(m) are used to estimate the minimal number of expected miRNA precursors. Define the quantity
Q(m)=1−∑n<mP(n)=∑n≥mP(n), (7)
MathType@MTEF@5@5@+=feaafiart1ev1aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacH8akY=wiFfYdH8Gipec8Eeeu0xXdbba9frFj0=OqFfea0dXdd9vqai=hGuQ8kuc9pgc9s8qqaq=dirpe0xb9q8qiLsFr0=vr0=vr0dc8meaabaqaciGacaGaaeqabaqabeGadaaakeaacqWGrbqucqGGOaakcqWGTbqBcqGGPaqkcqGH9aqpcqaIXaqmcqGHsisldaaeqbqaaiabdcfaqjabcIcaOiabd6gaUjabcMcaPiabg2da9maaqafabaGaemiuaaLaeiikaGIaemOBa4MaeiykaKIaeiilaWIaaCzcaiaaxMaadaqadaqaaiabiEda3aGaayjkaiaawMcaaaWcbaGaemOBa4MaeyyzImRaemyBa0gabeqdcqGHris5aaWcbaGaemOBa4MaeyipaWJaemyBa0gabeqdcqGHris5aaaa@4EA0@
then if n is the largest integer such that Q(n) ≥ 0.99, then n is the number of pre-miRNAs that we estimate to be encoded in the considered genome with 99% confidence. Figure 5 shows an illustration of these distributions for two viruses [17].
Figure 5 Estimation of the pre-miRNA content for HHV8 and HCV. Plot of the probability P(m) of a given virus to encode exactly m pre-miRNAs (Equation 6) and probability Q(m) to encode at least m pre-miRNAs (Equation 7). Equation 4 allows to calculate the corresponding number of expected pre-miRNAs, 5.8 ± 1.8 for HHV8 and 0.5 ± 0.7 for HCV. These values correspond to the mean and standard deviation of the distribution P(m) shown on the figure.
We perform a cross-check for the above approach by applying it to the full training set of the SVM, merging the positive and negative stems loop into one single set. The expected number of pre-miRNAs from Equation 4 is 157.2 ± 6.3. The number of pre-miRNAs present with 99% confidence in the training set is 143. These numbers are in total agreement with the sensitivity of the SVM, according to which 0.7 = 155 of the positives should be correctly classified.
Authors' contributions
AS and MZ performed the bioinformatic work. NP helped in setting up the web server. PL, AA, SP, MJB, and TT provided the cloning data. EvN developed the probabilistic pre-miRNA content estimation method. MZ supervised the collaboration between the various people.
Supplementary Material
Additional File 1
All predictions, human. The table contains exhaustive information about all the predicted miRNA precursors that have been assumed to co-transcribed with a known human miRNA present in the Rfam6.0 set. The latter may be the one given in the column "closest miRNA", characterized by the smallest genomic distance to the prediction and a common transcription sense. A row has a color background when it contains a prediction that has been confirmed by cloning. The genomic coordinates are given for the hg17 human genome assembly. In the columns about cloning and cross-species conservation the letters "h", "m", and "r" mean the corresponding property being satisfied for human, mouse, and rat, respectively. The last column show (not yet confirmed) pre-miRNA predictions obtained using methods based on cross-species conservation. The first letters "B", "L" and "X" indicate results by Berezikov et al. [10], Legendre et al. [15], and Xie et al. [21], respectively. The other letters are the identifiers given by these authors to the corresponding predictions. We also indicate the recent predictions by Altuvia et al., labeled by an "A", who studied human miRNA clusters using a different approach [23].
Click here for file
Additional File 2
All predictions, mouse. The same as the Additional file 1, but for the mouse predictions. The genome assembly use for the coordinates is mm5.
Click here for file
Additional File 3
All predictions, rat. The same as the Additional file 1, but for the rat predictions. The genome assembly used for the coordinates is rn3.
Click here for file
Additional File 4
All confirmed, human. Details of the confirmed miRNAs for human. The cloning frequency indicates the number of distinct small RNA cloned sequences found in the comprehensive cloning set that match our predicted pre-miRNA, "h", "m", and "r" corresponding to human, mouse, and rat, respectively. The secondary structure uses the text display from Mfold [32]. The last column gives the best homologs (within at most 5 mismatches) found in our confirmed predictions, in the false negative sets (Additional files 7, 8, and 9) and in Rfam7.1 (latest release at the time of publication). Notice MP-61 with good cloning evidence but an unusual position in the secondary structure.
Click here for file
Additional File 5
All confirmed, mouse. Same as the Additional file 4, but for mouse. Notice that the case MP-28 is identical to MP-61.
Click here for file
Additional File 6
All confirmed, rat. Same as the Additional file 4, but for rat. RP-66 is identical to HP-61 and MP-28.
Click here for file
Additional File 7
False negatives, human. Set of false negatives for human, i.e. stem loop candidates with a negative SVM score but which have a cloning evidence. The layout is identical to the table from the Additional files 4, 5, and 6. The Additional column "Other predictions" is filled as in the additional files 1, 2, and 3.
Click here for file
Additional File 8
False negatives, mouse. Set of false negatives for mouse, similar to the Additional file 7.
Click here for file
Additional File 9
False negative, rat. Set of false negatives for rat, similar to the Additional file 7. Notice the unusual position of RN-4 in the secondary structure.
Click here for file
Additional File 10
Additional files 1, 2, 3, 4, 5, 6, 7, 8, 9in text format. This archive contains the text file versions of the Additional files 1, 2, 3, 4, 5, 6, 7, 8, 9 as TAB-separated lists. They contain additionally the explicit genomic sequences of all the predictions.
Click here for file
Acknowledgements
AS was supported by the grant 205321-105945 from the Swiss National Science Foundation (to MZ). PL was supported by Dr. Mildred Scheel Stiftung für Krebsforschung of the Deutsche Krebshilfe. AA was supported by the FRAXA Research Foundation postdoctoral fellowship. SP was supported by the Lehman Brothers Foundation Fellowship through the Leukemia & Lymphoma Society. This research was also supported by the grant P01-GM073047-01 from the NIH (to TT).
==== Refs
Bartel D MicroRNAs: genomics, biogenesis, mechanism, and function Cell 2004 116 281 197 14744438 10.1016/S0092-8674(04)00045-5
He L Hannon G MicroRNAs: small RNAs with a big role in gene regulation Nat Rev Genet 2004 5 522 531 15211354 10.1038/nrg1379
Ambros V The functions of animal microRNAs Nature 2004 16 350 355 15372042 10.1038/nature02871
Lee Y Kim M Han J Yeom K Lee S Baek S Kim V MicroRNA genes are transcribed by RNA polymerase II EMBO J 2004 23 4051 4060 15372072 10.1038/sj.emboj.7600385
Lee Y Ahn C Han J Choi H Kim J Yim J Lee J Provost P Radmark O Kim S Kim V The nuclear RNase III Drosha initiates microRNA processing Nature 2003 25 415 419 14508493 10.1038/nature01957
Lund E Guttinger S A C Dahlberg J Kutay U Nuclear export of microRNA precursors Science 2004 303 95 98 14631048 10.1126/science.1090599
Hutvagner G McLachlan J Pasquinelli A Balint E Tuschl T Zamore P A cellular function for the RNA-interference enzyme Dicer in the maturation of the let-7 small temporal RNA Science 2001 293 834 838 11452083 10.1126/science.1062961
Krol J Sobczak K Wilczynska U Drath M Jasinska A Kaczynska D Krzyzosiak W Structural features of microRNA precursors and their relevance to miRNA biogenesis and small interfering RNA/short hairpin RNA design J Biol Chem 2004 279 42230 42239 15292246 10.1074/jbc.M404931200
Lim L Glasner M Yekta S Burge C Bartel D Vertebrate microRNA genes Science 2003 299 1540 12624257 10.1126/science.1080372
Berezikov E Guryev V van de Belt J Wienholds E Plasterk R Cuppen E Phylogenetic shadowing and computational identification of human microRNA genes Cell 2005 120 21 24 15652478 10.1016/j.cell.2004.12.031
Rodriguez A Griffiths-Jones S Ashurst J Bradley A Identification of mammalian microRNA host genes and transcription units Genome Res 2004 14 1902 1910 15364901 10.1101/gr.2722704
Baskerville S Bartel D Microarray profiling of microRNAs reveals frequent coexpression with neighboring miRNAs and host genes RNA 2005 11 241 247 15701730 10.1261/rna.7240905
Lim L Lau N Weinstein E Abdelhakim A Yekta S Rhoades M Burge C Bartel D The microRNAs of Caenorhabditis elegans Genes Dev 2003 17 991 1008 12672692 10.1101/gad.1074403
Lai E Tomancak P Williams R Rubin G Computational identification of Drosophila microRNA genes Genome Biol 2003 4 R42 12844358 10.1186/gb-2003-4-7-r42
Legendre M Lambert A Gautheret D Profile-based detection of microRNA precursors in animal genomes Bioinformatics 2005 21 841 845 15509608 10.1093/bioinformatics/bti073
Pfeffer S Zavolan M Grasser F Chien M Russo J Ju J John B Enright A Marks D Sander C Tuschl T Identification of virus-encoded microRNAs Science 2004 304 734 736 15118162 10.1126/science.1096781
Pfeffer S Sewer A Lagos-Quintana M Sheridan R Sander C Grasser F van Dyk L Ho C Shuman S Chien M Russo J Ju J Randall G Lindenbach B Rice C Simon V Ho D Zavolan M Tuschl T Identification of microRNAs of the herpesvirus family Nat Methods 2005 2 269 276 15782219 10.1038/nmeth746
Griffiths-Jones S The microRNA Registry Nucleic Acids Res 2004 1 D109 D111 14681370 10.1093/nar/gkh023
Tuschl T Comprehensive set of cloned small RNAs 2005
Seitz H Royo H Bortolin M Lin S Ferguson-Smith A Cavaille J A large imprinted microRNA gene cluster at the mouse Dlk1-Gtl2 domain Genome Res 2004 14 1741 1748 15310658 10.1101/gr.2743304
Xie X Lu J Kulbokas E Golub T Mootha V Lindblad-Toh K Lander E Kellis M Systematic discovery of regulatory motifs in human promoters and 3' UTRs by comparison of several mammals Nature 2005 434 338 345 15735639 10.1038/nature03441
Tanzer A Stadler P Molecular evolution of a microRNA cluster J Mol Biol 2004 339 327 335 15136036 10.1016/j.jmb.2004.03.065
Altuvia Y Landgraf P Lithwick G Elefant N Pfeffer S Aravin A Brownstein M Tuschl T Margalit H Clustering and conservation patterns of human microRNAs Nucleic Acids Res 2005 33 2697 2706 15891114 10.1093/nar/gki567
Hofacker I Priwitzer B Stadler P Prediction of locally stable RNA secondary structures for genome-wide surveys Bioinformatics 2004 20 186 190 14734309 10.1093/bioinformatics/btg388
Bonnet E Wuyts J Rouze P Van de Peer Y Evidence that microRNA precursors, unlike other non-coding RNAs, have lower folding free energies than random sequences Bioinformatics 2004 20 2911 2917 15217813 10.1093/bioinformatics/bth374
Wagner A Stadler P Viral RNA and evolved mutational robustness J Exp Zool 1999 285 119 127 10440723 10.1002/(SICI)1097-010X(19990815)285:2<119::AID-JEZ4>3.0.CO;2-D
Sempere L Freemantle S Pitha-Rowe I Moss E Dmitrovsky E Ambros V Expression profiling of mammalian microRNAs uncovers a subset of brain-expressed microRNAs with possible roles in murine and human neuronal differentiation Genome Biol 2004 5 R13 15003116 10.1186/gb-2004-5-3-r13
Regulatory RNA Gene Discovery
Hofacker I Fontana W Stadler P Bonhoeffer S Tacker M Schuster P Fast Folding and Comparison of RNA Secondary Structures Monatsh Chem 1994 125 167 188 10.1007/BF00818163
Zuker M Stiegler P Optimal computer folding of large RNA sequences using thermodynamics and auxiliary information Nucleic Acids Res 1984 9 133 148 6163133
Joachims T Schölkopf B, Burges C, Smola A Making large-Scale SVM Learning Practical Advances in Kernel Methods – Support Vector Learning 1993 Cambridge, USA: MIT-Press
Zuker M Mfold web server for nucleic acid folding and hybridization prediction Nucleic Acids Res 2003 31 3406 3415 12824337 10.1093/nar/gkg595
|
16274478
|
PMC1315341
|
CC BY
|
2021-01-04 16:27:46
|
no
|
BMC Bioinformatics. 2005 Nov 7; 6:267
|
utf-8
|
BMC Bioinformatics
| 2,005 |
10.1186/1471-2105-6-267
|
oa_comm
|
==== Front
Nutr JNutrition Journal1475-2891BioMed Central London 1475-2891-4-331628749910.1186/1475-2891-4-33ResearchSingle and repeated moderate consumption of native or dealcoholized red wine show different effects on antioxidant parameters in blood and DNA strand breaks in peripheral leukocytes in healthy volunteers: a randomized controlled trial [ISRCTN68505294] Arendt Bianca M [email protected] Sabine [email protected] Klaudia [email protected] Leonie [email protected] Rolf [email protected] Ulrich [email protected]üller Wolfgang-Ulrich [email protected] Roland [email protected] Department of Hemostasis and Transfusion Medicine, School of Medicine, University of Duesseldorf, Moorenstr. 5, 40225 Duesseldorf, Germany2 Institute for Medical Biometry, Informatics and Epidemiology, University Hospital of Bonn, Sigmund-Freud-Str. 25, 53105 Bonn, Germany3 Department of General Internal Medicine, University Hospital of Bonn, Sigmund-Freud-Str. 25, 53105 Bonn, Germany4 Institute for Medical Radiobiology, University Duisburg-Essen, 45122 Essen, Germany5 Institute for Molecular Biotechnology, RWTH Aachen, Worringerweg 1, 52074 Aachen, Germany2005 14 11 2005 4 33 33 4 6 2005 14 11 2005 Copyright © 2005 Arendt et al; licensee BioMed Central Ltd.2005Arendt 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
Red wine (RW) is rich in antioxidant polyphenols that might protect from oxidative stress related diseases, such as cardiovascular disease and cancer. Antioxidant effects after single ingestion of RW or dealcoholized RW (DRW) have been observed in several studies, but results after regular consumption are contradictory. Thus, we examined if single or repeated consumption of moderate amounts of RW or DRW exert antioxidant activity in vivo.
Methods
Total phenolic content and concentration of other antioxidants in plasma/serum, total antioxidant capacity (TEAC) in plasma as well as DNA strand breaks in peripheral leukocytes were measured in healthy non-smokers A) before, 90 and 360 min after ingestion of one glass of RW, DRW or water; B) before and after consumption of one glass of RW or DRW daily for 6 weeks. DNA strand breaks (SB) were determined by single cell gel electrophoresis (Comet Assay) in untreated cells and after induction of oxidative stress ex vivo with H2O2 (300 μM, 20 min).
Results
Both RW and DRW transiently increased total phenolic content in plasma after single consumption, but only RW lead to a sustained increase if consumed regularly. Plasma antioxidant capacity was not affected by single or regular consumption of RW or DRW. Effects of RW and DRW on DNA SB were conflicting. DNA strand breaks in untreated cells increased after a single dose of RW and DRW, whereas H2O2 induced SB were reduced after DRW. In contrast, regular RW consumption reduced SB in untreated cells but did not affect H2O2 induced SB.
Conclusion
The results suggest that consumption of both RW and DRW leads to an accumulation of phenolic compounds in plasma without increasing plasma antioxidant capacity. Red wine and DRW seem to affect the occurrence of DNA strand breaks, but this cannot be referred to antioxidant effects.
==== Body
Background
Polyphenols show antioxidant activity in vitro and probably also in vivo and therefore might protect from oxidative stress related diseases such as cardiovascular disease, cancer, or chronic inflammatory processes [1,2].
Red wine is rich in polyphenols [3,4] and is one of the main polyphenol sources in western countries [5,6]. Acute antioxidant effects after single ingestion of native and dealcoholized red wine as measured by increased antioxidant capacity in plasma or serum [7-12], reduced susceptibility of LDL to oxidation ex vivo [9], decreased plasma and urinary of F2-isoprostane concentrations [13] or protection from radiation-induced DNA damage [14] have been shown in several studies. Likewise, regular intake of red wine, dealcoholized red wine or red wine extract for 10 – 30 days induced antioxidant effects in some studies [15-20], but not in all [21-23]. Results of Cartron et al. [24] and van Golde et al. [25] even suggest prooxidant effects if red wine is consumed regularly. Hence it remains unclear if regular consumption of native or dealcoholized red wine can increase plasma antioxidant capacity and protects from oxidation in vivo. However, a sustained antioxidant response is necessary to achieve protection from oxidative stress related diseases. Hence, the aim of our study was to investigate if regular intake of moderate amounts of red wine could improve antioxidant parameters in healthy subjects, and if dealcoholized red wine could be an alternative source providing polyphenols without increasing the alcohol intake. We conducted two randomized, controlled intervention trials to examine the effects of single and repeated intake of native and dealcoholized red wine, respectively, on total phenolic content and antioxidant parameters in blood as well as on DNA strand breaks in peripheral leukocytes in healthy volunteers.
Methods
Subjects
Men and women, aged 18 – 50 y, free from any known diseases were included in the study. Smoking, excessive exercise (> 10 h/wk), pregnancy and lactation, present or former alcohol or drug abuse, medications interfering with antioxidants or intake of alcohol, supplementation of vitamins, minerals or probiotics (2 weeks before or during the study) were exclusion criteria. The study was performed according to the guidelines of the Declaration of Helsinki and was approved by the ethics committee of the University of Bonn. All subjects gave their informed written consent.
Study design
The study consisted of two parts, a single-dose analysis investigating the effects of a single ingestion of red wine (RW) versus dealcoholized red wine (DRW) and a long-term dietary intervention trial in which RW and DRW were consumed daily for 6 wk. Both trials were prospective, randomized and controlled studies.
Single-dose analysis
After an overnight fast, 27 healthy non-smokers (demographic data are shown in Table 1) were randomized to receive a single dose of 200 mL red wine (group RW; n = 9), 175 mL dealcoholized red wine (group DRW; n = 9), or 200 mL mineral water (controls; n = 9). Total phenolic content, total antioxidant capacity and concentrations of the major antioxidants (ascorbic acid, uric acid, albumin, bilirubin) in plasma, as well as DNA strand breaks in leukocytes (with and without ex vivo H2O2 treatment to induce oxidative stress) were measured before, 90 and 360 min after ingestion. Subjects were instructed to abstain from polyphenol rich foodstuff and from any alcoholic drinks 24 h before the first blood sampling until completion of the study. For this purpose they received a list of acceptable foods which are considered to be low in polyphenols. Compliance was controlled by a self-estimated 24-h dietary record. One hour after the first blood sampling, participants could choose from a breakfast buffet composed of foods low in polyphenols (white bread, butter, curd cheese, yogurt, cheese, honey, milk, fruit infusion, mineral water). All subjects were allowed to leave the study center at 90 min, when the initial blood samples had been drawn, and returned at 360 min for the follow-up blood samples. They were allowed to have lunch in between following the dietary restrictions concerning antioxidants.
Table 1 Characteristics of the subjects of the single-dose and the dietary intervention trial
Single-dose analysis Dietary intervention trial
Group RW DRW Controls RW DRW Controls
Subjects (f/m), n 9 (7/2) 9 (8/1) 9 (6/3) 24 (12/12) 25 (15/10) 25 (15/10)
Age, years 27.1 ± 9.0a 26.2 ± 3.5b 31.4 ± 5.8a,b 30.0 ± 8.3 26.7 ± 6.4 28.8 ± 7.1
BMI kg/m2 20.8 ± 1.1 21.8 ± 1.9 22.0 ± 2.8 23.5 ± 3.3c,d 21.9 ± 2.1c 21.4 ± 2.2d
Exercise h/week 3.4 ± 2.9 2.7 ± 3.3 1.9 ± 3.3 2.2 ± 2.2 2.0 ± 2.0 2.9 ± 2.4
RW: Red wine; DRW: dealcoholized red wine; f: female, m: male
Values given are means ± SD. Values marked with identical letters differ, p ≤ 0.05 by Mann-Whitney U-test
Dietary intervention trial
Seventy-eight healthy subjects were enrolled in the dietary intervention trial. They ingested 200 mL of red wine (group RW, 27 subjects) or 175 mL dealcoholized red wine (group DRW, 26 subjects) daily within one hour after dinner for 6 wk. The control group (25 subjects) did not receive any study drink. Subjects were instructed not to drink more than 2 cups (150 mL each) of coffee, black or green tea, and 2 glasses (200 mL each) of fruit juice per day, and to renounce from grape juice, multivitamin juices, and alcoholic beverages starting one week before the intervention throughout the whole study period.
Blood samples were drawn before and after intervention after an overnight fast (between 07.30 and 09.00 a.m.), and about 12 h after the last ingestion of RW or DRW. In addition to the laboratory parameters measured in the single-dose analysis, α-tocopherol concentration in serum was determined as changes are expected only in the long term [26]. To control compliance to dietary restrictions and assess possible changes of dietary patterns due to seasonal variations during the study period, self-estimated 7-day dietary records had to be completed in the week before and in the last week of intervention.
Study drinks
The red wine used in the present studies was Spätburgunder, 1999, Marienthaler Klostergarten, Ahr, Germany. Dealcoholized red wine was produced by vacuum extraction of alcohol from the same batch. The amounts of flavonoids and phenolic acids ingested from a single serving of RW (200 mL) and DRW (175 mL) are listed in Table 2. The application of 175 mL DRW based on the assumption that 12.5% of the volume (25 mL / 200 mL) would be lost due to alcohol extraction, which would increase polyphenol concentration in DRW. However, subsequent analysis revealed a lower polyphenol content in DRW due to the processing. Thus, intake of polyphenols, especially flavonoids, was slightly lower from 175 mL DRW compared to 200 mL RW. The water for the control group in the single-dose analysis was Markus Brunnen "Still" (Vereinte Mineral- und Heilquellen, Rosbach, Germany), a carbonated natural mineral water from which iron is removed.
Table 2 Polyphenol intake from a single serving of red wine or dealcoholized red wine
RW DRW
Serving, mL 200 175
Total phenolics,3 mg CE 293.2 271.6
TEAC, mmol/L 3.8 2.7
Phenolic acids
Gallic acid, mg 8.0 9.4
Caffeic acid,1 mg 3.7 3.1
p-Coumaric acid,2 mg 0.7 0.8
Flavonoids
Catechin, mg 26.5 10.8
Epicatechin, mg 14.4 8.5
Malvidin, mg 8.5 4.7
Peonidin, mg 1.0 0.5
RW: Red wine; DRW: dealcoholized red wine; CE: catechin equivalents; TEAC: trolox equivalent antioxidant capacity
1calculated from caftaric acid
2calculated from p-coumaroyl-glucosyl-tartrate
3Folin method
Dietary intake of polyphenols
The subjects received a standardized dietary record which they completed for 24 h (single-dose analysis) or 7 days (dietary intervention trial), respectively. To determine polyphenol intake as exactly as possible, polyphenol rich foods were listed in detail. Calculation of the intake of flavonoids (kaempferol, quercetin, myricetin, catechin, epicatechin, epigallocatechin, gallocatechin, naringenin, cyanidin, peonidin, petunidin, and malvidin) and phenolic acids (salicylic, p-hydroxy benzoic, gallic, syringic, and ellagic acid) was based on data of Linseisen et al. [27] and Radtke et al. [28], which were completed by data for quercetin and kaempferol in tomato products [29] and catechin and epicatechin in apples, red grapes [30], and black tea [31].
Collection of samples
Peripheral venous blood was collected in Vacutainer® tubes (Becton Dickinson, Heidelberg, Germany) containing Li-heparin or no anticoagulant. Samples were protected from light and stored on ice until centrifugation (3000 × g, 20 min, 4°C). Plasma samples for determination of total phenolic content and antioxidant capacity were stored at -70°C, and for determination of albumin, uric acid and bilirubin at -20°C. For measurement of ascorbic acid, plasma was mixed with 5% trichloro acetic acid and centrifuged (3 min, 12000 × g). Supernatants were stored at -70°C. Serum was frozen at -20°C until measurement of α-tocopherol. For determination of DNA damage, heparinized blood was kept in the dark at room temperature until processing 60 – 120 min after sampling.
Antioxidants in plasma and serum
Total phenolic content in plasma (TPP) was determined by the Folin-Ciocalteu method modified by Serafini et al. [10] to avoid plasma protein interference. Unlike Serafini et al. [10], we centrifuged the thawed plasma samples at 12000 × g for 5 min. To remove plasma proteins completely, 2 mol/L metaphosphoric acid (Merck, Darmstadt, Germany) was used for precipitation and an additional centrifugation step (2700 × g, 3 min) was introduced for the combined supernatants before adding Folin-Ciocalteu reagent (Fluka Chemie, Buchs, Switzerland). Experiments were performed in duplicate.
Plasma antioxidant capacity was determined by the Trolox equivalent antioxidant capacity (TEAC) assay as described previously [32]. The antioxidant capacity is given in comparison to a 1 mmol/L standard solution of 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox) (Sigma, Deisenhofen, Germany). Experiments were performed in triplicate.
To control, if antioxidants other than polyphenols could have an impact on TEAC, the following major antioxidants in blood were also measured. Vitamin C concentration in plasma was determined colorimetrically according to Speitling et al. [33]. Experiments were performed in duplicate. Uric acid, albumin and bilirubin in plasma were determined by routine procedures at the Institute for Clinical Chemistry and Laboratory Medicine, University Duesseldorf, Germany. Concentration of α-tocopherol in serum was measured by High Performance Liquid Chromatography at the Research Centre for Diabetes, Duesseldorf, Germany.
DNA strand breaks in peripheral leukocytes
Single Cell Gelelectrophoresis (Comet Assay) was performed to assess DNA strand breaks in peripheral leukocytes. In this assay, cells are embedded in agarose on slides, incubated in PBS (control cells) or H2O2 (induction of oxidative stress), and lysed. During subsequent electrophoresis, DNA regions that are relaxed due to strand breaks migrate towards the anode forming a typical comet structure.
This assay was performed according to the standard protocol of Bauch et al. [34] adapted to the use of hydrogen peroxide as a stressor. To simulate the in vivo situation as far as possible and to avoid artifacts by density centrifugation, whole blood was used instead of isolated mononuclear cells. Ten μL heparinized whole blood were diluted with 90 μL PBS (10 mmol/L, pH 7.4, without Ca2+/Mg2+) (Sigma) to obtain a single cell suspension. After preparation of the slides, they were incubated for 20 min at 4°C with either 300 μmol/L H2O2 (Sigma) in PBS to induce DNA strand breaks or in pure PBS (controls). Electrophoresis was performed at 4°C and 1 V/cm for 4 min on a flat-bed electrophoresis apparatus with electrode spacing of 39 cm (Mega Horizontal Gelbox Safety; Molecular Bio-Products, San Diego, USA). Experiments were performed in duplicate.
For analysis, gels were rehydrated in aqua bidest. (10 min, room temperature) and DNA was dyed with 150 μL propidium iodide (Sigma) (25 μmol/L, solved in 125 mmol/L Tris with 123 mmol/L NaCl in aqua bidest., pH 7.5). Measurements were performed with the image analysis software package Comet Assay II (Perceptive Instruments, Suffolk, UK) coupled to a fluorescence microscope (Leitz DM RB, Leica, Bensheim, Germany) with a CCD camera (Cohu, San Diego, USA). Fifty cells were measured on each slide and Tail Intensity (TI) and Tail Moment (TM) were calculated to quantify DNA strand breaks.
DNA strand breaks in untreated cells reflect the steady state between DNA damage and repair in vivo [35-37] and are therefore referred to as "endogenous" DNA damage in this paper. "Exogenous" DNA strand breaks are those induced by treatment with H2O2 ex vivo and were calculated as the difference between treated and untreated cells.
Sample size calculation
The sample size calculation was based on plasma antioxidant capacity (TEAC), which was considered as main outcome and based on data from a previous study with healthy subjects who ingested polyphenol rich fruit-juices or a fruit-vegetable concentrate [32]. With 6 subjects in each group (RW, ERW, controls) the single-dose analysis has a power of 90% to detect TEAC changes >0.033 mmol/L, which was the mean difference between two measurements in the previous study. To account for possible drop-outs, we included 9 subjects in each group, adding up to a total of 27 participants.
Mean TEAC difference in the previous long-term study was 0.075 mmol/L. To detect changes ≥ 0.075 mmol/L and differences between the groups with a power of 90%, a sample size of 25 subjects in each group (RW, DRW, controls) was calculated. Level of significance was 5% for all tests.
Statistical analysis
Values given are means ± SD unless indicated otherwise. Differences between the groups were examined using the Mann-Whitney U-test. The effects of the study drinks on the investigated parameters were evaluated by comparing the values obtained at different time intervals with baseline for each group using the Wilcoxon signed rank test. Differences indicated by p < 0.05 were considered statistically significant. Statistical tests were performed with SPSS 10.0 for Windows (SPSS Inc., Chicago, IL, USA).
Results
Single-dose analysis
The day before the study the subjects ingested 22.9 ± 47.4 mg polyphenols (Flavonoids: 2.1 ± 5.3 mg; Phenolic acids: 20.8 ± 46.2 mg). This corresponds to 8 – 16% of the average intake observed in another German cohort [27,28] and in our dietary intervention trial (see below) indicating good compliance to the dietary restrictions.
Results for the antioxidant parameters in plasma are presented in Table 3. Total phenolic content in plasma increased 90 min after ingestion of RW (p = 0.008) or DRW (p = 0.008). After 360 min values were higher than at baseline in groups RW (p = 0.02) and DRW (p = 0.008), but also in control subjects (p = 0.01). Vitamin C concentration in plasma decreased 360 min after RW consumption (p = 0.008), but increased in group DRW (p = 0.02). Significant changes of endogenous antioxidants, i.e. uric acid, albumin or bilirubin, have been observed after consumption of RW and DRW, and also in the control group, but TEAC was not altered significantly in any group.
Table 3 Antioxidant parameters in healthy volunteers after single ingestion of native or dealcoholized red wine
RW DRW Controls
Parameter
TPP, mg CE/L
Baseline 15.1 ± 2.0 15.9 ± 2.0 13.9 ± 2.8
90 min 17.7 ± 1.8*a 18.3 ± 2.5* b 14.3 ± 3.5a,b
360 min 15.5 ± 1.8* 16.6 ± 2.0* 14.3 ± 2.7*
TEAC, mmol/L
Baseline 1.22 ± 0.08 1.25 ± 0.08 1.23 ± 0.05
90 min 1.25 ± 0.07 1.28 ± 0.08 1.26 ± 0.05
360 min 1.23 ± 0.08 1.24 ± 0.07 1.24 ± 0.04
Vitamin C, mg/dL
Baseline 1.33 ± 0.26 1.15 ± 0.18 1.14 ± 0.30
90 min 1.27 ± 0.21 1.18 ± 0.18 1.17 ± 0.35
360 min 1.11 ± 0.22* 1.35 ± 0.32* 1.14 ± 0.26
Uric acid, mg/dL
Baseline 4.6 ± 1.0 4.8 ± 1.2 4.6 ± 1.1
90 min 5.3 ± 1.0* 4.9 ± 1.1 4.9 ± 1.2
360 min 4.5 ± 0.9 4.5 ± 1.0 4.4 ± 1.1
Albumin, g/dL
Baseline 4.1 ± 0.4 4.4 ± 0.4 4.1 ± 0.4
90 min 4.4 ± 0.4 4.4 ± 0.5 4.3 ± 0.3*
360 min 4.4 ± 0.4 4.4 ± 0.5 4.4 ± 0.5
Bilirubin, mg/dL
Baseline 0.73 ± 0.44 0.69 ± 0.27 0.54 ± 0.23
90 min 0.71 ± 0.37 0.68 ± 0.22 0.61 ± 0.27*
360 min 0.52 ± 0.29* 0.53 ± 0.26* 0.50 ± 0.22
TM0, arbitrary units
Baseline 1,86 ± 0,48 1,98 ± 0,33 2,19 ± 0,67
90 min 2,03 ± 0,43 2,36 ± 0,23 2,43 ± 0,43
360 min 2,61 ± 0,43* 2,67 ± 0,24* 2,33 ± 0,44
TM300, arbitrary units
Baseline 1,69 ± 0,92 1,43 ± 0,78 1,15 ± 1,07
90 min 1,39 ± 0,65 1,05 ± 0,87 0,78 ± 1,23
360 min 1,22 ± 0,67c 0,21 ± 0,45*c,d 1,21 ± 0,52d
Amounts of study drinks ingested were 200 mL red wine (RW), 175 mL dealcoholized red wine (DRW) or 200 mL water (Controls)
TPP: total phenolic content in plasma; CE: catechin equivalents; TEAC: trolox equivalent antioxidant capacity; TM0: Tail Moment in untreated cells (endogenous DNA strand breaks);
TM300: Tail Moment in cells treated with 300 μM H2O2 for 20 min (exogenous DNA strand breaks)
Values are means ± SD, n = 9 for all experiments except for control group at 90 min, n = 7 for all parameters; DRW, n = 8 for TM0, TM300, and control group at 360 min, n = 6 for TM0, TM300
*Values different from baseline, P < 0.05 by Wilcoxon signed rank test
a-dValues marked with identical letters differ between groups, P < 0.05 by Mann-Whitney U-test
Endogenous DNA strand breaks increased 360 min after consumption of either RW or DRW compared to baseline values. In group DRW an increase could be observed after 90 min, but only with borderline significance (p = 0.05). Hydrogen peroxide-induced DNA strand breaks decreased in group DRW at 360 min, whereas no alterations could be observed after RW consumption or in controls. Statistical analysis provides the same results for TI (Figure 1) and TM (Table 3).
Figure 1 DNA strand breaks in peripheral leukocytes after single ingestion of native or dealcoholized red wine. A) Tail Intensity in untreated cells (endogenous DNA strand breaks) B) Tail Intensity in cells treated with 300 μM H2O2 for 20 min (exogenous DNA strand breaks)Amounts of study drinks ingested were 200 mL red wine (RW), 175 mL dealcoholized red wine (DRW) or 200 mL water (Controls) The box represents the distribution falling between the 25th and 75th percentiles, with the median as the horizontal line within the box. The whiskers connect the largest and smallest values not categorized as outliers or extreme values, which are represented by single data points. *, ** Values different from baseline, *P < 0.05; ** P < 0.01 by Wilcoxon signed rank test a,bValues marked with identical letters differ between groups, P < 0.01 by Mann-Whitney U-test
Dietary intervention trial
Four subjects were lost during the study. One woman in group DRW did not tolerate the study drink. The other subjects were allocated to group RW. Of those, one woman fell ill, which was not related to the intervention. One man started smoking during the study and the other one was lost to follow-up. Demographic data of the 74 subjects who completed the study are summarized in Table 1.
Flavonoid intake increased significantly in group RW (28.9 ± 16.6 vs. 76.7 ± 24.5 mg/d; p < 0.001) and DRW (39.1 ± 20.5 vs. 64.2 ± 27.6 mg/d; p < 0.001), but not in controls (37.7 ± 21.3 vs. 41.0 ± 24.8 mg/d). During intervention, the flavonoid intake in group RW exceeded that in group DRW (p = 0.005), and it was higher in group RW (p < 0.001) and DRW (p = 0.005) than in controls. Intake of phenolic acids increased during supplementation of RW (107.8 ± 75.5 vs. 135.1 ± 82.5 mg/d; p = 0.01) and DRW (123.5 ± 83.2 vs. 158.3 ± 102.7 mg/d; p = 0.02), but also in controls (87.2 ± 69.7 vs. 135.6 ± 100.3 mg/d; p = 0.03). The amounts did not differ between the groups before or during intervention.
The difference to the average intake of 54 mg/d flavonoids [27] and 222 mg/d phenolic acids [28] reported by others could be explained by the restriction of polyphenol rich foods, especially coffee, which is a major source of phenolic acids in Germany [28].
Total phenolic content in plasma (Table 4) increased after 6 wk regular consumption of RW (p = 0.002), but did not change in group DRW or controls. A significant increase of plasma uric acid concentration was observed in controls (p = 0.04), and a decrease of bilirubin in group DRW (p = 0.02) (Table 4). In contrast, vitamin C and albumin in plasma, α-tocopherol in serum, and plasma antioxidant capacity did not change significantly in any group (Table 4). There was no significant difference between the groups before and after 6 wk of intervention considering the antioxidant parameters.
Table 4 Antioxidant parameters in healthy volunteers after regular consumption of native or dealcoholized red wine
RW DRW Controls
Subjects, n 24 25 25
Parameter
TPP, mg CE/L
Baseline 15.6 ± 1.6 15.6 ± 2.1 15.5 ± 1.4
6 wk 16.4 ± 1.4* 16.0 ± 2.0 15.3 ± 2.2
TEAC, mmol/L
Baseline 1.47 ± 0.06 1.44 ± 0.08 1.46 ± 0.08
6 wk 1.45 ± 0.04 1.42 ± 0.07 1.43 ± 0.07
Vitamin C, mg/dL
Baseline 1.30 ± 0.28 1.43 ± 0.36 1.48 ± 0.29
6 wk 1.35 ± 0.26 1.42 ± 0.39 1.57 ± 0.28
α-Tocopherol, mg/dL
Baseline 11.3 ± 3.0 11.2 ± 2.6 11.2 ± 2.9
6 wk 11.3 ± 3.0 11.5 ± 2.7 11.0 ± 2.7
Uric acid, mg/dL
Baseline 4.7 ± 1.2 4.4 ± 1.1 4.4 ± 1.2
6 wk 4.9 ± 1.2 4.5 ± 0.9 4.7 ± 1.5*
Albumin, g/dL
Baseline 4.2 ± 0.4 4.3 ± 0.4 4.2 ± 0.4
6 wk 4.3 ± 0.4 4.1 ± 0.6 4.2 ± 0.5
Bilirubin, mg/dL
Baseline 0.68 ± 0.28 0.68 ± 0.33 0.64 ± 0.33
6 wk 0.61 ± 0.35 0.59 ± 0.27* 0.62 ± 0.35
TM0, arbitrary units
Baseline 2,18 ± 0,56 1,97 ± 0,56 2,22 ± 0,79
6 wk 1,88 ± 0,48* 2,05 ± 0,70 1,95 ± 0,49
TM300, arbitrary units
Baseline 1,02 ± 0,69 0,87 ± 0,72 0,95 ± 0,54
6 wk 0,96 ± 0,49 0,92 ± 0,65 1,18 ± 0,50
Amounts ingested daily for 6 weeks were 200 mL red wine (RW) or 175 mL dealcoholized red wine (DRW). Control subjects did not receive any study drink.
TPP: total phenolic content in plasma; CE: catechin equivalents; TEAC: trolox equivalent antioxidant capacity; TM0: Tail Moment in untreated cells (endogenous DNA strand breaks);
TM300: Tail Moment in cells treated with 300 μM H2O2 for 20 min (exogenous DNA strand breaks)
Values are means ± SD
*Values different from baseline, P < 0.05 by Wilcoxon signed rank test
Endogenous DNA strand breaks (Figure 2, Table 4) decreased only after regular consumption of RW (TI: p = 0.03; TM: p = 0.04), but not in group DRW and controls. Susceptibility of leukocytes towards H2O2-induced DNA damage was not altered in any group. Differences in endogenous or ex vivo induced DNA strand breaks could not be observed between the groups, neither before nor after 6 wk intervention (Figure 2, Table 4).
Figure 2 DNA strand breaks in peripheral leukocytes after regular consumption of native or dealcoholized red wine. Amounts ingested daily for 6 weeks were 200 mL red wine (RW) or 175 mL dealcoholized red wine (DRW). Control subjects did not receive any study drink. TI0: Tail Intensity in untreated cells (endogenous DNA strand breaks); TI300: Tail Intensity in cells treated with 300 μM H2O2 for 20 min (exogenous DNA strand breaks)Boxes represent the distribution falling between the 25th and 75th percentiles, with the median as the horizontal line within the box. The whiskers connect the largest and smallest values not categorized as outliers or extreme values, which are represented by single data points. *Value different from baseline, P < 0.05 by Wilcoxon signed rank test.
Discussion
The present studies demonstrate that polyphenols from RW and DRW are bioavailable and accumulate in plasma after regular consumption. Even though total antioxidant capacity in plasma did not change after RW or DRW consumption, both drinks affected DNA strand breaks in peripheral leukocytes. Results are contradictory, as DNA damaging effects of RW and DRW occur after single ingestion, whereas regular RW consumption seems to protect from endogenous DNA strand breaks. Further investigations are required to clarify whether interactions of red wine polyphenols with cellular DNA result in genotoxic or anticarcinogenic effects.
The results of the single-dose analysis confirm that polyphenols are bioavailable from both RW and DRW [8,10,11], and that alcohol does not significantly affect polyphenol absorption in humans [38-40]. Accumulation of phenolic compounds in plasma could be measured after regular consumption of RW corresponding to previous findings [16,19] but not in group DRW. We assume that this is rather due to the slightly lower intake of polyphenols from DRW than to different alcohol content [38-40].
The impact of phenolic compounds from RW and DRW on the pro-/antioxidant balance in plasma seems to be low, as plasma antioxidant capacity was neither altered significantly in the short nor in the long term as also reported by others [7,20,22,23]. There was no effect of RW or DRW consumption on antioxidant vitamins in the dietary intervention trial corresponding to the findings of previous studies [17,20]. Interestingly, the concentration of vitamin C in plasma increased 360 min after consumption of one single dose of DRW whereas it decreased after RW suggesting that the polyphenols might act as antioxidants in vivo but this is counteracted by prooxidant effects of alcohol in native RW [13,23]. Changes in vitamin C levels did not have a significant impact on plasma antioxidant capacity but might have protected cellular DNA from oxidation ex vivo, which will be discussed later on.
Uric acid, one of the main antioxidants contributing to plasma antioxidant capacity [13,32], increased after single RW consumption as expected [13,41]. However, the increase of 15% in TEAC observed after 90 min, even in combination with a 17% increase of total phenolic content did not lead to a significant rise of plasma TEAC. This supports the observations of Cao et al. [7] implying that the TEAC method is not sensitive enough to measure small changes in the pro-antioxidant balance after ingestion of foods rich in polyphenols.
Further changes of albumin, uric acid and bilirubin concentrations occurred in both trials, probably due to normal biological variance, but they were not associated with changes of plasma TEAC.
In contrast to antioxidant capacity, there was a significant effect of RW and DRW on DNA strand breaks in peripheral leukocytes suggesting interactions of red wine polyphenols with cellular DNA. Regular consumption of RW protects cellular DNA from strand breaks in vivo, which corresponds to findings of previous studies [18,42], probably due to DNA stabilizing and/or antioxidant effects of RW polyphenols [43]. In contrast, single ingestion of RW or DRW increased endogenous DNA strand breaks. This was not expected as DNA damaging effects of RW polyphenols have been observed only in vitro and for concentrations of 25–1000 μM [44-46], whereas in human plasma polyphenols occur mainly as metabolites and in concentrations below 1 μM [6]. As the same outcome was found in groups RW and DRW, genotoxic effects of alcohol [47] can also be excluded. Since strand breaks occur during DNA repair via base or nucleotide excision [35], increased DNA strand breaks as observed in our study might also result from activation of DNA repair enzymes by polyphenols [48]. To clarify the mechanisms, a modified protocol for the single cell gel electrophoresis allowing to separately assessing oxidation of purine and pyrimidin bases [35] as well as DNA repair kinetics [49] should be used in future studies.
The discrepancy between increased DNA strand breaks after single and reduced DNA strand breaks after repeated consumption of RW might be due to different polyphenols which might have been present in plasma or leukocytes after single or regular consumption, respectively. This would be plausible regarding the different elimination half lives of the various RW polyphenols and their metabolites [6,39,50,51], but quantitative analysis of single polyphenolic compounds in plasma/cells would be necessary to confirm this. Furthermore, the different mode of RW consumption in the short- and the long-term study (after an overnight fast vs. regularly after dinner) could also provide an explanation. Since consumption of RW with meals reduces prooxidative effects occurring post-prandially [24,52,53], the DNA protective effects of RW in our dietary intervention trial might have resulted from reduced prooxidative effects of the meals, which were lacking in the single-dose analysis.
Exogenous DNA strand breaks, which reflect the activity of non-enzymatic antioxidants [35,37], have only been reduced after single consumption of DRW, which corresponds to the results of a recent study showing that consumption of 300 mL DRW protected lymphocytes from radiation-induced DNA damage ex vivo [14]. As in our study the plasma phenolic content was not different between groups RW and DRW, the increased vitaminC concentration in plasma observed in group DRW might have protected leukocyte DNA against oxidative stress ex vivo. Indeed, DNA protecting effects of vitamin C have been described in vitro [54,55] and are also supposed to occur in vivo [36,56,57]. This is also supported by Greenrod et al. [14] who observed that the contribution of catechin to DNA protecting effects or DRW is quite small, and that other factors seem to play an important role.
It has to be noted that in the dietary intervention trial consumption of RW and DRW lead to an increase in flavonoid intake compared to baseline and to controls, whereas phenolic acid consumption increased in RW, DRW and controls after 6 weeks compared to baseline without any differences between the groups. This is probably due to seasonal variations of the food pattern as our study period was between May and July, when consumption of fresh fruit rich in phenolic acids increased (data not shown). Hence, the observed effects of RW on DNA strand breaks in peripheral leukocytes can be attributed to the additional intake of flavonoids rather than to phenolic acids.
The major advantage of our study compared to others is the investigation of both single and regular consumption of RW and DRW which has only been done so far by Cartron et al. [24]. In both studies substantial differences between short- and long-term effects occurred implying that single-dose analysis alone are not appropriate to investigate potential health effects of red wine consumption.
However, our study has some constraints, regarding the different polyphenol intake with RW and DRW, which does not allow a direct comparison of both study drinks. In future studies, drinks with equal polyphenol composition should be applied. Moreover, in the single dose analysis TPP increased 360 min after intervention in groups RW and DRW, but also in controls. This could be due to 1) unknown amounts of polyphenols in "permitted" foods (e.g. white bread or pasta). These foods are generally considered to be low in polyphenols and thus, are not listed in the polyphenol database [27,28]. Nevertheless, they may have provided considerable amounts of polyphenols, especially phenolic acids [50] leading to an increase of TPP in the control group. 2) As food intake at lunchtime was not recorded, we cannot exclude that the volunteers failed to comply with the polyphenol poor diet in the time interval between blood sampling at 90 and 360 min. To avoid these confounding factors in future short-term studies we strongly recommend to provide the subjects with standardized amounts of polyphenol free formula diets instead of foods low in polyphenols ad libitum.
It is unlikely that random inhomogeneities of the groups (Table 1) would have influenced our results. In the single-dose analysis, control subjects were only slightly older (≈ 5 years) than those in group RW and DRW and therefore metabolic alterations are not expected. In the dietary intervention trial, the higher body weight and BMI in group RW could have reduced the response to RW, as the intake of polyphenols / kg body weight was lower in that group. However, effects on plasma phenolic content and DNA strand breaks occurred only with RW, and even more pronounced effects might have been observed in subjects with a lower body weight.
Conclusion
One glass of RW or DRW is sufficient to transiently increase concentration of phenolic compounds in plasma, but accumulation after regular consumption could only be shown for RW. Polyphenols from RW and DRW seem to affect DNA damage in leukocytes, but antioxidant effects probably play only a minor role. Endogenous DNA strand breaks increased after a single dose of RW or DRW but could be reduced after 6 wk daily consumption of RW. As DNA damage is involved in cancer initiation, the mechanisms and the physiological relevance of these findings are worth being considered in future studies.
Abbreviations
Red wine (RW); dealcoholized red wine (DRW); total antioxidant capacity (TEAC); total phenolic content in plasma (TPP).
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
BMA participated in study design and coordination, carried out the trolox equivalent antioxidant capacity assay and the single cell gel electrophoresis, did the statistical analysis and drafted the article. SE was involved in study design and realization and helped to draft the manuscript. KK measured total phenolic content and vitamin C concentration in plasma, and analysed the dietary records together with LG. RF did the sample size calculation and contributed to the statistical analysis. US supervised the clinical part of the study. W-UM supported the implementation, data analysis and interpretation of the single cell gel electrophoresis. RG had the initial idea and supervised the study. All authors read and approved the final manuscript.
Acknowledgements
We thank the State Winery Marienthal, Germany, for providing the wine, Carl Jung GmbH, Ruedesheim, Germany, for dealcoholization of the wine, Prof. Dr. Dietrich, Department of Wine Analysis and Beverage Research, Geisenheim, Germany, for polyphenol analysis of the study drinks, Dr. Susanto, Research Centre for Diabetes, Duesseldorf, Germany, for analysis of α-tocopherol in serum, and the Institute for Clinical Chemistry and Laboratory Medicine, University Duesseldorf, Germany, for analysis of uric acid in plasma. We thank Dr. Thomas Bauch, Institute for Medical Radiobiology, University Clinics of Essen, Germany, for introducing us to the comet assay and are grateful to the volunteers participating in these studies. The work was supported by scholarships from the Friedrich-Ebert-Foundation (BMA) and the Landesgraduiertenfoerderung North-Rhine-Westphalia (SE) and by a grant from the German Wine Academy Mainz, Germany.
==== Refs
Hollman PCH Evidence for health benefits of plant phenols: local or systemic effects? J Sci Food Agr 2001 81 842 852 10.1002/jsfa.900
Nijveldt RJ van Nood E van Hoorn DEC Boelens PG van Norren K van Leeuwen PAM Flavonoids: a review of probable mechanisms of action and potential applications Am J Clin Nutr 2001 74 418 425 11566638
Soleas GJ Diamandis EP Goldberg DM Wine as a biological fluid: history, production, and role in disease prevention J Clin Lab Analysis 1997 11 287 313 10.1002/(SICI)1098-2825(1997)11:5<287::AID-JCLA6>3.0.CO;2-4
Mattivi U Zulian C Nicolini G Valenti L Wine, biodiversity, technology, and antioxidants Ann N Y Acad Sci 2002 957 37 56 12074960
Hertog MGL Kromhout D Aravanis C Blackburn H Buzina R Fidanza F Giampaoli S Jansen A Menotti A Nedeljkovc S Flavonoid intake and long-term risk of coronary heart disease and cancer in the Seven Countries Study Arch Intern Med 1995 155 381 386 7848021 10.1001/archinte.155.4.381
Scalbert A Williamson G Dietary intake and bioavailability of polyphenols J Nutr 2000 130 2073S 2085S 10917926
Cao G Russel RM Lischner N Prior RL Serum antioxidant capacity is increased by consumption of strawberries, spinach, red wine or vitamin C in elderly women J Nutr 1998 128 2383 2390 9868185
Duthie GG Pedersen MW Gardner PT Morrice PC McJenkinson A McPhail DB Steele GM The effect of whisky and wine consumption on total phenol content and antioxidant capacity of plasma from healthy volunteers Eur J Clin Nutr 1998 52 733 736 9805220 10.1038/sj.ejcn.1600635
Serafini M Laranjinha JAN Almeida LM Maiani G Inhibition of human LDL lipid peroxidation by phenol-rich beverages and their impact on plasma total antioxidant capacity in humans J Nutr Biochem 2000 11 585 590 11137897 10.1016/S0955-2863(00)00124-8
Serafini M Maiani G Ferro-Luzzi A Alcohol-free red wine enhances plasma antioxidant capacity in humans J Nutr 1998 128 1003 1007 9614160
Simonetti P Gardana C Pietta P Plasma levels of caffeic acid and antioxidant status after red wine intake J Agric Food Chem 2001 49 5964 5968 11743793 10.1021/jf010546k
Whitehead TP Robinson D Allaway S Syms J Hale A Effect of red wine ingestion on the antioxidant capacity of serum Clin Chem 1995 41 32 35 7813078
Abu-Amsha Caccetta R Burke V Mori TA Beilin LJ Puddey IB Kroft KD Red wine polyphenols, in the absence of alcohol, reduce lipid peroxidative stress in smoking subjects Free Rad Biol Med 2001 30 636 642 11295361 10.1016/S0891-5849(00)00497-4
Greenrod W Stockley CS Burcham P Abbey M Fenech M Moderate acute intake of de-alcoholised red wine, but not alcohol, is protective against radiation-induced DNA damage ex vivo – Results of a comparative in vivo intervention study in younger men Mutat Res 2005 591 290 301 16083915
Carbonneau MA Leger CL Monnier L Bonnet C Michel F Fouret G Dedeieu F Descomps B Supplementation with wine phenolic compounds increases the antioxidant capacity of plasma and vitamin E of low-density lipoprotein without changing the lipoprotein Cu(2+)-oxidizability: possible explanation by phenolic location Eur J Clin Nutr 1997 51 682 690 9347289 10.1038/sj.ejcn.1600464
Fuhrman B Lavy A Aviram M Consumption of red wine with meals reduces the susceptibility of human plasma and low-density lipoprotein to lipid-peroxidation Am J Clin Nutr 1995 61 549 554 7872219
Chopra M Fitzsimons PEE Strain JJ Thurnham DI Howard AN Nonalcoholic red wine extract and quercetin inhibit LDL oxidation without affecting plasma antioxidant vitamin and carotenoid concentrations Clin Chem 2000 46 1162 1170 10926898
Leighton F Cuevas A Guasch V Pérez DD Strobel P San Martín A Urzua U Díez MS Foncea R Castillo O Plasma polyphenols and antioxidants, oxidative DNA damage and endothelial function in a diet and wine intervention study in humans Drugs Exp Clin Res 1999 XXV 133 141 10370876
Nigdikar SV Williams NR Griffin BA Howard AN Consumption of red wine polyphenols reduces the susceptibility of low-density lipoproteins to oxidation in vivo Am J Clin Nutr 1998 68 258 265 9701181
Tsang C Higgins S Duthie GG Duthie SJ Howie M Mullen W Lean MEJ Crozier A The influence of moderate red wine consumption on antioxidant status and indices of oxidative stress associated with CHD in healthy volunteers Br J Nutr 2005 93 233 240 15788107 10.1079/BJN20041311
de Rijke YB Demacker PN Assen NA Sloots LM Katan MB Stalenhoef AF Red wine consumption does not affect oxidizability of low-density lipoproteins in volunteers Am J Clin Nutr 1996 63 329 334 8602588
Sharpe PC McGrath LT McClean E Young IS Archbold GPR Effect of red wine consumption on lipoprotein (a) and other risk factors for atherosclerosis Q J Med 1995 88 101 108
van der Gaag MS van den Berg R van den Berg H Schaafsma G Hendriks HF Moderate consumption of beer, red wine and spirits has counteracting effects on plasma antioxidants in middle-aged men Eur J Clin Nutr 2000 54 586 591 10918470 10.1038/sj.ejcn.1601061
Cartron E Fouret G Carbonneau MA Lauret C Michel F Monnier L Descomps B Léger CL Red-wine beneficial long-term effect on lipids but not on antioxidant characteristics in plasma in a study comparing three types of wine – Description of two O-methylated derivatives of gallic acid in humans Free Radic Res 2003 37 1021 1035 14670010 10.1080/10715760310001598097
van Golde PHM Sloots LM Vermeulen WP Wielders JPM Hart HC Bouma BN van de Weil A The role of alcohol in the anti low density lipoprotein oxidation activity of red wine Atherosclerosis 1999 147 365 370 10559522 10.1016/S0021-9150(99)00206-3
Pietta P Simonetti P Gardana C Brusamolino A Morazzoni P Bombardelli E Relationship between rate and extent of catechin absorption and plasma antioxidant status Biochem Mol Biol Int 1998 46 895 903 9861443
Linseisen J Radtke J Wolfram G Flavonoidzufuhr Erwachsener in einem bayerischen Teilkollektiv der Nationalen Verzehrsstudie Z Ernährungswiss 1997 36 403 412 10.1007/BF01617836
Radtke J Linseisen J Wolfram G Phenolsäurezufuhr Erwachsener in einem bayerischen Teilkollektiv der Nationalen Verzehrsstudie Z Ernährungswissen 1998 37 190 197 10.1007/s003940050016
Stewart AJ Bozonnet S Mullen W Jenkins GI Lean ME Crozier A Occurrence of flavonols in tomatoes and tomato-based products J Agric Food Chem 2000 48 2663 2669 10898604 10.1021/jf000070p
Arts ICW van den Putte B Hollman PCH Catechin content of foods commonly consumed in the Netherlands. 1. Fruits, vegetables, staple foods, and processed foods J Agric Food Chem 2000 48 1746 1751 10820089 10.1021/jf000025h
Arts ICW Van den Putte B Hollman PCH Catechin content of foods commonly consumed in the Netherlands. 2. Tea, wine, fruit juices, and chocolate milk J Agric Food Chem 2000 48 1752 1757 10820090 10.1021/jf000026+
Arendt BM Boetzer AM Winkler P Lemoch H Rockstroh JK Berthold HK Spengler U Goerlich R Plasma antioxidant capacity of HIV-seropositive and healthy subjects during long-term ingestion of fruit juices or a fruit-vegetable-concentrate containing antioxidant polyphenols Eur J Clin Nutr 2001 55 786 792 11528495 10.1038/sj.ejcn.1601227
Speitling A Hüppe R Kohlmeier M Matiaske B Stelte W Thefeld W Wetzel S VERA-Schriftenreihe 1992 Niederkleen: Wiss. Fachverlag Dr. Fleck
Bauch T Böcker W Mallek U Müller W-U Streffer C Optimization and standardization of the "Comet Assay" for analyzing the repair of DNA damage in cells Strahlenther Onkol 1999 175 333 340 10432995 10.1007/s000660050020
Duthie SJ Ma A Ross MA Collins AR Antioxidant supplementation decreases oxidative DNA damage in human lymphocytes Cancer Res 1996 56 1291 1295 8640816
Møller P Loft S Oxidative DNA damage in human white blood cells in dietary antioxidant intervention studies Am J Clin Nutr 2002 76 303 310 12144999
Riso P Santangelo A Porrini M The comet assay for the evaluation of cell resistance to oxidative stress Nutr Res 1999 19 325 333 10.1016/S0271-5317(99)00002-0
Bell JRC Donovan JL Wong R Waterhouse AL German JB Walzem RL Kasim-Karakas SE (+)-Catechin in human plasma after ingestion of a single serving of reconstituted red wine Am J Clin Nutr 2000 71 103 108 10617953
Donovan JL Bell JRC Kasim-Karakas SE German JB Walzem RL Hansen RJ Waterhouse AL Catechin is present as metabolites in human plasma after consumption of red wine J of Nutrition 1999 129 1662 1668 10460201
Bub A Watzl B Heeb D Rechkemmer G Briviba K Malvidin-3-glucoside bioavailability in humans after ingestion of red wine, dealcoholized red wine and red grape juice Eur J Nutr 2001 40 113 120 11697443 10.1007/s003940170011
Abu-Amsha Caccetta R Croft KD Beilin LJ Puddey IB Ingestion of red wine significantly increases plasma phenolic acid concentrations but does not acutely affect ex vivo lipoprotein oxidizability Am J Clin Nutr 2000 71 67 74 10617948
Fenech M Stockley C Aitken C Moderate wine consumption protects against hydrogen peroxide-induced DNA damage Mutagenesis 1997 12 289 296 9237776
Duthie SJ Dobson VL Dietary flavonoids protect human colonocyte DNA from oxidative attack in vitro Eur J Nutr 1999 38 28 34 10338685 10.1007/s003940050043
Duthie SJ Johnson W Dobson VL The effect of dietary flavonoids on DNA damage (strand breaks and oxidised pyrimidines) and growth in human cells Mutat Res 1997 390 141 151 9150762
Johnson MK Loo G Effects of epigallocatechin gallate and quercetin on oxidative damage to cellular DNA Mutat Res 2000 459 211 218 10812333
Szeto YT Benzie IF Effects of dietary antioxidants on human DNA ex vivo Free Radic Res 2002 36 113 118 11999698 10.1080/10715760210161
Greenrod W Fenech M The principal phenolic and alcoholic components of wine protect human lymphocytes against hydrogen peroxide- and ionizing radiation-iduced DNA damage in vitro Mutagenesis 2003 18 119 126 12621066 10.1093/mutage/18.2.119
Ferguson LR Role of plant polyphenols in genomic stability Mutat Res 2001 475 89 111 11295156
Collins AR MA AG Duthie SJ The kinetics of repair of oxidative DNA damage (strand breaks and oxidized pyrimidines) in human cells Mutat Res 1995 336 69 77 7528897
Rechner AR Kuhnle G Bremner P Hubbard GP Moore KP Rice-Evans CA The metabolic fate of dietary polyphenols in humans Free Radic Biol Med 2002 33 220 235 12106818 10.1016/S0891-5849(02)00877-8
Boyle SP Dobson VL Duthie SJ Hinselwood DC Kyle JA Collins AR Bioavailability and efficiency of rutin as an antioxidant: a human supplementation study Eur J Clin Nutr 2000 54 774 782 11083486 10.1038/sj.ejcn.1601090
Natella F Ghiselli A Guidi A Ursini F Scaccini C Red wine mitigates the postprandial increase of LDL susceptibility to oxidation Free Radic Biol Med 2001 30 1036 1044 11316584 10.1016/S0891-5849(01)00504-4
Ursini F Zamburlini A Cazzolato G Maiorino M Bittolo Bon G Sevanian A Postprandial plasma lipid hydroperoxides: a possible link between diet and atherosclerosis Free Radic Biol Med 1998 25 250 252 9667503 10.1016/S0891-5849(98)00044-6
Fabiani R De Bartolomeo A Rosignoli P Morozzi G Antioxidants prevent the lymphocyte DNA damage induced by PMA-stimulated monocytes Nutr Cancer 2001 39 284 291 11759293 10.1207/S15327914nc392_19
Noroozi M Burns J Crozier A Kelly IE Lean MEJ Prediction of dietary flavonol consumption from fasting plasma concentration or urinary excretion Eur J Clin Nutr 2000 54 143 149 10694785 10.1038/sj.ejcn.1600908
Lenton KJ Therriault H Fulop T Payette H Wagner JR Glutathione and ascorbate are negatively correlated with oxidative DNA damage in human lymphocytes Carcinogenesis 1999 20 607 613 10223188 10.1093/carcin/20.4.607
Cadenas S Barja G Poulsen HE Loft S Oxidative DNA damage estimated by oxo8dG in the liver of guinea-pigs supplemented with graded dietary doses of ascorbic acid and alpha-tocopherol Carcinogenesis 1997 18 2373 2377 9450484 10.1093/carcin/18.12.2373
|
16287499
|
PMC1315342
|
CC BY
|
2021-01-04 16:39:30
|
no
|
Nutr J. 2005 Nov 14; 4:33
|
utf-8
|
Nutr J
| 2,005 |
10.1186/1475-2891-4-33
|
oa_comm
|
==== Front
Int J Health GeogrInternational Journal of Health Geographics1476-072XBioMed Central London 1476-072X-4-301628865610.1186/1476-072X-4-30ResearchAre associations between socio-economic characteristics and exposure to air pollution a question of study area size? An example from Scania, Sweden Stroh Emilie [email protected] Anna [email protected] Susanna [email protected]ö Petter [email protected] Lars [email protected]ömberg Ulf [email protected] Kristina [email protected] GIS Centre & The Department of Physical Geography and Ecosystem Analysis, Lund University, Sölvegatan 12, SE-223 62 Lund, Sweden2 Department of Occupational and Environmental Medicine, Lund University Hospital, SE-221 85 Lund, Sweden2005 16 11 2005 4 30 30 25 8 2005 16 11 2005 Copyright © 2005 Stroh et al; licensee BioMed Central Ltd.2005Stroh 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
Numerous studies have shown that exposure to air pollutants in the area of residence and the socio-economic status of an individual may be related. Therefore, when conducting an epidemiological study on the health effect of air pollution, socio-economy may act as a confounding factor. In this paper we examine to what extent socio-economic status and concentrations of NO2 in the county/region of Scania, southern Sweden, are associated and if such associations between these factors differ when studying them at county or city level. To perform this study we used high-resolution census data and modelled the annual exposure to NO2 using an emission database, a dispersion modelling program and a geographical information system (GIS).
Results
The results from this study confirm that socio-economic status and the levels of NO2 in the area of residence are associated in some cities. The associations vary considerably between cities within the same county (Scania). Even for cities of similar sizes and population bases the associations observed are different. Studying the cities together or separately yields contradictory results, especially when education is used as a socio-economic indicator.
Conclusion
Four conclusions have been drawn from the results of this study. 1) Adjusting for socio-economy is important when investigating the health effects of air pollution. 2) The county of Scania seems to be heterogeneous regarding the association between air pollution and socio-economy. 3) The relationship between air pollution and socio-economy differs in the five cities included in our study, depending on whether they are analysed separately or together. It is therefore inadvisable to determine and analyse associations between socio-economy and exposure to air pollutants on county level. This study indicates that the size and choice of study area is of great importance. 4) The selection of socio-economic indices (in this study: country of birth and education level) is important.
==== Body
Background
Humans are inevitably, although we constantly seem to forget this, part of the environment we live in. The way we interact with and affect the environment will have consequences not only for our surroundings but also ultimately on ourselves. Air pollution is an example of an anthropogenic effect that has become one of the major health hazards of our time. Airborne pollutants generated by traffic, industry, energy consumption, combustion, etc. are believed to cause not only respiratory and cardiovascular diseases in those exposed to them, but they may also cause premature death in certain groups of vulnerable people. Exposure to different air pollutants and associations to various respiratory and cardiovascular diseases have been the subject of intense study during recent years [1-5]. It has been shown that short-term variations in air pollution levels are related to lung function and asthma symptoms [6], as well as mortality and hospital admissions due to cerebrovascular disease and heart disease [7]. Particular matter ≤ 10 μm (PM10) and ozone are usually considered to be the air pollutants having the greatest effects on health but NO2 is a good indicator of air pollutants in general [8], and we have thus chosen this component as the indicator of exposure element in this study.
There is an element of discrimination in the way we are exposed to air pollutants. The area affected and the people exposed may be miles away from the source of the pollutants, which makes it difficult not only to control and regulate for pollution, but also to analyse its effects on health. The effect of the pollution may also differ depending on weather conditions or, for example, the socio-economic characteristics of the people exposed. Another difficulty is that the health effects of air pollution are similar to those of many other diseases caused by social factors such as poor dietary habits, lack of exercise and stress. Since both air pollutants and socio-economic status strongly influence health [9-11] socio-economic factors can act as confounders when studying the health effects of air pollutants [6], making it difficult to quantify the health effects of air pollutants alone.
Numerous studies have shown that groups with low socio-economic status tend to live in areas that are exposed to air pollutants to a greater extent than groups with high socio-economic status [12-16]. Several studies on the relation between geographical location and health have been conducted during the past 30 years; many of them confirming that socio-economic status can act as a confounder when investigating the exposure to different health hazards in the area of residence. According to Bowen (2002), who has reviewed 42 empirical research studies on the subject, many of these studies are based on such poor empirical foundations that the results should be viewed as unreliable [17]. Bowen also states that many of these studies fail to detect the underlying spatial process involved due to badly chosen geographical units. Willis, Jerrett, Burnett and Krewski (2003) studied the impact of the size of the study area (metropolitan areas versus county areas) on the results of a study on the relationship between long-term exposure to sulphate air pollution and mortality [18]. Their results clearly shows that the size of the study area can have an impact on the results of an epidemiological study, but that there are both advantages and disadvantages of large and small study areas.
It is important to be able to describe socio-economy on a group level when evaluating possible contextual health effects. A contextual effect is an effect where the group influences the individual, for example, when the socio-economic characteristics of a neighbourhood influence the effect on the health of an individual caused by air pollution. In a multi-level model, where data are analysed according to an individual's group affiliation, contextual effects are of particular interest. Here, group affiliation is determined by area of residence. Individuals from the same group are assumed to both influence, and be influenced by, group membership. This assumption might not be valid if a group is very heterogeneous. In the future we intend to analyse health effects of air pollution in Scania with multi-level models, as the level of socio-economy within an area of residence might have a contextual influence. Clearly, the definition of the area can affect the level, as well as the heterogeneity of socio-economy within that area. In this work, we investigated the relationship between air pollution and socio-economy in Scania, in order to define relevant contextual areas for further studies.
Improved technologies and more detailed data sources have made it possible to model personal exposure to air pollutants and socio-economic status. This study analyses the association between socio-economic status and mean annual concentrations of NO2 in Scania, as well as the possible influence of the choice of geographical level and study area on the results.
The present study was carried out on the population of the whole county of Scania in southern Sweden (described below) in 2001, separately for the five major cities in the region: Malmö, Helsingborg, Lund, Kristianstad and Trelleborg (Figure 1), and for the five cities grouped together in one data set. These cities differ in geographical location, infrastructure and population size, as well as in socio-economic structure. In order not to confound the analysis with a rural-urban gradient, the analyses of the five cities were performed strictly on residents within the city limits.
Figure 1 The county of Scania, Sweden, and five of the major cities in the region.
Scania
Scania is the southernmost county in Sweden. It covers around 11,350 km2, which is approximately 2% of the total Swedish land area. The region is densely populated, with more than 1.1 million people living in the area, which is approximately 11% of the total Swedish population. In Scania, approximately 67% of the population lives along the west coast. Most road traffic (passenger cars as well as trucks) from the European continent to Sweden and Norway passes through this area, and five motorways run through the region. There are also several harbours in the region and a considerable amount of cargo shipping and ferry transport along the coast. These factors, and the closeness to Copenhagen in Denmark, and the European continent, contribute to high concentrations of air pollutants in the region, compared with most other regions in Sweden.
Malmö
Malmö is the largest city in Scania, with a population of about 260,000 (the third largest city in Sweden) and an area of 66 km2. The city is the residential focus of the rich agricultural area of Scania, and used to be one of Sweden's most important industrial and trading cities. During the past 20 years industry and commerce in the area have decreased quite considerably, resulting in a sudden rise in the proportion of unemployed, to almost twice the national average [19]. During recent years, this number has begun to slowly decrease as the city has begun to adapt to new areas of the labour market. The completion of the Öresund bridge in 2000, connecting the mainland of Sweden with Denmark, has expanded the labour market for the residents in both Malmö and Copenhagen. The commuting between Sweden and Denmark now dominates commuter traffic, and this almost doubled between 2001 and 1997 [20,21]. Malmö is known for its high proportion of immigrants, with more than 20% of its residents born outside Sweden [22]. The city is considered to be one of Sweden's most segregated [23].
Helsingborg
Helsingborg is the second largest city in Scania and covers approximately 35 km2. It has a population of 89,000. The city has one of the most important harbours in Sweden and is of great importance as a transportation link for trains. Both the ferry traffic and commercial traffic are of great importance to the city. The distance to the Danish mainland across the strait is only 6 km, and the ferry traffic between the two countries is intense. The city is considered to be segregated, with a north/south socio-economic gradient [24].
Lund
Lund has a population of 76,000 residents and covers an area of 23 km2. The city is in many ways characterised by the university, and around one third of the residents are students. Thus the city's residents are younger than the national average, and especially the age group 20–29 is highly overrepresented. Due to the high number of students and young people, relative to the population size, migration is much higher than in the other cities. The university, the science park and the university teaching hospital are three of the major employers. The city is not as segregated as Malmö and Helsingborg which, in many ways, can be explained by the high migration to and from the city [25].
Kristianstad
Kristianstad is situated on the east coast of Scania. It has a population of 32,000 and an area of 19 km2. The city is the centre of Scania's eastern region and is situated in one of Sweden's largest fruit-growing districts. Kristianstad is an active centre of commerce in the region and most of its residents are involved in the trade or service sectors. Although the city is not as segregated as Malmö and Helsingborg, there are still distinct differences in socio-economy between the city's neighbourhoods [26].
Trelleborg
Trelleborg is the most southerly, as well as the smallest, of the cities included in the study. It has a population of only 24,000 inhabitants and an area of 10 km2. The city is highly affected by the busy trade and ferry traffic with the German harbours in Travemünde, Sassnitz and Rostock. Although the city is small, the fact that it is situated in a lively metropolitan area and the hectic trade traffic result in the same social problems as in the larger cities in the region. The proportion of residents with a high educational level is lower than the national average [27].
Aims
The main aim of this study was to describe associations between the levels of mean annual concentrations of air pollution (NO2) and two socio-economic indices ("country of birth" and "level of education") in the region of Scania in Sweden. These associations are of special interest for future studies of health effects resulting from exposure to air pollutants, since socio-economy might act as a confounding factor, both individually and contextually in such a study.
A secondary aim of this study was to investigate the possible influence of differences in size (or level) of the study area on any associations observed. Do the size and choice of the study area affect the associations seen, and if so, to what extent? The answer to this question is of utmost importance, not only for this study, but for health effect studies in general. It is important to establish the effects of socio-economy to ensure that the results of epidemiological studies in general are not biased. If the associations observed are dependent on the size and choice of the study area, generalising the contextual effect of socio-economy for too large an area may lead to erroneous results.
Results
The results of this study confirm the hypothesis that associations exist between socio-economic status and NO2 concentration in our study area (Table 1,2 and Figure 2, 3, 4, 5, 6, 7). According to the Spearman correlation analysis there is a statistically significant correlation between both country of birth and the level of education, not only in Scania as a whole, but also in all five cities included in the study, regardless of whether they are studied separately or together. The correlation coefficients are often low, however.
Table 1 Descriptive statistics for the population in Scania, Sweden. Statistics for the population in Scania, Sweden (2001) and the five cities in the region (Malmö, Helsingborg, Lund, Kristianstad and Trelleborg). Statistics for the socio-economic subgroup relating to educational level were calculated for the age group 25–64 years of age, while the subgroup "country of birth" includes all the habitants of the region/city.
Inhabitants Scania Malmö Helsingborg Lund Kristianstad Trelleborg
Number of inhabitants 1,165,411 258,140 88,992 76,334 32,295 24,254
Born in Sweden (%) 87 75 83 83 86 84
Born in OC (%) 10 20 12 12 12 11
Low education (%) 82 81 84 49 82 92
High education (%) 18 19 16 51 18 8
Concentrations of NO2 (μg/m3)
Median level 14 17 13 13 9 12
Quartiles 9/17 15/19 12/20 12/14 8/10 11/18
Min 3 10 10 9 6 7
Max 23 23 21 17 11 22
Table 2 Correlation between socio-economic subgroups and mean annual exposure to NO2 in Scania, Sweden (2001). The correlation coefficients for those born in OC (Other Country) and those with a low level of education are not presented since they are the exact inverse of the correlation coefficients for those born in Sweden, and those who are highly educated, respectively. The proportion with:"high level of education and concentration of NO2" is calculated for the age-group 25–64 years.
Spearman's correlation coefficient and p-value
Area Proportion born in Sweden and concentration of NO2 Proportion with high level of education and concentration of NO2
Scania Correlation coefficient -0.9 0.2
p-value <0.001 <0.001
All 5 cities together Correlation coefficient -0.5 -0.4
p-value <0.001 <0.001
Malmö Correlation coefficient -0.8 0.1
p-value <0.001 <0.001
Helsingborg Correlation coefficient 0.1 0.2
p-value <0.001 <0.001
Lund Correlation coefficient 0.6 0.8
p-value <0.001 <0.001
Kristianstad Correlation coefficient -0.3 0.2
p-value <0.001 <0.001
Trelleborg Correlation coefficient 0.1 0.5
p-value <0.001 <0.001
Figure 2 Bubble diagram for Scania; Percentage of population in Scania born in Sweden versus annual mean concentration of NO2 at their real estate. The bubbles are sized according to the weight of the estimate, which depends on the size of the population that the estimate is based on. The weight is thus a measure of the uncertainty in the estimate, and large bubbles represent more certain and influential estimates than small bubbles.
Figure 3 Bubble diagram for Scania; Percentage of population in Scania with high education versus annual mean concentration of NO2 at their real estate. The bubbles are sized according to the weight of the estimate, which depends on the size of the population that the estimate is based on. The weight is thus a measure of the uncertainty in the estimate, and large bubbles represent more certain and influential estimates than small bubbles.
Figure 4 Bubble diagram for Malmö; Percentage of population in Malmö born in Sweden versus annual mean concentration of NO2 at their real estate. The bubbles are sized according to the weight of the estimate, which depends on the size of the population that the estimate is based on. The weight is thus a measure of the uncertainty in the estimate, and large bubbles represent more certain and influential estimates than small bubbles.
Figure 5 Bubble diagram for Malmö; Percentage of population in Malmö with high education versus annual mean concentration of NO2 at their real estate. The bubbles are sized according to the weight of the estimate, which depends on the size of the population that the estimate is based on. The weight is thus a measure of the uncertainty in the estimate, and large bubbles represent more certain and influential estimates than small bubbles.
Figure 6 Bubble diagram for Lund; Percentage of population in Lund born in Sweden versus annual mean concentration of NO2 at their real estate. The bubbles are sized according to the weight of the estimate, which depends on the size of the population that the estimate is based on. The weight is thus a measure of the uncertainty in the estimate, and large bubbles represent more certain and influential estimates than small bubbles.
Figure 7 Bubble diagram for Lund; Percentage of population in Lund with high education versus annual mean concentration of NO2 at their real estate. The bubbles are sized according to the weight of the estimate, which depends on the size of the population that the estimate is based on. The weight is thus a measure of the uncertainty in the estimate, and large bubbles represent more certain and influential estimates than small bubbles.
The associations seen are not consistent between cities. Also, the sign of the associations differs between the two socio-economic indexes, i.e. "country of birth" and "level of education" show opposite correlations to the level of pollution.
The stronger correlation coefficients in the study implied that associations exists between country of birth and NO2 in Scania (negative), in all cities together (negative), in Malmö (negative), and in Lund (positive). The association between country of birth and level of NO2 implies that being an immigrant in Scania, Malmö or the five study cities (analysed as a group) is associated with elevated levels of NO2. In Lund, however, the association between NO2 concentrations and country of birth is the opposite, implying that being born in Sweden is associated with higher levels of NO2.
A correlation was observed between education level and concentration of NO2 in the cities together (negative), in Lund (positive) and Trelleborg (positive). This implies that less educated groups of people, living in the five cities studied and in Scania in general, are exposed to higher levels of NO2. This is, however, not the case in Lund and Trelleborg when studied separately, where highly educated residents seemed to be exposed to higher concentrations of NO2. We found that the sign of the correlation coefficient for education was positive in all cities when studied separately, but negative when analysed as one data set.
Discussion
This study confirms that associations exist between socio-economy and level of air pollution (Table 2). It is important to distinguish between a low p-value and a strong correlation. A low p-value does not imply that the association is strong, only that the association observed is not likely to occur due to chance, regardless of how weak the association actually is. The p-values in this study are all < 0.001 as a result of the many observations, while many of the correlation coefficients are below 0.5, which can be considered as indicating weak associations.
The results are not consistent; neither between cities nor between the socio-economic indexes, nor when analysing all cities together or separately. The consistency of socio-economic indexes will not be discussed further here, but we conclude that it appears important to gather as much information as possible about socio-economy to describe socio-economic status.
The results obtained when analysing the cities together in one data set were not at all consistent with the results when considering them separately, especially regarding level of education, where even the sign of the association differed between the combined data set and the separate cities (Table 2). Therefore, generalizing associations between socio-economy and air pollution from a regional level to a city level can give erroneous results. Although there seems to be a strong correlation between NO2 concentration and country of birth (Figure 2) this correspondence is probably strongly influenced by a rural-urban gradient, resulting in a biased relation between this socio-economic variable and the exposure level. Most immigrants in Sweden tend to move to cities, especially larger cities, rather than settle in the countryside [28]. Since the levels of NO2 are significantly lower outside cities, this gradient probably makes a major contribution to the strong correlation observed between the concentration of NO2 and "country of birth" in the whole county of Scania. On the regional level there may also be a risk of relations cancelling each other out, since they are not consistent between cities.
In Malmö, the negative correlation for the group born in Sweden implies that this group tends to live in areas with lower concentrations of NO2 than the group born in other countries (Figure 4). The proportion of immigrants in the different areas of Malmö is uneven which makes Malmö a segregated city. The neighbourhoods with high proportions of immigrants are located in the outskirts of the city. Since Malmö is surrounded by a ring road at which three of the five main motorways in the area converge, the emissions in these neighbourhoods tend to be higher than in other parts of Malmö.
In Helsingborg the correlation coefficients shows no correlation at all between the socio-economic variables analysed and levels of NO2. This is probably due to differences in directions between the socio-economic gradients and the concentration gradient for NO2. Helsingborg stretches along the coast of the strait between Sweden and Denmark, making the city quite long and narrow in a north/south direction, with a distinct socio-economic gradient in the same direction. The major source of air pollutants in the area is the lively ferry traffic along and across the strait, as well as from the three motorways that converge and run alongside the city. These factors cause the city to have the same variation in NO2 levels in the north south direction but vary in the west easter direction. This results in evening out of the exposure to air pollutants between the different socio-economic groups.
In Lund the correlation is positive for both the socio-economic variables, implying that people born in Sweden and individuals with a high level of education tend to live in areas with higher levels of NO2 than people born outside Sweden or with lower levels of education (Figure 6 and 7). This is probably the result of a phenomenon observed in many metropolitan areas, i.e. it is very desirable to live in the city centre where the cost of accommodation is high, and can usually only be afforded by people with high economic status. Since pollution levels are higher in the city centre these people tend to be more highly exposed. It is worth noting that this phenomenon is probably present in Malmö as well, although it does not compensate for the other socio-economic and NO2 gradients mentioned above.
In Kristianstad, only weak associations were seen between socio-economy and air pollution. Compared with the other cities in the study, the levels of NO2 in Kristianstad are, much lower, and less spread (Table 1). The small range in exposure decreases the possibilities of detecting an association
The positive correlation between the level of education and the exposure to NO2 seen in Trelleborg is weaker but implies a similar explanation to that in Lund, where highly educated people tend to live in the more central areas of the city. The fact that Trelleborg is located along the coast with a large, busy industrial harbour in the middle of the city also increases the risk of people living in the city centre of being highly exposed to NO2 and other air pollutants derived from the ferry traffic.
As already mentioned, the socio-economic indexes country of birth and educational level do not show the same associations to concentration of NO2 in our study. Strong associations are seen between country of birth and concentration of NO2 and regarding education and concentration of NO2, but not in the same cities (with Lund as an exception). The analysis regarding education was performed on the population between 25 and 64 years old only, while the analysis regarding country of birth was performed on the whole population. In order to investigate whether the difference in results between the two socio-economic indexes is a result of different age distributions, the analysis of the association between country of birth and concentration of NO2 was also performed on the age group 25–64 years. This did not alter the results. Thus the different age distributions do not explain why "country of birth" and "level of education" do not show the same associations with concentration of NO2.
We chose to analyse the data with simple correlation analysis. Geographical Weighted Regression (GWR) models provide a more sophisticated analysis, but such models are not suited to our purpose. GWR analyses spatially varying relationships by weighting them to form one surface for each relationship studied. Since one of the aims of our study was to investigate socio-economic associations with air pollution in separate cities, GWR is not suitable.
Valid exposure assignments are naturally of particular importance in reducing bias. Despite the fact that this study does not focus on this problem, we can not stress enough the importance of using a functional dispersion models and valid exposure data for exposure/concentration response relationship studies.
The inconsistency in the relationship between socio-economy and NO2 between cities, as well as over larger areas, may conceal existing associations [29]. Performing an analysis for a large area and then generalising the results to the whole country or region is not recommended since the socio-economic factors in larger cities in the area or the rural-urban gradient might dominate the results.
The results of this study clearly show that the size and type of the study area have an impact on the findings when investigating possible associations between socio-economy and exposure to air pollution. Relations observed on one area level may be almost the opposite on a larger scale. This is the case for the associations seen between country of birth and exposure to NO2 for the residents of Lund, where the correlation coefficient implies that being born in Sweden is associated with high levels of NO2. When Lund is included in the group of five cities studied the opposite association would, however, be implied namely that being an immigrant and living in any of these five cities would increase the risk of being highly exposed to NO2. This would be even more misleading if these cities were to be included in a regional study of the whole area of Scania. The smallest area studied was a city. It would have been interesting to divide the data further into districts, to see how such a division influenced the relationships.
Conclusion
The results of this study show a covariation between socio-economic status and the levels of NO2 in the area of residence in Scania, and thereby confirm that socio-economy could act as a confounding factor and create bias when investigating health effects of air pollution. In order to quantify the magnitude of bias, several other assumptions have to be made, such as risk levels for various groups of air pollution, smoking habits and other important variables (age, sex, etc).
In this study the exposure to air pollutants of different socio-economic groups varies considerably between areas. It has been shown that the relationships can differ, not only in magnitude but also in direction, whether splitting the study area into small areas or not, as seen for level of education. Relations can be the opposite in cities with similar sizes and populations. Larger areas, such as the county of Scania, are influenced by a large number of socio-economic gradients. The urban-rural gradient makes it especially difficult to determine and analyse associations between socio-economic variables and exposure to air pollutants on a regional level.
In epidemiological studies on the health effects of air pollution, incorporation of individual-level and area-level data on various socio-economic variables (i.e. country of birth, educational level, etc.) is important. Our results demonstrate that there is no consistency in how individuals in different socio-economic classes are exposed to air pollutants over a larger area, and that both the size of the area, and the choice of socio-economic indexes affect the associations observed.
Methods
The procedure used to analyse the covariation between socio-economic factors and exposure to air pollutants is described below:
1. Modelling the concentration of NO2
An emission database and a GIS program were used to model the levels of NO2 in Scania. This was done using a dispersion modelling program and emission data for the year 2001 together with meteorological data, resulting in a grid giving mean annual concentrations of NO2 for Scania in 2001.
2. Modelling the exposure to NO2
Mean annual NO2 exposure to each individual was modelled in a GIS program using the mean annual concentrations of NO2 and population data sets.
3. Division of the population into socio-economic groups
The population data were divided into socio-economic groups using the indexes country of birth and level of education.
4. Statistical analysis
The data were statistically analysed depending on each individual's socio-economic status and exposure to NO2 using a weighted Spearman rank correlation.
Modelling the concentration of NO2
To model the concentrations of NO2 for Scania an NOx emission database with approximately 24,000 sources for the area was used [30]. The different sources of emissions consist of roughly 23,500 line sources corresponding to the major roads and ship routes, 500 point sources corresponding to most of the larger industrial plants and approximately 50 area sources corresponding to emissions from heating, wood burning, farming machinery, construction machinery, etc. The emission database includes time variations for different pollutants depending on hour, day and month of the year. The emission database does not include any continental contribution, i.e. sources outside Sweden.
The software Enviman was used to model the concentrations of NOx (later converted into NO2 levels). Enviman uses a Gaussian model: AERMOD, a USEPA model [31]. The meteorological parameters in the calculation were based on climatological series for the period 1995–2003 gathered from a meteorological weather station in central Malmö [32].
In this study, the mean annual value of NOx was modelled with a spatial resolution of 250 × 250 m for the county of Scania. Both the levels and the spatial variation of the modelled concentrations of air pollutants are of interest. Therefore, the geographic resolution of the modelling grid is of vital importance. If the resolution is too low the model might generate a grid with lower variations in concentration between the cells. This might increase the uncertainty in the following exposure estimates. Modelling with higher resolution increases the opportunity to detect variations in space as well as extreme values. The capacity of the modelling tool and the computer, as well as the resolution and quality of the data in the emission database, set the limit for the highest resolution that is reasonable to model. To be able to model with this relatively high resolution the county was divided into three parts by splitting the region into three equally sized areas stretching from north to south. Each of these three areas had a 10-kilometre overlap with their neighbouring areas to prevent modelling errors that might otherwise appear along the borders (Figure 8). For each of the three areas two modelling operations were carried out: one to modelled the concentrations from local sources within that specific area, and a second to model the regional contribution of the NOx concentrations in each area contributed by the two neighbouring areas in the region. The local and regional concentrations for each cell were summed, and removing the 10-kilometre overlap between the areas and merging them together created a continuous raster.
Figure 8 Creating a continuous concentration grid of NO2 for Scania. The mean annual concentrations of NO2 in μg/m3 for Scania were modelled in three separate areas (1–3) each overlapping its neighbouring areas by 10 kilometres to prevent modelling errors in the border areas. These overlaps were later removed and the areas merged into a continuous concentration grid.
To convert the concentration of NOx to NO2 the following equation were used [32]:
NOx ≤ 13.85 μg/m3 ⇒ NO2 =NOx
NOx > 13.85 μg/m3 ⇒ NO2 =8.8·()
To take the continental contribution into account, 2.5 μg/m3 NO2 was added to the modelled concentration. The continental contribution was calculated as a regional mean from measurements from ten different reference stations in the area during the period 2000–2002 [30].
Finally, the concentration modelling was validated with measurements from 23 measuring stations in Scania. The yearly means for these were compared with the modelled annual concentrations of NO2 for the year 2001. The validation of the concentration modelling (Figure 9) gave a correlation coefficient of 0.69. For concentrations up to 10 μg/m3 the modelled concentrations are quite close to the measured values, while at concentrations above 10 μg/m3 the model seems to underestimate concentrations ranking from 10 to 15 μg/m3 and overestimate concentrations ranking from 15 to 25 μg/m3.
Figure 9 Validation of the concentration model. Modelled annual average concentration of NO2 (μg/m3) versus measured. The modelled concentration is plotted against measured concentration of the annual mean of NO2 from 23 measuring stations in Scania. y = 1.041x - 0.0605 R2 = 0.69
The concentration modelling (Figure 10) shows that there are considerably higher levels of NO2 in the western regions than in the less populated eastern part of the county. The emissions of NO2 from the three major cities in the region (Malmö, Helsingborg and Lund) can be seen clearly, as well as the major roads in the county. Also, the ferry traffic along the coast and some of the contributing emissions from Denmark in the west are visible. Areas with NO2 levels lower than 5 μg/m3 have a very low population density and consist mainly of agricultural land or forests.
Figure 10 Modelled annual mean concentrations of NO2 for Scania (2001).
Modelling the exposure to NO2
In this study, two different sets of population data were used (henceforth referred to as PD1 and PD2). Both of the data sets were obtained from the Regional Office of Scania, Sweden, and are based on the Swedish National Registry. The data sets are on individual level for the registered population in Scania during 2001, but they differ in spatial resolution and the amount of attribute data linked to them.
1. PD1
In PD1 individuals' location are represented by points at the centre coordinates of their real estate (listed in the National Registry). The only attributes linked to this data set were sex and age.
2. PD2
In PD2 the location of each individual's real estate is given as a centroid in a 1 km grid. PD2 was linked with socio-economic attributes from Statistics Sweden and contains the following attributes: sex, year of birth, country of birth, marital status, income for the previous year (2000) and highest educational level.
The task is to add exposure values to and all individuals in PD2. Since we cannot distinguish between the persons living in a 1 × 1 km cell all of them will get the same exposure value. This could be performed by only using only the information in PD2. However, the exposure value can be improved by using the detailed positions of the individuals in PD1 using the methodology described below.
The annual mean of NO2 in Scania for 2001 was modelled with a 250 × 250 m grid resolution (small coloured grid in Figure 11). Since the residences of the individuals within each square kilometre are not evenly distributed in space, it was decided to relocate the centre coordinate to a position within each square kilometre which better corresponded to the total population density within that specific square kilometre.
Figure 11 Calculation of the geographical centre of the residents' real estates within each 1 × 1 kilometre grid cell. Small grid cells: grid cells for modelled annual mean of NO2 (250 × 250 m). Large bold grid cell: 1 km2. Small black dots: the centre coordinates of the real estate (according to the National Registry) for each individual in population data set PD1. Black centre cross: the generalised position of the individual's real estate, within the specific km2, according to population data set PD2. White crossed circle: calculated geographical centre of the residents' real estate within the specific km2. White-edged grid cells: the cells within the concentration grid which, through bilinear interpolation were used to assign an NO2 exposure value to the individuals in the repositioned PD2.
To find the geographical centre of the individuals' residences in PD2 the data set was combined with PD1. PD1 contained the exact location of each individual's real estate (small black dots in Figure 11). By creating a grid (large bold grid cell in Figure 11) out of the kilometre points of census data obtained in PD2 and applying a ID number to each of these cells (consisting of each specific cells X and Y centre coordinates) this ID number could then be transferred to the points in PD1, depending on within which square kilometre grid cell the points in PD1 were positioned.
For each set of points with the same ID number (i.e. those that fell within the same square kilometre grid cell) in PD1, the geographical centre was then estimated by calculating the average X and Y coordinate for all the points. The coordinates for this new centre point were then transferred back to PD2 and the individuals within this data set were subsequently repositioned from their centroids to these new coordinates (white crossed circle in Figure 11).
An approximation of the individual's average annual exposure to NO2 in μg/m3 for these new locations was assigned to each individual in the repositioned population data set, PD2, by using bilinear interpolation (white edged grid cells in Figure 11) [33]. This relocation also increases the accuracy in the estimation of the NO2 concentration to which most of the individuals within each square kilometre were exposed [34].
Division of the population into socio-economic groups
From the repositioned population set, PD2, which was linked with socio-economic attributes and NO2 concentrations, different subsets were created based on the individual's country of birth and their educational level.
Country of birth
In the field of environmental justice it is common to study ethnicity and exposure to various pollutants. Sweden is not as segregated as some countries regarding race, and therefore race as a measure of socio-economic status is not applicable here. Nevertheless, segregation between immigrants, as a mixed ethnic group, and native Swedes exists to various extents. In 2000 around 11% of the Swedish population had been born outside the country, and most of these immigrants were concentrated in metropolitan areas or larger cities [35].
Since immigrants from other countries tend to have difficulties getting into the labour market they often belong to the lower social and income groups. Therefore, depending on country of birth, the individuals were classified into two different groups: Sweden – individuals born in Sweden (1,007,958 individuals) OC: Other Countries (110,214 individuals) The OC group includes individuals from countries outside the Scandinavia, countries and the European Union in 2001 and those not from: Australia, Canada, Japan, New Zealand or the USA. Individuals who were born in Scandinavian countries other than Sweden (2% of the total population in Scania), countries belonging to the European Union in 2001 (1% of the total population in Scania) and the major economies: Australia, Canada, Japan, New Zealand and the USA (0.2% of the total population in Scania) were excluded from the analysis. This was done as individuals from these countries might immigrate to Sweden under different conditions and for different reasons than immigrants from the other countries.
In a preliminary study it was examined whether or not this exclusion would affect the results of the study. It was found that leaving out this group did not significantly affect the results of the study, due to the small number of individuals in this group, compared to the other two categories,
Level of education
The Swedish educational system has changed over the years, and the definition of an individual with a high educational level has varied considerably. To reduce birth-cohort bias and to focus on the working population the analysis of this index was only carried out on individuals in the age group 25–64 years. Information on the highest level of education was missing for some individuals. In total around 70% of the total population of Scania was included in this analysis.
The attributes regarding education were divided into seven different categories:
1. Pre-secondary education for less than nine years (9% of the population)
2. Pre-secondary education for nine years (11% of the population)
3. Secondary education for maximum two years (19% of the population)
4. Secondary education for three years or longer (13% of the population)
5. Post-secondary education for less than three years (9% of the population)
6. Post-secondary education for three years or longer (9% of the)
7. Postgraduate studies (0.6 % of the population)
These seven different classes were grouped into two generalised classes representing individuals with a high level education (groups 6 and 7) and those with a low level of education (groups 1 to 5).
Statistical analysis
The data were analysed by calculating the Spearman rank correlation coefficients for the exposure to NO2 and the proportion of the population being highly educated and born in Sweden. A correlation coefficient is a number between -1 and 1 which measures the degree to which two variables are related. Spearman's correlation is based on ranks, and can be used to describe non-linear relationships. The levels of NO2 were rounded off to single-decimal numbers. The correlation coefficients were determined calculating the proportion with high education and the proportion born in Sweden for each NO2 level. The correlation coefficients for the association between level of NO2 and proportion being highly educated or born in Sweden were then calculated. Consequently, a positive correlation coefficient here shows that the level of NO2 increases with an increasing proportion of the population having a high socioeconomic status. Correlation coefficients in the range of -0.4 to 0.4 were regarded as weak correlations of minor importance.
The data were plotted in bubble diagrams, with the level of NO2 on the x-axis and the proportion belonging to each socio-economic group on the y-axis. The size of the bubbles represents the inverse variance of the proportion estimates, and thus represents the certainty in the estimate. A large bubble represents a proportion estimate that was calculated based on a larger group of people than a small bubble. The weights, wi, for NO2 concentration strata i was calculated according to:
where is the estimated population proportion and ni is the number of people in stratum i.
Table 1 presents the size of the population, the proportion belonging to the socio-economic subgroups, the median, range, and quartiles of the concentration of NO2 in the population for the whole of Scania and for each city in the study.
Authors' contributions
PP, KJ and US conceived the study and participated in its design and coordination.
US assisted with the statistical analysis.
LH helped to draft the manuscript.
SG carried out the NO2 concentration modelling.
AO performed the statistical analysis, the alignment and wrote part of the paper.
ES carried out the NO2 exposure modelling and wrote the final version of the paper.
Acknowledgements
We would like to acknowledge:
Our reviewers for helpful comments and suggestions.
The Swedish National Air Pollution and Health Effects Program (SNAP) for financial support.
Åke Boalt, Region Skåne, for providing us with population data for the region of Scania.
Niclas Olofsson, Department of Social and Economic Geography at Lund University, for providing us with useful help and information.
==== Refs
Chang C-CT S-S Ho S-C Yang C-Y Air pollution and hospital admissions for cardiovascular disease in Taipei, Taiwan Environmental Research 2005 98 114 119 15721891 10.1016/j.envres.2004.07.005
Chen LHK SF Beeson L Ghamsary M Shavlik D Petersen F Abbey D The association between ambient particulate air pollution and fatal coronary heart disease among persons with respiratory symptoms/disease Annals of Epidemiology 2005 15 642 10.1016/j.annepidem.2005.06.035
Iwai K Mizuno S Miyasaka Y Mori T Correlation between suspended particles in the environmental air and causes of disease among inhabitants: Cross-sectional studies using the vital statistics and air pollution data in Japan Environmental Research 2005 99 106 117 16053935 10.1016/j.envres.2004.11.004
Karakatsani AAS Katsouyanni K Dimitroulis I Trichopoulos D Benetou V Trichopoulou A Air pollution in relation to manifestations of chronic pulmonary disease: A nested case-control study in Athens, Greece European Journal of Epidemiology 2003 18 45 53 12705623 10.1023/A:1022576028603
Kraft M Eikmann T Kappos A Kunzli N Rapp R Schneider K Seitz H Voss J-H Wichmann H-E The German view: Effects of nitrogen dioxide on human health – derivation of health-related short-term and long-term values International Journal of Hygiene and Environmental Health 2005 208 305 318 16078645 10.1016/j.ijheh.2005.04.002
Hwang J-S Chang-Chuan Chan Effects of Air Pollution on Daily Clinic Visits for Lower Respiratory Tract Illness American journal of epidemiology 2002 155 1 11772777 10.1093/aje/155.1.1
WHO Health Aspects of air pollution. Results from the WHO project "Systematic review of health aspects of air pollution in Europe" 2004 30
WHO Health Aspects of Air Pollution with Particulate Matter, Ozone and Nitrogen Dioxide 2003 Bonn, Germany 98
Black A Murray LJ Cardwell C Davey Smith G McCarron P Lawlor DA Ebrahim S Secular trends in heart rate in young adults, 1949 to 2004: analyses of cross-sectional studies. Adverse socioeconomic position across the lifecourse increases coronary heart disease risk cumulatively: findings from the British women's heart and health study Heart 2005 59 785 793
Kivimaki M Davey Smith G Juonala M Ferrie JE Keltikangas-Jarvinen L Elovainio M Pulkki-Raback L Vahtera J Leino M Viikari JS Socioeconomic position in childhood and adult cardiovascular risk factors, vascular structure and function: the Cardiovascular Risk in Young Finns Study. Secular trends in heart rate in young adults, 1949 to 2004: analyses of cross-sectional studies. Adverse socioeconomic position across the lifecourse increases coronary heart disease risk cumulatively: findings from the British women's heart and health study Heart 2005 59 785 793
Lawlor DA Ebrahim S Davey Smith G Adverse socioeconomic position across the lifecourse increases coronary heart disease risk cumulatively: findings from the British women's heart and health study J Epidemiol Community Health 2005 59 785 793 16100318 10.1136/jech.2004.029991
Chaix BGS Jerrett M Kristersson K Lithman T Boalt A Merlo J Children's exposure to nitrogen dioxide in Sweden: investigating environmental injustice in an egalitarian country Journal of Epidemiology and Community Health 2005
Eroshina K Danishevski K Wilkinson P McKee M Environmental and social factors as determinants of respiratory dysfunction in junior schoolchildren in Moscow Journal of Public Health 2004 26 197 204 15284327 10.1093/pubmed/fdh144
Finkelstein MM Jerrett M Sears MR Environmental inequality and circulatory disease mortality gradients J Epidemiol Community Health 2005 59 481 487 15911644 10.1136/jech.2004.026203
Mitchell G Dorling D An environmental justice analysis of British air quality Environment and Planning A 2003 35 909 929
Neidell MJ Air pollution, health, and socio-economic status: the effect of outdoor air quality on childhood asthma Journal of Health Economics 2004 23 1209 15556243 10.1016/j.jhealeco.2004.05.002
Bowen W An Analytical Review of Environmental Justice. Research: What Do We Really Know? Environmental Management 2002 29 3 15 11740620 10.1007/s00267-001-0037-8
Willis A Jerrett M Burnett R Krewski D The Association between Sulfate Air Pollution and Mortality at the County Scale: An Exploration of the Impact of Scale on a Long-Term Exposure Study Journal of Toxicology and Environmental Health Part A 2003 66 1605 12959832 10.1080/15287390306432
Malmö Stad: Näringsliv & Arbetsmarknad
Databank Ö Öresund Databank 2005 Örestat, Statistiska Centralbyrån, EU, Danmarks Statistik
Resor och pendling över Öresund
Health NIoP Födelselandets betydelse – En rapport om hälsan hos olika invandrargrupper i Sverige Födelselandets betydelse – En rapport om hälsan hos olika invandrargrupperi Sverige 2002 Sandviken: Statens folkhälsoinstitut 146
Statistikrapport 2004 – statistisk uppföljning av Rapport Integration 2003
Exempel på segregationsanalys
Statistik om Lund
Kristianstad municipality information
Befolkning efter utbildningsnivå 31 december 1999
Martinson L Cars G Normann I NEHOM – National Report of Sweden 2002 NEHOM, Stockholm: Royal institute of Technology, Division of Urban Studies 140
Bowen W Salling M Haynes K Cyran E Toward Environmental Justice: Spatial Equity in Ohio and Cleveland Annals of the Associations of American Geographers 1995 85 641 663 10.1111/j.1467-8306.1995.tb01818.x
Gustafsson S Wellinder H Tinnerberg H Pilesjö P High resolution emissiondatabase – Scania 2001 Lund 2005
Cimorelli AJ Perry SG Venkatram A Weil JC Paine RJ Wilson RB Lee RF Peters WD Brode RW Paumier JO AERMOD: DESCRIPTION OF MODEL FORMULATION. U.S. ENVIRONMENTAL PROTECTION AGENCY OoAQPaS, Emissions, Monitoring, and Analysis Division 2004 North Carolina: U.S. ENVIRONMENTAL PROTECTION AGENCY, Office of Air Quality Planning and Standards, Emissions, Monitoring, and Analysis Division 91
Gustafsson S Wellinder H Tinnerberg H Pilesjö P Emissionsdatabas för luftföroreningar i Öresundsregionen med basår 2001 2005 Naturvårdsverket, Physical Geography and Ecosystem Analysis 80
Gonzalez R Woods R Digital Image Processing 1993 1 Massachusetts: Addison-Wesley Publishing Company
Larsson K Stroh E Pilesjö P Wellinder H Axmon A Strömberg U Nielsen J Assarsson E GIS för analys av hälsopåvrkan från luftföroreningar En studie i självrapporterade besvär i Växjö 2005 STEM, Lund: Lund University 47
SOU Sou: Tre städer – En storstadspolitik för hela landet 1998 Socialdepartementet, Regeringskansliets förvaltningsavdelning 184
|
16288656
|
PMC1315343
|
CC BY
|
2021-01-04 16:39:05
|
no
|
Int J Health Geogr. 2005 Nov 16; 4:30
|
utf-8
|
Int J Health Geogr
| 2,005 |
10.1186/1476-072X-4-30
|
oa_comm
|
==== Front
BMC Evol BiolBMC Evolutionary Biology1471-2148BioMed Central London 1471-2148-5-691632421410.1186/1471-2148-5-69Research ArticleEvolution and origin of merlin, the product of the Neurofibromatosis type 2 (NF2) tumor-suppressor gene Golovnina Kseniya [email protected] Alexander [email protected] Elena M [email protected] Leonid V [email protected] Long-Sheng [email protected] Institute of Cytology and Genetics, Russian Academy of Sciences, 10 Lavrent'ev Ave., 630090, Novosibirsk, Russia2 Center for Childhood Cancer, Children's Research Institute, Children's Hospital and Department of Pediatrics, The Ohio State University, 700 Children's Drive, Columbus, OH 43205-2696, USA2005 2 12 2005 5 69 69 18 3 2005 2 12 2005 Copyright © 2005 Golovnina et al; licensee BioMed Central Ltd.2005Golovnina 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
Merlin, the product of the Neurofibromatosis type 2 (NF2) tumor suppressor gene, belongs to the ezrin-radixin-moesin (ERM) subgroup of the protein 4.1 superfamily, which links cell surface glycoproteins to the actin cytoskeleton. While merlin's functional activity has been examined in mammalian and Drosophila models, little is understood about its evolution, diversity, and overall distribution among different taxa.
Results
By combining bioinformatic and phylogenetic approaches, we demonstrate that merlin homologs are present across a wide range of metazoan lineages. While the phylogenetic tree shows a monophyletic origin of the ERM family, the origin of the merlin proteins is robustly separated from that of the ERM proteins. The derivation of merlin is thought to be in early metazoa. We have also observed the expansion of the ERM-like proteins within the vertebrate clade, which occurred after its separation from Urochordata (Ciona intestinalis). Amino acid sequence alignment reveals the absence of an actin-binding site in the C-terminal region of all merlin proteins from various species but the presence of a conserved internal binding site in the N-terminal domain of the merlin and ERM proteins. In addition, a more conserved pattern of amino acid residues is found in the region containing the so-called "Blue Box," although some amino acid substitutions in this region exist in the merlin sequences of worms, fish, and Ciona. Examination of sequence variability at functionally significant sites, including the serine-518 residue, the phosphorylation of which modulates merlin's intra-molecular association and function as a tumor suppressor, identifies several potentially important sites that are conserved among all merlin proteins but divergent in the ERM proteins. Secondary structure prediction reveals the presence of a conserved α-helical domain in the central to C-terminal region of the merlin proteins of various species. The conserved residues and structures identified correspond to the important sites highlighted by the available crystal structures of the merlin and ERM proteins. Furthermore, analysis of the merlin gene structures from various organisms reveals the increase of gene length during evolution due to the expansion of introns; however, a reduction of intron number and length appears to occur in the merlin gene of the insect group.
Conclusion
Our results demonstrate a monophyletic origin of the merlin proteins with their root in the early metazoa. The overall similarity among the primary and secondary structures of all merlin proteins and the conservation of several functionally important residues suggest a universal role for merlin in a wide range of metazoa.
==== Body
Background
The advancement in genome sequencing projects, the accumulation of knowledge in bioinformatics, and the molecular genetic analysis of genes and their functions in a variety of model organisms provides us with an unprecedented opportunity to identify novel genes based on sequences related to characterized genes [1]. This process is conducted using pairwise sequence comparison with the understanding that genes form families wherein related sequences likely share similar functions. Although initial identification of new genes may not yield a clear indication of their respective functions, studies on their evolution may allow validation of their sequence identity and provide information on their putative functional characteristics. For genes evolved from duplication and/or adapted to different evolutionary niches during speciation, detailed sequence comparison can provide additional information regarding their biological and biochemical characteristics [2].
Neurofibromatosis type 2 (NF2) is a highly penetrant, autosomal dominant disorder, whose hallmark is the development of bilateral vestibular schwannomas [3,4]. The tumor suppressor gene associated with NF2 has been identified and termed the neurofibromatosis type 2 gene (NF2) [5,6]. The NF2 gene encodes a protein named merlin, for moesin-ezrin-radixin like protein, or schwannomin, a word derived from schwannoma, the most prevalent tumor seen in NF2. For simplicity, we refer to the NF2 gene product as merlin hereafter.
Merlin shares sequence similarity with the ezrin, radixin, and moesin (ERM) proteins, which belong to the protein 4.1 superfamily of cytoskeleton-associated proteins that link cell surface glycoproteins to the actin cytoskeleton [7,8]. Like ERM proteins, merlin consists of three predicted structural domains [5,6,9]. The N-terminal domain, termed the FERM (F for 4.1) domain, is highly conserved among all members of the ERM family and is important for interactions with cell surface glycoproteins, including CD44 and intercellular adhesion molecules [10-13]. Crystal structure analysis shows that the tertiary structure of the FERM domain of merlin closely resembles those of the FERM domain of moesin and radixin [14-18]. The FERM domain of merlin exists as a clover-shaped molecule consisting of three structural subdomains A, B, and C, which are homologous to lobes F1, F2, and F3 in moesin and radixin. Subdomain A, composed of residues 20–100, possesses a ubiquitin-like fold. Subdomain B, consisting of residues 101–215, folds itself into a topology like that of the acetyl-CoA-binding protein. Subdomain C, containing residues 216–313, adopts the pleckstrin homology/phosphotyrosine-binding fold found in a broad range of signaling molecules [14-16]. The second half of merlin contains a predicted α-helix domain, which is also present in the ERM proteins [19]. Although the unique C-terminus of merlin lacks the conventional actin-binding domain found in the ERM proteins [20,21], merlin can directly bind actin using the residues at the N-terminal domain and indirectly through its association with βII-spectrin or fodrin [22-24].
The merlin and ERM proteins are thought to be key regulators of interactions between the actin cytoskeleton and the plasma membrane in polarized cells. They act as important members of signal transduction pathways that control cell growth and participate in the sorting of membrane proteins during exocytic traffic [25,26]. However, unlike the ERM proteins, merlin has a distinct function as a tumor suppressor [27]. Growth suppression by merlin is dependent on its ability to form intramolecular associations [28,29]. In this regard, merlin exists in an "open" (inactive form) or "closed" (active growth-suppressive form) conformation that is regulated by phosphorylation [30-35].
While previous studies have focused primarily on the functional analysis of merlin, limited information is available about its overall distribution across eukaryotes and its evolution. A phylogenetic study indicates that the FERM domains of ERM homologs from sea urchins, Caenorhabditis elegans, Drosophila melanogaster, and vertebrates share 74–82% amino acid identity and have about 60% identity with those of merlin [25,36-42]. These levels of identity are exceptionally high, suggesting that the protein structures of the merlin and ERM proteins from these species may be well-conserved. The most divergent ERM proteins are found in tapeworms and schistosomes [36-39]. The FERM domains of these parasite proteins share only 44–58% similarity to their vertebrate homologs. The high degree of structural conservation among these proteins points to possible similarities or functional redundancies. Intriguingly, no FERM domain-encoding genes have been identified in the genome of the yeast Saccharomyces cerevisiae, implying that FERM domains evolved in response to multicellularity, rather than as a cytoskeletal component [25].
The goal of the present study was to expand our understanding of the taxonomic diversity of merlin and the phylogenetic relationships using experimentally annotated and predicted sequences. By the integration of the BLAST-based analysis using the available partial and whole genome sequences with phylogeny reconstruction, we have generated an evolutionary tree for the entire ERM-family members from various taxa and identified some interesting details about their phylogenetic origin. In addition, we compared sequence variability at functionally significant sites, including the major phosphorylation site of merlin, predicted the secondary structure of the merlin proteins of various species, and examined the exon-intron structural evolution of the NF2 gene.
Results and Discussion
BLAST identification of merlin sequences
To identify putative merlin and ERM sequences in a wide range of eukaryotes, we performed BLAST analysis of 15 available genome databases. By searching through all annotated proteins and genome sequences, we identified 50 sequences from 30 species. Table 1 summarizes the full list of the predicted and annotated merlin and ERM proteins identified, and their GenBank and available UniProtKB/Swiss-Prot accession numbers and related resources. No merlin-like sequences were found in the genomes of fungi, plants, and protozoa. While the sequencing projects of the hard ticks are still ongoing at The Institute for Genomic Research (TIGR), amino acid sequences deduced from partial cDNAs of salivary glands, which share a similarity with the FERM domain of merlin, have been noted from Rhipicephalus appendiculatus [43], Amblyomma variegatum [44], and Boophilus microplus [45].
Table 1 The list of the predicted and experimentally annotated merlin and ERM proteins included in this study.
Species Proteins UniProtKB/Swiss-Prot Identifiers GenBank Accession No. Entries from Genome Sequencing Projects Related Resources
Homo sapiens merlin (NF2) P35240 AAA36212
ezrin P15311 CAA35893
radixin P35241 AAA36541
moesin P26038 AAA36322
Pan troglodytes similar to NF2 XP_515061
Papio anubis merlin P59750 AAO23133
Bos taurus ezrin P31976 AAA30510
Sus scrofa radixin P26044 AAB02865
moesin P26042 AAB02864
Canis familiaris similar to NF2 XP_534729
Oryctolagus cuniculus ezrin Q8HZQ5 AAN06818
Mus musculus ezrin P26040 CAA43086
radixin P26043 CAA43087
merlin P46662 CAA52737
Rattus norvegicus ezrin P31977 AAR91694
NF2 XP_341249
Gallus gallus ezrin Q9YGW6 BAA75497
radixin Q9PU45 CAB59977
merlin NP_989828
Xenopus laevis unknown AAH77822
protein
Danio rerio nf2a Q6Q413 AAS66973
moesin Q503E6 AAH95359
Fugu rubripes radixin FRUP00000132603
moesin FRUP00000156313
merlin FRUP00000136298
Tetraodon nigroviridis unnamed CAG08868
protein 1 CAG08250
unnamed
protein 2
Ciona intestinalis erm-like ci0100149701
merlin-like ci0100130636
Ciona savignyi merlin-like paired_scaffold_109
Biomphalaria glabrata erm-like AAK61353
Lytechinus variegates moesin P52962 AAC46514
Apis mellifera similar to schwannomin XP_392673
Drosophila melanogaster merlin Q24564 AAB08449
moesin P46150 AAB48934
Drosophila yakuba merlin-like predicted in this work
Anopheles gambiae merlin-like fragment EAA07087
Caenorhabditis elegans erm1a P91015 AAB37643
erm1b P91016 AAB37642
nfm 1a Q20307 AAA19073
nfm 1b Q95QG5 AAK68385
Caenorhabditis briggsae erm-like BP:CBP03133
nfm1 BP:CBP05025
Caenorhabditis remanie merlin-like predicted in this work
erm-like
Brugia malayi merlin-like 316.m00022
Schistosoma japonicum JF2 AAB49033
Taenia saginata myosin-like Q94815 CAA65728
Echinococcus multilocularis EM10 A45620
Echinococcus granulosus EG10 Q24796 CAA82625
Phanerochaete chrysosporium ---
Aspergillus flavus ---
Arabidopsis thaliana ---
Oryza sativa ---
Trypanosoma brucei ---
Cryptosporidium parvum ---
Assembly of predicted merlin sequences from whole genome shotgun
To date, the genomes of Caenorhabditis remanei and Drosophila yakuba are represented by a set of contigs [46]. When contigs are ordered, oriented, and positioned with respect to each other by mate-pair reads, they are described as a scaffold. Scaffolds are the main product of the Whole Genome Shotgun strategy and can be assigned to chromosomes using chromosome-specific markers. Although the extensive scaffolds for the genomes of Caenorhabditis remanei and Drosophila yakuba are not currently available, we were able to assemble predictive protein sequences, which most resemble the merlin sequence of the closely-related organism, Caenorhabditis elegans or Drosophila melanogaster, respectively, using TBLASTN search across the available sets of contigs. In the Drosophila yakuba contig 49.37, we identified a predicted merlin sequence, which is nearly identical to that of the Drosophila melanogaster protein with the exception of three positions at the C-terminus, two substitutions at Glu468→Asp and Asn579→Ser and an insertion of Lys at position 575. Also, we found three Caenorhabditis remanei contigs, 564.6, 2151.1, and 2151.2, which contained merlin-like sequences with similarity, ranging from 81% to 100%, to its Caenorhabditis elegans counterpart. It should be noted that the deduced amino acid sequences were assembled manually, and in some cases, only partial or approximate amino acid sequences could be obtained. Nevertheless, they were useful for the identification of the definite gene in the respective genome and were valuable for the following phylogenetic reconstruction in order to validate the functional relationship and evolution of the definite gene.
Construction of a phylogenetic tree for the ERM family of proteins
To understand the origin and evolution of merlin, we conducted a phylogenetic analysis of the 50 proteins of the ERM family, which were identified from 30 different taxa (Table 1) using the neighbor-joining method [47,48] combined with the molecular evolutionary genetics analysis program MEGA2 [49]. Three protein 4.1 sequences from humans, mice, and zebrafish, respectively, were used as an outgroup. By comparing the bootstrap support values, which denote the number of times a grouping occurs out of 1,000 random samples from the alignment, we constructed a phylogenetic tree for the ERM family of proteins (Figure 1). Based on this phylogenetic analysis, the entire ERM family can be subdivided into the ERM clade and the merlin clade. While both clades show a strongly supported monophyletic origin, the merlin clade can be robustly defined and separated from the ERM clade (the bootstrap support value = 100). We identified a total of 22 sequences for the merlin clade and 28 sequences for the ERM clade. The topology of the phylogenetic tree within the merlin clade appears to agree with the general concept of evolutionary history of speciation.
Figure 1 The neighbor-joining tree of the ERM family. The diagram illustrates the basic resolution of the ERM-family members into two major clades, merlin and ERM. Bootstrap support values are shown above each node. Shaded boxes denote different subgroups of the ERM clade in vertebrates, which appeared after the expansion of the ERM-like ancestor. The Tetraodon nigroviridis "unnamed protein 1 and 2" sequences (GenBank Accession No. CAG08868 and CAG08250, respectively) and the Xenopus laevis "unknown protein" sequence (GenBank Accession No. AAH77822) were grouped based on their similarity to the merlin or ERM sequences.
The merlin clade can be further divided into three groups according to the order of derivation: worms, insects, and Chordata, with the earliest separated genus, Ciona, in the last taxonomic unit (Figure 1). The predicted merlin-like sequence from Caenorhabditis remanei branched from that of Caenorhabditis elegans, and similarly, Drosophila yakuba diverged from its Drosophila melanogaster counterpart. Both the "unnamed protein 1" of Tetraodon nigroviridis and the "unknown protein" of Xenopus laevis from the GenBank database are clustered in the Chordata merlin-like group with high bootstrap probabilities (Figure 1), which confirms their identity as merlin homologs. The protein fragment from Anopheles gambiae, which bears a sequence similarity to merlin, is grouped together with the Apis mellifera merlin-like protein by a bootstrap support value of 100.
Although the ERM-like proteins have been identified in Taenia saginata, Schistosoma japonicum, Echinococcus granulosus, and Echinococcus multilocularis [36-39], we did not find any merlin-like sequences in the genomes of these species. The lack of merlin-like sequences in these parasite genomes may be due to incomplete genome sequences in the database; however, this explanation is unlikely because the merlin-like sequence was also not observed in the genome of Schistosoma mansoni, which has been rigorously studied [50]. Another possibility is that the absence of merlin-like sequences in these organisms may reflect their adaptation to a parasitic lifestyle and the reduction of various organ systems. Alternatively, the merlin protein may emerge later during evolution. Similarly, no merlin-like sequence was found in the complete genomes of protozoa, fungi, and plants. Based on these results, we suppose that the derivation of merlin occurred in the early metazoa after its separation from flatworms.
As illustrated in the ERM clade in Figure 1, the ERM-like proteins found in parasites can be grouped together but form a separate branch from the rest of ERM proteins. Based on the phylogenetic analysis, the clustering of the "unnamed protein 2" of Tetraodon nigroviridis with the Fugu rubripes radixin protein defines it as a radixin-like protein. It should be noted that the two predicted ERM proteins, erm1a and erm1b of Caenorhabditis elegans [51], may represent different transcript variants of the same gene (also see below).
Furthermore, we have observed the evident expansion of the ERM-like ancestor in vertebrates (Figure 1). Since the ERM homolog of Ciona emerged prior to the vertebrate clade, it appears that the first duplication of the vertebrate ERM sequence occurred after its divergence from Ciona. Subsequent expansion within this sub-family has led to the present existence of three related groups of proteins, ezrin, radixin and moesin; among which, the ezrin group is the most ancient. Such an expanded complement may only be common to the ERM proteins of vertebrates because other metazoa have only one predicted ERM-like homolog [52-56]. Curiously, the increasing number of ERM members that occurred within the vertebrate clade paralleled the evolutionary complexity of the organism. It will be important to understand how these proteins evolved and how their functions coordinated because of the important and diverse functions of ERM proteins [8,25,26].
Evolution of the functionally important residues in merlin
Although initial identification of proteins via sequence similarities does not yield a clear indication of their respective functions, analysis of specific conserved regions and residues may provide important information regarding their putative functional characteristics. We conducted pairwise sequence comparison among all obtained merlin and ERM sequences, and identified several regions of interest. The results of the entire sequence alignment are provided in the Additional File 1 and are summarized in Figures 2, 3, and 4. Previously, the conservation of the N-terminal FERM domain among human ERM proteins and their functional importance were described [10-13]. In our alignment, we showed that this conservation extended to the merlin and ERM proteins of various species for which sequences were available to date. These data suggest a universal role for the presence of the FERM domain during evolution and further imply an existence of certain evolutionary constraints on the changes of their amino acid residues.
Figure 2 Sequence alignments of functionally important sites in the merlin and ERM proteins of various species. Comparison of the C-terminal region including the actin-binding site and two other predicted significant residues. Databank resources for the ERM-family proteins listed in Table 1 were used in the analysis, and only typical representatives from each group are displayed.
Figure 3 Alignment of the N-terminal domain, containing the Blue Box and the amino acid residue 204, conserved among the merlin proteins but divergent in the ERM proteins.
Figure 4 Sequence alignments reveal conservation of several functionally important residues, including the major phosphorylation site of the merlin group.
Although merlin lacks the C-terminal actin-binding site found in ERM proteins [7,20,21,57], it can directly interact with the actin cytoskeleton [22,58] or indirectly bind via the actin-binding protein βII spectrin/fodrin [23,24]. Sequence alignment showed extensive amino acid variability in the C-terminal region of the merlin proteins of various species, while a noncontiguous stretch of 25 amino acid residues, including the well-defined actin-binding site, was reliably aligned among all predicted ERM proteins with the exception of the "unnamed protein 2" of Tetraodon nigroviridis (Figure 2). According to the phylogenetic tree, the "unnamed protein 2" of Tetraodon nigroviridis is classified in the radixin group (Figure 1), and its sequence visibly differs from other radixin proteins only at the C-terminus. The reason for this sequence variability is presently unknown. It may be due to an inaccuracy in sequence assembly from the scaffold. Alternatively, the "unnamed 2 protein" may possess a unique characteristic and will be of considerable interest for functional comparison with other radixin proteins.
Sequence variability at the C-terminal domain of the merlin proteins of various species appears to be high, while some conservation can be found within separate taxonomic groups such as vertebrates, insects, and worms (Figure 2 and Additional File 1). A part of the C-terminal region is absent in Fugu rubripes, Danio rerio, C. briggsae, and Brugia malayi. This may be due to partial assembly of the protein sequences, as all of them were predicted by bioinformatics using the available genomes and cDNA sequences. Alternatively, the lack of conservation in the C-terminal region of merlin in these species may imply that this region does not share the same function. In the remaining organisms, the C-terminal amino acid residues have a specific charge distribution, in spite of decreased hydrophilicity, when compared with the C-terminal part of moesin [15]; however, they likely form structures similar to the B, C, and D helices found in moesin.
Unlike ERM proteins, two regions (residues 1–27 and 280–323) in the N-terminal half of merlin have been mapped that are sufficient for binding to F-actin [59,60]. The first 17 amino acids in the N-terminus of human merlin are present in the merlin proteins of various species but not in any ERM proteins (see Additional File 1). The merlin proteins of higher vertebrates contain these residues, eight of which are absent in the merlin proteins of other organisms. Crystal structure analysis suggests that the structure of these extreme N-terminal residues of merlin is disordered in solution but likely becomes ordered as merlin binds to some effector targets [17]. Our sequence alignment indicates that the conservation in the extreme N-terminus of merlin extends to the first 27 residues. The distribution of specific positively-charged residues also appears to be conserved in this N-terminal portion of the merlin proteins of lower vertebrates and insects. These results suggest that the first 27 amino acids of merlin serve as a common protein-binding motif. It is noteworthy that a similar sequence can be found in the ERM-like protein of Ciona; however, the N-terminal region of the Ciona protein contains ten positively-charged, basic amino acids, which may affect the binding to actin and/or other proteins (Additional File 1).
The internal actin-binding site, containing residues 280–323 in the N-terminal half of merlin, was found to be highly conserved among all merlin and ERM proteins analyzed, particularly the last 30 amino acid residues (Additional File 1). This region contains an extended helix at the beginning of the α-helical domain and its importance is supported by the identification of several disease-causing mutations (S315F, L316F, L316W, Q324L), which were predicted to destabilize the α-helical segment and disrupt its hydrogen bonding with subdomain A [16-18]. In addition, these residues have been shown to associate with F-actin in moesin [61,62] and to contribute to the ICAM-2-binding site in radixin [14].
Previously, LaJeunesse et al. [63] identified seven functionally important amino acid residues (170YQMTPEM177) in the N-terminal domain of Drosophila merlin, called the "Blue Box." These seven amino acids are identical between the human and Drosophila merlin proteins but differ from the ERM proteins. Sequence comparison revealed a more conserved pattern of the Blue Box; all seven amino acid residues of the Blue Box were found to be identical in the merlin sequences from vertebrates, fruit flies, and honeybees (Figure 3); however, several amino acid substitutions were found in the Blue Box of worms, fish, and Ciona. The most interesting substitutions were found in the merlin-like protein of Caenorhabditis from 174ThrProGlu176 to 174SerAlaAsp176. It is noteworthy that the methionine residue at position 177 in the Blue Box is conserved among all merlin proteins but not in the ERM proteins. These results further corroborate the functional importance of the seven amino acids in the Blue Box [63].
According to the crystal structure of the FERM domain in human merlin, the Blue Box residues are located in helix α3B of subdomain B [18] and form a defined area that is located on the surface of the protein [17]. Intriguingly, the three-dimensional conformation of merlin's Blue Box region is similar to that of the equivalent region in radixin [18], suggesting that regions in addition to the Blue Box are required for merlin to function as a tumor suppressor. Note that regions closely adjacent to the Blue Box-equivalent residues in human ERM proteins have been shown to participate in the N-terminal to C-terminal intramolecular interaction and ligand-binding, enabling increased mobility and structural changes in the activated FERM domain [14-16,64]. In light of the functional importance of the Blue Box in Drosophila merlin, its sequence conservation during evolution, and its location on the surface of merlin, the Blue Box probably participates in specific protein-protein interactions and contributes to other activities of merlin.
As in ERM proteins, phosphorylation affects the subcellular localization and intra- and inter-molecular associations of merlin [13,30-32]. In addition, it modulates the ability of merlin to suppress cell growth [34,35]. Two phosphorylation sites have been mapped to the Ser518 and Thr576 residues in the merlin protein. Phosphorylation on Ser518 has been shown to modulate the ability of merlin to form intramolecular associations and to bind to critical effectors important for growth suppression [34,35]. In contrast, phosphorylation on the Thr576 residue has no effect on merlin's functional activity, while phosphorylation on this residue is important for the function of ERM proteins [57,65-67]. Sequence alignment shows that the Ser518 residue is conserved across all merlin proteins of various species with the exception of the fruit fly and the worm, which contain a related threonine residue at the corresponding position (Figure 4). Since both the serine and threonine residues can be phosphorylated, we suggest that the corresponding threonine residue in merlin proteins of the fly and the worm may act as a phosphorylation site.
Gutmann et al. showed that mutations within the predicted α-helical region of the human merlin protein had little effect on its function, whereas those in its N- or C-terminus rendered the protein inactive as a negative growth regulator [28,29]. Specifically, five naturally occurring missense mutations, L64P, K79E, E106G, L535P and Q538P, were found to inactivate merlin function. Interestingly, we found that the Leu64 and Lys79 residues were conserved among the merlin and ERM proteins of various species (Figure 4). According to the crystal structure of the FERM domain of merlin, the L64P substitution would create a cavity in the hydrophobic core of subdomain A and affect its β-sheet structure [17,18]. The significance of this structural information was further supported by the finding that the L64P mutation impaired the ability of merlin to form an intramolecular complex between its two N-terminal interaction sites [28]. Moreover, the L64P mutant lost its ability to bind the cytoplasmic tail of CD44; this interaction correlates with the ability of merlin to function as a growth suppressor [29].
The Lys79 residue is situated at the end of helix α4A, and mutation at this residue (K79E) may cause the formation of a salt bridge with its neighboring Lys76 residue, which is normally hydrogen bonded to Tyr66 in helix α3A [17]. Two equivalent lysine residues, Lys60 and Lys63, were found in module F1 of moesin and were predicted to be involved in specific protein interactions, consequently changing the structure of an activated molecule [15,16]. Together with a group of positively-charged amino acids at the beginning of the helix module F3 (R275, K278, R279), these lysine residues interact with the negatively-charged residues (342-REKEE-346) in the C-terminal region [16]. Importantly, most of the homologous positively-charged residues located between lobes F1 and F3 in the radixin protein have been shown to bind to inositol 1,4,5-trisphosphate (IP3) [15]. In addition to Lys79, the Lys76 residue was also found to be highly conserved among various merlin and ERM proteins with the exception of the worm protein, which has a Gln76 instead of Lys76 (Figure 4 and Additional File 1). Also, the ERM-like proteins of parasites Taenia saginata, Echinococcus granulosus, and Echinococcus multilocularis contain an Arg76 residue, which is also a basic amino acid residue and may be capable of participating in interactions similar to those of the corresponding lysine residue. On the contrary, in the JF2 protein of Shistosoma mansoni, the position equivalent to Lys60 of moesin is occupied by a glutamic acid residue, and no conservation of residues 275, 278, and 279 in the JF2 protein was found, suggesting a unique structural feature for this Shistosoma protein.
Several other naturally-occurring missense mutations on human merlin, including E106G, L535P and Q538P, have also been found to affect its functional activity [29,68]. Our sequence alignment revealed that the Glu106, Leu535, and Gln538 residues were conserved among the merlin proteins of the Chordata group (Figure 4), highlighting the general importance of these residues for merlin function. Similar to the L64P mutation described above, the E106G mutation resulted in impaired intramolecular associations of merlin [29]. However, the Leu64 residue is highly conserved among all merlin and ERM proteins of various species, while Glu106 is conserved only in the merlin proteins of Chordata and worms. In the crystal structure of the FERM domain of merlin, the Glu106 residue is located at the linker A-B (α1'B) and participates in the inter-subdomain interaction by forming a hydrogen bond with the Lys289 residue [17,18]. This interaction may enable subdomain B to rotate closer to subdomain C. Intriguingly, Lys289 is conserved only among the merlin proteins of mammals, chickens, frogs, honeybees, and mosquitoes (Additional File 1). In the merlin proteins of other species, a negatively-charged aspartic or glutamic acid occupies this position, except in fish. Instead of lysine, an arginine residue was found in the homologous position of all ERM proteins (e.g., Arg273 for moesin), except for the ERM-like proteins in parasites (Additional File 1). This Arg273 residue, located at the beginning of the helix of subdomain F3, lies in a specific cleft between subdomains F1 and F3 with the positively-charged R275, K278, and R279 residues. According to the structure of radixin, the IP3-binding site is located at this basic cleft [14]. This region in the moesin protein has also been shown to interact with its C-terminal domain [16].
It should be mentioned that residues that are conserved in the merlin proteins, but not in the ERM proteins, of various species may be important for elucidating the functional difference between the merlin and ERM proteins. We found that the glutamic acid residue at position 204 of human merlin was conserved among all merlin proteins (Figure 3). In contrast, variable and uncharged amino acids were found at the corresponding position of the ERM proteins. Crystal structure of the FERM domain of human merlin shows that the Glu204 residue lies in the beginning of helix α4B and is included in a specific stretch of amino acids in subdomain B [17]. By sequence alignment of human merlin and ERM proteins, about 70 amino acids, including this specific stretch of residues, which are unique to merlin but different in ERM proteins, were identified (see Additional File 1). The majority of these amino acids can be subdivided into three clusters; each of them is specific to its corresponding subdomain and is located on the surface of the protein. These results suggest that these 70 amino acids likely take part in protein-protein interactions. Note that the conserved stretch of amino acids in subdomain B also includes the functionally important "Blue Box" discussed above.
Similarly, the isoleucine residue at position 546 was found to be conserved among the merlin proteins of various species, while a leucine residue was present in the corresponding position in all ERM proteins (Figure 2). The residue homologous to Leu529 in the C-terminal domain of moesin is located at the end of helix A with other hydrophobic residues, tightly contacting the hydrophobic faces of helices B and D of subdomain F2 [15]. Although the information about such an interaction in merlin is presently unavailable, additional crystal structure analysis should allow us to better understand the importance of this amino acid residue. In addition, it will be interesting to test whether mutations in the conserved amino acid residues identified in this study could affect protein function.
Predicted secondary structure of merlin and comparative analysis of the predicted α-helical region
Turunen et al. previously reported that the central region of ERM proteins contained approximately 200 residues that were predicted to be mostly α-helical [19]. To examine whether there was a similar structure present in all merlin proteins, we analyzed 21 merlin sequences from various organisms and predicted their secondary structures using the JPRED program [69]. The results of such an analysis for six representative species are presented in Figure 5. The predicted locations of α-helices and β-sheets in the N-terminal domain support the experimental findings from the structural analysis of the FERM domain of human merlin protein [18]. In addition, a predicted α-helical structure in the central-to-C-terminal region was found to be conserved among the merlin proteins of various species analyzed. Previously, it was shown that a truncated merlin protein of Drosophila, containing residues 1–600, lost the ability to localize to the cytoplasm and was concentrated at the plasma membrane [63]. However, two smaller truncated proteins, consisting of residues 1–350 or 1–375, were only loosely associated with the plasma membrane. These results suggest that the predicted α-helical region of merlin is important for its intracellular localization. Since almost the entire α-helical domain was absent in these truncated proteins, we suggest that it may contain a determinant for membrane association. This notion is further supported by the observation that additional truncated proteins, containing residues 1–300 or missing almost the entire α-helical domain, were diffusely localized to the cytoplasm.
Figure 5 The predicted secondary structures for the merlin proteins of various species. The region with a predicted β-sheet structure is shaded in grey, while the region with an α-helix structure is shaded in black. These predicted secondary structures correspond to the crystal structural data [18], which are shown above the alignment with the α-helix region indicated with a thick black bar and the β-sheet region with a thin black bar. The predicted α-helical domain in the central-to-C-terminal region of merlin is marked with an open bar. Asterisks denote known domains of the merlin protein with numbers pointing to the end of truncated Drosophila merlin protein discussed in the text. "+" denotes the beginning and the end of the predicted α-helical domain. The positions of specific residues in the FERM domain discussed in the text are denoted by black dots below the aligned sequences.
In human merlin, the predicted α-helical structure is situated between residues K312 and K478 (Figure 5 and Additional File 1). The N-terminal border of this structure was clearly recognized for 21 merlin sequences analyzed, whereas the C-terminal boundary could be traced only up to Urochordata (Ciona) and Nematoda (Caenorhabditis). This α-helical domain, predicted from all 21 merlin sequences, contains a high density of charged amino acids (from about 25% in Ciona to greater than 40% in vertebrates). Sequence alignment reveals 19 conserved amino acid residues in this predicted α-helical domain (Additional File 1). The amino acid identity for the predicted α-helical domain within each phylogenetic group is as follows: 1) Genus Drosophila (D. melanogaster and D. yakuba) – 99% (one amino acid substitution), 2) Genus Caenorhabditis (C. elegans, C. briggsae, and C. remanie) – 85.7%, 3) Genus Ciona (C. intestinalis and C. savignyi) – 71.7%, 4) vertebrates – 63.5%, and 5) mammals – 90%. Taken together, these results indicate that the merlin proteins of various species contain a conserved α-helical domain in the central to C-terminal region.
Exon-intron structural evolution of the merlin gene
Recent progress in automated computational analysis of partially and completely sequenced genomes using a gene prediction method and the analysis of expressed sequence tag (EST) has provided considerable opportunities not only to describe novel genes but also their exon-intron structures. Such an approach also allows the examination of the presence of different splicing variants/isoforms. To understand the evolution of the exon-intron structure of the merlin gene, we assembled all available NF2 gene-related sequences from various taxa. Using the sequences of proteins, mRNAs, and combined contigs [70], we established the structure of the merlin-like gene for Brugia malayi, which consists of 12 exons and 11 introns (Figure 6). Analogously, the homolog of the NF2 gene in Caenorhabditis elegans was found to contain 11 exons and 10 introns. It should be mentioned that the two NF2-like sequences, nfm-1a and nfm-1b of Caenorhabditis elegans, differ from each other only by the sequence of the last exon (Additional File 1), suggesting that they represent cDNA isoforms.
Figure 6 Schematic diagram of the exon and intron structures for the merlin genes of various species. The horizontal line depicts the merlin gene with its size indicated in base pairs (bp) on the right. The upright boxes represent exons. The lengths of the merlin mRNA sequences available in the database are shown in nucleotides (nt) and the lengths of the predicted merlin proteins are also indicated in amino acids (aa). The indicated human NF2 mRNA refers to the longest, full-length transcript identified, which contains a long 3' untranslated region [72]. Two major NF2 isoforms I and II are produced via alternative splicing and the lengths of their protein products are shown with that of isoform I indicated in the parentheses. It should be noted that Northern blot analysis detected the rat and mouse NF2 mRNAs of about 4.5 kb, indicating that the sizes of the rodent NF2 mRNAs shown here are not full-length. The asterisk (*) denotes the exon-intron structure of Brugia malayi predicted from this study.
As shown in Figure 6, the general arrangement of the merlin gene structure is conserved among mammalian species, especially at the region that encodes the N-terminal domain, albeit the length of the genes may differ slightly. The human NF2 gene consists of 17 exons and spans about 95 kb of DNA [5,6,71,72]. NF2 transcripts undergo alternative splicing, generating multiple isoforms [72-79]. Isoform I, missing exon 16, and isoform II, containing all 17 exons, are the two predominant species. As the result of alternative splicing, isoform 1 encodes a 595 amino acid protein. Isoform II differs from isoform I only at the C-terminus. Insertion of exon 16 into the mRNA provides a new stop codon, resulting in a 590 amino acid protein that is identical to isoform I over the first 579 residues. Because of the presence of a long 3' untranslated region, the longest NF2 isoform I RNA, containing the sequence from all 17 exons, is about 6.1 kb [72]. The merlin genes of other mammalian species, Rattus norvegicus, Canis familiaris, Mus musculus and Pan troglodytes, contain 16 exons (Figures 6 and 7). In addition, alternatively spliced merlin isoforms have been found in rodent species [80-82]. On the contrary, the structure of the merlin genes of Gallus gallus and Fugu rubripes are arranged differently from those of mammalian species, with 15 and 14 exons spanning much shorter DNAs of only about 25 kb and 12.3 kb, respectively (Figure 6). Although the NF2 gene of Fugu rubripes has not yet been completely assembled, we predict that it lacks the sequences of the first and the last exons of the mammalian NF2 gene based on our sequence alignment (see Additional File 1).
Figure 7 The alignment of exons with specific domains of merlin reveals the presence of homologous introns. Boxes represent the coding exons with numbers indicated accordingly. The locations of the three commonly discussed domains are marked with horizontal arrows under exons. The boundaries among these domains were defined according to the human merlin protein. The asterisk indicates that the exon structure shown is common to the merlin genes of all vertebrate species studied, including Homo sapiens, Pan troglodytes, Canis familaris, Mus musculus, Rattus norvegicus, Danio rerio, Fugu rubripes, and Xenopus laevis. The overall merlin gene structure of Gallus gallus is similar, except that exons 15 and 16 are fused together in this species.
In spite of the presence of 16 exons and the size of transcript similar to those found in some vertebrates, the gene for the merlin-like protein of Ciona intestinalis is relatively small with only about 4.3 kb (Figure 6). This tendency towards reduction of intron length and number continues to be seen in the merlin gene of worms and insects (Figures 6 and 7). The gene for the merlin-like protein of Caenorhabditis elegans, consisting of 11 exons, spans about 4.7-kb DNA, and that of Brugia malayi, containing 12 exons, is about 5.5 kb in length. The NF2 homolog of Drosophila melanogaster and the gene for the merlin-like protein of Apis mellifera are only about 2.9 kb, the smallest in the merlin clade, and consist of 5 and 8 exons, respectively (Figure 6). Interestingly, some conservation of the positions of homologous introns can be found in the NF2 gene from various species (Figure 7); however, the sizes of their introns are variable. Such an evolutionary conservation of homologous introns implies that the presence of regulatory sequences in the introns to regulate the transcriptional event.
Unlike the sizes and structures of the merlin or merlin-like genes in various organisms, the lengths of the merlin proteins and transcripts have not changed very much during evolution (Figure 6). Moreover, several functionally important domains of merlin also remain conserved. Since the merlin protein of the insect emerged after deviation from that of the worm, which was anciently derived from the common ancestor (Figure 1), it appears that the decrease in gene size and exon number occurred specifically within the insect group. This branch of merlin evolution is likely to develop independently and in the opposite direction from those more recently developed merlin proteins of Chordata. Parallel evolution towards the increase in merlin gene size and exon number between the worm and Chordata appears to be less likely. Thus, it is possible that the common ancestor for the merlin genes of the worm, the insect, and Chordata may contain a few more exons. During evolution, reduction of introns and/or fusion of exons occur within the insect group.
It is evident that the genome of the insect is more complicated than that of the worm. The simplification of the merlin gene structure in the insect is unique and may have a functional significance. This observation may explain the lack of splicing variants of the NF2 homolog in Drosophila, in contrast to those merlin isoforms found in mammals [72-82] and in Caenorhabditis elegans as we predicted in this study.
Conclusion
We have conducted the phylogenetic analysis of merlin diversity across metazoan genomes using the experimentally annotated and predicted sequences in conjunction with bioinformatic tools. We show that the merlin proteins have a monophyletic origin with a root in the early metazoa. We have also established the expansion of the ERM-like ancestors within the vertebrate clade that occurred after its separation from Urochordata. Several potentially important sites that are conserved among all merlin proteins but are divergent in the ERM members have been identified. As supported by the crystal structure data, these conserved residues are likely to be important for merlin function. Analysis of the evolution of the merlin gene structure reveals the existence of common splicing variants in human and Caenorhabditis elegans. While a trend toward the increase of gene length during evolution was observed, a reduction of intron number and length appears to occur in the merlin gene of the insect group. Taken together, our results have important implications for the evolution of the merlin proteins and their possible functional variability in different taxa.
Methods
BLAST search
Initial sequences of genes and proteins of interest from various organisms were identified from the National Center for Biotechnology Information (NCBI) database [83] using the BLAST algorithm [84]. We then searched the desirable sequences across genomic databases of completely or partially sequenced genomes available at The Sanger Institute [85] and The Institute for Genomic Research (TIGR; [86]). Also, we investigated other available sequence databases that contain information for specific organisms. The sources of sequences of the predicted or experimentally annotated merlin and ERM proteins are summarized in Table 1.
To obtain the entire amino acid sequence of an annotated protein, we used UniProt from Universal Protein Resource [87]. The erythrocyte membrane proteins 4.1 sequences of Homo sapiens (GenBank: CAI21970), Mus musculus (GenBank: NP_001006665), and Danio rerio (GenBank: AAQ97985) were also included in the analysis as an outgroup. Because of the presence of many non-conserved and large introns in the genes of interest, we conducted BLAST search using TBLASTN alignment algorithm in the cases where no protein sequences were available.
Alignments and phylogeny
The CLUSTAL_X program [88] was used to align the characterized or predicted protein sequences from different species. Phylogenetic analysis was carried out using the MEGA2.1 program [49].
Secondary structure prediction
The secondary structure prediction program JPRED [69] was used to predict the secondary structure of each merlin protein from various species. This program uses physical-chemical properties of the amino acid sequence and neural network for recognition of α-helices and β-sheets.
List of Abbreviations
NF2 – Neurofibromatosis type 2
NF2 – the Neurofibromatosis type 2 gene
ERM – ezrin, radixin, and moesin
FERM – 4.1, ezrin, radixin, and moesin
IP3 – inositol 1,4,5-trisphosphate
TIGR – The Institute for Genomic Research
EST, expressed sequence tag
NCBI – National Center for Biotechnology Information
Authors' contributions
KG and AB carried out the phylogenetic analysis of merlin diversity across metazoan genomes and drafted the manuscript. EMA and LVO helped with the design of the study and preparation of data for the figures. LSC was the principal investigator of the project and participated in the design, coordination, and writing of the manuscript. All authors read and approved the final manuscript
Supplementary Material
Additional File 1
Complete amino acid sequence alignment of the merlin and ERM proteins. Letters shaded in grey illustrate the conservation pattern of aligned sequences. The names of the merlin proteins from various species are shown in yellow. The conserved residues of the 'Blue Box' are also shaded in yellow. The positions of residues discussed in the text are colored in red. Blue letters denote the conserved residues within the predicted α-helical domain. Numbers indicated at the end of each sequence refer to the positions of the last residue within each protein analyzed.
Click here for file
Acknowledgements
We sincerely thank Sarah S. Burns, Maria Mihaylova, and D. Bradley Welling for critical reading of the manuscript, and Hui-Chun Hsiao and Kiselev Arkadyi for the help with the diagrams. This study was supported by grants from the US Department of Defense Neurofibromatosis Research Program and Russian Fund of Fundamental Investigations.
==== Refs
Dacks JB Doolittle WF Novel syntaxin gene sequences from Giardia, Trypanosoma and algae: implications for the ancient evolution of the eukaryotic endomembrane system J Cell Sci 2002 115 1635 1642 11950882
Hsu S Bioinformatics in reproductive biology – functional annotation based on comparative sequence analysis J Rep Immunol 2004 63 75 83 10.1016/j.jri.2004.01.008
NIH Consensus Statement on Acoustic Neuroma Neurofibromatosis Res Newsletter 1992 8 1 7
Bull. World Health Org Prevention and control of neurofibromatosis: Memorandum from a joint WHO/NNFF meeting 1992 70 173 182
Rouleau GA Merel P Luchtman M Sanson M Zucman J Marineau C Hoang-Xuan K Demczuk S Desmaze C Plougastel B Pulst SM Le noir G Bijlsma E Fashold R Dumanski J de Jong P Parry D Eldridge R Aurias A Delattre O Thomas G Alteration in a new gene encoding a putative membrane-organizing protein causes neuro-fibromatosis type 2 Nature 1993 363 515 521 8379998 10.1038/363515a0
Trofatter JA MacCollin MM Rutter JL Murrell JR Duyai MP Parry DM Eldridge RE Kley N Menon AG Pulaski K Haase VH Ambrose CM Munroe D Bove C Haines JL Martuza RL MacDonald ME Seizinger BR Short MP Buckner AJ Gusella JF A novel moezin-, ezrin-, radixin-like gene is a candidate for the neurofibromatosis 2 tumor suppressor Cell 1993 72 791 800 8453669 10.1016/0092-8674(93)90406-G
Algrain M Arpin M Louvard D Wizardry at the cell cortex Current Biol 1993 3 451 454 10.1016/0960-9822(93)90354-Q
Tsukita S Oishi K Sato N Sagara J Kawai A Tsukita S ERM family members as molecular linkers between the cell surface glycoprotein CD44 and actin-based cytoskeleton J Cell Biol 1994 126 391 401 7518464 10.1083/jcb.126.2.391
Takeuchi K Kawashima A Nagafuchi A Tsukita S Structural diversity of band 4.1 superfamily members J Cell Sci 1994 107 1921 1928 7983158
Chishti AH Kim AC Marfatia SM Lutchman M Hanspal M Jindal H Liu SC Low PS Rouleau GA Mohandas N Chasis JA Conboy JG Gaskard P Takakuwa Y Huang SC Benz EJ JrBretcher A Fenon RG Gusella JF Ramesh V Solomon F Marchesi VT Tsukita S Hoover KB The FERM domain: a unique module involved in the linkage of the cytoplasmic proteins to the membrane Trends Biochem Sci 1998 23 281 282 9757824 10.1016/S0968-0004(98)01237-7
Hamada K Shimizu T Matsui T Tsukita S Hakoshima T Structural basis of the membrane-targeting and unmasking mechanisms of the radixin FERM domain EMBO J 2000 19 4449 4462 10970839 10.1093/emboj/19.17.4449
Herrlich P Morrison H Sleeman J Orian-Rousseau V Konig H Weg-Remers S Ponta H CD44 acts both as a growth- and invasiveness-promoting molecule and as a tumor-suppressing cofactor Ann NY Acad Sci 2000 910 106 118 10911909
Morrison H Sherman LS Legg J Banine F Isacke C Haipek CA Gutmann DH Ponta H Herrlich P The NF2 tumor suppressor gene product, merlin, mediates contact inhibition of growth through interactions with CD44 Genes Dev 2001 15 968 980 11316791 10.1101/gad.189601
Hamada K Shimizu T Yonemura S Tsukita Sh Tsukita S Hakoshima T Structural basis of adhesion – molecule recognition by ERM proteins revealed by the crystal structure of the radixin – ICAM-2 complex EMBO J 2003 22 502 514 12554651 10.1093/emboj/cdg039
Pearson MA Reczek D Bretscher A Karplus PA Structure of the ERM protein moesin reveals the FERM domain fold masked by an extended actin binding tail domain Cell 2000 101 259 270 10847681 10.1016/S0092-8674(00)80836-3
Edwards SD Keep NH The 2.7 Å crystal structure of the activated FERM domain of moesin: an analysis of structural changes on activation Biochemistry 2001 40 7061 7068 11401550
Kang BS Cooper DR Devedjiev Y Derewenda U Derewenda ZS The structure of the FERM domain of merlin, the neurofibromatosis type 2 gene product Acta Crystollogr Sec D 2002 58 381 391 10.1107/S0907444901021175
Shimizu T Seto A Maita N Hamada K Tsukita Sh Tsukita S Hakoshima T Structural basis for neurofibromatosis type2 J Biol Chem 2002 277 10332 10336 11756419 10.1074/jbc.M109979200
Turunen O Sainio M Jaaskelainen J Carpen O Vaheri A Structure-function relationships in the ezrin family and the effect of tumor-associated point mutations in neurofibromatosis 2 protein Biochim Biophys Acta 1998 1387 1 16 9748471
Turunen O Wahlström T Vaheri A Ezrin has a COOH-terminal acting-binding site that is conserved in the ezrin protein family J Cell Biol 1994 126 1445 1453 8089177 10.1083/jcb.126.6.1445
Gary R Bretscher A Ezrin self-association involves binding of an N-terminal domain to a normally masked C-terminal domain that includes the F-actin binding site Mol Biol Cell 1995 6 1061 1075 7579708
Sainio M Zhao F Heiska L Turunen O den Bakker M Zwarthoff E Lutchman M Rouleau GA Jaaskelainen J Vaheri A Carpen O Neurofibromatosis 2 tumor suppressor protein colocalizes with ezrin and CD44 and associates with actin-containing cytoskeleton J Cell Sci 1997 110 2249 2260 9378774
Scoles DR Huynh DP Morcos PA Coulsell ER Robinson NG Tamanoi F Pulst SM Neurofibromatosis 2 tumor suppressor schwannomin interacts with βII-spectrin Nature Genet 1998 18 354 359 9537418 10.1038/ng0498-354
Neill GW Crompton MR Binding of the merlin-I product of the neurofibromatosis type 2 tumour suppressor gene to a novel site in beta-fodrin is regulated by association between merlin domains Biochem J 2001 358 727 735 11535133 10.1042/0264-6021:3580727
Bretscher A Edwards K Fehon RG ERM proteins and merlin: integrators at the cell cortex Nature Rev Mol Cell Biol 2002 3 586 599 12154370 10.1038/nrm882
Ramesh V Merlin and ERM proteins in Schwann cells, neurons and growth cones Nature Rev Neurosci 2004 5 462 470 15152196 10.1038/nrn1407
McClatchey AI Merlin and ERM proteins: unappreciated roles in cancer development? Nature Rev Cancer 2003 3 877 883 14668818 10.1038/nrc1213
Gutmann DH Haipek CA Hoang Lu K Neurofibromatosis 2 tumor suppressor protein, merlin, forms two functionally important intramolecular associations J Neurosci Res 1999 58 706 716 10561699 10.1002/(SICI)1097-4547(19991201)58:5<706::AID-JNR12>3.0.CO;2-Z
Gutmann DH Hirbe AC Haipek CA Functional analysis of neurofibromatosis 2 (NF2) missense mutations Hum Mol Genet 2001 10 1519 1529 11448944 10.1093/hmg/10.14.1519
Shaw RJ Paez JG Curto M Yaktine A Pruitt WM Saotome I O'Bryan JP Gupta V Ratner N Der CJ Jacks T McClatchey AI The Nf2 tumor suppressor, merlin, functions in Rac-dependent signaling Dev Cell 2001 1 63 72 11703924 10.1016/S1534-5807(01)00009-0
Kissil JL Johnson KC Eckman MS Jacks T Merlin phosphorylation by p21-activated kinase 2 and effects of phosphorylation on merlin localization J Biol Chem 2002 277 10394 10399 11782491 10.1074/jbc.M200083200
Xiao GH Beeser A Chernoff J Testa JR p21-activated kinase links Rac/Cdc42 signaling to merlin J Biol Chem 2002 277 883 886 11719502 10.1074/jbc.C100553200
Alfthan K Heiska L Gronholm M Renkema GH Carpen O Cyclic AMP-dependent protein kinase phosphorylates merlin at serine 518 independently of p21-activated kinase and promotes merlin-ezrin heterodimerization J Biol Chem 2004 279 18559 18566 14981079 10.1074/jbc.M313916200
Surace EI Haipek CA Gutmann DH Effect of merlin phosphorylation on neurofibromatosis 2 (NF2) gene function Oncogene 2004 23 580 587 14724586 10.1038/sj.onc.1207142
Rong R Surace EI Haipek CA Gutmann DH Ye K Serine 518 phosphorylation modulates merlin intramolecular association and binding to critical effectors important for NF2 growth suppression Oncogene 2004 23 8447 8454 15378014 10.1038/sj.onc.1207794
Frosch PM Frosch M Pfister T Schaad V Bitter-Suermann D Cloning and characterisation of an immunodominant major surface antigen of Echinococcus multilocularis Mol Biochem Parasitol 1991 48 121 130 1762625 10.1016/0166-6851(91)90108-I
Frosch PM Muhlschlegel F Sygulla L Hartmann M Frosch M Identification of a cDNA clone from the larval stage of Echinococcus granulosus with homologies to the E. multilocularis antigen EM10-expressing cDNA clone Parasitol Res 1994 80 703 705 7886043 10.1007/BF00932958
Kurtis JD Ramirez BL Wiest PM Dong KL El-Meanawy A Petzke MM Johnson JH Edmison J Maier RA JrOlds GR Identification and molecular cloning of a 67-kilodalton protein in Schistosoma japonicum homologous to a family of actin-binding proteins Infect Immun 1997 65 344 347 8975937
Hubert K Cordero E Frosch M Solomon F Activities of the EM10 protein from Echinococcus multilocularis in cultured mammalian cells demonstrate functional relationships to ERM family members Cell Motil Cytoskeleton 1999 42 178 188 10098932 10.1002/(SICI)1097-0169(1999)42:3<178::AID-CM2>3.0.CO;2-3
McCartney BM Fehon RG Distinct cellular and subcellular patterns of expression imply distinct functions for the Drosophila homologues of moesin and the neurofibromatosis 2 tumor suppressor, merlin J Cell Biol 1996 133 843 852 8666669 10.1083/jcb.133.4.843
Hansson CM Ali H Bruder CE Fransson I Kluge S Andersson B Roe BA Menzel U Dumanski JP Strong conservation of the human NF2 locus based on sequence comparison in five species Mamm Genome 2003 14 526 536 12925885 10.1007/s00335-003-3011-3
Chen Y-X Gutmann DH Haipek CA Martinsen BJ Bronner-Fraser M Krull CE Characterization of chicken Nf2/merlin indicates regulatory roles in cell proliferation and migration Dev Dyn 2004 229 541 54 14991710 10.1002/dvdy.20002
Saitou N Nei M The neighbor-joining method: A new method for reconstructing phylogenetic trees Mol Biol Evol 1987 4 406 425 3447015
Kumar S Gadagkar SR Efficiency of the neighbor-joining method in reconstructing deep and shallow evolutionary relationships in large phylogenies J Mol Evol 2000 51 544 53 11116328
Kumar S Tamura K Jakobsen IB Nei M MEGA2: molecular evolutionary genetics analysis software Bioinformatics 2001 17 1244 1245 11751241 10.1093/bioinformatics/17.12.1244
The C. elegans Sequencing Consortium Genome sequence of the nematode C. elegans: a platform for investigating biology Science 1998 282 2012 2018 9851916 10.1126/science.282.5396.2012
Adams MD Celniker SE Holt RA Evans CA The genome sequence of Drosophila melanogaster Science 2000 287 2185 2195 10731132 10.1126/science.287.5461.2185
Dehal P Satou Y Campbell RK Chapman J The draft genome of Ciona intestinalis: insights into chordate and vertebrate origins Science 2002 298 2157 2167 12481130 10.1126/science.1080049
Stein LD Bao Z Blasiar D Blumenthal T The genome sequence of Caenorhabditis briggsae: a platform for comparative genomics PLoS Biol 2003 1 E45 14624247 10.1371/journal.pbio.0000045
Foster JM Kumar S Ganatra MB Kamal IH Ware J Ingram J Pope-Chappell J Guiliano D Whitton C Daub J Blaxter ML Slatko BE Construction of bacterial artificial chromosome libraries from the parasitic nematode Brugia malayi and physical mapping of the genome of its Wolbachia endosymbiont Int J Parasitol 2004 34 733 746 15111095 10.1016/j.ijpara.2004.02.001
Matsui T Maeda M Doi Y Yonemura S Amano M Kaibuchi K Tsukita S Rho-kinase phosphorylates COOH-terminal threonines of ezrin/radixin/moezin (ERM) proteins and regulates their head-to-tail association J Cell Biol 1998 140 647 657 9456324 10.1083/jcb.140.3.647
Deguen B Merel P Goutebroze L Giovannini M Reggio H Arpin M Thomas G Impaired interaction of naturally occurring mutant NF2 protein with actin-based cytoskeleton and membrane Hum Mol Genet 1998 7 217 226 9425229 10.1093/hmg/7.2.217
Xu H Gutmann DH Merlin differentially associates with the microtubule and acting cytoskeleton J Neurosci Res 1998 51 403 415 9486775 10.1002/(SICI)1097-4547(19980201)51:3<403::AID-JNR13>3.0.CO;2-7
Brault E Gautreau A Lamarine M Callebaut I Gilles T Normal membrane localization and actin association of the NF2 tumor suppressor protein are dependent on folding of its N-terminal domain J Cell Sci 2001 114 1901 1912 11329377
Roy C Martin M Mangeat P A dual involvement of the amino terminal domain of ezrin in F- and G-actin binding J Biol Chem 1997 272 20088 20095 9242682 10.1074/jbc.272.32.20088
Martin M Roy C Montcourrier P Sahuquet A Mangeat P Three determinants in ezrin are responsible for cell extension activity Mol Cell Biol 1997 8 1543 1557
LaJeunesse DR McCartney BM Fehon RG Structural analysis of Drosophila merlin reveals functional domains important for growth control and subcellular localization J Cell Biol 1998 141 1589 1599 9647651 10.1083/jcb.141.7.1589
Smith WJ Nassar N Bretscher A Cerione RA Karplus PA Structure of the active N-terminal domain of ezrin J Biol Chem 2003 278 4949 4956 12429733 10.1074/jbc.M210601200
Nakamura F Amieva MR Furthmayr H Phosphorylation of threonine 558 in the carboxyl-terminal actin-binding domain of moesin by thrombin activation of human platelets J Biol Chem 1995 270 31377 31385 8537411 10.1074/jbc.270.20.12319
Oshiro N Fukata Y Kaibuchi K Phosphorylation of moesin by rho-associated kinase (Rho-kinase) plays a crucial role in the formation of microvilli-like structures J Biol Chem 1998 273 34663 34666 9856983 10.1074/jbc.273.52.34663
Dard N Louvet-Vallee S Santa-Maria A Maro B Phosphorylation of ezrin on threonine T567 plays a crucial role during compaction in the mouse early embryo Dev Biol 2004 271 87 97 15196952 10.1016/j.ydbio.2004.03.024
Stokowski RP Cox DR Functional analysis of the neurofibromatosis type 2 protein by means of disease – causing point mutations Am J Hum Genet 2000 66 873 891 10712203 10.1086/302812
Zucman-Rossi J Legoix P Der Sarkissian H Cheret G Sor F Bernardi A Cazes L Giraud S Ollagnon E Lenoir G Thomas G NF2 gene in neurofibromatosis type 2 patients Hum Mol Genet 1998 7 2095 2101 9817927 10.1093/hmg/7.13.2095
Chang L-S Akhmametyeva EM Wu Y Zhu L Welling DB Multiple transcription initiation sites, alternative splicing, and differential polyadenylation contribute to the complexity of human neurofibromatosis 2 transcripts Genomics 2002 79 63 76 11827459 10.1006/geno.2001.6672
Arakawa H Hayashi N Nagase H Ogawa M Nakamura Y Alternative splicing of the NF2 gene and its mutation analysis of breast and colorectal cancers Hum Mol Genet 1994 3 565 568 8069299
Bianchi AB Hara T Ramesh V Gao J Mutations in transcript isoforms of the neurofibromatosis 2 gene in multiple human tumour types Nature Genet 1994 6 185 192 8162073 10.1038/ng0294-185
Hara T Bianchi AB Seizinger BR Kley N Molecular cloning and characterization of alternatively spliced transcripts of the mouse neurofibromatosis 2 gene Cancer Res 1994 54 330 335 7506121
Hitotsumatsu T Kitamoto T Iwaki T Fukui M Tateishi J An exon 8-spliced out transcript of neurofibromatosis 2 gene is constitutively expressed in various human tissues J Biochem (Tokyo) 1994 116 1205 1207 7706208
Pykett MJ Murphy M Harnish PR George DL The neurofibromatosis 2 (NF2) tumor suppressor gene encodes multiple alternatively spliced transcripts Hum Mol Genet 1994 3 559 564 8069298
Nishi T Neurofibromatosis 2 gene has novel alternative splicing which control intracellular protein binding Int J Oncol 1997 10 1025 1029
Schmucker B Tang Y Kressel M Novel alternatively spliced isoforms of the neurofibromatosis type 2 tumor suppressor are targeted to the nucleus and cytoplasmic granules Hum Mol Genet 1999 8 1561 1570 10401006 10.1093/hmg/8.8.1561
Haase VH Trofatter JA MacCollin M Tarttelin E Gusella JF Ramesh V The murine NF2 homologue encodes a highly conserved merlin protein with alternative forms Hum Mol Genet 1994 3 407 411 8012352
Huynh DP Nechiporuk T Pulst SM Alternative transcripts in the mouse neurofibromatosis type 2 (NF2) gene are conserved and code for schwannomins with distinct C-terminal domains Hum Mol Genet 1994 3 1075 1079 7981675
Gutmann DH Wright DE Geist RT Snider WD Expression of the neurofibromatosis 2 (NF2) gene isoforms during rat embryonic development Hum Mol Genet 1995 4 471 478 7795605
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
Thompson JD Gibson TJ Plewniak F Jeanmougin F Higgins DG The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools Nucleic Acids Res 1997 15 4876 4882 9396791 10.1093/nar/25.24.4876
|
16324214
|
PMC1315344
|
CC BY
|
2021-01-04 16:29:16
|
no
|
BMC Evol Biol. 2005 Dec 2; 5:69
|
utf-8
|
BMC Evol Biol
| 2,005 |
10.1186/1471-2148-5-69
|
oa_comm
|
==== Front
Immun AgeingImmunity & ageing : I & A1742-4933BioMed Central London 1742-4933-2-161630768110.1186/1742-4933-2-16CommentaryImmunosenescence and Vaccination Pawelec Graham [email protected] Center for Medical Research (ZMF), University of Tubingen, Tubingen, Germany2005 24 11 2005 2 16 16 2 11 2005 24 11 2005 Copyright © 2005 Pawelec; licensee BioMed Central Ltd.2005Pawelec; 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.
The problems associated with the ageing immune system and vaccination were discussed recently at an international workshop at the Jenner Institute for Vaccine Research, Compton, UK, 6–7 October, 2005. This is a commentary on that session. The meeting included discussions on T and B cell differentiation and ageing, as well as dendritic cell and neutrophil data, with the emphasis on T cell immunosenescence, perceived as the most important hindrance to satisfactory responses to vaccines in the elderly. The main questions to be addressed in this context are the reasons for dysfunctionality of T cells in the elderly and what to do to improve T cell function. Several of the major reasons for poor T cell responses in the elderly were discussed; however, many important questions remain: The next meeting at the Jenner Institute may already be able to provide some of the answers to these questions, which have serious implications for public health issues in increasingly elderly populations.
==== Body
Introduction
An international workshop on Immunosenescence and Vaccination was organised at the Jenner Institute for Vaccine Research, Compton, UK, 6–7 October, 2005 by Peter Beverley (Jenner Institute), Arne Akbar (University College, London, UK) and Don Palmer (Royal Veterinary College, London, UK). The most profound clinical impact of age on the immune system concerns the response of the elderly to vaccination. The meeting included discussions on T and B cell differentiation and ageing, as well as dendritic cell and neutrophil data, with the emphasis on T cell immunosenescence, perceived as the most important hindrance to satisfactory responses to vaccines in the elderly. The main questions to be addressed in this context are the reasons for dysfunctionality of T cells in the elderly and what to do to improve T cell function.
What is the problem?
Several of the major reasons for poor T cell responses in the elderly were discussed: is it a problem of stem cells with decreased potential for differentiation into naïve T cells? Is it a problem of the thymic environment not supporting this differentiation sufficiently any more? Is it a problem of naïve T cell ageing in the periphery and/or memory cell ageing and loss? The answers as they are emerging of course suggest that all these problems, and more, apply, as outlined below.
Diana Wallace and Peter Beverley [1] at the Jenner Institute, together with Derek Macallan in London, measured T cell turnover during human ageing by assessing the rate of uptake of deuterated glucose into DNA and its dilution with cell division. In healthy young people, CD45RO+ cells divide more rapidly than CD45RA+ cells in both the CD4 and CD8 subsets, whereby CD45RO+ cells have a 26 day half-life, with 2.7% of cells dividing every day, whereas the RA+ cells have a 154 day half-life and only 0.5% divide every day. Rather surprisingly, perhaps, this is the same in young and old, so these findings already suggest that memory T cells are turning over faster than naïve T cells. As the naïve cell division rate is the same in the elderly and the young, it is likely that naïve cells in the elderly have accumulated a larger number of population doublings during the life of the individual. This is consistent with and indeed explains the data on naïve cell ageing in mice from Laura Haynes [2] (Trudeau Institute, Saranac Lake, NY). However, it also suggests that if memory cell division is more rapid, replicative senescence of this population should occur faster. However, the rate of attrition of CD8+ CD45RA+ primed T cells was lower in the elderly than the young (by a factor > 10), and these persistent cells contained large clonal populations. This probably reflects acquisition of apoptosis-resistance by senescent CD8 cells and their accumulation in a dysfunctional state, as later described by Graham Pawelec [3] (Center for Medical Research, University of Tübingen, Germany). As these clonal expansions were to a great degree associated with seropositivity for Cytomegalovirus in the young, such data are consistent with many others presented here and elsewhere suggesting that CMV is a major driving force behind many of the measured manifestations of immunosenescence in humans. During this talk and others, there was much discussion on the impact of CMV on the human immune system and whether CMV in some way regulated (prevented) cell death in virus-specific, or even bystander, CD8 populations.
Role of persistent antigenic stimulation
Chronic antigenic stimulation over a lifetime via a source resistant to elimination (persistent Herpes viruses, parasites, cancer, even autoantigens) may result in deleterious effects on the immune system. There is now a good body of evidence regarding the impact of CMV, and to a lesser extent, EBV infection on CD8 cells in young and elderly humans, but knowledge of its effects on CD4 cells is less well-developed, and on NK cells in the elderly almost lacking. However, Paul Moss [4] (Dept Cancer Studies, University of Birmingham, UK) reported preliminary studies showing little effect of CMV on NK cells, B cells or regulatory T cells in the elderly. He emphasised that CMV negatively influenced the number of naïve CD8 cells in the elderly and suggested that CMV accelerated age-associated processes which occur anyway, probably driven by other "subdominant" antigens. This is consistent with the much-discussed idea that chronic antigenic stimulation of any kind at any age can drive clonal expansions resulting in overall deleterious effects on immune function. Advanced age merely results in the accumulation of multiple such problems. Along these lines, evidence presented by Janko Nikolich-Zugich [5] (Oregon Health & Science Unversity, OR, USA) strongly suggests that even sustained proliferation not dependent on exogenous antigen (including that induced by lymphopenia or adjuvants) over a lifetime can induce persistent T cell clonal expansions in specific pathogen-free mice. In addition, he showed that persistent HSV-1 infection also led to the accumulation of clones specific for this virus, which also became dysregulated over time. These events were prevented by continuous antiviral treatment of infected mice, showing directly that subclinical herpesvirus reactivation is necessary to drive expansions of memory cells. G. Pawelec [3] pursued this line of thought by comparing chronic antigenic stimulation caused by persistent activating viruses in the elderly with that caused by tumour antigens in (mostly younger) cancer patients, as well as changes that can be observed in culture models in vitro. Functionally and phenotypically, many similarities in CD8 responses can be observed in the elderly and in cancer patients. The accumulation of CMV-specific CD8 cells bearing the hallmarks of anergic cells represents an important part of the "Immune Risk Phenotype" (IRP) predictive of mortality in the very elderly [6], and similarly dysfunctional cells in cancer patients may compromise the response to the tumour in an analogous fashion. The emergence of the IRP concept from the longitudinal ageing studies carried out by Anders Wikby and others in Sweden, and studied as a long-term collaborative project supported by the European Commission (currently under contract QLK6-CT-2002-02283, coordinated by G. Pawelec, see ) will hopefully also prove useful for assessing "biomarkers of immunosenescence" in younger individuals, and possibly cancer patients, as well. Recent work from the group has shown that CD8 clonal expansions driven predominantly by CMV (but also with EBV playing a part) begin already in early middle-age. We now know that remaining life expectancy in the very elderly (>85 yr) correlates with having greater numbers of different clonal expansions compared to those with few clonal expansions (but note that all individuals possess very large overall numbers of CMV-specific CD8 cells) [7]. We explain this by hypothesising that CMV drives clonal expansions of multiple CD8 cells recognising different CMV epitopes (in the young, up to 10% of the T cell repertoire is already obsessed with CMV, as shown by P. Moss [8] in his presentation). The number of different clones expanded increases with age, as does the overall number of CD8 cells recognising CMV epitopes. However, in very advanced age, this clonal diversity starts to shrink, as apoptosis-resistant senescent cells are more slowly lost by attrition (via a process not yet clarified) and individuals possessing only a small number of clonal expansions are now more at risk of death than those still retaining a larger repertoire. Why this is so, is also not yet clear, but it is felt unlikely that they die of CMV disease.
The presumably important role of CMV-specific CD4 cells in this context has not yet been assessed. In this direction, Arne Akbar [9] presented his studies on the functional status of CD4 cells in the elderly using assays which only reveal non-anergic cells, unlike the studies on CD8 cells, which revealed larger numbers of dysfunctional cells. This is because the production of equivalent MHC class II multimers is technically much more challenging than the class I reagents used for CD8 cells. Nonetheless, the data on those CMV-specific CD4 cells in the elderly which did remain functional were extremely informative. In seropositive donors, CMV-specific cells had shorter telomeres, a sign of increased proliferative history. CMV lysates stimulated secretion of type I interferon (IFN-α) from dendritic cells, which acted to inhibit telomerase function in all virus-specific cells (not just the CMV-specific cells) and to increase the fraction of CD27- CD28- cells. This may explain the unexpected finding that CTL in CMV-seropositives possess more differentiated phenotypes, whether they are CMV-specific or specific for unrelated antigens.
What can be done to rejuvenate the elderly immune system?
Targeted neutralisation of CMV-stimulated IFN-α production might be beneficial in the CMV-positive elderly, according to the above-mentioned results. Other possibilities include the use of anti-viral agents or therapeutic vaccination, but neither of these approaches is likely to be available in the near future for use in the elderly. Nonetheless, agents such as valacyclovir have been developed, which have a good safety record in younger people and might be applicable to the elderly. Similarly, it seems unlikely that adoptive immunotherapy with ex-vivo-generated CMV-specific CD8 cells, despite being semi-routine in stem cell transplant recipients, would ever be easily transferable to the elderly. The elimination of the dysfunctional CMV-specific CD8 cells might be beneficial in removing potentially suppressive cells, or at the very least, simply making "space" for naïve cells that could in theory still be generated from residual thymic islands present in most elderly. How to target these, while leaving the functional CMV-specific cells intact, however, is a critical issue. Perhaps combining the use of KLRG-1 and CD57 to eliminate double positive cells might be a possibility, as proposed by G. Pawelec, because it is the small fraction of KLRG1+ but CD57- cells which seems to retain functionality in the elderly. One other possibility might be positive selection of the CMV-specific cells with a naïve phenotype (a very small number of which are present in most elderly, it seems) and expanding these ex vivo using IL 15. However, this would also require radical treatment involving ex vivo culture and adoptive immunotherapy, which might be practically difficult in the elderly. Along these lines, P. Beverley reported that naïve CD8 cells maintained in long-term culture in this way retained their naïve phenotype. These cultured cells upregulated telomerase and actually increased telomere length. This approach clearly also relies on the presence of naïve cells in the elderly, which could be isolated and expanded in culture. However, according to Beatrix Grubeck-Loebenstein [10] (Institute for Biology of Ageing, Innsbruck, Austria), there may not actually be any. She reported her search for truly naïve cells in the elderly, obtained by isolating CD8+ CD28+ CD45RA+ CD62L+ cells. Theoretically, these should be naïve, but she found that this subset also had shorter telomeres in the old than in the young, and, moreover, clonal spectratyping revealed a smaller repertoire in the former. So these cells have undergone considerable division, despite being phenotypially naïve. The implication is that even naïve cells in the elderly have "aged" in the same way as has been investigated in detail by L. Haynes [11] in the mouse model. She described that (in this case CD4+) naïve cells in old mice (CD28+, CD134+, CXCR5+) had low levels of CD40L expression (although this could be enhanced by IL 2 – use of which might also represent a potential avenue to remediation). When all the CD4 cells were depleted by antibody treatment, and the animals then left for 2 months to repopulate, new naïve cells did develop and even in old animals were now perfectly functional. These results do suggest that deleting accumulated dysfunctional cells, and allowing repopulation, can indeed "rejuvenate" the T cell system. Nonetheless, once again, it does not seem very likely that such radical approaches would be permitted in elderly people. However, focussing on the thymus and attempting to increase output of naïve cells in the elderly, whether or not combined with peripheral depletion, may still be a viable method for ameliorating immunosenescence. To this end, Richard Aspinall (Imperial College, London, UK) developed the strategy of targeting IL 7 [12] to the thymus only, to avoid cytokine side effects, by producing a fusion protein with CCR9, the receptors for which are expressed exclusively by thymic stromal cells. On challenge with influenza, mice had a decreased lung viral load when given the fusion protein, and fewer TNF-α-producing CD8 cells. The success of this approach requires not only that naïve cells are still present, but that in addition to the thymus, T cell progenitors are also still present and functional in the elderly. However, according to Ken Dorshkind [13] (UCLA, Los Angeles, USA), this may not be the case. He showed that "early T cell progenitors" (ETP; CD44+ CD25- c-kithigh CD127-) from old animals are fewer in number, more prone to apoptosis and have less proliferative capacity than in the young. Thus, if there is a block in the potential of hematopoietic stem cells to generate ETP, it would be difficult to see how IL 7 supplementation would succeed. One final possibility is the well-established approach of caloric restriction, discussed by J. Nikolich-Zugich [14] in the rhesus monkey. Calorically-restricted CMV+ monkeys have maintained numbers of naïve CD8 cells better than ad lib fed animals; they also had greater numbers of T cell receptor-excision circle-bearing cells (ie those with a lesser proliferative history) and lower levels of pro-inflammatory factors (TNF-α, IFN-γ). They also had fewer effector-memory cells. So perhaps even regarding the deleterious effects of CMV, the advice to eat less might still be some of the best available.
Conclusion
Many questions remain. If CMV is really having such a disastrous effect in "accelerating" immunosenescence in the elderly, is it doing the same in the young? Are CMV-seronegative elderly donors healthier and do they live longer (and are centenarians CMV-seronegative?)? If CD4 and CD8 cells are equally affected, but B cells, dendritic cells and NK cells are really not (to be confirmed), are all types of the former equally affected (specifically, what about T-regulatory cells?)? Does CMV reactivate more frequently in the at-risk elderly, or does simply the duration of infection determine the level of immunosenescence? Do at-risk elderly ever die of disease which is related to CMV reactivation? Do CMV+ elderly people who have entered the IRP group ever leave it, and is this associated with survival benefit? Does it matter whether the source of chronic antigenic stress is CMV or can other antigens do the same? The next meeting at the Jenner Institute may already be able to provide some of the answers to these important questions which have serious implications for public health issues in increasingly elderly populations.
==== Refs
Wallace DL Zhang Y Ghattas H Worth A Irvine A Bennett AR Griffin GE Beverley PC Tough DF Macallan DC Direct measurement of T cell subset kinetics in vivo in elderly men and women J Immunol 2004 173 1787 1794 15265909
Haynes L The effect of aging on cognate function and development of immune memory Curr Opin Immunol 2005 17 476 9 16054352
Pawelec G Akbar A Caruso C Solana R Grubeck-Loebenstein B Wikby A Human immunosenescence: is it infectious? Immunol Rev 2005 205 257 68 15882359 10.1111/j.0105-2896.2005.00271.x
Cook M Briggs D Craddock C Mahendra P Milligan D Fegan C Darbyshire P Lawson S Boxall E Moss P Donor KIR genotype has a major influence on the rate of cytomegalovirus reactivation following T-cell replete stem cell transplantation Blood 2005 Oct 20
Lang A Nikolich-Zugich J Development and migration of protective CD8+ T cells into the nervous system following ocular herpes simplex virus-1 infection J Immunol 2005 174 2919 25 15728503
Pawelec G Ouyang Q Colonna-Romano G Candore G Lio D Caruso C Is human immunosenescence clinically relevant? Looking for 'immunological risk phenotypes Trends Immunol 2002 23 330 332 12103341 10.1016/S1471-4906(02)02255-X
Wikby A Ferguson F Forsey R Thompson J Strindhall J Lofgren S Nilsson BO Ernerudh J Pawelec G Johansson B An immune risk phenotype, cognitive impairment, and survival in very late life: impact of allostatic load in Swedish octogenarian and nonagenarian humans J Gerontol A Biol Sci Med Sci 2005 60 556 65 15972602
Khan N Hislop A Gudgeon N Cobbold M Khanna R Nayak L Rickinson AB Moss PA Herpesvirus-specific CD8 T cell immunity in old age: cytomegalovirus impairs the response to a coresident EBV infection J Immunol 2004 173 7481 9 15585874
Akbar AN Fletcher JM Memory T cell homeostasis and senescence during aging Curr Opin Immunol 2005 17 480 5 16098721
Schwaiger S Wolf AM Robatscher P Jenewein B Grubeck-Loebenstein B IL-4-producing CD8+ T cells with a CD62L++(bright) phenotype accumulate in a subgroup of older adults and are associated with the maintenance of intact humoral immunity in old age J Immunol 2003 170 613 9 12496450
Haynes L Eaton SM The effect of age on the cognate function of CD4+ T cells Immunol Rev 2005 205 220 8 15882356 10.1111/j.0105-2896.2005.00255.x
Aspinall R Ageing and the immune system in vivo Commentary on the 16th session of British Society for Immunology Annual Congress, Harrogate, December 2004 Immun Ageing 2005 2 5
Min H Montecino-Rodriguez E Dorshkind K Reduction in the developmental potential of intrathymic T cell progenitors with age J Immunol 2004 173 245 50 15210781
Nikolich-Zugich J Messaoudi I Mice and flies and monkeys too: Caloric restriction rejuvenates the aging immune system of non-human primates Exp Gerontol 2005 Aug 5 16087306
|
16307681
|
PMC1315345
|
CC BY
|
2021-01-04 16:36:31
|
no
|
Immun Ageing. 2005 Nov 24; 2:16
|
utf-8
|
Immun Ageing
| 2,005 |
10.1186/1742-4933-2-16
|
oa_comm
|
==== Front
J Negat Results BiomedJournal of Negative Results in Biomedicine1477-5751BioMed Central London 1477-5751-4-81631862910.1186/1477-5751-4-8ResearchOn the genetic involvement of apoptosis-related genes in Crohn's disease as revealed by an extended association screen using 245 markers: no evidence for new predisposing factors Wagenleiter Sonja EN [email protected] Peter [email protected] Denis A [email protected] Larissa [email protected] Thomas [email protected] Wolfram [email protected] Jörg T [email protected] Department of Human Genetics, Ruhr-University, Bochum, Germany2 Institute for Clinical Molecular Biology, University Schleswig-Holstein, Kiel, Germany3 Department of Gastroenterology, University Hospital Bergmannsheil, Bochum, Germany2005 30 11 2005 4 8 8 7 7 2005 30 11 2005 Copyright © 2005 Wagenleiter et al; licensee BioMed Central Ltd.2005Wagenleiter 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.
Crohn's disease (CD) presents as an inflammatory barrier disease with characteristic destructive processes in the intestinal wall. Although the pathomechanisms of CD are still not exactly understood, there is evidence that, in addition to e.g. bacterial colonisation, genetic predisposition contributes to the development of CD. In order to search for predisposing genetic factors we scrutinised 245 microsatellite markers in a population-based linkage mapping study. These microsatellites cover gene loci the encoded protein of which take part in the regulation of apoptosis and (innate) immune processes. Respective loci contribute to the activation/suppression of apoptosis, are involved in signal transduction and cell cycle regulators or they belong to the tumor necrosis factor superfamily, caspase related genes or the BCL2 family. Furthermore, several cytokines as well as chemokines were included. The approach is based on three steps: analyzing pooled DNAs of patients and controls, verification of significantly differing microsatellite markers by genotyping individual DNA samples and, finally, additional reinvestigation of the respective gene in the region covered by the associated microsatellite by analysing single-nucleotide polymorphisms (SNPs). Using this step-wise process we were unable to demonstrate evidence for genetic predisposition of the chosen apoptosis- and immunity-related genes with respect to susceptibility for CD.
==== Body
Introduction
Crohn's disease (CD) is a chronic inflammatory disorder characterized by destructive processes in the intestinal wall. Interactions between genetic and environmental factors potentially lead to an imbalance between the luminal bacterial flora, and the innate as well as the adaptive immune systems [1,2]. Epidemiological and genome wide studies have lead to the identification of factors establishing genetic involvement in CD [1,3,4]. Despite of fundamental findings, namely the variation in the CARD15 receptor and their association with CD, the causative instances regulating the exaggerated mucosal response remained elusive. The proposed pathomechanisms of CD are manifold. The dysregulated response of the innate immune system is supposed to present a crucial step in the pathogenesis of CD [5]. This fact has been confirmed genetically by several CD associations of genes such as CD14, TLR4 and in some instances the interaction of their variations with CARD15 [6,7]. In regard to the polarized T helper (Th) response, the adaptive immune system appears affected in CD as well [8-10]. Moreover, several studies implicated a role of programmed cell death in CD [11-15]. Apoptosis mediates 'self-tolerance', the elimination of autoreactive immune compartments. In addition, the thoroughly controlled termination of a physiological immune response is due to the process of programmed cell death. In CD mucosal T cells show less susceptibility to apoptosis [16]. In this context TNFα protein exerts multiple physiological effects, and anti-TNFα therapeutic strategies (e.g. infliximab) are effective in (maintaining) remission of CD [17]. In several studies it has been revealed that treatment of CD patients with infliximab leads to an activation of T cells rendering them susceptible for apoptosis [18,19]. Interestingly, the effect of this treatment may not be due to neutralisation of soluble TNFα (and its binding to the TNFRs), but rather it may be caused by its affinity to membrane-bound TNFα putatively changing the ratio of anti- and pro-apoptotic mediators towards induction of apoptosis [18,20]. Although the mechanisms of the causal role of T cells responses in CD remain to be determined in detail, there is substantial clinical evidence that suggests a role for uncontrolled activated T lymphocytes in the pathogenic process of CD [21-24]. Nevertheless, it is uncertain, whether a genetic basis for a decreased activation/apoptosis of T lymphocytes in CD patients exists, and whether increased anti-apoptotic markers, found in T cells of these patients are due to the mucosal inflammation, secondarily [18].
In such a complex situation we used extended association screening (EAS) with markers representing 245 apoptosis- and (innate) immunity-related genes. The majority of the investigated markers have been successfully utilized in respective studies before [25,26]. Our population based linkage mapping comprises a 3-stage analysis with pooled DNA in the initial phase and subsequently individual genotyping. In order to confirm such results, several tagging SNPs of the adjacent gene represented by the marker were analysed. Here, we investigated the role of distinct biological pathways for the susceptibility of CD.
Materials and methods
Patients
One hundred and fifty eight well-characterized patients with a clinical, endoscopical and histological diagnosis of CD were included. This patient cohort has been reported before [27,28]. All patients were of German origin and the diagnosis of CD was adjusted according to the diagnostic criteria of the European Community Workshop on Inflammatory Bowel Diseases (IBD). As controls a group of healthy northern German (NoG) and western German (WeG) origin were analysed. In the initial step a group of ~100 NoG individuals were used. In order to exclude population stratification, genotyping of chosen SNPs was performed in 180–460 NoG and WeG individuals.
Pooling of DNA
The DNA concentration from each individual of the patient and control cohorts was quantified by spectrophotometry, carried out four times, and then diluted accordingly to 100 ng/μl. In a second step the DNA was diluted to a concentration of 65 ng/μl and once more measured by spectrophotometry. Finally, DNA diluted to 50 ng/μl was adjusted to a final amount of 1000 ng for each individual in a pool of 50 persons. In the initial stage, marker analyses were performed with two patient and two control subpools, respectively.
Tailed primer PCR
Tailed primer PCR was performed as described before [25]: An 18 bp-tail was added to each sense oligonucleotide. PCR reaction included three oligonulceotides, two of which were target specific. The third one consists of the same sequence as the abovementioned tail that was additionally fluorescence-labelled.
Microsatellite markers
Intragenic microsatellite or markers located in the immediate vicinity (<50 kb) of the specific gene were included. Information on the oligonucleotide sequences and location of markers are given at the website (Additional file 1; see also Tab. 1). As reported before, only markers with equal "intra-subgroup" allele distributions with ≥ 2 alleles were considered in subsequent analyses [25]. Significantly associated markers were genotyped individually in order to exclude false-positive results due to possible pooling artefacts. All in all, 245 microsatellite markers representing distinct genes were analysed on an ABI377 slab-gel system (Applied Biosystems, Darmstadt, Germany).
Table 1 Genes investigated for CD association as represented by an intra- or juxtagenic microsatellite marker (for additional information see URL: )
apoptosis related REQ TNFSF12 CTLA4 Casp10 IL4
RNF7 TNFSF14 DAP Casp14 IL4R
SMAC TNFSF15 DAPK1 Casp2 IL6
AIF TIAF1 TNFSF18 FADD Casp3 IL8
APR3 TIAL1 TNFSF4 IKBKG Casp4 IRF1
BCLG TP73 TNFSF5 MADD Casp5 NRG1B
BFAR VDR TNFSF6 MAP2K6 Casp6 PRL
CIDEB TNFSF7 MAP3K14 Casp7 PRLR
CYBB Bcl2 related TNFSF8 MAP3K5 Casp8
CYP51 TNFSF9 MAP4K4 CASP8AP2 chromosome 6
DAD1 BCL2A1 TOSO NFKB1 Casp9
DAP3 Bag1 NFKB2 No.1
DATF1 BAK innate immunity NSMAF apoptosis suppressor No.4
DAXX BAX PAWR No.5
DEDD BCL2 BPI PIAS3 No.6
DHCR24 BCL2L1 CD14 PTEN API5 No.7
EIF4G2 BCL2L11 CD5L RARB BIRC1 No.8
FASTK BCL2L13 DEFB119/ DEFB121 RIPK1 BIRC2 D6S1014
FLIP BID DEFB127 RIPK2 BIRC3 D6S1959
FRZB BIK HBD1 RIPK3 BIRC4 D6S273
GSK3B BNIP3L IFNB1 RXRB BIRC6
GSR MCL1 LY64 STK17A BIRC8 others
GZMA LY86 STK17B
GZMB TNF superfamily LY96 TANK cytokine chemokines BPHL/TUBB
HLCS NCF1 TRADD TAPBPR
NME3 LTB (TNFSF3) NCF4 Traf3 VEGF
NOL3 LTBR (TNFRSF3) PGLYRP Traf4 AXL LGALS3
NOS1 TNFa PLA2G4A Traf5 CSF1R BDNF
NOS2A TNFRSF10A PLUNC Traf6 CSF2 NGFB
NOX1 TNFRSF10B SerpinA1 CSF2RB NGFR
NOX3 TNFRSF10C SerpinB1 cell cycle regulators CSF3 TrkC
NOX4 TNFRSF10D SFTPA1 Dtk
P2RX1 TNFRSF11A SLPI CCND2 erbB3 positive control
P53AIP1 TNFRSF11B STAT3 CDC2 GAS1 CARD15
PDCD10 TNFRSF12 TGFB1 CDKN1A IGF1
PDCD2 TNFRSF17 TLR1 CDKN2A IGF2R
PDCD5 TNFRSF18 TLR2 PAK1B IL10
PDCD6 TNFRSF19 TLR3 RbAp48 IL10RA
PDCD6IP TNFRSF19L TLR4 Rb2/p130 IL10RB
PDCD8 TNFRSF1A TLR5 RBP1 IL11RA
PLA2G10 TNFRSF1B TLR7 RBP2 IL12A
PLA2G1B TNFRSF21 TLR8 RBQ-1 IL12B
PLA2G6 TNFRSF4 TLR9 RBQ-3 IL12RB2
PTGS1 TNFRSF5 TLR10 TP53 IL13RA2
REQ TNFRSF6 (FAS) TP53INP1 IL18
RNF7 TNFRSF6B signal transduction IL18R
SMAC TNFRSF7 caspase related IL1RL1
TIAF1 TNFRSF8 Traf1 IL1B
TIAL1 TNFRSF9 BCL10 ADPRT IL2
TP73 TNFSF10 CHUK CARD4 IL24
VDR TNFSF11 CRADD Casp1 IL2RA
Statistics for initial comparisons of allele frequencies
Raw data from ABI377 profiles were analysed by the Genotyper software (ABI) producing a marker specific allele image profile (AIP) which includes different heights of peaks reflecting the allele frequencies. In order to test differences of the AIPs between CD patients and the controls, all peak heights were summarized for each pool and set to 100 %. The total allele count for each distinct allele was then estimated. Thereupon, the AIPs of the case and control pools were compared statistically by means of contingency tables. Hence, P values are nominal and approximate, because estimated rather than observed counts were used for allele frequencies. The significance level was set at p = 0.05. In order to focus the statistics on major alleles, all minor alleles with a frequency of less than 0.05 were summarized to a virtual allele. Subsequently, a second statistical analysis by means of contingency tables was undertaken. A third step for statistical testing each allele individually was accomplished (and the summation of all other marker alleles), whereby the respective value of the patient group was compared with those of the controls and subsequent χ2 analyses. Despite of evidence that correction for multiple comparisons might eliminate 'real positive' results [26], Q value correction was performed with a cut off of 5% for the initial screening procedure [29].
Nevertheless, for selecting markers for further investigations, non-corrected P values were simply ranked according to their evidence for association including all performed statistical procedures.
Individual genotyping
Markers with significantly different allele distributions between patients and controls were controlled by genotyping individual DNA samples of patients and controls in order to exclude false-positive results due to pooling artefacts. Individual genotyping was performed by capillary gel electrophoresis by using the BeckmanCoulter CEQ8000 genetic analysis system (Beckman Coulter, Germany). Results were analysed by comparing each microsatellite allele frequency from the CD cohort with the corresponding allele frequency of the control group by χ2 testing and corrected by the number of marker specific alleles according to Bonferroni (see Tab. 2 and URL: ). Hardy-Weinberg equilibrium (HWE) was tested using the Genepop program .
Table 2 P values for microsatellite markers located intragenically or in the immediate vicinity of represented genes after the initial step and individual genotyping.
p values
gene (as represented by the respective marker) after analysis with pooled DNA after summation of alleles beneath 5% after analyses of each single allele (most significant allele) after individual genotyping1 (pc value) after correction by Q-value of pooled data
FLIP 0.2871 0.1936 0.0100 0.0044 (pc > 0.05; c = 9) n.s.
BCL2A1 0.0948 0.0948 0.0275 n.s. n.s.
BAG1 0.2541 0.2541 0.0163 n.s. n.s.
BPI 0.0011 0.0011 0.0031 n.s. n.s.
erbB3 0.0760 0.0932 0.0100 n.s. n.s.
TP73 0.5928 0.3535 0.0302 n.s. n.s.
TLR9 0.3004 0.3004 0.0300 n.s. n.s.
TNFRSF17 0.0012 0.0014 0.0014 0.0012 (pc < 0.01; c = 6) n.s.
CARD15 0.0083 0.0247 0.0054 0.0050 (pc < 0.04; c = 7) n.s.
P values were generated using three different procedures as described in the methods' section. Briefly, data were analysed by means of contingency tables, initially comparing allele distributions represented by the AIF (after analyses with pooled DNA), then after summation of alleles < 5% in order to focus on the major alleles and, finally, after comparison of each single allele between the control and patient cohorts. For analysing the results of the individual genotyping χ2 testing was utilised.
1Genotyping was performed with the same individuals used in the pooling procedure, and, when remaining significant, further individuals were added to the analyses (FLIP: CD = 134, controls = 150; TNFRSF17: CD = 147, controls = 135; CARD15: CD = 144, controls = 165).
SNP genotyping
SNPs in genes as represented by significantly associated markers after individual genotyping were investigated by analysis of restriction fragment length polymorphisms (RFLP; see Tab. 3). As the marker representing the TNFRSF17 gene is located in ~1 MBp distance to the MHC class II transactivator (MHC2TA) gene, a functional variation (rs3087456, [30]) of MHC2TA was genotyped by RFLP analyses in 147 CD patients and 463 healthy controls from the abovementioned control populations (see Tab. 3). The results were evaluated by means of χ2 -and HWE testing. Linkage disequilibrium (LD) between the marker alleles and the polymorphism was calculated by the Genepop program.
Table 3 Investigated SNPs in genes as represented by significantly differing microsatellites of the individual genotyping step.
Gene rs# Allele 01/02 Oligonucleotides (sense/antisense) RE TM (°C) Allele: fragment length (bp)
FLIP Rs7583529 A/C GGTGATTATTCGGACCCCA/AACTACAGATCCCGTGTGGAG TseI 57 01: 155
02: 103/52
Rs2041765 T/C GAACAAGGAGAGAACCTGGAC/GAGCTGGAAGGCACAGTACA MboII 56 01: 309
02: 188/121
TNFRSF17 Rs3743591 A/G ATAAGCAGTTTCTGTTTCAGATGT/CTCTACAAGAATTCCAGAGCA BceAI 55 01: 223
02: 147/76
Rs11570139 C/T GCCCTGATATTTACACCCTGT/CAGCCATCTGCAACATGAT CaiI 54 01: 269
02: 161/108
Rs373496 T/C AGGAACTGAAACTCACAATAGC/CAGCTCATTATCTGTCTGATGTT AluI 55 01: 247
02: 100/90/54/3
MHC2TA Rs3087456 G/A * 1 GTGAAATTAATTTCAGAGCTGT/CTCAGCTTCCCCAAGGAT BfmI 58 01: 268
02: 231/37
Analyses were performed by using the RFLP method. The table depicts information on the used SNPs as well as RFLP/PCR conditions. * 1 A 5'-tail was added to the mismatch (bold letter) sense primer (5'-CATCGCTGATTCGCACAT-3'). PCR was performed with a third oligonucleotide with the equal sequence as the tail. RE: restriction enzyme; TM: melting temperature (used for annealing in PCR).
Results
Initial step
Microsatellites representing 245 genes involved in apoptosis regulation (see Tab. 1) were investigated by using EAS. None of the markers presented with significant intra-subgroup differences confirming the homogeneity of the pools. The statistical evaluation of the microsatellite frequencies in the CD patient and the control cohorts revealed 9 significantly different allele distributions of intra- or juxtagenic markers for FLIP, BCL2A1, BAG1, BPI, erbB3, TP73, TLR9, TNFRSF17 and CARD15 (summarized in Tab. 2).
Individual genotyping
Individual genotyping confirmed significant P values only for the 3 markers FLIP (p = 0.0044, pc > 0.05, in HWE), TNFRSF17 (p = 0.0012, pc < 0.01, in HWE) and the positive control CARD15 (p = 0.0050, pc < 0.04, in HWE). The additional associations for the other markers were rejected (see Tab. 2 and Additional file 1). There were no differences analysing CARD15+ and CARD15- patients.
SNP genotyping
SNP markers (Tab. 3) were genotyped located in the respective genes in the vicinity of the microsatellites representing TNFRSF17 and FLIP. Thus, SNPs were analyzed spread across the genes representing haplotypes as predisposed by the 'LD Select' method reported before [31]. RFLP analyses did not reveal any association of the selected SNPs, neither by comparing the CARD15+ nor the CARD15- patients with the control group.
Comparison of TNFRSF17 microsatellite alleles
The genotypes of the TNFRSF17 microsatellite alleles were compared between the patient and control cohorts. Analyses revealed evidence either for a predisposing (allele 3) and a protective allele (2) or linkage between these alleles and the marker alleles, respectively.
Genotypes including allele 2 are overrepresented in the control cohort, whereas those with the apparently predisposing allele 3 are more frequent in the CD cohort, thus confirming the results of individual genotyping (see Fig. 1).
Figure 1 Genotype frequencies of the microsatellite representing the TNFRSF17 gene. Only genotypes with a frequency of > 0.01 are included. Alleles of the respective microsatellite are indicated as numbers in the X-axis according to their length in bp. For example: 1–1 (read from the number below the numerical series and the first number of the numerical series) means homozygous genotype for microsatellite allele number one and 1–4 heterozygous genotype for allele 1 and 4. Genotypes comprising allele 2 are over-represented within the control group (47% vs. 29%; pc = 0.0042 with c = 2), whereas allele 3 genotypes are more frequent in the patient cohort (58% vs. 52% pc = 0.3130; c = 2). Therefore, allele 2 might imply a protective effect and/or allele 3 a predisposing effect on CD. Interestingly, the genotype 2–3 is more prevalent in the control group. This result can be interpreted by a different effect size of allele 2 (↑) as compared to allele 3, or the significant difference of this microsatellite is only due to linkage of allele 2 with a protective factor.
MHC2TA analyses
The analyses of the functionally significant polymorphism rs3087456 revealed a marginal association in our CD patients when allele or genotype frequencies were compared between the combined control (WeG and NoG did not differ in allele frequencies) and the patient cohorts (see Tab. 4). Analyses for LD between TNFRSF17 and MHC2TA alleles, however, did not reveal any significant deviations from equilibrium.
Table 4 Allele and genotype frequencies of the functional MHC2TA polymorphism (rs3087456).
Allele frequencies p value OR (CI) Genotype frequencies p value
CD (n = 147) C 0.32 0.05 1.33 (0.90–2.01) CC 0.08 0.54
T 0.68 CT 0.48 0.06
TT 0.44 0.03
controls (n = 463) C 0.26 CC 0.07
T 0.74 CT 0.39
TT 0.54
OR: odds ratio; CI: 95% confidence interval
Discussion
The pathomechanisms of CD are still not exactly understood, albeit certain CARD15 variations appear especially frequent in CD patients; thus genetic involvement is proven. These genetic predisposition factors, however, are neither sufficient nor explain they the pathogenesis in all CD patients. In this study we present an association screen mainly for apoptosis and immunity related genes by microsatellite markers as investigated in a 3-step approach.
Our initial analyses revealed 9 significantly different allele distributions of intra- or juxtagenic markers for FLIP, BCL2A1, BAG1, BPI, erbB3, TP73, TLR9, TNFRSF17 and CARD15 (see Tab. 2). Yet, after correction by Q-value, none of those markers remained significant. On the other hand, a recent study raised the question, whether the correction for multiple comparisons should be applied at all in EAS [26]. For example, in these analyses a previously significantly associated microsatellite (representing the TNFα gene), which has been used as a positive control such as CARD15, would have been rejected by the correction procedure. Therefore, it remains conceivable that the abovementioned markers represent rather hints for additional predisposing factors/loci with low effect size.
The most promising markers (reflected by a significant p-value) were included in further analyses regardless of the correction procedure. Individual genotyping rejected most markers found to be significantly different in the initial step of our approach and only three markers remained significant representing the TNFRSF17, FLIP, CARD15 genes (Tab. 2). Obviously, pooled and individual genotyping yield somewhat contradictory results. Eight microsatellites revealed significantly differences between the patient and control cohorts after the pooling procedure, whereas individual genotyping results in the confirmation of 'only' 2 markers. These conspicuous differences might be due to several artefacts caused by analyses with pooled DNA. For example, a typical artefact is the length-dependent amplification of short alleles or the presence of null-alleles. Additionally, consistency of the analyses by a slab-gel system might reflect a further hindrance in this subtle procedure. Nevertheless, individual genotyping eliminates false positive results due to pooling artefacts and, in case of significant results, enables the thorough analyses of the marker alleles in detail (see Fig.1). In order to confirm the aforementioned positive results further markers (SNPs, Tab. 3) were genotyped located in the respective genes in the vicinity of the microsatellites representing TNFRSF17 and FLIP. Yet, RFLP analyses did not reveal any association of the selected SNPs and, therefore, the microsatellite data were not confirmed. On the other hand, these SNPs might not represent regions properly that encompass regulatory elements.
In some instances, the LD of distinct microsatellite alleles covers long genetic distances, thus further gene variations might be in linkage with these alleles. Since the significantly associated 'TNFRSF17' marker is located at the IBD8 region with 1MBp distance to the major histocompatibility class (MHC) II transactivator (MHCIITA), a previously reported functional variation of the MHC2TA gene was analysed (see Tab. 4; [30]). MHC2TA regulates the expression of human leukocyte antigen (HLA) genes regulating the adaptive immune system by presenting antigens to CD4+ T cells, thereby re-activating these cells. The HLA region has been implicated in IBD [32]. In addition to the localisation of MHC2TA at IBD8 and the associated marker in the adjacent region, the putative biological relevance of the functional rs3087456 polymorphism for CD motivated us to genotype this variation. The analyses did reveal a marginal association in our CD patients when allele or genotype frequencies were compared between the combined control and patient cohorts (see Tab. 4). Yet there was no evidence for LD between TNFRSF17 and MHC2TA alleles. In order to validate these data further patient cohorts comprising more individuals must be scrutinised. In addition, other genes that might be linked with the 'TNFRSF17' marker must be analysed (at least 15 RefSeq genes in the region are encompassed by the microsatellite marker and MHC2TA).
In conclusion, this study did not reveal overt evidence for CD predisposition factors in apoptotic (and immune) pathways. Certainly, our approach depends on the LD between the investigated microsatellites and putative predisposing or protective alleles, depending on functional relevance to the disease. Thus, in some instances microsatellites might not be entirely representative for the adjacent genes. Furthermore, the investigated genes only cover part of the factors which coordinate programmed cell death. Yet, future information about haplotype blocks may facilitate more far-fetched interpretations of our analyses.
Supplementary Material
Additional File 1
This file provides detailed information on the sequence of used oligonucleotides, represented gene, marker distance to gene and kind of nucleotide repeat (di, tri, etc.). Furthermore, the file includes graphical information on individually genotyped microsatellites markers with significant differences in allele distributions.
Click here for file
==== Refs
Russel MG Pastoor CJ Janssen KM van Deursen CT Muris JW van Wijlick EH Stockbrugger RW Familial aggregation of inflammatory bowel disease: a population-based study in South Limburg, The Netherlands. The South Limburg IBD Study Group Scand J Gastroenterol Suppl 1997 223 88 91 9200312
Hugot JP Thomas G Genome-wide scanning in inflammatory bowel diseases Dig Dis 1998 16 364 369 10207223 10.1159/000016893
Hugot JP Chamaillard M Zouali H Lesage S Cezard JP Belaiche J Almer S Tysk C O'Morain CA Gassull M Association of NOD2 leucine-rich repeat variants with susceptibility to Crohn's disease Nature 2001 411 599 603 11385576 10.1038/35079107
Ogura Y Bonen DK Inohara N Nicolae DL Chen FF Ramos R Britton H Moran T Karaliuskas R Duerr RH A frameshift mutation in NOD2 associated with susceptibility to Crohn's disease Nature 2001 411 603 606 11385577 10.1038/35079114
Rogler G Update in inflammatory bowel disease pathogenesis Curr Opin Gastroenterol 2004 20 311 317 15703658 10.1097/00001574-200407000-00003
Franchimont D Vermeire S El Housni H Pierik M Van Steen K Gustot T Quertinmont E Abramowicz M Van Gossum A Deviere J Deficient host-bacteria interactions in inflammatory bowel disease? The toll-like receptor (TLR)-4 Asp299gly polymorphism is associated with Crohn's disease and ulcerative colitis Gut 2004 53 987 992 15194649 10.1136/gut.2003.030205
Klein W Tromm A Griga T Folwaczny C Hocke M Eitner K Marx M Duerig N Epplen JT Interaction of polymorphisms in the CARD15 and CD14 genes in patients with Crohn disease Scand J Gastroenterol 2003 38 834 836 12940436 10.1080/00365520310003147
Pallone F Blanco Gdel V Vavassori P Monteleone I Fina D Monteleone G Genetic and pathogenetic insights into inflammatory bowel disease Curr Gastroenterol Rep 2003 5 487 492 14602058
Okazawa A Kanai T Watanabe M Yamazaki M Inoue N Ikeda M Kurimoto M Ishii H Hibi T Th1-mediated intestinal inflammation in Crohn's disease may be induced by activation of lamina propria lymphocytes through synergistic stimulation of interleukin-12 and interleukin-18 without T cell receptor engagement Am J Gastroenterol 2002 97 3108 3117 12492197 10.1111/j.1572-0241.2002.07107.x
Singh UP Singh S Iqbal N Weaver CT McGhee JR Lillard JW Jr IFN-gamma-inducible chemokines enhance adaptive immunity and colitis J Interferon Cytokine Res 2003 23 591 600 14585199 10.1089/107999003322485099
Souza HS Tortori CJ Castelo-Branco MT Carvalho AT Margallo VS Delgado CF Dines I Elia CC Apoptosis in the intestinal mucosa of patients with inflammatory bowel disease: evidence of altered expression of FasL and perforin cytotoxic pathways Int J Colorectal Dis 2005 20 277 286 15503066 10.1007/s00384-004-0639-8
Xia B Yu YH Guo QS Li XY Jiang L Li J Association of Fas-670 gene polymorphism with inflammatory bowel disease in Chinese patients World J Gastroenterol 2005 11 415 417 15637757
Peppelenbosch MP van Deventer SJ T cell apoptosis and inflammatory bowel disease Gut 2004 53 1556 1558 15479669 10.1136/gut.2004.040824
Brannigan AE O'Connell PR Hurley H O'Neill A Brady HR Fitzpatrick JM Watson RW Neutrophil apoptosis is delayed in patients with inflammatory bowel disease Shock 2000 13 361 366 10807010
Seidelin JB Nielsen OH [Apoptosis in chronic inflammatory bowel disease. The importance for pathogenesis and treatment] Ugeskr Laeger 2003 165 790 792 12625119
Ina K Itoh J Fukushima K Kusugami K Yamaguchi T Kyokane K Imada A Binion DG Musso A West GA Resistance of Crohn's disease T cells to multiple apoptotic signals is associated with a Bcl-2/Bax mucosal imbalance J Immunol 1999 163 1081 1090 10395708
Hommes DW van Deventer SJ Targeting tumor necrosis factor-alpha in inflammatory bowel disease: why, how, and when? Curr Opin Gastroenterol 2003 19 350 357 15703576 10.1097/00001574-200307000-00005
ten Hove T van Montfrans C Peppelenbosch MP van Deventer SJ Infliximab treatment induces apoptosis of lamina propria T lymphocytes in Crohn's disease Gut 2002 50 206 211 11788561 10.1136/gut.50.2.206
Van den Brande JM Braat H van den Brink GR Versteeg HH Bauer CA Hoedemaeker I van Montfrans C Hommes DW Peppelenbosch MP van Deventer SJ Infliximab but not etanercept induces apoptosis in lamina propria T-lymphocytes from patients with Crohn's disease Gastroenterology 2003 124 1774 1785 12806611 10.1016/S0016-5085(03)00382-2
Sands BE Why Do Anti-Tumor Necrosis Factor Antibodies Work in Crohn's Disease? Rev Gastroenterol Disord 2004 4 S10 S17 15580148
Sturm A Leite AZ Danese S Krivacic KA West GA Mohr S Jacobberger JW Fiocchi C Divergent cell cycle kinetics underlie the distinct functional capacity of mucosal T cells in Crohn's disease and ulcerative colitis Gut 2004 53 1624 1631 15479683 10.1136/gut.2003.033613
Sartor RB Current concepts of the etiology and pathogenesis of ulcerative colitis and Crohn's disease Gastroenterol Clin North Am 1995 24 475 507 8809232
Doering J Begue B Lentze MJ Rieux-Laucat F Goulet O Schmitz J Cerf-Bensussan N Ruemmele FM Induction of T lymphocyte apoptosis by sulphasalazine in patients with Crohn's disease Gut 2004 53 1632 1638 15479684 10.1136/gut.2003.037911
Elson CO Sartor RB Tennyson GS Riddell RH Experimental models of inflammatory bowel disease Gastroenterology 1995 109 1344 1367 7557106 10.1016/0016-5085(95)90599-5
Jagiello P Gencik M Arning L Wieczorek S Kunstmann E Csernok E Gross WL Epplen JT New genomic region for Wegener's granulomatosis as revealed by an extended association screen with 202 apoptosis-related genes Hum Genet 2004 114 468 477 14968360 10.1007/s00439-004-1092-z
Wieczorek S Jagiello P Arning L Dahmen N Epplen JT Screening for candidate gene regions in narcolepsy using a microsatellite based approach and pooled DNA J Mol Med 2004 82 696 705 15309313 10.1007/s00109-004-0569-5
Klein W Tromm A Griga T Fricke H Folwaczny C Hocke M Eitner K Marx M Duerig N Epplen JT A polymorphism in the CD14 gene is associated with Crohn disease Scand J Gastroenterol 2002 37 189 191 11843056 10.1080/003655202753416867
Klein W Tromm A Griga T Fricke H Folwaczny C Hocke M Eitner K Marx M Duerig N Epplen JT Interleukin-4 and interleukin-4 receptor gene polymorphisms in inflammatory bowel diseases Genes Immun 2001 2 287 289 11528525 10.1038/sj.gene.6363779
Storey JD Tibshirani R Statistical significance for genomewide studies Proc Natl Acad Sci U S A 2003 100 9440 9445 12883005 10.1073/pnas.1530509100
Swanberg M Lidman O Padyukov L Eriksson P Akesson E Jagodic M Lobell A Khademi M Borjesson O Lindgren CM MHC2TA is associated with differential MHC molecule expression and susceptibility to rheumatoid arthritis, multiple sclerosis and myocardial infarction Nat Genet 2005 37 486 494 15821736 10.1038/ng1544
Carlson CS Eberle MA Rieder MJ Yi Q Kruglyak L Nickerson DA Selecting a maximally informative set of single-nucleotide polymorphisms for association analyses using linkage disequilibrium Am J Hum Genet 2004 74 106 120 14681826 10.1086/381000
Yap LM Ahmad T Jewell DP The contribution of HLA genes to IBD susceptibility and phenotype Best Pract Res Clin Gastroenterol 2004 18 577 596 15157829 10.1016/j.bpg.2004.01.003
|
16318629
|
PMC1315346
|
CC BY
|
2021-01-04 16:37:38
|
no
|
J Negat Results Biomed. 2005 Nov 30; 4:8
|
utf-8
|
J Negat Results Biomed
| 2,005 |
10.1186/1477-5751-4-8
|
oa_comm
|
==== Front
Thromb JThrombosis Journal1477-9560BioMed Central London 1477-9560-3-191631646810.1186/1477-9560-3-19Case ReportLarge mobile thrombus in non-atherosclerotic thoracic aorta as the source of peripheral arterial embolism Malyar Nasser M [email protected] Rolf A [email protected] Zoran [email protected] Raimund [email protected] Department of Cardiology, West German Heart Centre, University, Duisburg-Essen, Germany2 Department of Angiology, University, Duisburg-Essen, Germany2005 29 11 2005 3 19 19 7 10 2005 29 11 2005 Copyright © 2005 Malyar et al; licensee BioMed Central Ltd.2005Malyar 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.
The presence of thrombi in the atherosclerotic and/or aneurysmatic aorta with peripheral arterial embolism is a common scenario. Thrombus formation in a morphologically normal aorta, however, is a rare event. A 50 years old woman was admitted to the mergency department for pain, coldness, and anesthesia in the the left foot. She had a 25 years history of cigarette smoking, a history of postmenopausal hormone replacement therapy (HRT), hypercholesterolemia and hyperfibrinogenemia. An extensive serologic survey for hypercoagulability, including antiphospholipid antibodies, and vasculitis disorders was negative. Transesophageal echocardiography revealed a large, pedunculated and hypermobile thrombus attached to the aortic wall 5 cm distal of the left subclavian artery. The patient was admitted to the surgery department, where a 15 cm long fresh, parietal thrombus could be removed from the aorta showing no macroscopic wall lesions or any other morphologic abnormalities.
This case report demonstrates the possibility of evolving a large, pedunculated thrombus in a morphologically intact aorta in a postmenopausal woman with thrombogenic conditions such as hyperfibrinogenemia, hypercholesterolemia, smoking and HRT. For these patients, profiling the individual risk and weighing the benefits against the potential risks is warranted before prescribing HRT.
==== Body
Background
The cardiac cavities are in >85 % of cases the primary source of peripheral arterial embolism (PAE). Due to new and sophisticated imaging techniques in recent years, such as spiral computed tomography (CT) scan, magnetic resonance imaging (MRI) and transesophageal echography (TEE), noncardiac sources of PAE has been detected with increasing frequency. Among these non-cardiac sources, the aorta has been reported in up to 5 % of cases to be the origin of PAE [1]. However, while mural thrombus in the aneurysmatic or atherosclerotic aorta with protruding atheromas may be the source of major arterial embolism [2], emboli originating from a non-atheromatous and non-aneurysmatic aorta is a rare event [3]. We report a case of pedunculated, large and highly mobile thrombus, formed within a morphologically normal descending thoracic aorta, resulting in recurrent peripheral embolism that caused acute ischemia of the lower extremities.
Case report, diagnostic workup and therapeutic intervention
A 50 years old woman was admitted to the emergency department because of acute pain, coldness, and anesthesia in the left foot. Pulses at the left popliteal, tibialis posterior and at the dorsalis pedis artery were not palpable. For cardiovascular and prothrombotic risk factors, she had a 25 years history of cigarette smoking (30–35 cigarettes/day), a history of 16 months of postmenopausal hormone replacement therapy (HRT) by oral intake of an estrogen-gestagen combination (Presomen, 0.6 mg/d for 16 months) and untreated hypercholesterolemia (5.90 mmol/L, reference: <5.2 mmol/L). She had no previous medical history of arrhythmia, ischemic heart disease, diabetes mellitus, or stroke. Duplex sonographic examination of the arterial system of the limbs revealed the acute thromboembolic occlusion of the left popliteal artery. Diagnostic work up was initiated to determine the source of the embolism. The resting ECG and a holter monitoring for 24 h disclosed normal sinus rhythm without any pathologic findings. The transthoracic echocardiographic examination (TTE) showed no structural or functional cardiac abnormalities. Subsequently, a TEE was performed for evaluation of the cardiac cavities and of the thoracic aorta, revealing a large, pedunculated and hypermobile thrombus (Figure 1, Panl A and B) attached to a non-atherosclerotic aortic wall caudal of the left subclavian artery. CT scans of the thoraco-abdominal aorta (Figure 2) confirmed the presence of the thrombus seen in TEE. The entire thoracic aorta had normal dimensions with no visible signs of atherosclerosis. For biochemical laboratory parameters, thrombocytes (530*109/L, reference: <410*109/L), fibrinogen (5.5 g/L, normal: <3.50 g/L) and total Cholesterol (5.90 mmol/L, reference: <5.2 mmol/L) were elevated. All other parameters including glucose (5.8 mmol/L) and homocysteine (8.6 μmol/L) levels were within the reference limit. An extensive serologic survey for hypercoagulability, including antiphospholipid antibodies, and vasculitis disorders was negative.
Figure 1 Transesophageal echocardiographic cross-sectional (Panl A) and longitudinal (Panel B) images of the descending thoracic aorta showing the highly mobile, floating thrombus. The original lumen of the aorta (21 mm) is reduced by the thrombus to a circumferential patent lumen of 4–6 mm.
Figure 2 Thoraco-abdominal CT-scan image showing the thrombus in the thoracic aorta (arrow).
Because of the high risk for recurrent peripheral embolization due to the size and the hypermobility of the thrombus, the patient was admitted to the vascular surgery unit, where a 15 cm long thrombus in the thoracic aorta with its origin 5 cm distal of the left subclavian artery could be removed. During surgery no macroscopical signs of intimal lesions of the aorta could be detected. Histological evaluation of the thrombus revealed a red, parietal thrombus without any evidence of malignancy. The postoperative course was uneventfull and the patient was discharged 10 days after the surgical intervention. The patient was treated with vitamin-k-antagonist phenprocoumon (INR: 2.5–3.5) for secondary prevention of thromboembolism and with 10 mg/d of atorvastatin for hypercholesterolemia additional to cessation of smoking and oral intake of HRT. On this therapeutic regimen, the patient was asymptomatic with no pathologic findings in the imaging workup at six months follow up.
Discussion
In this case report we demonstrate the presence of a large, mobile thrombus in a morphologically normal aorta as the source of PAE. While thrombi in aneurysmatic or atherosclerotic aorta, especially in combination with pro-thrombotic risk factors such as hypercoagulability, antiphospholipid syndrome, disorders of protein C and S and vasculitis, have previously been described, the presence of thrombi in a non-atherosclerotic aorta is a rare event. As the aorta of the patient described in this case report had normal dimensions and was free of any visible atherosclerotic intimal lesions, the additive prothrombotic effects of hypercoagulability, due to increased level of fibrinogen, thrombocytes and cholesterol, and of oral HRT and excessive smoking seem most likely to be the cause of the thrombus formation.
Plasma fibrinogen is an important component of the coagulation cascade, as well as a major determinant of blood viscosity and blood flow. Increasing evidence from epidemiological studies suggests that elevated plasma fibrinogen levels are associated with an increased risk of cardiovascular disorders, including ischemic heart disease, stroke and thromboembolism [4]. The increase in plasma fibrinogen levels may promote a prothrombotic or hypercoagulable state, and may in part explain the risk of stroke and thromboembolism. The independent and close relation between fibrinogen and cardiovascular risk including thromboembolic complications in the peripheral arterial system has been documented in a substantial number of studies. Fibrinogen strongly affects blood coagulation, blood rheology and platelet aggregation. In addition, it has direct effects on the vascular wall and is a prominent acute phase reactant [5,6].
The adverse cardiovascular effects of HRT have been well documented in the Women's Health Initiative (WHI) trial [7-11]. The combined estrogens-progestin arm of the WHI trial [8] demonstrated that some risks, such as for thromboembolism, coronary heart disease, and stroke, arise within the first 1 to 2 years of therapy, while other risks such as the risk for breast cancer appear to increase with longer-term hormone therapy. In our patient, the thromboembolic event occurred 16 months after the initiation of the HRT, which is consistent with the observation in the WHI study. The recently published clinical recommendations and guidelines[12] by the US Preventive Services Task Force (USPSTF) on hormone therapy for the prevention of chronic conditions in postmenopausal women consider the harmful effects of combined estrogens and progestin likely to exceed the benefits of chronic disease prevention. Therefore, it is recommended that the decision of HRT for postmenopausal complaints should be based on weighing the possible harm arising from the individual risk factors and the benefit.
The combination of smoking and HRT is a substantial risk factor for cardiovascular complications and a burden for the public health care, regrading the fact that products for HRT in the postmenopause are the second most frequently prescribed drugs in the USA [13,14] and 50 % of women on HRT are smokers [8]. Smoking per se not only increaes the incidence of thrombotic events, it has also been shown to decrease the level of serum estrogens by 50 % [15], thereby reducing, or even completely abolishing, the well-established beneficial effects of estrogen such as reducing hot flushes, prevention of osteoporosis and the positive effect on lipid metabolism [13,14]. Cigarette smoking induces global changes in both peripheral and central vascular function [16]. Some authors postulate that the most important effects of cigarette smoking in promoting atherosclerosis and thromboembolic complications may be endothelial disturbance and fibrin formation [17]. Consequently, cessation of smoking is a crucial preventive step towards minimizing the thromboembolic risks.
Generally, thromboembolic events are associated with advanced age, complex and ulcerated atherosclerotic plaques to which the thrombus is attached. Interestingly, the region around the left subclavian artery seem to be one of the predisposed localization for thrombus formation [18]. However, as this case report in accordance with few previously published studies demonstrate, it can also affect younger patients. In the study by Laperche et al. [18] the mean age of the 23 patients with aortic thrombus was 45 ± 8.4 years, with the youngest patient being 34 years old. In their study, smoking was the leading risk factor (n = 16) followed by hypercholesterolemia (n = 11) and elevated fibrinogen levels (n = 10). In our patient all 3 risk factors were present. In the study by Laperche et al. [18] histopathological examination revealed microscopic features of atherosclerosis limited to the insertion site in all patients undergoing surgery for thrombectomy. Considering the high thrombogenic potential of atherosclerotic plaques [19], it is likely that the thrombus in our patient was attached to an atherosclerotic plaque, even when it was not visible on TEE. Perler et al. [20] reported previously about thromboembolic events originating from a morphologically normal aorta in two young women. Both patients were taking oral steroidal medications, and both patients were heavy cigarette smokers. These cases and a review of the previous literature suggest that the development of aortic mural thrombi, at least in some patients, may not always result from diffuse and advanced atheromas, but may constitute a separate and distinct clinical entity of a premature atherosclerosis.
Detection of mobile thrombus of the thoracic aorta has become increasingly common thanks to routine exploration using TEE after any embolic events. TEE has been shown to detect reliably aortic thrombi [21]. Using high-frequency, multiplane probes, TEE sufficiently allows not only the detection, it also provides such important information as the size, the location and the implantation base of the thrombus. It has previously been demonstrated that these information obtained from TEE correlate well with information obtained from macroscopic examination [18]. The high spatial resolution allows detailed examination of the entire thoracic aorta as well as the intimal surfaces. Additionally, the real-time imaging provides information such as the mobility of the thrombus, which is not possible with CT. Such additional information, however, is helpful for immediate and long-term therapeutic decision making. However, it must be emphasized that the resolution capability of TEE is not sufficient to detect accurately atherosclerotic changes in the aortic wall at the microscopic level, as is possible with intravascular ultrasound.
The abdominal aorta also constitutes a frequent origin for thromboembolism [1,22]. This region, however, is not accessible to TEE. Therefore, optimal approach for a complete diagnostic workup for search of the thromboembolic sources requires the synergistic use of at least two modalities, i.e. TEE (for exploration of the cardiac and thoracic aorta origins) and CT or MRI (for exploration of the abdominal origins). The ubiquitaer availability, the excellent sensitivity and specificity and immediate results make TEE as the preferred and primarily used imaging modality for diagnostic evaluation in such cases.
Several treatments have been used in different forms with variable success for management of aortic thrombus, including anticoagulant therapy alone [18,23], thrombolysis [24], thromboaspiration and surgery [25]. At our institution, we initiate immediate anticoagulant therapy with heparin for 2 weeks after the thrombus is specified. At the end of 2 weeks TEE (alternatively MRI) is performed. If the thrombus shows reduction in size, the heparin therapy is continued until the thrombus resolves. If anticoagulant therapy fails, surgical intervention is considered. Surgical intervention is preferred to anticoagulant regimen in young patients, in the presence of a large hypermobile thrombus (as in this case report) and in patients with recurrent embolic events. Similar therapeutic strategies have also been suggested by others [18,25,26]. Endovascular stent-grafting provides a new minimally invasive therapeutic option in the treatment of symptomatic mobile thoracic aortic thrombus [27], however, its role in the setting of aortic thrombosis regarding the long term outcome is not established yet.
Evidence-based recommendations and guidelines regarding the therapeutic management and the long-term surveillance are lacking. Prospective studies are required to address the issue of evidence based guidelines and optimal approaches for the diagnostic, therapeutic and follow up management of these patients.
Conclusion
Embolic events arising from thrombi within a non-aneurysmatic, non-atherosclerotic aorta is a rare but possible emergency event. Physicians have to be aware of the higher risk for developing thromboembolic complication in women under HRT in presence of additional prothrombotic risk factors such as smoking and hypercoagulability, not only in the venous but also in the arterial vascular system; therefore, profiling the individual risk and weighing the benefits against the potential risks is warranted before prescribing HRT. Transesophageal echocardiography serves as the most appropriate initial imaging modality for diagnostic work up and therapeutic management in patients with peripheral embolic events.
List of abbreviations
CT: computed tomography
HRT: hormone replacement therapy
MRI: magnetic resonance imaging
PAE: peripheral arterial embolism
TTE: transthoracic echocardiography
TEE: transesophageal echocardiography
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
N.M.M.: tacked out the diagnostic workup of the patient, carried out the TTE and TEE examinations and drafted the manuscript; R.A.J: tacked the MRI and CT examinations and contributed to the writing of the manuscript; Z.B.: carried out the therapy evolution of the patient; R.E.: revised the article critically for important intellectual content and have given final approval of the version to be published. All authors read and approved the final manuscript.
Acknowledgements
Written consent was obtained from the patient for publication of the study.
==== Refs
Gagliardi JM Batt M Khodja RH Le bas P Mural thrombus of the aorta Ann Vasc Surg 1988 2 201 204 3191002
Aldrich HR Girardi L Bush HLJ Devereux RB Rosengart TK Recurrent systemic embolization caused by aortic thrombi Ann Thorac Surg 1994 57 466 468 8311614
Panetta T Thompson JE Talkington CM Garrett WV Smith BL Arterial embolectomy: a 34-year experience with 400 cases Surg Clin North Am 1986 66 339 353 3952607
Kamath S Lip GY Fibrinogen: biochemistry, epidemiology and determinants Qjm 2003 96 711 729 14500858 10.1093/qjmed/hcg129
Kerlin B Cooley BC Isermann BH Hernandez I Sood R Zogg M Hendrickson SB Mosesson MW Lord S Weiler H Cause-effect relation between hyperfibrinogenemia and vascular disease Blood 2004 103 1728 1734 14615369 10.1182/blood-2003-08-2886
Yarnell JW Patterson CC Sweetnam PM Lowe GD Haemostatic/inflammatory markers predict 10-year risk of IHD at least as well as lipids: the Caerphilly collaborative studies Eur Heart J 2004 25 1049 1056 15191776 10.1016/j.ehj.2004.04.011
Langer RD Pradhan AD Lewis CE Manson JE Rossouw JE Hendrix SL LaCroix AZ Ridker PM Baseline associations between postmenopausal hormone therapy and inflammatory, haemostatic, and lipid biomarkers of coronary heart disease. The Women's Health Initiative Observational Study Thromb Haemost 2005 93 1108 1116 15968396
Anderson GL Limacher M Assaf AR Bassford T Beresford SA Black H Bonds D Brunner R Brzyski R Caan B Chlebowski R Curb D Gass M Hays J Heiss G Hendrix S Howard BV Hsia J Hubbell A Jackson R Johnson KC Judd H Kotchen JM Kuller L LaCroix AZ Lane D Langer RD Lasser N Lewis CE Manson J Margolis K Ockene J O'Sullivan MJ Phillips L Prentice RL Ritenbaugh C Robbins J Rossouw JE Sarto G Stefanick ML Van Horn L Wactawski-Wende J Wallace R Wassertheil-Smoller S Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women's Health Initiative randomized controlled trial Jama 2004 291 1701 1712 15082697 10.1001/jama.291.14.1775
Wassertheil-Smoller S Hendrix SL Limacher M Heiss G Kooperberg C Baird A Kotchen T Curb JD Black H Rossouw JE Aragaki A Safford M Stein E Laowattana S Mysiw WJ Effect of estrogen plus progestin on stroke in postmenopausal women: the Women's Health Initiative: a randomized trial Jama 2003 289 2673 2684 12771114 10.1001/jama.289.20.2673
Rossouw JE Effect of postmenopausal hormone therapy on cardiovascular risk J Hypertens Suppl 2002 20 S62 5 12183856
Rossouw JE Anderson GL Prentice RL LaCroix AZ Kooperberg C Stefanick ML Jackson RD Beresford SA Howard BV Johnson KC Kotchen JM Ockene J Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results From the Women's Health Initiative randomized controlled trial Jama 2002 288 321 333 12117397 10.1001/jama.288.3.321
Hormone therapy for the prevention of chronic conditions in postmenopausal women: recommendations from the U.S. Preventive Services Task Force Ann Intern Med 2005 142 855 860 15897536
Mueck AO Seeger H Smoking, estradiol metabolism and hormone replacement therapy Arzneimittelforschung 2003 53 1 11 12608008
Mueck AO Seeger H Smoking, estradiol metabolism and hormone replacement therapy Curr Med Chem Cardiovasc Hematol Agents 2005 3 45 54 15638743 10.2174/1568016052773270
Jensen J Christiansen C Rodbro P Cigarette smoking, serum estrogens, and bone loss during hormone-replacement therapy early after menopause N Engl J Med 1985 313 973 975 4047104
Wiesmann F Petersen SE Leeson PM Francis JM Robson MD Wang Q Choudhury R Channon KM Neubauer S Global impairment of brachial, carotid, and aortic vascular function in young smokers: direct quantification by high-resolution magnetic resonance imaging J Am Coll Cardiol 2004 44 2056 2064 15542292 10.1016/j.jacc.2004.08.033
Smith FB Lowe GD Fowkes FG Rumley A Rumley AG Donnan PT Housley E Smoking, haemostatic factors and lipid peroxides in a population case control study of peripheral arterial disease Atherosclerosis 1993 102 155 162 8251001 10.1016/0021-9150(93)90157-P
Laperche T Laurian C Roudaut R Steg PG Mobile thromboses of the aortic arch without aortic debris. A transesophageal echocardiographic finding associated with unexplained arterial embolism. The Filiale Echocardiographie de la Societe Francaise de Cardiologie Circulation 1997 96 288 294 9236447
Fernandez-Ortiz A Badimon JJ Falk E Fuster V Meyer B Mailhac A Weng D Shah PK Badimon L Characterization of the relative thrombogenicity of atherosclerotic plaque components: implications for consequences of plaque rupture J Am Coll Cardiol 1994 23 1562 1569 8195515
Perler BA Kadir S Williams GM Aortic mural thrombus in young women: premature arteriosclerosis or separate clinical entity? Surgery 1991 110 912 916 1948663
Davila-Roman VG Westerhausen D Hopkins WE Sicard GA Barzilai B Transesophageal echocardiography in the detection of cardiovascular sources of peripheral vascular embolism Ann Vasc Surg 1995 9 252 260 7632553 10.1007/BF02135284
Hahn TL Dalsing MC Sawchuk AP Cikrit DF Lalka SG Primary aortic mural thrombus: presentation and treatment Ann Vasc Surg 1999 13 52 59 9878657 10.1007/s100169900220
Dee W Geibel A Kasper W Konstantinides S Just H Mobile thrombi in atherosclerotic lesions of the thoracic aorta: the diagnostic impact of transesophageal echocardiography Am Heart J 1993 126 707 710 8362729 10.1016/0002-8703(93)90424-8
Hausmann D Gulba D Bargheer K Niedermeyer J Comess KA Daniel WG Successful thrombolysis of an aortic-arch thrombus in a patient after mesenteric embolism N Engl J Med 1992 327 500 501 1625748
Choukroun EM Labrousse LM Madonna FP Deville C Mobile thrombus of the thoracic aorta: diagnosis and treatment in 9 cases Ann Vasc Surg 2002 16 714 722 12417931 10.1007/s10016-001-0314-2
Lau LS Blanchard DG Hye RJ Diagnosis and management of patients with peripheral macroemboli from thoracic aortic pathology Ann Vasc Surg 1997 11 348 353 9236989 10.1007/s100169900059
Fueglistaler P Wolff T Guerke L Stierli P Eugster T Endovascular stent graft for symptomatic mobile thrombus of the thoracic aorta J Vasc Surg 2005 42 781 783 16242568 10.1016/j.jvs.2005.05.054
|
16316468
|
PMC1315347
|
CC BY
|
2021-01-04 16:36:24
|
no
|
Thromb J. 2005 Nov 29; 3:19
|
utf-8
|
Thromb J
| 2,005 |
10.1186/1477-9560-3-19
|
oa_comm
|
==== Front
Mol PainMolecular Pain1744-8069BioMed Central London 1744-8069-1-341629724210.1186/1744-8069-1-34CommentaryFeed-forward inhibition: a novel cellular mechanism for the analgesic effect of substance P Wu Long-Jun [email protected] Hui [email protected] Shanelle W [email protected] Megumu [email protected] Min [email protected] Department of Physiology, Faculty of Medicine, University of Toronto, University of Toronto Centre for the Study of Pain, Medical Science Building, 1 King's College Circle, Toronto, Ontario M5S 1A8, Canada2005 18 11 2005 1 34 34 29 9 2005 18 11 2005 Copyright © 2005 Wu et al; licensee BioMed Central Ltd.2005Wu 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.
Substance P (SP) is a neuropeptide well known for its contribution to pain transmission in the spinal cord, however, less is known about the possible modulatory effects of SP. A new study by Gu and colleagues, published in Molecular Pain (2005, 1:20), describes its potential role in feed-forward inhibition in lamina V of the dorsal horn of the spinal cord. This inhibition seems to function through a direct excitation of GABAergic interneurons by substance P released from primary afferent fibers and has a distinct temporal phase of action from the well-described glutamate-dependent feed-forward inhibition. It is believed that through this inhibition, substance P can balance nociceptive output from the spinal cord.
==== Body
The spinal cord dorsal horn is one of the relay stations for nociceptive information transmitted by peripheral sensory afferents. Some of these sensory afferents are substance P (SP) and glutamate-containing Aδ- and C- fibers. Upon noxious stimulation, particularly intense stimulation, tachykinins such as SP and neurokinin A (NKA) are released from primary afferent fibers and excite dorsal horn neurons via activation of the neurokinin-1 and neurokinin-2 receptors (NK1R and NK2R), respectively [1,2]. A series of studies have established a role for SP in the transmission of pain information [3,4]. Mice genetically engineered not to express the precursor of SP [4] and mice that do not express SP's target, the NK1R [3], both display reduced responses to painful stimuli. Despite these promising initial findings, NK1R antagonists have failed to produce analgesia in a variety of clinical pain models [5]. One possible explanation for these inconsistent results is that SP may produce mixed effects in sensory-related transmission and modulation. Indeed, Mohrland and Gebhart reported that an intrathecal injection of SP had antinociceptive effects [6]. Similarly, a study has found an analgesic effect mediated by SP and further suggested that it might be mediated by μ-2 opioid receptors [7]. Other studies demonstrate an interaction between tachykinin and opioid systems, lending support for a role of opioid receptors in SP-mediated antinociception [8].
Although many studies have highlighted the importance of SP in pain transmission, the synaptic mechanisms underlying the antinociceptive effect of SP remain unclear. In a recent study published in Molecular Pain, Gu and colleagues used a combination of electrophysiological, pharmacological, genetic and behavioral techniques in rats and found that SP can modulate inhibitory transmission in lamina V of the spinal cord dorsal horn, thereby exerting an analgesic effect on nociceptive sensory processing [9]. Through this finding, the study provides novel insight into the role of SP in the spinal cord dorsal horn, which has important implications for the therapeutic control of pathological pain.
SP drives glutamate-independent feed-forward inhibition
The neuronal network in the spinal dorsal horn is extremely complex. Within this circuit, inhibitory interneurons containing GABA and/or glycine play important roles in controlling network excitability. Activation of these neurons can initiate feedback inhibition and feed-forward inhibition, which are believed to be critical for the fine tuning of sensory information at the spinal level [10]. SP and glutamate, present in some primary afferent fibers that respond to painful stimuli, can mediate excitatory responses in postsynaptic dorsal horn neurons [1,11]. Although it had been shown that glutamate can bind to receptors on GABA and glycine-releasing interneurons to decrease nociceptive transmission, the possibility that SP could also drive inhibitory activity in the spinal cord dorsal horn had not been explored.
Gu and colleagues stimulated the dorsal root of the spinal cord, which contains primary afferent fibers, with high frequency stimulation to cause the release of both SP and glutamate in the presence of N-methyl-D-aspartate (NMDA) receptors and non-NMDA receptors antagonists. They found a robust and long-lasting increase in both the frequency and amplitude of spontaneous inhibitory postsynaptic currents (sIPSCs). Similar results were obtained with chemical stimulation using capsaicin, which also excites primary afferent fibers to release both glutamate and SP [9]. This increase in sIPSCs when glutamate-mediated excitatory responses were blocked suggests that glutamate-independent feed-forward inhibition exists in the dorsal horn.
Could SP be responsible for glutamate-independent feed-forward inhibition? First, among different neuropeptides such as galanin, neuropeptide Y, somatostatin, calcitonin gene-related peptide and SP tested, only SP enhanced the frequency and amplitude of sIPSCs. These results demonstrate that the effect of SP is selective. Second, application of NK1R antagonists blocked both electrical and chemical stimulation-induced increases of sIPSCs. Third, exogenously applied SP- or stimulation-induced increases of sIPSCs were abolished in NK1R knockout (NK1R-/-) mice. Taken together, these results demonstrate that increases of SP can drive inhibitory activity in the dorsal horn of spinal cord.
To further explore the neuronal mechanisms underlying SP-driven feed-forward inhibition, pharmacological tools were used to show that pertussis toxin-sensitive G proteins such as Gi and Go are involved in SP-driven feed-forward inhibition. Moreover, GIN mice, a strain of transgenic mice that express enhanced green fluorescent protein in GABAergic neurons, were used to show that SP can directly excite some GABAergic neurons by inducing prolonged depolarization, firing of action potentials, and increases of intracellular Ca2+. Since SP had no effect on both miniature inhibitory postsynaptic currents (mIPSC) and evoked inhibitory postsynaptic currents (eIPSC), SP-driven feed-forward inhibition is more likely due to the direct excitation of GABAergic interneurons by SP releasing primary afferent fibers than to action potential-independent modulation of activity within the synapse.
SP-driven feed-forward inhibition is summarized in Figure 1. The intense painful stimulation of primary afferents, mostly Aδ- and C- fibers, induces the release of both glutamate and SP, mainly in lamina I and V. In lamina I, SP release activates projection neurons that relay pain-related information to higher centers in the nervous system to process the different qualities of the stimulus. In lamina V, SP binds pertussis toxin-sensitive Gi/Go-coupled NK1R on inhibitory interneurons, which leads to a cascade of downstream events that may result in changes in the activity of non-selective cation channels or inwardly rectifying potassium channels [12] ultimately leading to GABA and/or glycine release. This could initiate feed-forward inhibition in the projection neurons and probably feed-back inhibition on the primary afferent fibers.
Figure 1 Schematic diagram of SP-driven feed-forward inhibition in lamina V of the spinal cord dorsal horn. Sensory information starts from dorsal root ganglion (DRG) neurons, is relayed by spinal cord dorsal horn neurons and then is projecting to the brain. The intense painful stimulation of primary afferent, mostly Aδ- and C- fibers, induced the release of SP in lamina I and V. On the one hand, SP directly excites projection neurons in laminar I, thereby inducing pronociceptive response. On the other hand, SP in laminar V excites inhibitory interneurons in lamina V, through NK1 receptor (NK1R) and the following signaling pathway involved pertussis toxin-sensitive Gi/Go protein and possible downstream targets Ca2+ or K+ channels. The firing of these interneurons releases GABA and/or glycine, activate GABAA receptor (GABAAR) and/or glycine receptor (GlyR), and initiates feed-forward inhibition in the projection neurons ascending to the brain. The inhibitory interneuron is in green and the projection neuron is in red.
Different temporal phases between SP- and glutamate-driven feed-forward inhibition
Even though SP coexists with glutamate in primary nociceptive afferent synaptic terminals, their roles in various pain conditions may be distinct considering that higher frequencies of primary afferent stimulation are required to evoke the release of SP compared to glutamate [13]. Consistent with this idea, previous electrophysiological studies reported that activation of these fibers produced two phases of excitatory postsynaptic potentials (EPSPs), a fast EPSP mediated by non-NMDA receptors and a slow EPSP possibly mediated by the NK1R and NK2R [1,14]. Therefore, it is conceivable that SP-driven feed-forward inhibition and glutamate-driven feed-forward inhibition have different temporal phases.
To address this issue, Gu and colleagues first studied SP-driven feed-forward inhibition in the absence of glutamate receptor antagonists [9]. Their results showed that high frequency stimulation induced a robust and long-lasting increase in sIPSCs in wild-type mice but not in NK1R-/- mice. However, electrical stimulation-induced immediate eIPSCs, which could be blocked by glutamate receptor antagonists and are believed to be glutamate-driven feed-forward inhibition, were similar between wild-type and NK1R-/- mice. Taken together, these results provide evidence for a long-lasting, glutamate-independent SP-driven feed-forward inhibition that is distinct from the immediate pulse-by-pulse glutamate-driven feed-forward inhibition.
Functional significance of SP-driven feed-forward inhibition
After confirming the existence of the SP dependent feed-forward inhibition as well as delineating its distinct temporal phase, Gu and colleagues searched for the functional significance of the SP-driven feed-forward inhibition in the spinal cord dorsal horn. They hypothesized that the SP-driven feed-forward inhibition may serve to balance neuronal activity by counteracting SP-mediated excitatory nociceptive responses. This hypothesis challenges the traditional role of SP as a purely "pro-pain" substance in favor of a more homeostatic role.
A clever set of experiments were devised to address this question. First, SP-mediated nociceptive transmission in the superficial dorsal horn was blocked with a selective lesion of NK1R-expressing neurons using an intrathecal injection of SP-conjugated saporin (SP-SAP) [17,18]. It is important to note that the SP-SAP injection does not cause cytotoxicity in lamina V neurons so the SP-driven feed-forward inhibition remained intact. If an antinociceptive role for SP-driven feed-forward inhibition exists, the activation of SP release would have an analgesic effect in SP-SAP injected animals but blocking SP transmission would result in behavioral sensitization in SP-SAP injected animals.
Behavior responses to nociceptive heat stimuli were studied in SP-SAP treated animals after capsaicin treatment and intrathecal application of NK1R antagonists, resulting in two major findings [9]. First, capsaicin-induced behavioral sensitization in control animals was attenuated in SP-SAP treated animals consistent with the role of SP transmission in the superficial laminae as being "pro-pain". Second, NK1R antagonists reduced behavioral sensitization in the control group whereas the opposite effect was found in SP-SAP treated rats that support a possible antinociceptive role for SP in lamina V, probably through feed-forward inhibition. These results may explain hyperstimulation-induced analgesia, in which pain is controlled through additional painful stimulation or through counter irritation [9,19], and the lack of an analgesic effect of NK1R antagonists administered in clinical trials [5]. This study portrays SP as a double-edged sword for pain transmission, SP's actions may depend both on the location of release (lamina I vs lamina V) and an the type of neuron it synapses with (projection neurons vs inhibitory neurons).
Future directions
The elegant study by Gu and colleagues proposes a novel cellular mechanism for the neurokinin system in pain transmission and modulation. Dissecting its intricacies will not only contribute to a better understanding of how somatosensory inputs, including pain information, are coded within the spinal dorsal horn, but may also foster the development of more efficacious treatments for pain control.
Due to the important implications of SP-driven feed-forward inhibition in nociceptive transmission, some limitations of these findings need to be addressed. (1) SP can be released from at least three different sources: primary afferent fibers, descending projections fibers and local neurons. The current study uses dorsal root stimulation so it is likely that the primary afferents are the cause of the SP release [20,21]. Since it is possible that SP has a different function depending on its source, this study cannot rule out the other potential actions of SP when it is released from local neurons or descending terminals. (2) Activation of the NK1R in lamina V induced interneuronal depolarization and firing that was not due to action potential-independent synaptic transmission. Therefore, these receptors are unlikely to be localized in the presynaptic terminal. It would be interesting to investigate two related issues. One is the exact location of NK1Rs on the neuron, since some studies report that NK1Rs are postsynaptically or extrasynaptically located [22]. The other issue is the molecular mechanism for SP induction of interneuronal excitability. It is thought that SP might directly induce a postsynaptic current or indirectly modulate various receptors and channels. The Gi/Go-coupled NK1R is reported to activate phospholipase A2 and mobilize arachidonic acid [23]. The final targets for the signaling pathway initiated by SP binding to NK1R to result in SP-driven feed-forward inhibition in the spinal cord dorsal horn are still a mystery. (3) Although SP and NKA are synthesized together, they affect spinal nociception in different ways [24, 25]. Still, it cannot be ruled out that NKA might play a role in SP-driven feed-forward inhibition. The study excludes the possible involvement of neurokinin B (NKB) in SP-driven feed-forward inhibition since the NK3R (the receptor for NKB) antagonist did not significantly attenuate capsaicin-induced increases of sIPSCs. (4) This study showed that a long lasting increase in sIPSCs is SP-dependent but glutamate-independent. This conclusion is based on pharmacological studies using ionotropic glutamate receptor antagonists. However, whether metabotropic glutamate receptors are involved in SP-driven inhibition remains unknown. The authors cannot prove without a doubt that SP-driven feed-forward inhibition is glutamate independent without first excluding the potential role of metabotropic glutamate receptors.
Although the complexity of nociceptive transmission and modulation in the spinal cord make it difficult to address all questions in a single model, we believe the description of a SP-driven feed-forward inhibition to be a crucial finding towards the understanding of the role of SP in sensory transmission, modulation and plasticity.
==== Refs
Li P Zhuo M Substance P and neurokinin A mediate sensory synaptic transmission in young rat dorsal horn neurons Brain Res Bull 2001 55 521 531 11543953 10.1016/S0361-9230(01)00553-6
Yoshimura M Jessell TM Primary afferent-evoked synaptic responses and slow potential generation in rat substantia gelatinosa neurons in vitro J Neurophysiol 1989 62 96 108 2754484
De Felipe C Herrero JF O'Brien JA Palmer JA Doyle CA Smith AJ Laird JM Belmonte C Cervero F Hunt SP Altered nociception, analgesia and aggression in mice lacking the receptor for substance P Nature 1998 392 394 397 9537323 10.1038/32904
Cao YQ Mantyh PW Carlson EJ Gillespie AM Epstein CJ Basbaum AI Primary afferent tachykinins are required to experience moderate to intense pain Nature 1998 392 390 394 9537322 10.1038/32897
Rupniak NM Kramer MS Discovery of the antidepressant and anti-emetic efficacy of substance P receptor (NK1) antagonists Trends Pharmacol Sci 1999 20 485 490 10671176 10.1016/S0165-6147(99)01396-6
Mohrland JS Gebhart GF Substance P-induced analgesia in the rat Brain Res 1979 171 556 559 476489 10.1016/0006-8993(79)91062-X
Goettl VM Larson AA Activity at phencyclidine and mu opioid sites mediates the hyperalgesic and antinociceptive properties of the N-terminus of substance P in a model of visceral pain Neuroscience 1994 60 375 382 7521022 10.1016/0306-4522(94)90251-8
Guan JS Xu ZZ Gao H He SQ Ma GQ Sun T Wang LH Zhang ZN Lena I Kitchen I Elde R Zimmer A He C Pei G Bao L Zhang X Interaction with vesicle luminal protachykinin regulates surface expression of delta-opioid receptors and opioid analgesia Cell 2005 122 619 631 16122428 10.1016/j.cell.2005.06.010
Nakatsuka T Chen M Takeda D King C Ling J Xing H Ataka T Vierck C Yezierski R Gu JG Substance P-driven feed-forward inhibitory activity in the mammalian spinal cord Mol Pain 2005 1 20 15987503 10.1186/1744-8069-1-20
Melzack R Wall PD Pain mechanisms: a new theory Science 1965 150 971 979 5320816
De Biasi S Rustioni A Glutamate and substance P coexist in primary afferent terminals in the superficial laminae of spinal cord Proc Natl Acad Sci U S A 1988 85 7820 7824 2459717
Quartara L Maggi CA The tachykinin NK1 receptor. Part I: ligands and mechanisms of cellular activation Neuropeptides 1997 31 537 563 9574822 10.1016/S0143-4179(97)90001-9
Duggan AW Release of neuropeptides in the spinal cord Prog Brain Res 1995 104 197 223 8552769
Urban L Randic M Slow excitatory transmission in rat dorsal horn: possible mediation by peptides Brain Res 1984 290 336 341 6198043 10.1016/0006-8993(84)90952-1
Mantyh PW Rogers SD Honore P Allen BJ Ghilardi JR Li J Daughters RS Lappi DA Wiley RG Simone DA Inhibition of hyperalgesia by ablation of lamina I spinal neurons expressing the substance P receptor Science 1997 278 275 279 9323204 10.1126/science.278.5336.275
Nichols ML Allen BJ Rogers SD Ghilardi JR Honore P Luger NM Finke MP Li J Lappi DA Simone DA Mantyh PW Transmission of chronic nociception by spinal neurons expressing the substance P receptor Science 1999 286 1558 1561 10567262 10.1126/science.286.5444.1558
Cheing GL Luk ML Transcutaneous electrical nerve stimulation for neuropathic pain J Hand Surg [Br] 2005 30 50 55 15620492 10.1016/j.jhsb.2004.08.007
Warden MK Young WS Distribution of cells containing mRNAs encoding substance P and neurokinin B in the rat central nervous system J Comp Neurol 1988 272 90 113 2454979 10.1002/cne.902720107
Johansson O Hokfelt T Pernow B Jeffcoate SL White N Steinbusch HW Verhofstad AA Emson PC Spindel E Immunohistochemical support for three putative transmitters in one neuron: coexistence of 5-hydroxytryptamine, substance P- and thyrotropin releasing hormone-like immunoreactivity in medullary neurons projecting to the spinal cord Neuroscience 1981 6 1857 1881 6170907 10.1016/0306-4522(81)90028-2
Li JL Wang D Kaneko T Shigemoto R Nomura S Mizuno N The relationship between neurokinin-1 receptor and substance P in the medullary dorsal horn: a light and electron microscopic immunohistochemical study in the rat Neurosci Res 2000 36 327 334 10771111 10.1016/S0168-0102(00)00095-X
Garcia M Sakamoto K Shigekawa M Nakanishi S Ito S Multiple mechanisms of arachidonic acid release in Chinese hamster ovary cells transfected with cDNA of substance P receptor Biochem Pharmacol 1994 48 1735 1741 7526867 10.1016/0006-2952(94)90459-6
Cumberbatch MJ Chizh BA Headley PM Modulation of excitatory amino acid responses by tachykinins and selective tachykinin receptor agonists in the rat spinal cord Br J Pharmacol 1995 115 1005 1012 7582496
Trafton JA Abbadie C Basbaum AI Differential contribution of substance P and neurokinin A to spinal cord neurokinin-1 receptor signaling in the rat J Neurosci 2001 21 3656 3664 11331395
|
16297242
|
PMC1315348
|
CC BY
|
2021-01-04 16:40:08
|
no
|
Mol Pain. 2005 Nov 18; 1:34
|
utf-8
|
Mol Pain
| 2,005 |
10.1186/1744-8069-1-34
|
oa_comm
|
==== Front
J Ethnobiol EthnomedJournal of Ethnobiology and Ethnomedicine1746-4269BioMed Central London 1746-4269-1-111628865710.1186/1746-4269-1-11ResearchEthnomedicinal botany of the Apatani in the Eastern Himalayan region of India Kala Chandra Prakash [email protected] GB Pant Institute of Himalayan Environment & Development, Kosi-Katarmal, Almora, Uttaranchal- 263 643, India2005 16 11 2005 1 11 11 21 9 2005 16 11 2005 Copyright © 2005 Kala; licensee BioMed Central Ltd.2005Kala; 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.
This paper investigates the wealth of medicinal plants used by the Apatani tribe of Arunachal Pradesh. Apatani have traditionally settled in seven villages in the Ziro valley of Lower Subansiri district of Arunachal Pradesh in the Eastern Himalayan region of India. The present study has resulted in the documentation of 158 medicinal plant species used by the Apatani group of villages. These medicinal plant species were distributed across 73 families and 124 genera. Asteraceae was the most dominant family (19 species, 11 genera) of medicinal plants, followed by Zingiberaceae, Solanaceae, Lamiaceae and Araceae. For curing ailments, the use of aboveground plant parts was higher (80%) than the belowground plant parts in the Apatani group of villages. Of the aboveground plant parts, leaf was used in the majority of cases (56 species), followed by fruit. Different belowground plant forms such as root, tuber, rhizome, bulb and pseudo-bulb were used by Apatani as a medicine. About 52 types of ailments were cured by using these 158 medicinal plant species. The results of this study are further discussed in the changing socio-economic contexts.
Apatani tribeArunachal PradeshEastern Himalayaindigenous usesethnomedicinal plants
==== Body
Introduction
Tribal communities are mainly the forest dwellers who have accumulated a rich knowledge on the uses of various forests and forest products over the centuries. India possesses a total of 427 tribal communities, of these more than 130 major tribal communities live in North East India, which is comprised of the 8 states Meghalaya, Mizoram, Manipur, Tripura, Sikkim, Assam, Nagaland and Arunachal Pradesh. The major tribal communities of the North East India have been categorized into sub-tribes and if these sub-tribes are taken into account the total number of tribal groups reaches up to 300. In general, the tribes of North East India have been categorized into two broad ethnic communities, such as the Khasi and the Jaintia tribe of Meghalaya, who belong to 'Monkhemar' culture of Austoic dialect, and the rest of the tribal groups are basically Mongoloid, who belongs to Tibeto-Burman subfamily of Tibeto-Chinese group [1-3].
In the North East India, each state contains a number of tribal groups. Arunachal Pradesh is one of the states in North East India inhabited by 28 major tribes and 110 sub-tribes [4]. Arunachal Pradesh is the 12th mega biodiversity region of the world [5]. More than 545 species of orchids have been reported from the state, which is the highest number of orchid species known from any single state of India [6]. Such a rich biodiversity in the state has provided an initial advantage to its inhabitants for observing, and scrutinizing the rich flora and fauna for developing their own traditional knowledge. Most of the tribe economies have been historically engaged in subsistence agriculture or hunting and gathering. Over the years, they have developed a great deal of knowledge on the use of plants and plant products in curing various ailments.
A review of the literature reveals that many tribal areas and tribal communities in the eastern Himalayan region of India are either under explored or unexplored with regard to their floral wealth used in curing diseases. The Apatani is one such little studied tribe of Arunachal Pradesh [3]. Therefore, a need was felt to gather in-depth information on the plant species used by this tribal group and suggest that similar studies need to be carried out across the various groups of tribes for comparison as well as for documenting the knowledge which may be under threat due to the influence of modernization. The present paper thus aims to highlight and record in detail the traditional knowledge of the Apatani tribe on the use of medicinal plant species growing in and around their settlements.
Study area: Apatani group of villages
Literally, the word Apatani is composed of two words- 'Apa' and 'Tani'. According to the local language, 'Apa' means display of affection and 'Tani' stands for human race. The Apatani, generally, speak in their own language which has no script. Traditionally, they had settled in seven villages (e.g. Hong, Hari, Billa, Dutta, Hija, Mudang-Tage, and Michi Bamin) those were organized in accordance with the traditional lines of the three forefathers called Nichi-Nitii, Tinii-Diibo-Dre-Hija, and Talyang-Hao. These 7 villages are located in the Ziro valley of Lower Subansiri district in the central western part of Arunachal Pradesh in India between 26°55' – 28°21' N and 92°40' – 94°21' E. The Ziro valley (often called the Apatani valley) lies between the Panior and Kamla (Kuru) rivers at an altitude of 1524 to 2738 m a.m.s.l. The Apatani group of villages is located at 2200 m elevation. The Apatani migrated to the present location from the Talle Valley located in south eastern region [7]. The pattern of Apatani villages is that of string settlements (village houses are in a straight line) and homes are made of bamboo and timber. The Apatani belong to the Tibeto-Mongoloid stock, and trace their descent from one legendary ancestor, the Abotani.
The Apatani believe in indigenous religion called as 'Donyi-Polo' and are patriarchal in social system. The traditional village council, which regulates and administers the community, consists of three bodies namely Akha Buliyang, Yapha Buliyang and Ajang Buliyang. In each of these traditional institutions has one or two persons represent from each clan. Earlier, the Apatani had prominent tattoo marks on the face to distinguish themselves from other communities settled nearby. However, the practice of tattooing has been discouraged in the recent past and now is on the verge of extinction.
In 1991 census the population of Apatani was 22,526 (Table 1). The decadal (1991–2001) growth rate of the Apatani is 8.62%, which is much lower than that of the state (26.21%). The Ziro valley has an area of 1058 km2 of which 43 km2 is under agriculture, and remaining under forests, plantations and settlements. It is bounded with the areas traditionally belonging to neighbouring Nishi tribe. The land holding size varies from 0.02 to 10.00 ha with over 93% holdings consisting of 0.026–3.00 ha. The Ziro valley exhibits a humid sub-tropical to temperate type of climate with 108.1 cm rainfall and a temperature ranges from a maximum of 30.6°C to minimum of 1.1°C [8]. The climatic, altitudinal and geomorphological variations have shaped the two major vegetation types in and around the study area- sub-tropical forests and temperate forests. Sub-tropical forests in the study area are represented by Castanopsis indica, Acer sp., Pinus wallichiana and Pinus roxburghii, whereas, the temperate forests are represented by Quercus glauca, Alnus nepalensis, Castanopsis indica, Pyrus sp., Prunus sp., Populus sp. and Acer sp [9,10]. Many shrub species such as Berberis wallichiana, Viburnum foetidum, Prunus sp., Rubus sp., Spirea sp. and Symplocos sp. occur in the forested areas.
Table 1 Demographic profile of the Apatani in Ziro valley of Arunachal Pradesh
Year Population % to state
Apatani Arunachal Pradesh
1961 10,793 3,36,588 3.21
1971 12,888 4,68,511 2.75
1981 16,580 6,31,839 2.62
1991 22,526 8,64,558 2.61
2001 24,650 10,91,117 2.26
Source: Census of India, Part IX-B, Government of India.
Medicinal plants survey
A literature survey was carried out for compilation of existing information on the medicinal plants used by Apatani villagers [2,3,10-14]. In addition, field surveys in Apatani villages were undertaken during May and June 2005 to gather data on the indigenous uses of medicinal plant species by the Apatani. During the survey period, information was also gathered using semi-structured questionnaires on types of ailments cured by the traditional use of medicinal plants and plant parts used in curing different ailments. Cross-checking of data was made with the help of group discussions among different age classes of Apatani villagers that include both the genders of the society. The participant observation method was also employed to understand the methods and techniques adopted by the Apatani in curing diseases. The surrounding forested area and agricultural land of the Apatani villages were also surveyed with local youths and knowledgeable elders for the identification of various medicinal plant species and their indigenous uses. Since there is lack of comprehensive records on floral diversity of North East Himalaya including Arunachal Pradesh, the plant specimens were identified through various floral inventories [10,13,15]. The collected information was analyzed, and correlation was made between different genera and species of the medicinal plants in order to understand the pattern in medicinal plant uses and occurrences.
Results and discussion
The Apatani mainly subsist on agriculture and animal husbandry. Wet-rice cultivation is their most important agriculture practice. One of the Apatani proverbs reads "Tanii hii jebi danii", which means the Apatani depend on wet-rice cultivation. The Apatani have also developed a unique system of fish farming in their wet-rice croplands. They use available natural resources such as bamboo, cane, pine, Phragmites sp. and Castanopsis sp. in order to check the soil erosion, to conserve the soil fertility, to cultivate varieties of rice landraces, and to culture the fish in an integrated manner. Two species of bamboo (Phyllostachys bambusoides and Dendrocalamus hamiltonii) are also cultivated in private lands by the Apatani for construction of houses and other domestic uses. Bamboo shoots are also consumed by the Apatani as a vegetable. Apong, a locally prepared beer by fermenting rice, finger millet and barley, is an important beverage of the Apatani, which they prefer to consume with mutton. Domestic and semi-domestic cattle also play important role in maintaining the economic status of the Apatani. Possessing a large number of domestic animals is an indication of the prosperity of their respective owner [16]. Mithun (Bos frontalis) is preferred mostly for the meat. In addition, pigs, cows, and multiple varieties of birds and fish are consumed by the Apatani. A number of wild edible fruits and vegetables are also collected by the Apatani from the nearby forested areas to supplement the domestic nutritional requirements.
Traditionally, the Apatani group of villages was not only familiar with the knowledge of medicinal plants but they were also expert traders and met their necessities in exchange of paddy, which was always in excess of their requirements [17]. Earlier, they had no connection with the plains of Assam due to obstructions created by the Nishi who were earning a lot by acting as middlemen between the Apatani and the people residing in the plains. However, the Apatani had occupied a compact area in Ziro valley and were one of the self-sufficient tribes in North East India [8]. Their immediate dependence on nature had developed knowledge which ultimately is reflected in their traditional culture, religion, local belief, folklore, taboos language and dialects. For many centuries, the Apatani had kept alive a self-managed system of folk medicine that was mainly based on herbal remedies [10]. Their ingenuity still reflects their traditional management and sharing of natural resources in a way that there is optimum utilization of such resources [8,18]. The Nishi are one of the neighbours of the Apatani who live at lower elevations and are the most populous tribe in the state. Over the past few decades, the interaction between the Apatani and the Nishi has increased many fold due to migration of Apatani people in search of better education in Itanagar, a capital of Arunachal Pradesh. The availability of motor roads and the invasion of modern civilization have also enhanced the day to day interaction and the exposure of the Apatani to the rest of the world. Such interaction has provided a possible sharing of traditional knowledge of the Apatani with their neighbouring community.
During the present course of investigations, a total of 158 medicinal plant species used by the Apatani group of villages were documented. These medicinal plant species were distributed across 73 families and 124 genera (Table 2). In terms of number of medicinal plant species, Asteraceae was the most dominant family (19 species, 11 genera) of medicinal plants, followed by Zingiberaceae, Solanaceae, Lamiaceae, Araceae, and Verbanaceae (Table 3). There was a significant positive correlation (r = 0.92, p > 0.01) between the number of genera and number of species used as medicine by the Apatani (Figure 1). The invention of maximum number of uses of Asteraceae by the Apatani tribe demonstrates the dominance of Asteraceae around the Apatani group of villages. Asteraceae is the most dominant family of medicinal plants across the North Eastern States of India [13].
Table 2 Medicinal plant species, plant parts used and ailments cured by the Apatani of Ziro valley in Arunachal Pradesh
Sl No. Species Family Part used Uses
1 Acorus calamus L. Araceae Root Cut, wounds, skin diseases, bone fracture
2 Ageratum conyzoides L. Asteraceae Leaf Cut, wounds
3 Allium cepa L. Liliaceae Bulb Eye pain
4 Allium hookeri Thwait. Liliaceae Bulb Eruption of skin, cough, cold, wounds
5 Alocasia forniculata (Roxb.) Schott. Araceae Root Crack of heels
6 Alstonia scholaris (L.) Br. Apocynaceae Leaf, bark Headache, stomach disorder, menstrual disorder
7 Amomum aromaticum Roxb. Zingiberaceae Leaf, seed Fever, abortion
8 Amorphophallus paeoniifolius (Dennst.) Nicolson Araceae Corn Piles
9 Andrographis paniculata (Burm. f.) Wall. ex Nees Acanthaceae Leaf Dysentery
10 Anisomeles indica (L.) O.K. Lamiaceae Shoot Bodyache
11 Angiopteris evecta (Forst.) Hoffm. Angiopteridaceae Stem Health tonic
12 Antidesma acidum Retz. Euphorbiaceae Leaf Wounds
13 Argemone mexicana L. Papaveraceae Shoot Skin diseases
14 Artemisia indica Willd. Asteraceae Leaf Bodyache, asthma, skin diseases
15 Artemisia maritima L. Asteraceae Shoot Blood purification
16 Artemisia nilagirica (Cl.) Pamp. Asteraceae Leaf Cough, headache, sores
17 Asplenium nidus L. Aspleniaceae Leaf Ulcer
18 Barleria prionitis L. Acanthaceae Leaf Cough
19 Begonia roxburghii (Miq.) DC. Begoniaceae Leaf Indigestion
20 Berberis wallichiana (Wall.) Brongn. Berberidaceae Fruit, root Indigestion, bodyache
21 Bergenia ciliata (Haw.) Sternb. Saxifragaceae Root, leaf Cut, wounds
22 Brassiopsis glomarulata (Bl.) Regel. Araliaceae Fruit Cough
23 Buddleja asiatica Lour. Buddlejaceae Leaf Inflammation
24 Callicarpa macrophylla Vahl Verbenaceae Leaf Headache
25 Callicarpa vastita Roxb. Verbenaceae Leaf Indigestion
26 Calotropis gigantea (L.) Br. Asclepiadaceae Root Dog bite
27 Canarium resiniferum Brace ex King Burseraceae Fruit Urinary complaints
28 Capparis spinosa Lam. Capparaceae Root Rheumatic pain
29 Cardamine hirsuta L. Brassicaceae Leaf Indigestion
30 Castanopsis tribuloides (Sm.) DC. Fagaceae Stem Cough, goiter, indigestion
31 Centella asiatica L. Apiaceae Shoot Constipation, gastritis, blood purification
32 Chenopodium ambrosioides L. Chenopodiaceae Leaf Toothache
33 Christella parasitica (L.) Lev. Thelypteridaceae Fronds Cut, wounds
34 Chromolaena odorata (L.) King & Robinson Asteraceae Leaf Cut, wounds, headache, fever
35 Cirsium lapskyle Petral. Asteraceae Shoot Indigestion
36 Cissampelos pareira L. Menispermaceae Tuber Health tonic
37 Clerodendrum glandulosum Coleb. ex Wall. Verbenaceae Leaf Blood pressure, fever, cough
38 Clerodendrum serratum (L.) Moonb Verbenaceae Leaf Eye disorders
39 Coelogyne pectata Lindl. Orchidaceae Pseudobulb Burns
40 Colocasia affinis Schott Araceae Leaf Fever, respiratory disorder
41 Crassocephalum crepidioides (Benth.) Moore Asteraceae Leaf Indigestion, headache, stomachache, cut, wounds
42 Crotolaria pallida Ait. Fabaceae Root Bodyache
43 Croton roxburghii Balak Euphorbiaceae Fruit Indigestion
44 Curcuma caesia Roxb. Zingiberaceae Rhizome Cough, asthma
45 Curcuma aromatica Salisb. Zingiberaceae Whole plant Blood purification
46 Curcuma zedoaria Rosc. Zingiberaceae Rhizome Cold, cough
47 Cuscuta reflexa Roxb. Cuscutaceae Whole plant Purgative
48 Cyathea gigantea (Wall. ex Hk. f.) Holt. Cyatheaceae Leaf Bodyache
49 Cyathula prostrata (L.) Bl. Amaranthaceae Shoot Appetizer, dysentery, skin diseases
50 Cymbidium aloifolium (L.) Sw. Orchidaceae Tuber Wounds
51 Dendrocnide sinuta (Bl.) Chew. Urticaceae Leaf Urogenital disorder, toothache, dysentery
52 Dicranopteris linearis (Burm. f.) Und. Gleicheniaceae Whole plant Indigestion
53 Dicrocephala bicolor (Roth) Sch. Asteraceae Shoot Digestive problems
54 Dillenia indica L. Dilleniaceae Furit Stomachache
55 Dioscorea alata L. Dioscoraceae Tuber Indigestion
56 Dioscorea bulbifera L. Dioscoraceae Tuber Indigestion
57 Dioscorea hamiltonii Hk. f. Dioscoraceae Tuber Dysentery
58 Diplazium esculentum (Retz.) Sw. Athyriaceae Fronds Constipation
59 Ecbolium viride (Forsk) Alston Meliaceae Root Rheumatism
60 Eclipta prostrata (L.) L. Asteraceae Shoot Cut, wounds
61 Elaeagnus caudata Sch. ex Momiyama Elaeagnaceae Fruit Health tonic
62 Elaeagnus pyriformis Hk. f. Elaeagnaceae Fruit Constipation
63 Elatostema platyphyllum Wedd. Urticaceae Root Vomiting
64 Elsholzia blanda (Benth.) Benth. Lamiaceae Leaf Itching
65 Eluesine coracana (L.) Gaertn. Poaceae Grains Stomach disorder, tonic, cold
66 Eupatorium odoratum L. Asteraceae Leaf Wounds, cut
67 Erigeron bonariensis L. Asteraceae Leaf Nose block
68 Eryngium foetidum L. Apiaceae Seed Madness, headache
69 Ficus benjamina L. Moraceae Stem Stomach disorder
70 Ficus hirta Vahl Moraceae Fruit Wounds, cut
71 Gerbera pilosellioides (L.) Cass. Asteraceae Leaf Rheumatic pain
72 Gloriosa superba L. Liliaceae Tuber Killing lice in hairs
73 Gmelina arborea Roxb. Verbenaceae Leaf Stomach disorder
74 Gynostemma pedata Bl. Cucurbitaceae Leaf Throat ache
75 Gynura biscolor (Roxb. ex Willd.) DC. Asteraceae Leaf Intestinal worms
76 Gynura nepalensis DC. Asteraceae Leaf Indigestion
77 Hedychium coronarium Koen. Zingiberaceae Rhizome Bodyache
78 Hedychium dekianum Rao & Verma Zingiberaceae Rhizome Cut, wounds
79 Hedychium spicatum Buch.-Ham. ex Sm. Zingiberaceae Rhizome Stomach disorder
80 Hibiscus rosa-sinensis L. Malvaceae Flower Reproductive disorders
81 Houttuynia cordata Thunb. Saururaceae Shoot Freshness, good sleep, heart disorders
82 Hyptis suaveolens (L.) Poit. Lamiaceae Leaf Itching, cough, cold
83 Hypericum japonicum Thunb. ex Murr. Hypericaceae Stem Cut, wounds
84 Impatiens latifolia L. Balsaminaceae Leaf Headache, digestive disorder
85 Impatiens racemosa DC. Balsaminaceae Leaf Digestive disorder
86 Indigofera tinctoria L. Fabaceae Root Wound
87 Jasminum humile L. Oleaceae Root Ringworm
88 Laginaria siceraria (Molina) Standl. Cucurbitaceae Fruit Burns
89 Leonotis nepetifolia R. Br. Lamiaceae Seed Burns
90 Lithocarpus dealbatus (Miq.) Rehder Fagaceae Fruit Indigestion
91 Litsea cubeba (Lour.) Pers. Lauraceae Fruit Cough, cold, hair tonic, indigestion, good sleep
92 Litsea salicifolia (Nees) Hk.f. Lauraceae Fruit Bone fracture, stomach disorder
93 Lygodium scandens (L.) Sw. Schizaeaceae Leaf Skin diseases
94 Mahonia napalensis DC. Berberidaceae Stem Itching
95 Measa indica (Roxb.) DC. Myrsinaceae Fruit Indigestion
96 Mikania micrantha Kunth. Asteraceae Leaf Itching, skin diseases, headache
97 Miliusa roxburghiana (Wall. ex Griff.) Hk. f. & Th. Annonaceae Leaf Headache
98 Molineria crassifolia Baker Hypoxidaceae Fruit Diarrhoea
99 Molineria recurveta (Dryand) Hebbert. Hypoxidaceae Leaf Bodyache
100 Mucuna pruriens (L.) DC. Fabaceae Stem Eye disorder
101 Murraya koenigii (L.) Spr. Rutaceae Leaf Stomach trouble
102 Musa paradissica L. Musaceae Fruit Indigestion
103 Myrica esculenta Ham. ex D. Don. Myricaceae Fruit, bark Indigestion, skin eruption
104 Myrsine semiserrata Wall. Myrsinaceae Seed Skin diseases
105 Oenanthe javanica (Bl.) DC. Apiaceae Shoot Indigestion
106 Oroxylum indicum (L.) Vent. Bignoniaceae Seed Purgative, headache
107 Osbeckia stellata Buch.-Ham. ex D. Don Melastomataceae Leaf Toothache
108 Oxalis corniculata L. Oxalidaceae Shoot Appetizer, headache
109 Paedaria foetida L Rubiaceae Stem Gastritis, diarrhea, stomach disorder
110 Passiflora foetida L. Passifloraceae Fruit Respiratory disorder
111 Photinia integrifolia Lindl. Rosaceae Fruit Indigestion
112 Perilla frutescens (L.) Britt. Lamiaceae Seed Fever, headache
113 Physalis angulata L. Solanaceae Fruit Gastric trouble
114 Physalis minima L. Solanaceae Fruit Gastric trouble
115 Physalis peruviana L. Solanaceae Leaf Pain in pregnancy
116 Picrorhiza kurrooa Benth. Scrophulariaceae Root Cold, fever
117 Pinus roxburghii Sarg. Pinaceae Seed Indigestion
118 Piper brachystachyum Wall. Piperaceae Seed Cough
119 Piper trioicum Roxb. Piperaceae Root Cough
120 Plantago major L. Plantaginaceae Leaf Constipation
121 Plectranthus japonicus (Burm. f.) Koidz. Acanthaceae Leaf Fever
122 Polygonum nepalense Meissn. Polygonaceae Leaf Indigestion
123 Polygonum perfoliatum L. Polygonaceae Leaf Indigestion
124 Portulaca oleracea L. Portulacaceae Stem, leaf Appetizer
125 Pouzolzia hirta (Bl.) Hassk. Urticaceae Root Constipation
126 Pterospermum acerifolium Willd. Sterculiaceae Flower Earache
127 Rhus chinensis Miller Anacardiaceae Fruit Blood dysentery
128 Rubia cordifolia L. Rubiaceae Shoot Stomachache
129 Rubus calycinus Wall. Rosaceae Fruit Stomach disorder
130 Rubus ellipticus Sm. Rosaceae Fruit Indigestion
131 Rubus paniculatus Sm. Rosaceae Fruit Stomach disorder
132 Rubus roseafolius Sm. Rosaceae Fruit Indigestion
133 Rumex nepalensis Spr. Polygonaceae Leaf Indigestion
134 Saurauria roxburghii Wall. Saurauriaceae Leaf Constipation
135 Schefflera glomerata L. Araliaceae Fruit Indigestion
136 Schizostachium capitatum (Munro) Majumdar Poaceae Shoot Diarrhea, dysentery, stomach disorder
137 Senna alata (L.) Roxb. Caesalpiniaceae Leaf Skin diseases
138 Senna tora (L.) Roxb. Ceasalpiniaceae Leaf Low blood pressure
139 Sphenomeris chinensis (L.) Maxon Lindsaeceae Fronds Sprains
140 Solanum kurzii Brace ex Prain Solanaceae Fruit Cough, worms infestation
141 Solanum myriacanthum Dunal Solanaceae Seeds Toothache
142 Solanum nigrum L. Solanaceae Leaf Liver tonic, indigestion
143 Solanum torvum Sm. Solanaceae Fruit Cough, skin diseases
144 Sonchus asper (L.) Hill Asteraceae Shoot Indigestion
145 Sonchus arvensis L. Asteraceae Shoot Stomachache, gastritis
146 Spilanthus clava L. Asteraceae Leaf Throat pain
147 Spilanthes paniculata DC. Asteraceae Leaf Constipation
148 Stellaria media (L.) Vill. Caryophyllaceae Leaf Itching
149 Stereospermum chelonoides (L. f.) DC. Bignoniaceae Leaf Sprain
150 Strobilanthus helictus T. Anders. Acanthaceae Shoot Indigestion
151 Terminalia chebula Retz. Combretaceae Fruit Cough
152 Toddalia aculeata Pers. Rutaceae Fruit Throat pain
153 Urtica dioica L. Urticaceae Leaf Bone fracture
154 Vernonia cinerea (L.) Less Asteraceae Leaf Indigestion
155 Zanthoxylum acanthopodium DC. Rutaceae Fruit Dysentery
156 Zanthoxylum armatum DC. Rutaceae Fruit Cold, cough, fever, appetizer
157 Zanthoxylum oxyphyllum Edgew. Rutaceae Fruit Stomach disorder
158 Zingiber officinale Rosc. Zingiberaceae Rhizome Cough
Table 3 Dominant families of medicinal plants used by the Apatani in terms of number of species occupied
Family Genera Species
Asteraceae 11 19
Zingiberaceae 4 8
Solanaceae 2 7
Lamiaceae 5 5
Araceae 5 5
Verbenaceae 3 5
Rutaceae 3 5
Rosaceae 2 5
Urticaceae 4 4
Acanthaceae 4 4
Figure 1 Relationship between genera and species richness of medicinal plants used by the Apatani in Ziro valley of Arunachal Pradesh.
Different parts of medicinal plant species were used by the Apatani as a medicine. For curing ailments, the use of aboveground plant parts was higher (80%) than the belowground plant parts. Of the aboveground plant parts, leaf was used in the majority of cases (56 species), followed by fruits. Different belowground plant forms such as root, tuber, rhizome, bulb and pseudo-bulb were also used by the Apatani as a source of curing ailments (Table 4). The whole plant of 3 species [e.g. Curcuma aromatica Salisb., Cuscuta reflexa Roxb. and Dicranopteris linearis (Burm. f.) Und.] was used as medicine. These 158 medicinal plant species were used in curing about 52 types of ailments, of which the highest numbers of plant species (40 species) were used for the treatment of gastrointestinal disorders such as indigestion and constipation. About 19 medicinal plant species were used in curing cough and cold, and 15 medicinal plant species were used for healing cuts and wounds (Table 5).
Table 4 Patterns in Apatani use of medicinal plant parts
Aboveground plant parts used Number of Species Belowground plant parts used Number of Species
Whole shoot 15 Root 14
Leaf 56 Tuber 6
Fruit 31 Rhizome 6
Seed 10 Bulb 2
Stem 6 Pseudo-bulb 1
Fronds 4 Corn 1
Bark 2
Flower 2
Total 126 Total 30
Table 5 Major ailments cured by the Apatani in terms of using the plant species
Ailments Number of plants used
Indigestion 40
Cough and cold 19
Cut and wounds 15
Headache 12
Stomach disorder 11
Skin diseases 11
Fever 8
Body-ache 6
Dysentery 6
Throat-ache 5
Previous studies carried out in North East India have reported 41 medicinal plant species used by the Apatani of Arunachal Pradesh [13]. However, they had selected many North Eastern States and 12 different tribal communities for investigations. Based on their experiences, they had suggested the need of carrying out detailed investigations of each tribe. So far different authors have reported 1350 species of plants used in ethnomedicinal preparations, 665 species of food plants and 899 species for miscellaneous uses from the entire North East India [3]. The present inventory of 158 medicinal plant species as used by the Apatani is one of its kinds in terms of the highest number of species recorded so far used by a single tribe of the North East India. This fact provides a strength to the statements of earlier researchers that North East India is still under-explored and certain areas in the district of Subansiri still remain unexplored [3,10]. Hence, a need for detailed investigations of ethnobotanical knowledge held by each tribal community in North East India is required before such valuable knowledge vanishes. In spite of the rich wealth of bio-resources and potential, development is far from meeting the expectations of local people in Arunachal Pradesh mainly in terms of existing health care facilities and herbal industries. Ethnomedicinal knowledge is also important from a humanitarian point of view in that in long run as this knowledge may help to identify important medicinal uses that can help in curing and healthcare around the world. Attempts should be made to share the benefits arising from such knowledge with its holders. The present inventory of medicinal plants used by the Apatani opens new avenues to scrutinize such a rich natural resource for further analysis in order to develop the potential of herbal medicine.
Acknowledgements
I thank Director, G.B. Pant Institute of Himalayan Environment & Development, for providing logistic support. I extend my thanks to Dr. Pitamber P. Dhyani, Dr. Drupad C. Choudhury, Dr. Rakesh C. Sundriyal, Dr. Nehal A. Farooquee, Dr. Manju Sundriyal, Mr. Mihin Dollo, and Mr. Rubu Bukar for helping in various ways during the course of this study. Dr. M.S. Rawat is thanked for discussions on the subject matter. I acknowledge the help provided by all the Apatani group of villages during the fieldwork. Dr. Andrea Pieroni and three anonymous referees are thanked for their constructive comments on earlier drafts of the manuscript.
==== Refs
Ramakrishnan PS Tropical forests, exploitation, conservation and management Impact of Science on Society 1992 42 149 162
Anonymous Editorial Indian Journal of Traditional Knowledge 2005 4 3 4
Dutta BK Dutta PK Potential of ethnobotanical studies in North East India: an overview Indian Journal of Traditional Knowledge 2005 4 7 14
Tag H Das AK Kalita P Plants used by the hill Miri of Arunachal Pradesh in ethnofisheries Indian Journal of Traditional Knowledge 2005 4 57 64
Agarwal KC Biodiversity 1999 Bikaner: Agrobotanica
Chowdhery HJ Orchid flora of Arunachal Pradesh 1998 Calcutta: Botanical Survey of India
Haimendorf CVF The Apatanis and their neighbours: A primitive society of the Eastern Himalayas 1962 London: Routledge and Keagan Paul
Saikia SK Das DN 'Aji gnui asonii'- a practice of organic hill farming among the Apatani tribe of Eastern Himalaya International Journal of Sustainable Development and World Ecology 2004 11 211 217
Kaul RN Haridasan K Forest types of Arunachal Pradesh- A preliminary study Journal of Economic and Taxonomic Botany 1987 9 378 389
Rawat MS Chowdhury S Ethno-medico-botany of Arunachal Pradesh 1998 Dehradun: Bishen Singh Mahendra Pal Singh
Pal GD Observations on ethnobotany of tribals of Subansiri district, Arunachal Pradesh Bulletin of Botanical Survey of India 1984 26 26 37
Pal GD Observations on less known interesting tribal uses of plants in Lower Subansiri district, Arunachal Pradesh Journal of Economic and Taxonomic Botany 1992 10 198 203
Saklani A Jain SK Cross cultural ethnobotany of Northeast India 1994 New Delhi: Deep Publications
Sundriyal RC Singh T Sinha GN Arunachal Pradesh: Environmental planning and sustainable development – Opportunities and challenges 2002 Almora: G.B. Pant Institute of Himalayan Environment and Development
Hooker JD The Flora of British India Kent: L. Reeve & Co. Ltd., NR 1872 94
Dhyani PP Kala CP Current research on medicinal plants: Five lesser known but valuable aspects Current Science 2005 88 335
Chakravarty LN Glimpses of the early history of Arunachal 1973 Jorhat: Annada Printing House
Rai SC Apatani paddy-cum fish cultivation: An indigenous hill farming system of North East India Indian Journal of Traditional Knowledge 2005 4 65 71
|
16288657
|
PMC1315349
|
CC BY
|
2021-01-04 16:40:12
|
no
|
J Ethnobiol Ethnomed. 2005 Nov 16; 1:11
|
utf-8
|
J Ethnobiol Ethnomed
| 2,005 |
10.1186/1746-4269-1-11
|
oa_comm
|
==== Front
Health Qual Life OutcomesHealth and Quality of Life Outcomes1477-7525BioMed Central London 1477-7525-3-761631367810.1186/1477-7525-3-76ReviewA review of the psychometric properties of the Health of the Nation Outcome Scales (HoNOS) family of measures Pirkis Jane E [email protected] Philip M [email protected] Pia K [email protected] Sarity [email protected] Tim J [email protected] Michelle K [email protected] School of Population Health, The University of Melbourne, Melbourne, Australia2 School of Population Health, The University of Queensland, Brisbane, Australia3 New South Wales Institute of Psychiatry, Sydney, Australia2005 28 11 2005 3 76 76 4 11 2005 28 11 2005 Copyright © 2005 Pirkis et al; licensee BioMed Central Ltd.2005Pirkis 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 Health of the Nation Outcome Scales was developed to routinely measure outcomes for adults with mental illness. Comparable instruments were also developed for children and adolescents (the Health of the Nation Outcome Scales for Children and Adolescents) and older people (the Health of the Nation Outcome Scales 65+). All three are being widely used as outcome measures in the United Kingdom, Australia and New Zealand. There is, however, no comprehensive review of these instruments. This paper fills this gap by reviewing the psychometric properties of each.
Method
Articles and reports relating to the instruments were retrieved, and their findings synthesised to assess the instruments' validity (content, construct, concurrent, predictive), reliability (test-retest, inter-rater), sensitivity to change, and feasibility/utility.
Results
Mostly, the instruments perform adequately or better on most dimensions, although some of their psychometric properties warrant closer examination.
Conclusion
Collectively, the Health of the Nation Outcome Scales family of measures can assess outcomes for different groups on a range of mental health-related constructs, and can be regarded as appropriate for routinely monitoring outcomes.
Mental healthoutcome measurementHealth of the Nation Outcome Scales (HoNOS)Health of the Nation Outcome Scales for Children and Adolescents (HoNOSCA)Health of the Nation Outcome Scales 65+ (HoNOS65+)
==== Body
The Health of the Nation Outcome Scales (HoNOS) arose out of the UK's Health of the Nation Strategy, and was created by Wing and colleagues as an instrument that could be routinely used to measure outcomes for adults with mental illness [1,2]. Comparable measures for children and adolescents (HoNOSCA) and older people (HoNOS65+) were later developed by Gowers and colleagues [3,4] and Burns et al [5], respectively.
All three instruments measure mental health and social/behavioural functioning (see Table 1), and are being used increasingly as routine clinical outcome measures against which the quality and effectiveness of mental health services can be monitored, judged and improved. They are the most widely used routine outcome measures in British mental health services [6], and they are being used at admission, review and discharge in inpatient and ambulatory public-sector mental health services in all Australian states/territories [7]. They are also being used widely in New Zealand, and, to a greater or lesser degree, in other countries, including Canada, Denmark, France, Italy, Germany and Norway.
Table 1 Items, structure and scoring for the HoNOS family of measures
Instrument Item Subscales/sections Scoring
HoNOS 1. Overactive, aggressive, disruptive or agitated behaviour
2. Non-accidental self-injury
3. Problem drinking or drug taking
4. Cognitive problems
5. Physical illness or disability problems
6. Problems associated with hallucinations and delusions
7. Problems with depressed mood
8. Other mental and behavioural problems
9. Problems with relationships
10. Problems with activities of daily living
11. Problems with living conditions
12. Problems with occupation and activities Behaviour (1–3)
Impairment (4–5)
Symptoms (6–8)
Social (9–12) Each item rated on a 5-point scale:
0. no problem
1. minor problem requiring no action
2. mild problem but definitely present
3. moderately severe problem
4. severe to very severe problem.
Scoring yields individual item scores, subscale scores and a total score.
HoNOSCA 1. Problems with disruptive, antisocial or aggressive behaviour
2. Problems with over-activity, attention or concentration
3. Non-accidental self-injury
4. Problems with alcohol, substance or solvent misuse
5. Problems with scholastic or language skills
6. Physical illness or disability problems
7. Problems associated with hallucinations, delusions or abnormal perceptions
8. Problems with non-organic somatic symptoms
9. Problems with emotional and related symptoms
10. Problems with peer relationships
11. Problems with self-care and independence
12. Problems with family life and relationships
13. Poor school attendance
14. Problems with knowledge or understanding about the nature of the child or adolescent's difficulties
15. Problems with lack of information about services or management of the child or adolescent's difficulties Section A (1–13)
Behaviour (1–4)
Impairment (5–6)
Symptoms (7–9)
Social (10–13)
Section B (14–15) Each item rated on a 5-point scale:
0. no problem
1. minor problem requiring no action
2. mild problem but definitely present
3. moderately severe problem
4. severe to very severe problem.
Scoring yields individual item scores, subscale scores and a total score (derived from Section A only).
HoNOS65+ 1. Behavioural disturbance (e.g., overactive, aggressive, disruptive or agitated behaviour, uncooperative or resistive behaviour);
2. Non-accidental self-injury;
3. Problem drinking or drug taking;
4. Cognitive problems;
5. Physical illness or disability problems;
6. Problems associated with hallucinations and delusions;
7. Problems with depressive symptoms;
8. Other mental and behavioural problems;
9. Problems with relationships;
10. Problems with activities of daily living;
11. Problems with living conditions; and
12. Problems with occupation and activities Behaviour (1–3)
Impairment (4–5)
Symptoms (6–8)
Social (9–12) Each item rated on a 5-point scale:
0. no problem
1. minor problem requiring no action
2. mild problem but definitely present
3. moderately severe problem
4. severe to very severe problem.
Scoring yields individual item scores, subscale scores and a total score.
Despite their relative widespread use as outcome measures, there is some reported concern – particularly among clinicians who are using the instruments. Anecdotally, some clinicians question the psychometric soundness of the instruments, and argue that they do not have good clinical utility [7]. With the exception of a specific review of the applicability of the HoNOS and the HoNOS65+ for older people [8], there has been no comprehensive review of these instruments that can inform this debate. The current paper fills this gap, by appraising the psychometric properties of each.
Methods
The review could best be described as a qualitative systematic review [9]. It involved a comprehensive search of all potentially relevant articles, using explicit search criteria. However, because it assessed the psychometric properties of three different instruments on eight different dimensions, it was beyond its scope to statistically combine the results of different studies. Instead, the results were summarised in a narrative fashion.
Article retrieval
Searches of the electronic databases MEDLINE and PSYCINFO were conducted from their respective years of inception to November 2005. The search was retrieved articles using the following search terms:
• MENTAL HEALTH or PSYCHIATR*
• OUTCOME MEASURE* or ROUTINE OUTCOME MEASURE*;
• HEALTH OF THE NATION OUTCOME SCALES or HONOS;
• HEALTH OF THE NATION OUTCOME SCALES 65+ or HONOS65+; and
• HEALTH OF THE NATION OUTCOME SCALES FOR CHILDREN AND ADOLESCENTS or HONOSCA.
Potentially relevant peer-reviewed journal articles were retrieved by this means, and their reference lists scanned for further pertinent articles. Efforts were also made to retrieve government and other reports, both from within Australia and overseas, largely by conducting Internet searches using the above terms. Greatest weight was given to the peer-reviewed articles for two reasons. Firstly, it was possible to be confident that they had undergone some academic checking for scientific merit. Secondly, this approach created a relatively 'level playing field' for all instruments. It is acknowledged, however, that the relative standing of the given journal was not taken into account, and the individual studies were not systematically rated for quality (although consideration was given to the strength of their design).
In addition, the review primarily concerned itself with articles (and reports) that involved explicit testing of the psychometric properties of a given instrument (e.g., a study that examined the validity and reliability of the HoNOS). Articles that described the use of a given instrument in a study of some other kind (e.g., a randomised controlled trial that used the HoNOS as an outcome measure in assessing the relative merits of two different types of treatment) were given less weight. This decision was made on the grounds that the latter type of study, by design, implicitly accepted the psychometric value of the given instrument and to use the findings as evidence for the psychometric robustness of that instrument would create a somewhat circular argument.
Critical appraisal of the instruments
Evidence from the above articles and reports was used to critically appraise each of the instruments. The critical appraisal exercise was guided by a checklist that drew on the work of Greenhalgh et al [10], Green and Gracely [11], McDowell and Newell [12] and Chronbach and Meehl [13].
Specifically, the checklist elicited evaluative information on each instrument, namely its:
• Content validity, which refers to the instrument's comprehensiveness (i.e., how adequately the sampling of items reflects its aims), and is commonly ascertained by asking stakeholders to review the content of the instrument;
• Construct validity, which involves conceptually defining the construct to be measured by the instrument, and assessing the internal structure of its components and the theoretical relationship of its item and subscale scores;
• Concurrent validity, which pits the instrument against 'gold standards' (e.g., scores on more established instruments);
• Predictive validity, which assesses the instrument's ability to predict future outcomes (e.g., resource use or treatment response);
• Test-retest reliability, or the degree of agreement when the same instrument is applied to the same consumer by the same rater at two different time points;
• Inter-rater reliability, or the degree of agreement when the same instrument is applied to the same consumer by different raters at the same time point;
• Sensitivity to change, or the degree to which the instrument demonstrates change over time, as measured against 'gold standards' (e.g., change assessed by more established instruments); and
• Feasibility/utility, or the degree to which the instrument is acceptable to and useful for stakeholders.
Results
HoNOS
Content validity
Shergill et al [14], Orrell et al [15] and McClelland et al [16] explored the content validity of the HoNOS by asking consumer/carer advocacy groups and mental health professionals to comment on whether its items reflected areas of concern for them. In the main, respondents in these studies were positive, suggesting that the HoNOS was appropriate, well-designed and thorough.
However, respondents were concerned about the restriction imposed by the rater being forced to indicate only one problem in Item 8 (Other mental and behavioural problems) [14,16], and questioned the ability of Item 6 (Problems associated with hallucinations and delusions) to accurately describe the symptoms and role performance of a person with schizophrenia [15]. They also felt that the social items (Items 10, 11 and 12) were problematic because the complexity of information needed to rate them [15,16].
Respondents also noted that, for some items, anchor points and their associated terminology were subjective [14,15]. They commented on difficulties with knowing which item to use for rating some symptoms, such as elated mood. In addition, they observed the failure of the instrument to take into account factors such as culture, poverty, abuse, safety and risk, bereavement and medication compliance [14,15]. Some respondents suggested that the HoNOS was open to human error and misinterpretation [16].
Construct validity
In studies of the internal consistency of the HoNOS, Cronbach's alpha has ranged from 0.59 to 0.76, indicating moderately high internal consistency and low item redundancy, and supporting the instrument's use as a meaningful summary of severity of symptoms [1,14-20]. That said, Trauer [18,21] has argued that the HoNOS does not measure a single, underlying construct of mental health status.
McClelland et al [16] examined the relative contribution of each of the HoNOS items to the total score, and found that Item 7 (Problems with depressed mood), Item 8 (Other mental and behavioural problems) and Item 9 (Problems with relationships) had the greatest weight, contributing 15%, 19% and 14% to the total, respectively. By contrast, Item 11 (Problems with living conditions) and Item 12 (Problems with occupation and activities) contributed only 3% each.
Preston [22], Trauer [18] and McClelland [16] examined the subscale structure of the HoNOS. In his study, Preston found that the four factor model defined by the original subscales had good fit, but that the contribution of individual items to their respective subscales varied in two separate mental health services, indicating differentiation in construct interpretation. Trauer's examination of the subscales revealed a poorer fit than Preston's, leading him to propose an alternative five factor structure which has been supported in later studies [20]. McClelland's study also identified alternative factors.
Concurrent validity
Numerous studies have considered the concurrent validity of the HoNOS, assessing its performance against more established instruments that have been shown to validly measure related constructs. In the main, the HoNOS has been shown to perform well against clinician-rated instrument such as the Role Functioning Scale [1], Brief Psychiatric Rating Scale [1,14-16], Global Assessment Scale [14-16,23-25], Life Skills Profile [20,23], Manchester Audit Tool [26], Clifton Assessment Procedures for the Elderly – Behaviour Rating Scale [14], Clinical Dementia Rating [14], Mini-Mental State Examination [14], Schedules for Clinical Assessment in Neuropsychiatry [25,27], Broad Rating Schedule [25], Disability Assessment Schedule [25], Social Adjustment Scale [25], Location of Community Support Scale [15], Social Behaviour Schedule [15,27], Hamilton Rating Scale for Depression [28] and Positive and Negative Symptoms Scale [28]. There are some exceptions, with low correlations being found between the HoNOS and the Brief Psychiatric Rating Scale in one study [29] and the Beck Depression Inventory in another [19].
By contrast, the HoNOS has shown poor or mixed performance against consumer-rated instruments such as the Symptom Check List 90 – Revised [29,30], Social Adjustment Scale [29], Medical Outcomes Study Short Form 36 [30], Camberwell Assessment of Need Short Appraisal Schedule [31], Quality of Life Scale [14], Avon Mental Health Measure [32], Outcome of Problems of Users of Services [32], an instrument adapted from the Quality of Life Index for Mental Health [23] and even a self-rating version of the HoNOS with a similar question structure [33]. As with the clinician-rated measures, there are exceptions to the general rule, but even where studies have reported correlations between the HoNOS and consumer-rated measures – e.g., the Camberwell Assessment of Need Short Appraisal Schedule [34-36], Medical Outcomes Study Short Form 36 [15,28], General Health Questionnaire [15] and Comprehensive Quality of Life Scale [28] – they tend to vary across domains and be lower than those between the HoNOS and clinician-rated measures. These findings are not surprising, given that poorer correspondence is typically found between instruments that rely on information from informants of different classes than those which rely on information from informants of the same class, since different informants have access to different information.
The ability of the HoNOS to discriminate between consumer groups differentiated on a range of treatment- and service-based indicators has also been used to test its concurrent validity. Several studies have found high total scores on the HoNOS to be associated with diagnoses of drug and alcohol, psychotic and bipolar disorders, high scores on items relating to hallucinations/delusions and social and cognitive problems to be associated with a diagnosis of schizophrenia, high scores on items relating to aggressive behaviour, drinking/drug taking and anxiety to be associated with a diagnosis of mania, and high scores on items relating to suicidal thoughts/behaviours, physical illness and depressed mood to be associated with a diagnosis of depression [16,20,24,26,37,38]. Similarly, a number of studies have found that the HoNOS can discriminate between consumers with differing levels of need or disability, as indicated by their current or expected location of treatment – e.g., those receiving standard case management versus those assertive case management [39], those in residential/nursing home, day patient, outpatient and inpatient settings [14,15,28], and those in long-stay settings with low, medium and high expectations of discharge [40].
Predictive validity
Several studies have examined the predictive validity of the HoNOS. Most have found it to have reasonably good predictive validity, explaining a significant proportion of the variance in resource use (e.g., as measured by service contacts, length of stay and costs) and treatment outcome (e.g., as measured by readmission rates, retention in the community, treatment response and death) [23,28,41-43]. There have been exceptions, however, with some studies finding limited correspondence between HoNOS total scores and resource use [44,45].
Test-retest reliability
Few studies have examined the test-retest reliability of the HoNOS, but those that have generally report fair to moderate overall reliability scores [14,15,30]. Particularly low reliability scores have been reported for Item 1 (Overactive, aggressive, disruptive or agitated behaviour), Item 3 (Problem drinking or drug taking), Item 7 (Problems with depressed mood), and Item 10 (Problems with activities of daily living).
Inter-rater reliability
Most studies of the inter-rater reliability of the HoNOS total score have found that the overall agreement between pairs of raters is fair to moderate [14,27,30], or even moderate to good [1,15,25,28], but that agreement is poor on particular items. Items identified as problematic include Item 4 (Cognitive problems) [27], Item 7 (Problems with depressed mood) [27], Item 8 (Other mental and behavioural problems) [1,27], Item 9 (Problems with relationships) [15], Item 11 (Problems with living conditions) [15,46] and Item 12 (Problems with occupation and activities) [1,27,46].
Sensitivity to change
The sensitivity of the HoNOS to change has been assessed in a number of studies which have examined the extent to which the direction and magnitude of movement in HoNOS total or item scores correlates with some external measure of change.
The simplest of these studies have examined change in HoNOS over time in given settings, hypothesising that there should be a decrease in severity as the consumer nears the end of an episode. Generally, these studies have found decreases of the greatest magnitude in inpatient settings and of lesser magnitude in community settings [16,46-48]. That said, there is some evidence that there may be an interaction between setting, diagnosis and severity, and that the HoNOS may be able to detect change in the community for those with depression and anxiety [26] and those with higher HoNOS total scores at episode start [49]. Particular items may also interact with setting, with one study that considered the range of inpatient and community settings finding that scores on all items except Item 11 (Problems with living conditions) showed decreases over time [16], and another that concentrated on a community setting only finding that only Items 7 (Problems with depressed mood), 8 (Other mental and behavioural problems) and 9 (Problems with relationships) had sufficient relevance and variability to change over time [48].
Other studies have used clinician or consumer judgement as the 'gold standard' against which to evaluate whether change has occurred and, if so, whether the HoNOS is capable of detecting it. In separate studies, Taylor and Wilkinson [50] and Gallagher and Teesson [39] found correlations between changes in consumers' HoNOS total scores and clinical judgements about whether they had improved, remained stable or deteriorated made by GPs and case managers, respectively. Likewise, Hunter et al [32] found that significant decreases in HoNOS total scores between initial and repeat ratings corresponded with consumers' self report of their goals having been met.
Still other studies have compared the HoNOS's dynamic properties and capacity to detect change against other, more established measures of outcome. Using these criteria, McClelland et al [16] found the HoNOS to perform commensurately with the Global Assessment Scale and the Brief Psychiatric Rating Scale. Sharma et al [51] found it performed well against the Modified Clinical Global Impressions Scale, although the correlations were greatest for those with extreme improvement or deterioration. Ashaye et al [43] found the HoNOS was correlated with the Clifton Assessment of Strengths, Interests and Goals and two quality of life scales in elderly consumers, particularly those with dementia and depression. By contrast, Bebbington et al [27] found the HoNOS performed poorly by comparison with the Schedules for Clinical Assessment in Neuropsychiatry and the Social Behaviour Schedule.
A final approach to examining sensitivity to change has involved assessing whether improvements in HoNOS total scores are observed for consumers who receive evidence-based therapies and therefore would be expected to show reductions in symptom severity. Bech et al [37], for example, hypothesised that consumers who received lithium and/or ECT would show greater improvement on the HoNOS than consumers who did not, and found this to be the case, at least for the Behaviour and Symptoms subscales.
Feasibility/utility
There has been considerable debate about the feasibility/utility of the HoNOS. The least enthusiastic authors have argued that it is of limited value in informing care planning [24,51-55]. More positive authors have suggested it is a comprehensive, user-friendly tool that is likely to have utility in routine outcome measurement [1,16,19,28,38,39,56], and, with other evidence, could make a valuable contribution in informing clinical judgements [2].
Audits of the extent to which the HoNOS is being used in particular settings have generally lent support to the latter view. Glover and Sinclair-Smith [57] found that 60% of mental health care provider trusts in Britain had implemented routine outcome measurement (with the majority using the HoNOS), and James and Kehoe [58] found that 77% of consumers in a UK district service had HoNOS scores recorded in their care plans. The latter finding was supported by Broadbent [41], who found that the HoNOS was completed for the majority of consumers on an electronic case register in the UK. In a trial in New Zealand, Eagar, Trauer and Mellsop [20] found that 95% of episodes of care had at least one HoNOS completed (and that the majority had few missing items), although only 58% had one completed at the beginning and the end of the episode.
Reports of clinicians' experiences with using the HoNOS have been more mixed. James and Kehoe [58], Broadbent [41] and Milne et al [59] found that UK clinicians were relatively positive about the HoNOS, viewing it as potentially useful, but insisting that its ongoing use would depend on adequate resourcing, infrastructure, training and feeback. By contrast, Gilbody [54] found that many UK psychiatrists questioned the instrument's usefulness. In field trials conducted in Australia, Trauer [60] found that clinicians at one site were extremely positive about the HoNOS, whereas those at four others were more ambivalent, believing that it contributed only minimally to their treatment practices.
HoNOSCA
Content validity
No studies available.
Construct validity
Gowers et al [3,4] and Harnett et al [61] examined the internal structure of the HoNOSCA during its development, considering both individual items and subscales. They considered the correlations between the individual items and found them to be low, which they took as evidence that each item carried independent weight. They then examined the factor structure of the HoNOSCA, and found that it generally mirrored the instrument's subscales. Brann [62], by contrast, also examined the factor structure of the HoNOSCA and produced preliminary evidence for a different set of factors. Neither Gowers et al nor Brann found support for the instrument's sections.
Gowers et al [3,4] also considered the extent to which the HoNOSCA total score accurately reflected clinical severity, arguing that high total scores should more frequently be associated with high scores on a few items than on mild to moderate scores on a number of items. They found that the total score increased as a linear function of high individual item scores, a finding confirmed by Brann et al [63] in a subsequent study.
Concurrent validity
Several studies have weighed up the HoNOSCA's performance against other measures. Studies that have examined the correlation between the HoNOSCA total score and scores on other clinician-rated measures have typically reported moderate correlations (r = 0.6 or above). This was the case when the HoNOSCA was compared with the Children's Global Assessment Scale [64], the Paddington Complexity Scale [61,64], and the Global Assessment of Psychosocial Disability [65].
Studies that have evaluated the HoNOSCA against parent- and child/adolescent-rated instruments have typically produced lower correlations. Yates et al [64] found only modest correlations between the HoNOSCA and the Behaviour Check List, Strengths and Difficulties Questionnaire, Child Health Related Quality of Life Questionnaires and Modified Harter Self-Esteem Questionnaire. Gowers et al [66] found overall low levels of agreement between the HoNOSCA and the HoNOSCA-SR (a consumer-rated version of the instrument for adolescents) at an individual level, although some groups (e.g., outpatients with eating disorders) were exceptions. Again, these findings are to be expected, given that instruments that rely on information from different classes of informants are likely to demonstrate lower levels of correspondence than those that rely on informants from the same class.
Other studies have assessed the ability of the HoNOSCA to discriminate between groups of consumers based on their clinical and/or treatment profile. Gowers et al [3,4] and Yates et al [64] found that the HoNOSCA could distinguish between consumers in inpatient and outpatient settings and between consumers presenting to clinics with different areas of focus, respectively. Harnett et al [61] found that HoNOSCA total scores were associated with the number of critical incidents in which adolescent consumers were involved. Manderson and McCune [67], Brann et al [63] and Harnett et al [61] found that the HoNOSCA yielded coherent age/sex results – e.g., boys scored higher than girls on Item 1 (Problems with disruptive, antisocial or aggressive behaviour) but lower on Item 9 (Problems with emotional and related symptoms), and younger children scored higher than older children on Item 5 (Problems with scholastic or language skills) but lower on Item 3 (Non-accidental self-injury). Brann et al [63] also reported that the HoNOSCA yielded intuitive results when they considered diagnosis – e.g., consumers with attention deficit and conduct disorders scored highest on Items 1 and 2 (Problems with disruptive, antisocial or aggressive behaviour, and Problems with over-activity, attention or concentration). Similarly, Bilenberg [65] found that high HoNOSCA total scores were associated with comorbidity.
Predictive validity
Brann [62] found that HoNOSCA total scores at community assessment could discriminate between adolescents who later received treatment from intensive outreach teams and their counterparts who progressed to other forms of community care.
Test-retest reliability
There are few published studies on the test-retest reliability of the HoNOSCA, and those which do exist are arguably studies of the sensitivity to change (or lack of change) of the instrument, since they cover considerable time periods and consider stability in relation to other measures. Garralda et al [68] examined the test-retest reliability of the instrument over a six month period, for consumers for whom clinicians indicated there had been no change on a global rating scale, and reported a figure of 0.69. Similarly, Brann [62] reported correlations of 0.80 over three months and 0.76 over five months when he examined the instrument's test-retest reliability, again in a group of consumers who were judged not to have changed over the given period. Likewise, Harnett et al [61] reported a correlation of 0.80 between initial and subsequent HoNOSCA total scores assessed over a 2–4 week period for inpatient adolescents, whom the authors suggested would be likely to remain relatively stable after a 'settling in' period.
Inter-rater reliability
Studies have consistently found that the majority of Section A items demonstrate good or very good inter-rater reliability. However, there is less agreement about which items perform poorly. For example, Brann et al [63] reported a particularly low intra-class correlation (0.06) for Item 10 (Problems with peer relationships), but Gowers et al [3,4] found that this item achieved an intra-class correlation of 0.77.
There is also debate about the inter-rater reliability of Section B. Gowers et al [3,4] found that the two items comprising this section each had good inter-rater reliability: Item 14 (Problems with knowledge or understanding about the nature of the child or adolescent's difficulties) and Item 15 (problems with lack of information about services or management of the child or adolescent's difficulties) had intra-class correlations of 0.73 and 0.78, respectively. By contrast, the equivalent figures in a later study by Garralda et al [69] were 0.27 and 0.03.
Sensitivity to change
Three approaches have been taken to assessing the ability of the HoNOSCA to detect change. The first and methodologically weakest approach involves simply determining whether HoNOSCA total scores change over time, with no reference to whether this reflects real change. In the original field work associated with the development of the HoNOS, for example, Gowers et al [3,4] noted that 'the HoNOSCA demonstrated satisfactory sensitivity to change, with a mean overall reduction in total scores of 38% between rating points, on average nearly three months apart'. Manderson and McCune [67] made a similar observation, as did Harnett et al [61].
The second approach examines the correspondence between change as assessed by the HoNOSCA and change as defined by the difference between scores on other measures. Studies by Gowers et al [66], Garralda et al [68] and Bilenberg [65] have reported changes in HoNOSCA total scores that are comparable in direction and magnitude with other clinician-rated measures, such as the Children's Global Assessment Scale and the Global Assessment of Psychosocial Disability, and, to a lesser extent with parent- and/or consumer-rated measures such as the HoNOSCA-SR, the Behaviour Check List and the Strengths and Difficulties Questionnaire.
The third approach uses global outcome judgements as the 'gold standard'. Typically, these require clinicians (or parents/referrers) to indicate whether the consumer has improved, deteriorated or remained stable, via some sort of Likert scale. Studies by Gowers et al [3,4], Garralda et al [68], Brann et al [62,63] and Bilenberg [65] have all reported close correspondence between change (or lack of change) recorded on the HoNOSCA and such global judgements.
Feasibility/utility
Studies that have questioned clinicians about the feasibility/utility of the HoNOSCA have generally found them to be positive about its brevity and ease of use, its clinical utility, and its ability to be incorporated into routine practice. Their main concerns have related to the instrument's applicability to children aged under five, its emphasis on child/adolescent symptoms and functioning, and its failure to take into account context. Some clinicians have also questioned whether it may be less useful in the case of particular disorders [3,4,65,67,69].
These and other studies have further considered feasibility/utility by examining the behaviour of services and individual clinicians. For example, Gowers et al [3,4] reported that in the original HoNOSCA field trial none of the sites dropped out and 71% of consumers were rated at both Time 1 and Time 2. They continued to report optimal completion rates in their later work [4].
HoNOS65+
Content validity
During initial HoNOS65+ development, Burns et al [5] asked mental health professionals working with older consumers to review the content of the HoNOS. This process resulted in modifications to the glossary to address their concerns regarding the comprehensiveness of the instrument for older consumers [70]. Since this time, ongoing issues have been noted anecdotally, and further refinements to the glossary have been made [71-73].
Construct validity
There is a paucity of evidence on the construct validity of the HoNOS65+. The only relevant data come from the original pilot work by Burns et al [70], where a factor analysis revealed that four factors accounted for 57.4% of the variance in HoNOS65+ item scores.
Concurrent validity
Studies by Burns et al [70], Mozley et al [74], Spear et al [75] and Bagley et al [76] have examined the correlations between the HoNOS65+ and more established clinician-rated measures that assess similar domains. Reasonable correlations have been observed between the HoNOS65+ total score and the Mini-Mental State Examination [70,74,75], Crighton Royal Behaviour Rating Scale [70], and Barthel Activities of Daily Living Index [70].
As a general rule, however, stronger correlations have been observed between specific HoNOS65+ items and other instruments:
• Item 4 (Cognitive problems) with the Mini-Mental State Examination [70,75];
• Item 6 (Problems associated with hallucinations and delusions), Item 7 (Problems with depressive symptoms), Item 8 (Other mental and behavioural problems) and Item 9 (Problems with relationships) with the Brief Psychiatric Rating Scale [70];
• Item 4 (Cognitive problems), Item 5 (Physical illness or disability problems) and Item 12 (Problems with occupation and activities) with the Barthel Activities of Daily Living Index [70];
• Item 1 (Behavioural disturbance), Item 4 (Cognitive problems), Item 5 (Physical illness or disability problems), Item 7 (Problems with depressive symptoms), Item 8 (Other mental and behavioural problems), Item 10 (Problems with activities of daily living), Item 11 (Problems with living conditions) and Item 12 (Problems with occupation and activities) with the Crighton Royal Behaviour Rating Scale [70]; and
• Item 1 (Behavioural disturbance), Item 4 (Cognitive problems), Item 9 (Problems with relationships) with the Brief Agitation Rating Scale [75].
There are exceptions, however. Equivocal findings have been reported regarding the relationship between HoNOS65+ Item 7 (Problems with depressive symptoms) and the Geriatric Depression Scale. The original pilot found the correlations between Item 7 and individual items on the Geriatric Depression Scale were good, but that there was no significant correlation between it and the total score [70]. Later studies have produced conflicting results, with one finding a good correlation between Item 7 and the Geriatric Depression Scale [75] and the other finding that the former detected only a minority of the consumers identified as depressed by the latter [76].
A few studies have investigated the ability of the HoNOS65+ to discriminate between different consumer groups. Burns et al [70] found the instrument was able to discriminate between consumers with dementia and those with functional psychiatric disorders, with the former scoring higher on Item 1 (Behavioural disturbance), Item 4 (Cognitive problems) and Item 10 (Problems with activities of daily living), and the latter scoring higher on Item 2 (Non-accidental self injury), Item 7 (Problems with depressive symptoms), Item 8 (Other mental and behavioural problems). Spear et al [75] reported similar findings, demonstrating that consumers with dementia generally had higher HoNOS65+ total scores than those with mood disorders, but had lower scores on the symptoms subscale.
Predictive validity
No studies available.
Test-retest reliability
No studies available.
Inter-rater reliability
Burns et al [70] and Spear et al [75] both found inter-rater reliability to be good to very good for most items. Burns et al found that only Item 2 (Non-accidental self-injury), Item 10 (Problems with activities of daily living), Item 11 (Problems with living conditions) and Item 12 (Problems with occupation and activities) did not consistently perform well. In Spear et al's study, Item 4 (Cognitive problems), Item 5 (Physical illness or disability problems) and Item 9 (Problems with relationships) demonstrated only poor to moderate inter-rater reliability. Allen et al [71], by contrast, found problems with a broader range of items, largely related to difficulties in interpretation.
Sensitivity to change
Spear et al [75] found that consumers showed improvement on all HoNOS65+ subscales and on the HoNOS65+ total score between assessment and discharge from inpatient and community services, and that the discharge HoNOS65+ total score and the change in HoNOS65+ total scores showed moderate but significant correlations with the Clinician's Interview Based Impression of Change Scale.
Feasibility/utility
In the original pilot, Burns et al [70] assessed the feasibility/utility of the HoNOS65+ by asking raters whether or not they would find the instrument helpful in working with individual consumers; 39% indicated it would be very useful and 50% that it would be of some use. Spear et al [75] reported similar findings. In both studies, almost all respondents reported that it was easy to administer.
Feasibility/utility have also been considered in terms of uptake, both at a national level and at a service level. Reilly et al [77] conducted a survey of old age psychiatrists across the UK, and found that 18% reported that the HoNOS65+ was being used in their service. Spear et al examined the proportion of episodes of care at which the HoNOS65+ was administered within a single service, and found completion rates of 96%.
Other studies have examined the feasibility/utility of the HoNOS65+ more generally, considering issues that have arisen during implementation. Allen et al [71], for example, observed that clinical leadership and timely feedback were crucial, as were minimising the paperwork burden and clarifying analysis and reporting issues. In a similar vein, MacDonald [78] argued that suitable infrastructure must be in place, the data must be managed appropriately, and analysis and reporting should be guided by clinicians' requirements.
Discussion
Table 2 summarises the review's findings. Mostly, the members of the HoNOS family have adequate or good validity, reliability, sensitivity to change and feasibility/utility. That said, some of the psychometric properties of the instruments are under-investigated and therefore warrant closer examination. There may also be scope for additional work on particular psychometric properties, even where some studies have already been conducted, given that the instruments are being used in the context of routine outcome measurement – e.g., inter-rater reliability (given that a number of raters may be involved in administering measures for the same consumer) and sensitivity to change (given that outcome measurement requires a valid and reliable assessment of improvement, deterioration or stability over time).
Table 2 Psychometric properties of the HoNOS family of measures
HoNOS HoNOSCA HoNOS65+
Validity Content Good Insufficient evidence Insufficient evidence
Construct Good Adequate Insufficient evidence
Concurrent Good Good Good
Predictive Good Adequate Insufficient evidence
Reliability Test-retest Adequate Adequate Insufficient evidence
Inter-rater Adequate Adequate Good
Sensitivity to change Adequate Good Adequate
Feasibility and utility Adequate Adequate Adequate
One caveat should be considered when interpreting these findings. The majority of studies considered in the review examined the psychometric properties of the original instruments, used as per standard instructions. It must be acknowledged that various modifications have been made to the instruments, to cater for the local context. So, for example, in Australia when the instruments are being used at discharge from an acute inpatient setting, the rating period is the last three days rather than the last two weeks (in recognition of the brevity of such admissions). As yet, no formal psychometric testing has been applied to the modified instruments, and there is a question about the extent to which the findings as they relate to the standard instruments can be generalised.
Conclusion
This caveat aside, it can be concluded that that, collectively, the HoNOS family of measures can assess outcomes for different groups on a range of mental health-related constructs. Where tested, their psychometric performance is adequate or better. This is important, because it means they can be regarded as appropriate for routinely monitoring consumer outcomes, with a view to improving treatment quality and effectiveness.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
JP, PB and TC devised the conceptual framework for the review. JP, PB, PK and MW identified and retrieved all references. JP, PK, SD and MW extracted relevant information from the references, reviewed the measures, and drafted the report upon which the paper is based. All authors contributed to drafting and re-drafting the paper.
Acknowledgements
The authors would like to acknowledge Alan Morris-Yates, Bill Buckingham and the members of the Information Strategy Committee Expert Groups who provided comments on the report upon which this paper is based. They would also like to thank Mike Slade for commenting on an earlier draft of the paper.
==== Refs
Wing JK Beevor AS Curtis RH Park SB Hadden S Burns A Health of the Nation Outcome Scales (HoNOS). Research and development British Journal of Psychiatry 1998 172 11 18 9534825
Wing JK Lelliott P Beevor AS Progress on HoNOS British Journal of Psychiatry 2000 176 392 393 10827890 10.1192/bjp.176.4.392
Gowers SG Harrington RC Whitton A Lelliott P Beevor A Wing J Jezzard R Brief scale for measuring the outcomes of emotional and behavioural disorders in children. Health of the Nation Outcome Scales for children and Adolescents (HoNOSCA) British Journal of Psychiatry 1999 174 413 416 10616607
Gowers S Bailey-Rogers SJ Shore A Levine W The Health of the Nation Outcome Scales for Child and Adolescent Mental Health (HoNOSCA) Child Psychology and Psychiatry Review 2000 5 50 56 10.1017/S1360641700002148
Burns A Beevor A Lelliott P Wing J Blakey A Orrell M Mulinga J Hadden S Health of the Nation Outcome Scales for elderly people (HoNOS 65+). Glossary for HoNOS 65+ score sheet British Journal of Psychiatry 1999 174 435 438 10616612
Royal College of Psychiatrists http://www.rcpsych.ac.uk/cru/honoscales/what.htm
Pirkis J Burgess P Coombs T Clarke A Jones-Ellis D Dickson R Routine measurement of outcomes in Australian public sector mental health services Australia and New Zealand Health Policy 2005 2 8 15840170 10.1186/1743-8462-2-8
Turner S Are the health of the Nation Outcome Scales (HoNOS) useful for measuring outcomes in older people's mental health services? Ageing and Mental Health 2004 8 387 396 10.1080/13607860410001725063
Cook DJ Mulrow CD Haynes RB Systematic reviews: Synthesis of best evidence for clinical decisions Annals of Internal Medicine 1997 126 376 380 9054282
Greenhalgh J Long AF Brettle AJ Grant MJ Reviewing and selecting outcome measures for use in routine practice Journal of Evaluation in Clinical Practice 1998 4 339 350 9927249
Green RS Gracely EJ Selecting a rating scale for evaluating services to the chronically mentally ill Community Mental Health Journal 1987 23 91 102 3652669 10.1007/BF00757163
McDowell I Newell C Measuring Health: A Guide to Rating Scales and Questionnaires 1996 Oxford, Oxford University Press
Chronbach LJ Meehl PE Construct validity in psychological tests Psychological Bulletin 1955 52 281 302 13245896
Shergill SS Shankar KK Seneviratna K Orrell MW The validity and reliability of the Health of the Nation Outcome Scales (HoNOS) in the elderly Journal of Mental Health (UK) 1999 8 511 521
Orrell M Yard P Handysides J Schapira R Validity and reliability of the Health of the Nation Outcome Scales in psychiatric patients in the community British Journal of Psychiatry 1999 174 409 412 10616606
McClelland R Trimble P Fox ML Stevenson MR Bell B Validation of an outcome scale for use in adult psychiatric practice Quality in Health Care 2000 9 98 105 11067258 10.1136/qhc.9.2.98
Stedman T Yellowlees P Mellsop G Clarke R Drake S Measuring Consumer Outcomes In Mental Health: Field Testing of Selected Measures of Consumer Outcome in Mental Health 1997 Canberra, Department of Health and Family Services
Trauer T The subscale structure of the Health of the Nation Outcome Scales (HoNOS) Journal of Mental Health (UK) 1999 8 499 509 10.1080/09638239917193
Page AC Hooke GR Rutherford EM Measuring mental health outcomes in a private psychiatric clinic: Health of the Nation Outcome Scales and Medical Outcomes Short Form SF-36 Australian and New Zealand Journal Psychiatry 2001 35 377 381 10.1046/j.1440-1614.2001.00908.x
Eagar K Trauer T Mellsop G Performance of routine outcome measures in adult mental health care Australian and New Zealand Journal of Psychiatry 2005 39 713 718 16050925 10.1111/j.1440-1614.2005.01655.x
Trauer T Comment Australian and New Zealand Journal of Psychiatry 2000 34 520 521 10.1046/j.1440-1614.2000.00757.x
Preston NJ The Health of the Nation Outcome Scales: Validating factorial structure and invariance across two health services Australian and New Zealand Journal Psychiatry 2000 34 512 519 10.1046/j.1440-1614.2000.00726.x
Parker G O'Donnell M Hadzi-Pavlovic D Proberts M Assessing outcome in community mental health patients: A comparative analysis of measures International Journal of Social Psychiatry 2002 48 11 19 12008903 10.1177/002076402128783046
Browne S Doran M McGauran S Health of the Nation Outcome Scales (HoNOS): Use in an Irish psychiatric outpatient population Irish Journal of Psychological Medicine 2000 17 17 19
Amin S Singh SP Croudace T Jones P Medley I Harrison G Evaluating the Health of the Nation Outcome Scales. Reliability and validity in a three-year follow-up of first-onset psychosis British Journal of Psychiatry 1999 174 399 403 10616604
Rees A Richards A Shapiro DA Utility of the HoNOS in measuring change in a community mental health care population Journal of Mental Health 2004 13 295 304 10.1080/09638230410001700925
Bebbington P Brugha T Hill T Marsden L Window S Validation of the Health of the Nation Outcome Scales British Journal of Psychiatry 1999 174 389 394 10616602
Hope JD Trauer T Keks NA Reliability, validity and utility of the Health of the Nation Outcomes Scale (HoNOS) in Australian adult psychiatric services Schizophrenia Research 1998 29 9 10 10.1016/S0920-9964(97)88307-7
Adams M Palmer A O'Brien JT Crook W Health of the Nation Outcome Scales for psychiatry: Are they valid? Journal of Mental Health 2000 9 193 198 10.1080/09638230050009186
Brooks R The reliability and validity of the Health of the Nation Outcome Scales: Validation in relation to patient derived measures Australian and New Zealand Journal Psychiatry 2000 34 504 511 10.1046/j.1440-1614.2000.00755.x
Issakidis C Teesson M Measurement of need for care: A trial of the Camberwell Assessment of Need and the Health of the Nation Outcome Scales Australian and New Zealand Journal Psychiatry 1999 33 754 759 10.1046/j.1440-1614.1999.00598.x
Hunter R McLean J Peck D Pullen I Greenfield A McArthur W Quinn C Eaglesham J Hagen S Norrie J The Scottish 700 Outcomes Study: A comparative evaluation of the Health of the Nation Outcome Scale (HoNOS), the Avon Mental Health Measure (AVON), and an Idiographic Scale (OPUS) in adult mental health Journal of Mental Health 2004 13 93 105 10.1080/09638230410001654594
Trauer T Callaly T Concordance between mentally ill patients and their case managers using the Health of the Nation Outcome Scales (HoNOS) Australasian Psychiatry 2002 10 24 28 10.1046/j.1440-1665.2002.00387.x
Slade M Beck A Bindman J Thornicroft G Wright S Routine clinical outcome measures for patients with severe mental illness: CANSAS and HoNOS British Journal of Psychiatry 1999 174 404 408 10616605
Salvi G Leese M Slade M Routine use of mental health outcome assessments: Choosing the measure British Journal of Psychiatry 2005 186 146 152 15684239 10.1192/bjp.186.2.146
Craig RJ Measures for mental health outcomes British Journal of Psychiatry 2005 187 90 91 15994580 10.1192/bjp.187.1.90-a
Bech P Bille J Schutze T Sondergaard S Wiese M Waarst S Health of the Nation Outcome Scales (HoNOS): Implementability, subscale structure and responsiveness in the daily psychiatric hospital routine over the first 18 months Nordic Journal of Psychiatry 2003 57 285 290 12888403 10.1080/08039480310002156
Bonsack C Borgeat F Lesage A Measuring patients' problems severity and outcomes in a psychiatric sector: A field study with the French version of the Health of Nation Outcome Scales (HoNOS-F)/Mesurer la severite des problemes des patients et leur evolution dans un secteur psychiatrique: Une etude sur le terrain du Health of Nation Outcome Scales en francais: (HoNOS-F) Annales Medico-Psychologiques 2002 160 483 488 10.1016/S0003-4487(02)00208-1
Gallagher J Teesson M Measuring disability, need and outcome in Australian community mental health services Australian and New Zealand Journal Psychiatry 2000 34 850 855 10.1046/j.1440-1614.2000.00815.x
Allan S McGonagle I A comparison of HoNOS with the Social Behaviour Schedule in three settings Journal of Mental Health 1997 6 117 124 10.1080/09638239718888
Broadbent M Reconciling the information needs of clinicians, managers and commissioners: A pilot project Psychiatric Bulletin 2001 25 423 425 10.1192/pb.25.11.423
Schneider J Wooff D Carpenter J Brandon T McNiven F Service organisation, service use and costs of community mental health care Journal of Mental Health Policy and Economics 2002 5 79 87 12529563
Ashaye K Seneviratna K Shergill S Orrell M Do the Health of the Nation Outcome Scales predict outcome in the elderly mentally ill? A 1-year follow-up study Journal of Mental Health (UK) 1999 8 615 620
Goldney RD Fisher LJ Walmsley SH The Health of the Nation Outcome Scales in psychiatric hospitalisation: A multicentre study examining outcome and prediction of length of stay Australian and New Zealand Journal Psychiatry 1998 32 199 205
Boot B Hall W Andrews G Disability, outcome and case-mix in acute psychiatric in-patient units British Journal of Psychiatry 1997 171 242 246 9337977
Trauer T Callaly T Hantz P Little J Shields R Smith J Health of the Nation Outcome Scales. Results of the Victorian field trial British Journal of Psychiatry 1999 174 380 388 10616601
Goldney RD Fisher LJ Walmsley SH A pilot study of the Health of the Nation Outcome Scales as a measurement of outcome in a private psychiatric hospital Australasian Psychiatry 1996 4 319 321
Audin K Margison FR Clark JM Barkham M Value of HoNOS in assessing patient change in NHS psychotherapy and psychological treatment services British Journal of Psychiatry 2001 178 561 566 11388975 10.1192/bjp.178.6.561
Parabiaghi A Barbato A D'Avanzo B Erlicher A Lora A Assessing reliable and clinically significant change on the Health of the Nation Outcome Scales: Method for displaying longitudinal data Australian and New Zealand Journal of Psychiatry 2005 39 719 725 16050926 10.1111/j.1440-1614.2005.01656.x
Taylor JR Wilkinson G HoNOS v. GP opinion in a shifted out-patient clinic Psychiatric Bulletin 1997 21 483 485
Sharma VK Wilkinson G Fear S Health of the Nation Outcome Scales: A case study in general psychiatry British Journal of Psychiatry 1999 174 395 398 10616603
Stafrace S Doubts about HoNOS Australian and New Zealand Journal Psychiatry 2002 36 270 10.1046/j.1440-1614.2002.t01-5-01016.x
Stein GS Usefulness of the Health of the Nation Outcome Scales British Journal of Psychiatry 1999 174 375 377 10616599
Gilbody SM House AO Sheldon TA Psychiatrists in the UK do not use outcomes measures: National survey British Journal of Psychiatry 2002 180 101 103 11823316 10.1192/bjp.180.2.101
Andrews G Page AC Outcome measurement, outcome management and monitoring Australian and New Zealand Journal of Psychiatry 2005 39 649 651 16050918 10.1111/j.1440-1614.2005.01648.x
Ashaye O Mathew G Dhadphale M A comparison of older longstay psychiatric and learning disability inpatients using the Health of the Nation Outcome Scales International Journal of Geriatric Psychiatry 1997 12 548 552 9193963 10.1002/(SICI)1099-1166(199705)12:5<548::AID-GPS543>3.0.CO;2-S
Glover GR Sinclair Smith H Computerised information systems in English mental health care providers in 1998 Social Psychiatry and Psychiatric Epidemiology 2000 35 518 522 11197928 10.1007/s001270050274
James M Kehoe R Using the Health of the Nation Outcome Scales in clinical practice Psychiatric Bulletin 1999 23 536 538
Milne D Reichelt K Wood EI Implementing HoNOS: An eight stage approach Clinical Psychology and Psychotherapy 2001 8 106 116 10.1002/cpp.252
Trauer T The Health of the Nation Outcome Scales in outcome measurement: A critical review Australasian Psychiatry 1998 6 11 14
Harnett PH Loxton NJ Sadler T Hides L Baldwin A The Health of the Nation Outcome Scales for Children and Adolescents in an adolescent inpatient sample Australian and New Zealand Journal of Psychiatry 2005 39 129 135 15701060 10.1111/j.1440-1614.2005.01533.x
Brann P Routine Outcome Measurement in Child/Adolescent Mental Health: HoNOSCA - Valid Enough? Feasible Enough? Melbourne, Monash University Forthcoming
Brann P Coleman G Luk E Routine outcome measurement in a child and adolescent mental health service: An evaluation of HoNOSCA Australian and New Zealand Journal of Psychiatry 2001 35 370 376 11437812 10.1046/j.1440-1614.2001.00890.x
Yates P Garralda ME Higginson I Paddington Complexity Scale and Health of the Nation Outcome Scales for Children and Adolescents British Journal of Psychiatry 1999 174 417 423 10616608
Bilenberg N Health of the Nation Outcome Scales for Children and Adolescents (HoNOSCA): Results of a Danish field trial European Child and Adolescent Psychiatry 2003 12 298 302 14689262 10.1007/s00787-003-0343-1
Gowers S Levine W Bailey-Rogers S Shore A Burhouse E Use of a routine, self-report outcome measure (HoNOSCA-SR) in two adolescent mental services British Journal of Psychiatry 2002 180 266 269 11872520 10.1192/bjp.180.3.266
Manderson J McCune N The use of HoNOSCA in a child and adolescent mental health service Irish Journal of Psychological Medicine 2003 20 52 55
Garralda ME Yates P Higginson I Child and adolescent mental health service use. HoNOSCA as an outcome measure British Journal of Psychiatry 2000 177 52 58 10945089 10.1192/bjp.177.1.52
Garralda E Yates P HoNOSCA: Uses and limitations Child Psychology and Psychiatry Review 2000 5 131 132 10.1017/S1360641700002306
Burns A Beevor A Lelliott P Wing J Blakey A Orrell M Mulinga J Hadden S Health of the Nation Outcome Scales for elderly people (HoNOS 65+) British Journal of Psychiatry 1999 174 424 427 10616609
Allen L Bala S Carthew R Daley S Doyle E Driscoll P Grey B Macdonald A Experience and application of HoNOS65+ Psychiatric Bulletin 1999 23 203 206
Macdonald AJ HoNOS 65+ glossary British Journal of Psychiatry 1999 175 192 10627814
College Research Unit HoNOS 65+: A Tabulated Glossary for Use with HoNOS65+ (Version 3) 2002 London, Royal College of Psychiatrists
Mozley CG Huxley P Sutcliffe C Bagley H Burns A Challis D Cordingley L 'Not knowing where I am doesn't mean I don't know what I like': Cognitive impairment and quality of life responses in elderly people International Journal of Geriatric Psychiatry 1999 14 776 783 10479750 10.1002/(SICI)1099-1166(199909)14:9<776::AID-GPS13>3.0.CO;2-C
Spear J Chawla S O'Reilly M Rock D Does the HoNOS 65+ meet the criteria for a clinical outcome indicator for mental health services for older people? International Journal of Geriatric Psychiatry 2002 17 226 230 11921150 10.1002/gps.592
Bagley H Cordingley L Burns A Mozley CG Sutcliffe C Challis D Huxley P Recognition of depression by staff in nursing and residential homes Journal of Clinical Nursing 2000 9 445 450 11235320 10.1046/j.1365-2702.2000.00390.x
Reilly S Challis D Burns A Hughes J The use of assessment scales in Old Age Psychiatry Services in England and Northern Ireland Aging Mental Health 2004 8 249 255 15203406 10.1080/13607860410001669787
Macdonald AJD The usefulness of aggregate routine clinical outcomes data: The example of HoNOS65+ Journal of Mental Health (UK) 2002 11 645 656 10.1080/09638230020023
|
16313678
|
PMC1315350
|
CC BY
|
2021-01-04 16:38:13
|
no
|
Health Qual Life Outcomes. 2005 Nov 28; 3:76
|
utf-8
|
Health Qual Life Outcomes
| 2,005 |
10.1186/1477-7525-3-76
|
oa_comm
|
==== Front
Head Face MedHead & Face Medicine1746-160XBioMed Central London 1746-160X-1-121631646410.1186/1746-160X-1-12ResearchComparison of laboratory and immediate diagnosis of coagulation for patients under oral anticoagulation therapy before dental surgery Kruse-Loesler Birgit [email protected] Matthias [email protected] Johannes [email protected] Department of Cranio-Maxillofacial Surgery, University of Muenster, Waldeyerstr. 30, D-48149 Muenster, Germany2005 29 11 2005 1 12 12 24 8 2005 29 11 2005 Copyright © 2005 Kruse-Loesler et al; licensee BioMed Central Ltd.2005Kruse-Loesler 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
Dental surgery can be carried out on patients under oral anticoagulation therapy by using haemostyptic measures. The aim of the study was a comparative analysis of coagulation by laboratory methods and immediate patient diagnosis on the day of the planned procedure.
Methods
On the planned day of treatment, diagnoses were carried out on 298 patients for Prothrombin Time (PT), the International Normalised Ratio (INR), and Partial Thromboplastin Time (PTT). The decision to proceed with treatment was made with an INR < 4.0 according to laboratory results.
Results
Planned treatment did not go ahead in 2.7% of cases. Postoperatively, 14.8% resulted in secondary bleeding, but were able to be treated as out-patients. 1.7% had to be treated as in-patients. The average error between the immediate diagnosis and the laboratory method: 95% confidence interval was -5.8 ± 15.2% for PT, -2.7 ± 17.9 s for PTT and 0.23 ± 0.80 for INR. The limits for concordance were 9.4 and -21.1% for PT, 15.2 and -20.5 s for PTT, and 1.03 and -0.57 for INR.
Conclusion
This study showed a clinically acceptable concordance between laboratory and immediate diagnosis for INR. Concordance for PT and PTT did not meet clinical requirements. For patients under oral anticoagulation therapy, patient INR diagnosis enabled optimisation of the treatment procedure when planning dental surgery.
==== Body
Background
To prevent thrombosis and embolism, patients are treated increasingly as out-patients and are sometimes given anticoagulant therapy for many years. The anticoagulant with coumarin derivatives is widely distributed. As vitamin K antagonists, these derivatives inhibit the g-carboxylation of glutamic acids in the synthesis of coagulation factors II, VII, IX and X in the liver. With regards to comparability and standardisation of test results for oral anticoagulation therapy, the International Normalised Ratio (INR) is recommended by WHO for monitoring of patients' coagulation physiology. INR values between 2.0 and 3.0 are recommended for thromboembolic illnesses, atrial fibrillation, heart valve diseases and myocardial infarct. For mechanical heart valves and recurring embolism, an INR value between 2.5 and 3.5 is set.
Haemorrhages, which can occur after teeth extractions, for example, only present a significant risk for patients under anticoagulation therapy in exceptional cases [7]. In contrast, stopping anticoagulation therapy independently before dental surgery procedures can present patients with unnecessary life-threatening risks from thromboembolism [27]. With INR values for therapy between 2.0 and 3.5, extractions of one or more teeth and uncomplicated osteotomies, taking into account relevant local haemostasis methods without danger of haemorrhaging, are possible [12], whereby the INR value is to be determined pre-operatively on the day of operation [21]. For immediate patient diagnosis of the coagulation status, instruments have been developed which are used in operating theatres and in intensive care [18]. The same systems are used by patients to self-monitor oral anticoagulation [1,17]. The aim of this study was a comparative analysis of coagulation and treatment planning for immediate diagnosis of patients and diagnosis in the main laboratory.
Methods
Patients
A total of 298 patients who received dental treatment under anticoagulation therapy were included in this study. The period of acquisition reached from 10.10.2002 to 16.04.2004. The group tested consisted of 106 women and 192 men with an age range of 11–91 years and an average age of 60.1 years. Anticoagulation treatment was carried out after thromboembolic illnesses (30.5%), atrial fibrillation (27.4%), myocardial infarct (15.1%) and heart valve replacement (34.0%). Coumarin therapy was carried out for 92.4% of patients. 6.6% were treated under general anaesthesia and oral anticoagulation was administered by intravenous heparin. Additional inhibition of thrombocyte aggregation by acetylsalicylic acid or ticlopidine occurred for 2.7%. The decision to proceed with treatment was made with an INR < 4.0 according to laboratory results [2]. 83.0% of patients underwent surgery, where predominantly a single tooth was extracted and small osteotomies were carried out. 28.3% received preservation treatment, 9.4% prosthetic treatment and 10.4% periodontal treatment. To protect against an increased risk of intra- and postoperative complications with cardiovascular diseases, 56.7% were administered intravenously and 39.7% were put under ECG monitoring. Endocarditis prophylaxis was carried out for 43.4% in accordance with the guidelines of the German Society for Heart Diseases.
Local haemostyptic measures consisted of sealing the thick wound in a collagen dressing (66.0%), a fibrin adhesive (27.4%), and the use of a protective plate (18.9%). A local flap coverage was used in 15.1% of cases.
Examination methods
Patients' intra- and postoperative trends were recorded and complications were documented. Before treatment, capillary blood was extracted. The diagnosis for Prothrombin Time (PT), International Normalised Ration (INR) and Partial Thromboplastin Time (PTT) was carried out with a CoaguChek Pro device (Roche Diagnostics, Mannheim, Germany). CoaguChek Pro PT cassettes contain thromboplastin from rabbit brains (ISI 2.04). As a PTT reagent, cow brain sulphatide was used as an activator, and soya phosphatide as a platelet substitute. The measuring principle of the machine is based on laser-photometric detection of erythrocyte movement, which is suspended when coagulation begins to set in. Measurement and quality control were carried out, in accordance with the manufacturer's instructions, by 3 experienced assistant medical technicians. At the same time, blood tests were taken in the coagulation tubes (0.106 mol/l citrate, 10%) and analysed in the clinic's main laboratory. Both the PT and PTT diagnoses were carried out with a BCS coagulation analyser (Dade Behring, Marburg, Germany). Thromborel S (tissue factor from a human placenta, ISI 1.09) and Pathromtin SL (Dade Behring, Marburg, Germany) were used as reagents.
Analysis and statistics
The test results were analysed by a thorough data analysis with determination of average values and standard deviations. As a cohesive measurement between the parameters, the correlation coefficient r with the corresponding test (HO: r = 0) was used. With p values < 0.01 the correlation was accepted. Selected parameter combinations were depicted in range diagrams with the corresponding regression line. The concordance between measurement methods was analysed in accordance with Bland and Altman [3]. Average and relative errors and the absolute and relative limits of concordance were calculated and depicted in diagram form. When comparing methods, a relative error of ± 10% and relative limits of concordance at ± 25% were assessed to be clinically acceptable [19].
Results
Planned treatment did not go ahead in 2.7% of cases. 2.3% also had an INR value > 4.0 in immediate diagnosis. 2.3% of immediate diagnoses were in the same measurement range; laboratory diagnoses gave an INR value > 4.0. Intra-operatively, there was the complication of one major haemorrhage. Postoperatively, 14.8% resulted in secondary bleeding, but were able to be treated as out-patients. 1.7% had to be treated as in-patients.
No significant difference was determined in coagulation values for patients who did and did not suffer secondary bleeding (Fig. 1). On analysis of postoperative secondary bleeding within the category of the "additional operative measures" variable, a high proportion (33.3%) was treated with a local flap reconstruction (Tab. 1).
Figure 1 Immediate (CoaguChek) and laboratory diagnosis of INR concerning secondary bleeding.
Table 1 Frequency of secondary bleeding concerning additional surgical treatment options
no bleeding secondary bleeding
suture 88.9% 11.1%
collagen dressing 92.3% 7.7%
fibrin adhaesive 85.0% 15.0%
local flap coverage 66.7% 33.3%
Fig. 2 shows the range diagram of PT diagnoses from CoaguChek and the laboratory with linear regression and confidence interval. The results for PTT and INR are shown in the same way in Fig. 3 and Fig. 4. The correlation of CoaguChek PT and INR values show a charge-dependent difference in the curve progression (Fig. 5).
Figure 2 Correlation of immediate (CoaguChek) and laboratory diagnosis of Prothrombin Time (PT) [%] (Pearsons coefficient r = 0.93).
Figure 3 Correlation of immediate (CoaguChek) and laboratory diagnosis of Partial Thromboplastin Time (PTT) [s] (Pearsons coefficient r = 0.74).
Figure 4 Correlation of immediate (CoaguChek) and laboratory diagnosis of the International Normalised Ratio (INR) (Pearsons coefficient r = 0.86)
Figure 5 Correlation of immediate diagnosis (CoaguChek) of the Prothrombin Time (PT) [%] and the International Normalised Ratio (INR) (Spearman coefficient r = -0.89)
The relative error between the laboratory and immediate determination method ± 95% confidence interval was -5.8 ± 15.2% for PT (Fig. 6), -2.7 ± 17.9 s for PTT (Fig. 7) and 0.23 ± 0.80 for INR (Fig. 8). The limits for concordance were 9.4 and -21.1% for PT, 15.2 and -20.5 s for PTT, and 1.03 and -0.57 for INR.
Figure 6 Relative error between the laboratory and immediate determination method concerning Prothrombin time (PT)
Figure 7 Relative error between the laboratory and immediate determination method concerning Partial Thromboplastin Time (PTT).
Figure 8 Relative error between the laboratory and immediate determination method concerning International Normalised Ratio (INR).
The relative error between the laboratory methods and immediate diagnosis was 21.3% with an average PT of 34.5%, 6.3% with an average PTT of 47.2 s, and 7.6% with an average INR of 2.77. The relative limit for concordance between laboratory methods and immediate diagnosis was 48.6% for PT, 35.7% for PTT and 24.3% for INR.
Discussion
The proportion of patients with postoperative secondary bleeding is comparable with the literature [2,6,16] as regards local haemostyptic measures used. Good results were established from the collagen dressing and the use of fibrin adhesive or tranexamic acid in various studies [4,6,15,22,26]. The plastic cover doubled the frequency of secondary bleeding and should therefore be reserved for the cover of maxillary sinus connection for the treatment of patients under oral anticoagulation therapy. Studies on the precision of measurement devices for patient diagnosis of the coagulation status established a firm concordance with laboratory results [5,25], whereby correlation analyses were mostly used for assessment of the concordance. Neither the determination of the correlation coefficient nor the implementation of a regression analysis are considered to be suitable measures for comparing measurement methods [3]. A lower average error with INR diagnosis with a partially more marked distribution of values was established [1,24]. Measurement devices that used thromboplastin with an ISI value > 2.0 showed a decrease in concordance with comparative tests [11,13,23].
Concordance for PTT diagnoses are also the subject of controversial discussion. There are also studies on good concordance [20], as well as studies on large deviations between laboratory and immediate diagnoses [9,18,19]. In this study, a relative error of 10% between methods was regarded as clinically acceptable for PTT and INR. The PT diagnosis was clearly above the required value, which can partly be traced back to the charge-specific differences in calibration. The relative limits of concordance at 25% could only be maintained for INR. For PT and PTT the limit was exceeded. Comparable results for INR diagnoses as regards the average error and the limits of concordance were published, whereby clear differences were given at four research centres [8]. In previous studies, a positive average error with an increase in rising INR values was established from the difference between immediate and laboratory diagnoses [14]. A short treatment time using oral anticoagulation was given as a possible cause, which must not be assumed in the current study regarding predominantly long-term treatment. There is a possible error in the use of thromboplastins of differing sensitivity and the lack of calibration of the ISI with immediate diagnosis [10].
Conclusion
In conclusion, patient INR diagnosis with the CoaguChek Pro device allows for clinically acceptable optimisation of the treatment procedure when planning dental surgery for patients under oral anticoagulation therapy.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
BKL set up the design of the study, performed the surgical part, and helped to draft the manuscript. MK carried out the statistical analysis. JK performed the statistical analysis and participated in the design of the study, the coordination of the patients and helped to draft the manuscript. All authors read and approved the final version of the manuscript.
==== Refs
Anderson OR Harrison L Hirsh J Evaluation of a portable prothrombin time monitor for horne use by patients who require long-term oral anticoagulant therapy Arch Intern Med 1993 153 1441 1447 8512435 10.1001/archinte.153.12.1441
Beirne OR Koehler JR Surgical management of patients on warfarin sodium J Oral Maxillofac Surg 1996 54 1115 1118 8811824 10.1016/S0278-2391(96)90172-X
Bland JM Altman OG Statistical methods for assessing agreement between two methods of clinical measurement Lancet 1986 1 307 310 2868172
Bodner L Weinstein JM Baumgarten AK Efficacy of fibrin sealant in patients on various levels of oral anticoagulant undergoing oral surgery Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1998 86 421 424 9798225 10.1016/S1079-2104(98)90367-5
Boldt J Walz G Triem J Suttner S Kumle B Point-of-care (POC) measurement of coagulation after cardiac surgery Intensive Care Med 1998 24 1187 1193 9876982 10.1007/s001340050743
Bublitz R Sommer S Weingart D Bauerle K Hämostyptische Wundversorgung bei Marcumarpatienten; Kollagenvlies vs. Tranexamsäure Mund Kiefer Gesichtschir 2000 4 240 244 10994324 10.1007/s100060000155
Campbell JH Alvarado F Murray RA Anticoagulation and minor oral surgery: should the anticoagulation regimen be altered? J Oral Maxillofac Surg 2000 58 135 136 10.1016/S0278-2391(00)90324-0
Cosmi B Palareti G Moia M Carpenedo M Pengo V Biasiolo A Rampazzo P Morstabilini G Testa S Accuracy of a portable prothrombin time monitor (CoaguChek) in patients on chronic oral anticoagulant therapy: a prospective multicenter study Thromb Res 2000 100 279 286 11113271 10.1016/S0049-3848(00)00323-6
Ferring M Reber G de Moerloose P Merlani P Diby M Ricou B Point of care and central laboratory determinations of the aPTT are not inter changeable in surgical intensive care patients Can J Anesth 2000 48 1155 1160 11744594
Gosselin R Owings JT White RH Hutchinson R Branch J Mahackian K A comparison of point-ot-care instruments designed for monitoring oral anticoagulation with standard laboratory methods Thromb Haemost 2000 83 698 703 10823265
Hirsh J Poller L The international normalized ratio. A guide to understanding and correcting its problems Arch Intern Med 1994 154 282 288 8297194 10.1001/archinte.154.3.282
Joos U Krekeler G Kollagenvlies als lokales Hämostyptikum bei gerinnungsgestörten Patienten ZWR 1978 87 810 812 309696
Kaatz SS White RH Hili J Mascha E Humphries JE Becker DM Accuracy of laboratory and portable monitor international normalized ratio determinations; Comparison with a criterion standard Arch Intern Med 1995 155 1861 1867 7677552 10.1001/archinte.155.17.1861
Kemme MJ Faaij RA Schoemaker RC Kluft C Meijer P Cohen AF Burggraaf J Disagreement between bedside and laboratory activated partial thromboplastin time and international normalized ratio for various novel anticoagulants Blood Coagul Fibrinolysis 2001 12 583 591 11685048 10.1097/00001721-200110000-00012
Martinowitz U Mazar AL Taicher S Varon D Gitel SN Ramot B Rakocz M Dental extraction for patients on oral anticoagulant therapy Oral Surg Oral Med Oral Pathol 1990 70 274 277 2145537 10.1016/0030-4220(90)90139-J
Merten HA Springorum H Wieding JU Giesen K Köstering H Homologe Kollagenvlies-Implantation in der zahnärztlichen Chirurgie bei blutungsgefährdeten Patienten Dtsch Z Mund Kiefer Gesichtschir 1989 13 301 304 2700135
Müller-Beißenhirtz W Deickert F Lang H Schöndorf T Spanuth E Taborski U Witt K Selbstkontrolle der oralen Antikoagulation: Standortbestimmung J Lab Med 1997 21 558 562
Nuttall GA Oliver WC JrBeynen FM Dull JJ Murray MJ Nichols WL Intraoperative measurement of activated partial thromboplastin time and prothrombin time by a portable laser photometer in patients following cardiopulmonary bypass J Cardiothorac Vasc Anesth 1993 7 402 409 8400094 10.1016/1053-0770(93)90160-M
Reich OL Yanakakis MJ Vela-Cantos FP OePerio M Jacobs E Comparison of bedside coagulation monitoring tests with standard laboratory tests in patients after cardiac surgery Anesth Analg 1993 77 673 679 8214648
Ruzicka K Kapiotis S Quehenberger P Handler S Hornykewycz S Michitsch A Huber K Oejaco C Meghdadi S Pabinger I Eichinger S Jilma B Speiser W Evaluation of bedside prothrombin time and activated partial thromboplastin time measurement by coagulation analyzer CoaguChek Plus in various clinical settings Thromb Res 1997 87 431 440 9306617 10.1016/S0049-3848(97)00159-X
Schmelzeisen R Stellungnahme der OGZMK. Zahnärztliche Chirurgie bei Patienten mit Antikoagulantientherapie ZM 1997 92 54 55
Sindet-Pederson S Ramstrom G Bernvil S BIomback M Hemostatic effect of tranexamic acid mouthwash in anticoagulant-treated patients undergoing oral surgery N Engl J Med 1989 320 840 843 2648144
Tripodi A Arbini AA Chantarangkul V Bettega 0 Mannucci PM Are capillary whole blood coagulation monitors suitable for the control of oral anticoagulant treatment by the international normalizes ratio? Thromb Haemost 1993 70 921 924 8165612
van den Besselaar AMHP A comparison of INRs determined with a whole blood prothrombin time device and two international reference preparations for thromboplastin Thromb Haemost 2000 84 410 412 11019963
van den Besselaar AMHP Breddin K Lutze G Parker-Williams J Taborski U Vogel G Tritschler W Zerback R Leinberger R Multicenter evaluation of a new capillary blood prothrombin time monitoring system Blood Coagul Fibrinolysis 1995 6 726 732 8825222
Vinckier F Vermylen J Blood loss following dental extractions in anticoagulated rabbits: effects of tranexamic acid and socket packing Oral Surg Oral Med Oral Pathol 1985 59 2 5 3871930 10.1016/0030-4220(85)90105-7
Wahl MJ Dental surgery in anticoagulated patients Arch Intern Med 1998 158 1610 1616 9701094 10.1001/archinte.158.15.1610
|
16316464
|
PMC1315351
|
CC BY
|
2021-01-04 16:37:14
|
no
|
Head Face Med. 2005 Nov 29; 1:12
|
utf-8
|
Head Face Med
| 2,005 |
10.1186/1746-160X-1-12
|
oa_comm
|
==== Front
BMC Dev BiolBMC Developmental Biology1471-213XBioMed Central London 1471-213X-5-261631646510.1186/1471-213X-5-26Methodology ArticleDevelopment of antibodies to human embryonic stem cell antigens Cai Jingli [email protected] Judith M [email protected] Mahendra S [email protected] Marisa [email protected] Eva [email protected] Hsiao-Tzu [email protected] Stem Cell Biology Unit, Laboratory of Neurosciences, National Institute on Aging, 333 Cassell Dr, Rm406A, Baltimore, MD 21224, USA2 Stem Cell Department, R&D Systems, Inc., 614 McKinley Place. Minneapolis, MN 55413, USA2005 29 11 2005 5 26 26 26 4 2005 29 11 2005 Copyright © 2005 Cai et al; licensee BioMed Central Ltd.2005Cai 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
Using antibodies to specific protein antigens is the method of choice to assign and identify cell lineage through simultaneous analysis of surface molecules and intracellular markers. Embryonic stem cell research can be benefited from using antibodies specific to transcriptional factors/markers that contribute to the "stemness" phenotype or critical for cell lineage.
Results
In this report, we have developed and validated antibodies (either monoclonal or polyclonal) specific to human embryonic stem cell antigens and early differentiation transcriptional factors/markers that are critical for cell differentiation into definite lineage.
Conclusion
These antibodies enable stem cell biologists to conveniently identify stem cell characteristics and to quantitatively assess differentiation.
==== Body
Background
Although the stem cell concept was introduced decades ago, to date, stem cells can only be defined functionally, not morphologically or phenotypically. Two functions define stem cells. Firstly, they are self-renewing, thus able to propagate to generate additional stem cells. Secondly they can differentiate into various progenitor cells, which commit to further maturation along a specific lineage. While stem cells can be best defined functionally, a good number of molecular markers have been used to prospectively identify various stem cell populations. Although the functional importance of many of these antigens remains unknown, their unique expression pattern and timing of expression provide a useful tool for scientists to identify as well as isolate stem cells.
Embryonic stem cells (ESC), derived from the inner cell mass of pre-implantation embryos, have been recognized as the earliest stem cell population [1,2]. This pluripotent population can differentiate into all somatic tissue including germ cells. In the case of human ESC, they can differentiate into some extra-embryonic derivatives as well. Like mouse ESC, human ES cells can be maintained and propagated on mouse fibroblast feeders for extended periods in media containing basic fibroblast growth factor (bFGF) [3]. Gene expression of undifferentiated human ES cells has been investigated among several ES cell lines by a variety of techniques. They include comparison with databases, reverse transcriptase-polymerase chain reaction, focused cDNA microarrays, and immunocytochemistry. A list of molecules comprised of known ES-specific or -highly expressed genes and candidates that can serve as markers for human ESCs and may also contribute to the "stemness" phenotype has been established [3-11]. For example, pluripotent ESC can be characterized by high level expression of Oct3/4 (POU domain, class 5, transcription factor 1, Pou5f1) and Nanog, which are a member of POU domain and homeobox transcription factors respectively. A critical amount of Oct3/4 and Nanog expression is required to sustain stem-cell pluripotency and both of these markers are downregulated as cells differentiate in vitro and in vivo [4-9]. Antibodies to Oct3/4 which cross react with human Oct 3/4 have been widely used to monitor the presence of undifferentiated ESC.
No single marker however is sufficient or unique for identifying ESCs. Oct3/4 for example is expressed by germ cells and may be expressed by specific populations later in development. Likewise, Nanog has been shown to express in other tissues. We and other have noted however, that while no single marker is sufficient a constellation of positive and negative markers can in concert unambiguously allow one to define the state of ESC cultures and that surface markers in combination can be used to prospectively sort for ESC.
Based on published data at the level of gene expression, we have cloned a number of candidate marker genes. We have also expressed the recombinant protein and generated a panel of monoclonal or polyclonal antibodies to these proteins. Using these antibodies we have confirmed the specificity and selectivity of these antibodies on several ESC lines and established their utility as stem cells markers. Our results confirm the expression pattern and timing of these cell markers at the protein level, whereas previous data reported at the level of gene expression.
Results and discussion
Characterization of undifferentiated human ES cells and differentiated EBs by antibodies
All monoclonal antibodies were initially selected for their abilities to recognize recombinant proteins in direct ELISAs. A subset were also tested by Western Blot analysis using recombinant proteins and cell lysate to confirm binding to a single epitope. The best clone was later screened for its applications for immunocytochemistry and flow cytometry using various cell lines. Human peripheral blood platelets were used for screening mouse anti-human CD9 antibody. MCF-7 cells were used for screening mouse anti-human E-Cadherin and PODXL (podocalyxin-like) antibodies. MG-63 cells were used for screening mouse anti-human GATA1 (GATA binding protein 1) antibody. Beta-TC6 cells were used for screening for mouse anti-human/mouse PDX-1 (pancreatic duodenal homeobox-1) antibody. NTERA-2 cells were used for screening mouse anti-human Oct3/4 and SOX2 (sex-determining region Y-box 2) antibodies. All polyclonal antibodies were affinity-purified using recombinant proteins and validated by direct ELISAs and Western. Caco-2 cells were used for validation of goat anti-human GATA6 antibody and NTERA-2 cells were used for validation of goat anti-human Nanog and anti-human Oct3/4 antibodies (Summarized in Table 1).
Table 1 Summary list of antibody verification by western blot.
Antibody Sample used for analysis Mol. Wt. (KD)
Gt × hBrachyury mouse ES-derived EB lysate 48
Ms × hDPPA5 N/A N/A
Gt × hGATA6 Caco2 cell lysate 65
Gt × hNanog NTERA-2 cell lysate 33
Gt × hOct 3/4 NTERA-2 cell lysate 39
Gt × hPDX1 beta-TC 6 cell lysate 32
Gt × hSOX17 mouse ES-derived EB lysate 45
Ms × hCD9 PBMC 25
Rt × hGATA-1 N/A N/A
Ms × hE-Cadherin MCF-7 cell lysate 97
Ms × hPODXL MCF-7 cell lysate 57
Ms × hSOX2 NTERA-2 cell lysate 36
N/A: 1. DPPA5 is still being subcloned. Only Elisa verification is available.
2. The clone for GATA-1 (MAB1779) does not work for Western blot application but is useful for IHC, The clone picked for Western blot analysis does not work for IHC (MAB17791, see data in ).
After antibodies were validated in direct ELISAs, Western blot or cell lines (Fig. 1 and data not shown), they were used to examine the expression of individual molecules in undifferentiated human ES cells and differentiated EBs. When examined by immunohistochemistry, high level of expressions of Oct3/4, SOX2, E-Cadherin, PODXL and Nanog were observed in undifferentiated human ES cells (Fig. 2A, 2B and 2C). DPPA5 (developmental pluripotency associated 5) expression was also observed in undifferentiated human ES cells (data not shown). We noted that a subset of the proteins used were membrane bound proteins. To test if any of the antibodies generated could recognize an extracellular epitope and thus be used for live cell sorting, we repeated staining of live cells as previously described. The CD9, E-Cadherin and PODXL antibodies recognized an extracellular epitope and their ability to select cells by FACS was confirmed (Fig. 3). Minimal or no expressions of Oct3/4, E-Cadherin, PODXL and Nanog were detected in the differentiated EBs (Fig. 2D, 2E and 2F). However, SOX2 expression, which is observed in neural progenitor cells, is persistent in subsets of EBs.
Figure 1 Western blot analysis for Gt × hOct3/4 (A), Gt × hNanog (B) and Ms × hSOX2 (C) in NTERA-2 cell lysate, Ms × hE-Cadherin (D) in MCF-7 cell lysate, Ms × hCD9 (E) in PBMC lysate and Ms × hPDX-1(F) in β-TC-6 cell lysate. Numbers indicate the positions of molecular weight markers.
Figure 2 Undifferentiated human ES cells (A, B, and C) and differentiated EBs (D, E and F) were analyzed using antibodies to indicated molecular markers. Immunostaining with goat anti-human Oct3/4 (Red in A and D), mouse anti-human SOX2 (Green in A and D), goat anti-human E-Cadherin (Red in B and E), mouse anti-human PODXL (Green in B and E), and goat anti-human Nanog (Red in C and F), are contrasted with DAPI nuclear staining (Blue in C-F). Note the dramatic downregulation of ESC specific markers (Oct3/4, E-Cadherin, PODXL, and Nanog) in EBs. However, SOX2 expression is persistent in subsets of EB cells. Scale bars = 100 μm.
Figure 3 Human embryonic stem cells stained with anti-CD9 (A), anti-E-Cadherin (B), and anti-PODXL (C) and antigen expression detected by a flow cytometer. The specific staining is indicated by green histogram and corresponding isotype control is indicated by black histogram.
Suspension culture with FGF withdrawal is known to induce differentiation of ES cells to all three germ layer precursors [12]. The differentiation status of the EB used here was detected to contain all germ cell markers by RT-PCR (Fig. 4). In order to examine how more antibodies can be used for characterization of early differentiation events from human ES cells, we examined the expressions of endodermal markers, SOX17, GATA6 and PDX-1, and mesodermal markers, Brachyury and GATA1, in the undifferentiated human ES cells and differentiated EBs. Expressions of SOX17, GATA6, PDX-1, Brachyury and GATA1 were not detected in undifferentiated human ES cells (data not shown). In contrast to the undifferentiated ES cells, subpopulations of SOX17-, GATA6-, Brachyury- and GATA1-positive cells were observed (Fig 4). These results suggest that both endodermal and mesodermal precursors exist in EBs with FGF withdrawal for 8 days. However, no PDX-1-positive cells were seen in EBs differentiated with the same treatment (data not shown).
Figure 4 Differentiated EBs were analyzed by either immunocytochemistry or RT-PCR to the indicated molecular markers. (A) Immunostaining with goat anti-human SOX17 (Red), is contrasted with Fluoro Nissl nuclear staining (Green). (B) Immunostaining with goat anti-human GATA6 (Red), is contrasted with DAPI nuclear staining (Blue). (C) Immunostaining with goat anti-human brachyury (Red), is contrasted with DAPI nuclear staining (Blue). (D) Immunostaining with mouse anti-human GATA1 (Red). Note that each antibody recognizes subsets of EB cells. Scale bars = 100 μm. (E) The differentiation status of EB is detected by RT-PCR using different germ layer cell markers. Selected endoderm markers AFP, FoxA2; mesoderm markers Hand1, MSX1 and ectoderm marker Msl1 were all highly expressed in the EB samples while their expression was either undetectable or at low level in the ES samples. G3PDH was a positive control showing similar amount of RNA samples were used for analysis.
Examination of cross-reactivity of antibodies on mouse ES and differentiated cells
We have also examined the cross-reactivities of these antibodies to mouse ES cells using mouse D3 ES cell line and mouse fetal endodermal tissue. Cross-reactivity to mouse of goat anti-Oct3/4, goat anti-PDX-1, goat anti-SOX17 and mouse anti-SOX2 was detected. Minimal cross-reactivity to mouse, measured by 10% intensity to human by higher than control cells, was observed in mouse anti-CD9 and mouse anti-E-cadherin antibodies. Goat anti-Nanog and mouse anti-PODXL antibodies appear to be human-specific as well (data not shown). The subtypes of monoclonal antibodies were also identified in the best clones. These results are summarized in Table 2.
Table 2 Summary of antibodies detection in ES and EB samples.
Antibody ES EB Reactivity to mouse Isotype of monoclonal antibody (Clone No.)
Gt × hBrachyury No Yes NT*
Ms × hDPPA5 Yes NT* NT* ND*
Gt × hGATA6 No Yes NT*
Gt × hNanog Yes Down No
Gt × hOct 3/4 Yes Down Yes
Gt × hPDX-1 No No Yes
Gt × hSOX17 No Yes Yes
Ms × hCD9 Yes No Minimal Mouse IgG2B (clone 209306)
Ms × hE-cadherin Yes No Minimal Mouse IgG2B (clone 180224)
Ms × hGATA1 No Yes NT* Rat IgG2B (clone 234732)
Ms × hPODXL Yes No No Mouse IgG2A (clone 222328)
Ms × hSOX2 Yes Yes Yes Mouse IgG2A (clone 245610)
*NT, Not tested; ND, Not determined.
Conclusion
The expression patterns detected using antibodies developed in our facility are consistent with data reported using reverse transcriptase-polymerase chain reaction or cDNA microarrays. Moreover several of the monoclonal antibodies have differing heavy chain subunits allowing double labeling using subtype specific markers to be performed.
In summary, we have developed a useful collection of antibodies that would be useful for identification of stem cell characteristics and assessment of differentiation. Several additional antibodies to the molecules that have been identified as potential cell lineage markers [13] are currently under development using the same approach.
Methods
Cloning and expression of Brachyury, DPPA5, CD9, E-Cadherin, GATA1, GATA6, Nanog, Oct3/4, PDX-1, PODXL, SOX2 and SOX17
Brachyury (aa. 1–202), DPPA5 (a.a. 1–116), GATA1 (a.a. 1–413), GATA6 (aa. 1–449), Nanog (aa. 153–305), Oct3/4 (aa. 1–265), PDX-1 (aa. 1–283), SOX2 (aa. 135–317) and SOX17 (aa. 177–414) were expressed in E. Coli and extracellular domains of CD9, E-Cadherin, PODXL were expressed in mouse NSO cells. All proteins were purified and sequenced before they were used as antigens for immunizations and as substrate for antibody screening and subcloning.
Production and purification of antibodies
All monoclonal antibodies were derived from fusions of mouse myeloma with B cells obtained from BALB/c mice which had been immunized with purified antigen. The IgG fraction of the culture supernatant was purified by Protein G affinity chromatography (Sigma). Each panel of antibodies was screened and selected for their abilities to detect purified recombinant antigen in direct ELISA and Western blot. All polyclonal antibodies were derived from sera of goats which had been immunized and boost it with purified antigen. Antibody was purified from the sera by an antigen-affinity chromatography.
Cells and cell culture
Human Caco-2, MG-63, MCF-7, NTERA-2 and mouse D3 cells were purchased from American Type Culture Collection (ATCC). Cells were cultured according to the ATCC instructions. Information regarding human ES cell line HSF-6 (NIH code UC06) can be obtained at the website [14]. Undifferentiated human ES cells were cultured according to the protocol provided by the University of California, San Francisco in human ES culture medium [DMEM supplemented with 20% KnockOut Serum Replacement (Invitrogen) and 5 ng/mL of bFGF (R&D Systems)]. To induce formation of embryoid bodies (EBs), ES colonies were harvested, separated from the MEF feeder cells by gravity, gently resuspended in ES culture medium and transferred to non-adherent suspension culture dishes (Corning). Unless otherwise noted, EBs derived from human ES cell aggregates were cultured for 8 days in ES culture medium deprived of bFGF and used for analysis by immunohistochemistry as described.
Western blot
Cells are solubilized in hot 2× SDS gel sample buffer (20 mM dithiothreitol, 6% SDS, 0.25 M Tris, pH 6.8, 10% glycerol, 10 mM NaF and bromophenyl blue) at 2 × 106 per mL. The extracts are heated in a boiling water bath for 5 minutes and sonicated with a probe sonicator with 3–4 bursts of 5–10 seconds each. Samples are diluted with 1× SDS sample buffer to the desired loading of 1–5 × 103 per lane. Lysates were resolved by SDS-PAGE, transferred to Immobilon-P membrane, and immunoblotted with 0.5 μg/mL primary Abs as described in R&D Systems Website [15].
Immunohistochemistry
Antibodies were used with the appropriate secondary reagents at a concentration of 5 to 10 μg/ml. Cells or sections of EBs were fixed with 4% paraformaldehyde in PBS at room temperature for 20 min, then blocked and permeabilized with 0.1% Triton X-100, 1% BSA, 10% normal donkey serum in PBS at room temperature for 45 min. After blocking, cells were incubated with diluted primary antibody overnight at 4°C followed by coupled anti-mouse or anti-goat IgG (Molecular Probes) at room temperature in the dark for an hour. Between each step cells were washed with PBS with 0.1% BSA.
RT-PCR
Total RNA was extracted from EBs using Trizol LS (Invitrogen). cDNA was synthesized by using Superscript II reverse transcriptase (Invitrogen) according to the manufacturer's recommendations. The PCR primers are available upon request.
Flow cytometry
Antibodies were prepared at the concentration of 0.1 mg/mL. 10 μL of the stock solution was added to 1 – 2.5 × 105 cells in a total reaction volume not exceeding 200 μL. The sample was then incubated for 20 min at 2–8 °C. Following incubation, excess antibody was removed by washing cells twice with FACS buffer (2% FCS and 0.1% sodium azide in Hank's buffer). After wash, cells were resuspend in 200 μL of FACS buffer and the binding of unlabeled monoclonal antibodies was visualized by adding 10 μL of a 25 μg/mL stock solution of a secondary developing reagent such as goat anti-mouse IgG conjugated to a fluorochrome for 20 min at 2–8°C. Following incubation, cells were washed once with FACS buffer, once with PBS. After wash, cells were resuspend in 400 μL of PBS and analyzed on a FACScant flow cytometer (Becton-Dickinson, Mountain View, CA). Five thousand events were collected and analyzed using CELL Quest software.
Authors' contributions
Dr. Cai contributed significantly in validating antibodies in human ES cells and human EBs. Ms. Olson performed initial screening of antibodies in various cell lines. Dr. Rao initiated the project, supervised Dr. Cai, and participated in all discussions for this report. Ms. Stanley and Ms. Taylor performed Western blot analysis. Dr. Ni coordinated collaborative work between two labs, monitored the generation of the antibodies, and directed the project at R&D Systems.
Acknowledgements
We are grateful to Drs. M. Tsang and J.H. Kim for their valuable advice and assistance.
==== Refs
Evans MJ Kaufman MH Establishment in culture of pluripotential cells from mouse embryos Nature 1981 292 154 156 7242681 10.1038/292154a0
Martin GR Isolation of a pluripotent cell line from early mouse embryos cultured in medium conditioned by teratocarcinoma stem cells Proc Natl Acad Sci U S A 1981 78 7634 7638 6950406
Thomson JA Itskovitz-Eldor J Shapiro SS Waknitz MA Swiergiel JJ Marshall VS Jones JM Embryonic stem cell lines derived from human blastocysts Science 1998 282 1145 1147 9804556 10.1126/science.282.5391.1145
Rosner MH Vigano MA Ozato K Timmons PM Poirier F Rigby PW Staudt LM A POU-domain transcription factor in early stem cells and germ cells of the mammalian embryo Nature 1990 345 686 692 1972777 10.1038/345686a0
Niwa H Miyazaki J Smith AG Quantitative expression of Oct-3/4 defines differentiation, dedifferentiation or self-renewal of ES cells Nat Genet 2000 24 372 376 10742100 10.1038/74199
Pesce M Scholer HR Oct-4: gatekeeper in the beginnings of mammalian development Stem Cells 2001 19 271 278 11463946 10.1634/stemcells.19-4-271
Chambers I Colby D Robertson M Nichols J Lee S Tweedie S Smith A Functional expression cloning of Nanog, a pluripotency sustaining factor in embryonic stem cells Cell 2003 113 643 655 12787505 10.1016/S0092-8674(03)00392-1
Mitsui K Tokuzawa Y Itoh H Segawa K Murakami M Takahashi K Maruyama M Maeda M Yamanaka S The homeoprotein Nanog is required for maintenance of pluripotency in mouse epiblast and ES cells Cell 2003 113 631 642 12787504 10.1016/S0092-8674(03)00393-3
Cavaleri F Scholer HR Nanog: a new recruit to the embryonic stem cell orchestra Cell 2003 113 551 552 12787492 10.1016/S0092-8674(03)00394-5
Pera MF Filipczyk AA Hawes SM Laslett AL Isolation, characterization, and differentiation of human embryonic stem cells Methods Enzymol 2003 365 429 446 14696363
Bhattacharya B Miura T Brandenberger R Mejido J Luo Y Yang AX Joshi BH Ginis I Thies RS Amit M Lyons I Condie BG Itskovitz-Eldor J Rao MS Puri RK Gene expression in human embryonic stem cell lines: unique molecular signature Blood 2004 103 2956 2964 15070671 10.1182/blood-2003-09-3314
Itskovitz-Eldor J Schuldiner M Karsenti D Eden A Yanuka O Amit M Soreq H Benvenisty N Differentiation of human embryonic stem cells into embryoid bodies compromising the three embryonic germ layers Mol Med 2000 6 88 95 10859025
Brandenberger R Wei H Zhang S Lei S Murage J Fisk GJ Li Y Xu C Fang R Guegler K Rao MS Mandalam R Lebkowski J Stanton LW Transcriptome characterization elucidates signaling networks that control human ES cell growth and differentiation Nat Biotechnol 2004 22 707 716 15146197 10.1038/nbt971
|
16316465
|
PMC1315352
|
CC BY
|
2021-01-04 16:30:31
|
no
|
BMC Dev Biol. 2005 Nov 29; 5:26
|
utf-8
|
BMC Dev Biol
| 2,005 |
10.1186/1471-213X-5-26
|
oa_comm
|
==== Front
Virol JVirology Journal1743-422XBioMed Central London 1743-422X-2-861630067810.1186/1743-422X-2-86ResearchStimulation of poliovirus RNA synthesis and virus maturation in a HeLa cell-free in vitro translation-RNA replication system by viral protein 3CDpro Franco David [email protected] Harsh B [email protected] Craig E [email protected] Bart [email protected] Eckard [email protected] Aniko V [email protected] Department of Molecular Genetics and Microbiology, School of Medicine, Stony Brook University, Stony Brook, N. Y. 11790, USA2 Department of Biochemistry and Molecular Biology, Pennsylvania State University, University Park, Pennsylvania 16802, USA3 Department of Microbiology and Hygiene, Vrije Universiteit Brussel, B-1090 Brussels, Belgium2005 21 11 2005 2 86 86 30 6 2005 21 11 2005 Copyright © 2005 Franco et al; licensee BioMed Central Ltd.2005Franco 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.
Poliovirus protein 3CDpro possesses both proteinase and RNA binding activities, which are located in the 3Cpro domain of the protein. The RNA polymerase (3Dpol) domain of 3CDpro modulates these activities of the protein. We have recently shown that the level of 3CDpro in HeLa cell-free in vitro translation-RNA replication reactions is suboptimal for efficient virus production. However, the addition of either 3CDpro mRNA or of purified 3CDpro protein to in vitro reactions, programmed with viral RNA, results in a 100-fold increase in virus yield. Mutational analyses of 3CDpro indicated that RNA binding by the 3Cpro domain and the integrity of interface I in the 3Dpol domain of the protein are both required for function. The aim of these studies was to determine the exact step or steps at which 3CDpro enhances virus yield and to determine the mechanism by which this occurs. Our results suggest that the addition of extra 3CDpro to in vitro translation RNA-replication reactions results in a mild enhancement of both minus and plus strand RNA synthesis. By examining the viral particles formed in the in vitro reactions on sucrose gradients we determined that 3CDpro has only a slight stimulating effect on the synthesis of capsid precursors but it strikingly enhances the maturation of virus particles. Both the stimulation of RNA synthesis and the maturation of the virus particles are dependent on the presence of an intact RNA binding site within the 3Cpro domain of 3CDpro. In addition, the integrity of interface I in the 3Dpol domain of 3CDpro is required for efficient production of mature virus. Surprisingly, plus strand RNA synthesis and virus production in in vitro reactions, programmed with full-length transcript RNA, are not enhanced by the addition of extra 3CDpro. Our results indicate that the stimulation of RNA synthesis and virus maturation by 3CDpro in vitro is dependent on the presence of a VPg-linked RNA template.
PoliovirusRNA replicationvirus maturationHeLa cell-free translation-RNA replication system
==== Body
Introduction
The HeLa cell-free in vitro translation-RNA replication system [1] offers a novel and useful method for studies of the individual steps in the life cycle of poliovirus. These processes include the translation of the input RNA, processing of the polyprotein, formation of membranous replication complexes, uridylylation of the terminal protein VPg, synthesis of minus and plus strand RNA, and encapsidation of the progeny RNA genomes to yield authentic progeny virions [1-4]. Although these processes occurring in vitro represent, in large part, what happens in virus-infected cells, there are also differences between virus production in vivo and in vitro. In the in vitro system a large amount of viral RNA (~1 × 1011 RNA molecules) has to be used, as template for translation and replication, in order to obtain infectious viral particles and the yield of virus is still relatively low. This has been attributed to insufficient concentrations of viral proteins for RNA synthesis or encapsidation, to differences in membranous structures or the instability of viral particles in vitro [3,5]. With the large amount of input RNA the level of translation in vitro is relatively high from the beginning of incubation and hence complementation between viral proteins is more efficient than in vivo [6,7]. We have recently observed that in vitro translation-RNA replication reactions, programmed with viral RNA, contain suboptimal concentrations of the important viral precursor protein 3CDpro for efficient virus production. By supplying the in vitro reactions at the beginning of incubation either with 3CDpro mRNA or purified 3CDpro protein the virus yield could be enhanced 100 fold [8,9]. Our results also indicated that both the 3Cpro proteinase and 3Dpol polymerase domains of the protein are required for its enhancing activity.
Poliovirus (PV), a member of the Picornaviridae virus family, replicates its plus strand genomic RNA within replication complexes contained in the cytoplasm of the infected cell. These complexes provide a suitable environment for increased local concentration of all the viral and cellular proteins needed for RNA replication and encapsidation of the progeny RNA genomes. Translation of the incoming plus strand RNA genome of PV yields a polyprotein, which is cleaved into functional precursors and mature structural and nonstructural proteins (Fig. 1). This is followed by the synthesis of a complementary minus strand RNA, which is used as template for the production of the progeny plus strands [reviewed in [10]]. Although the process of viral particle assembly is not fully understood it is believed to occur by the following pathway: The P1 precursor of the structural proteins is cleaved into VP0, VP1 and VP3, which form a noncovalent complex, the protomer [11]. The protomers associate into pentamers and six pentamers form an icosahedral particle (empty capsid) enclosing the progeny plus strand RNA yielding provirions. It is unclear whether the progeny RNA is inserted into the empty capsid or whether the pentamers condense around the RNA [12,13]. Maturation is completed by the cleavage of VP0 into VP2 and VP4, possibly by a RNA-dependent autocatalytic mechanism [11]. From the nonstructural viral proteins 2CATPase [14] and VPg [15] have been proposed to have a role in encapsidation but their functions are not yet known.
Figure 1 Genomic structure of poliovirus and processing of the P3 domain of the polyprotein. The plus strand RNA genome of poliovirus is illustrated with the terminal protein VPg covalently linked to the 5' end of the RNA. The 5' nontranslated region (NTR) and 3' NTR are shown with single lines. The genome is terminated with a poly(A) tail. The polyprotein (open box) contains structural (P1) and nonstructural (P2 and P3 domains) that are processed into precursor and mature proteins. Processing of the P3 domain by 3Cpro/3CDpro is shown enlarged.
The viral proteins most directly involved in RNA replication include protein 3AB, the precursor of 3A, which is a small membrane binding and RNA binding protein, the terminal protein VPg, RNA polymerase 3Dpol and proteinase 3Cpro/3CDpro. As a proteinase 3CDpro is responsible for the processing of the capsid precursor [16] but it also has very important functions as an RNA binding protein [17-21]. It forms complexes with the 5' cloverleaf structure in PV RNA either in the presence of cellular protein PCBP2 [18,22] or viral protein 3AB [19]. The interaction between PCBP2, 3CDpro and the cloverleaf has been proposed to mediate the switch from translation to RNA replication [23] and the circularization of PV RNA through interaction with poly(A) binding protein bound to the poly(A) tail of the genome [24]. In addition, 3CDpro binds to the cre(2C) element [20,21], and to the 3'NTR in a complex with 3AB [19]. Polypeptide 3CDpro is also a precursor of proteinase 3Cpro and RNA polymerase 3Dpol. The 3Cpro domain of the polypetide contains both the proteinase active site and the primary RNA binding domain [25,26]. The function of the 3Dpol domain appears to be to modulate these activities of the protein [27,28] and it also contains RNA binding determinants [27]. By itself 3Dpol is the RNA dependent RNA polymerase, which possesses two distinct synthetic activities. It elongates oligonucleotide primers on a suitable template [29] and it links UMP to the hydroxyl group of a tyrosine in the terminal protein VPg [20]. The 3Dpol polypeptide possesses a structure similar to other nucleic acid polymerases of a right hand with palm, thumb and finger subdomains [30]. Interaction between polymerase molecules along interface I results in a head to tail oligomerization of the protein, which is important for its biological functions [31].
The aim of these studies was to determine how the addition of extra 3CDpro protein to in vitro translation RNA-replication reactions, programmed with viral RNA, stimulates virus synthesis by 100 fold. In the presence of extra 3CDpro we have observed a mild stimulation of both minus and plus strand RNA synthesis. The primary effect of 3CDpro, however, is the enhancement of virus maturation resulting in a striking increase in the specific infectivity of the virus particles produced. Both of these processes are dependent on the RNA binding activity of the protein in the 3Cpro domain. Mutational analysis of 3CDpro suggests that the formation of 155S mature virions also requires an intact interface I in the 3Dpol domain of the protein. Interestingly, plus strand RNA synthesis and virus production in translation RNA-replication reactions, programmed with PV transcript RNA, are not stimulated by 3CDpro.
Results
Effect of 3CDpro(3CproH40A) on viral RNA synthesis in in vitro translation-RNA replication reactions
We have previously shown that translation of 3CDpro mRNA along with the viral RNA template in in vitro translation-RNA replication reactions, programmed with viral RNA, enhances total RNA synthesis about 3 fold [9]. The addition of 3CDpro, however, had no effect on the translation of the input viral RNA or processing of the polyprotein [8,9]. We have now extended these results by testing the effect of mutations in 3CDpro on the ability of the protein to stimulate RNA synthesis. Translation-RNA replication reactions were incubated at 34°C either in the absence or presence of extra purified 3CDpro(3CproH40A). This protein, which contains a proteinase active site mutation, H40A, served as the positive control in all of our experiments. Samples were taken at 2-hour intervals and these were incubated with [α-35S]CTP for 1 hour. RNA synthesis was measured by the incorporation of label into polymer using a filter-binding assay. As shown in Fig. 2A, RNA synthesis is maximal 8 hrs after the start of translation and by 16 hr the total amount of RNA present in the reaction decreases. At the peak of RNA synthesis there is a 3-fold difference between reactions containing extra 3CDpro(3CproH40A) and those to which no additional protein has been added.
Figure 2 Effect of 3CDpro(3CproH40A) on viral RNA synthesis in the translation-RNA replication system. (A) Comparison of the stimulating activities of purified 3CDpro(3CproH40A) with mutant 3CDpro(3CproR84S/I86A) or 3CDpro(3CproH40G; 3DpolR455A/R456A) on total viral RNA synthesis. Translation-RNA replication reactions were carried out in the presence of [α-35S]CTP. Where indicated purified 3CDpro proteins (5.5 nM) or mRNA (1.4 μg/ml) was added at t = 0 hr. Samples were taken at the indicated time points (Method I) and the total amount of label incorporated into polymer was determined with a filter-binding assay, as described in Materials and Methods. (B), (C) Comparison of the stimulating activities of purified 3CDpro(3CproH40A) with that of mutants 3CDpro(3CproH40G, 3DpolR455A/R456A) and 3CDpro(3CproR84S/I86A), respectively, on plus strand RNA synthesis. Translation-RNA replication reactions were carried out for 4 hr and the replication complexes were isolated by centrifugation (Materials and Methods). The pellets were resuspended in translation reactions lacking viral RNA in the presence of [α-32P]CTP and the samples were incubated for 1 hr at 34°C. Following extraction and purification the RNA products were applied to a nondenaturing agarose gel (Materials and Methods). A [32P]UMP-labeled PV transcript RNA was used as a size marker for full length PV RNA.
Protein 3CDpro is the precursor of both proteinase 3Cpro and polymerase 3Dpol. The 3Cpro domain contains both the proteinase and the RNA binding site [25,26]. While the primary RNA binding determinant of 3CDpro lies in 3Cpro, lower affinity binding determinants are located in the 3Dpol domain [27,28]. We have recently shown that a mutation (3CproR84A/I86A) in the RNA binding domain of 3CDpro abolishes that ability of the protein to stimulate virus production in the in vitro system [8]. To examine the effect of these mutations on RNA synthesis we have carried out translation-RNA replication reactions in the presence 3CDpro(3CproR84S/I86A) mRNA. As shown in Fig. 2A, the mutation totally abolished the stimulatory activity of 3CDpro(3CproH40A) in RNA synthesis suggesting that RNA binding is required for participation of the extra 3CDpro(3CproH40A) in genome replication.
Our previous results indicated that the 3Dpol domain of 3CDpro is also required for the ability of 3CDpro to stimulate virus synthesis in the in vitro system [8]. This conclusion was based on the observation that two groups of mutations R455A/R456A [32] and D339A/S341A/D349A [33] in the 3Dpol domain of the protein abolished the enhancement of virus yield in the in vitro system [8]. These complementary mutations in the thumb and palm subdomains of the protein, respectively, are located at interface I of the 3Dpol protein structure and have been found to disrupt the oligomerization of the polypeptide [32,33]. Previous studies have indicated that oligomeric forms of the 3Dpol polypeptide are required for enzyme function [31]. To determine the effect of 3CDpro(3CproH40G, 3DpolR455A/R456A) on RNA synthesis we added the purified mutant protein to translation RNA-replication reactions. This mutant protein exhibited a 2-fold stimulation in RNA synthesis, only slightly lower than what is obtained with 3CDpro(3CproH40A) (Fig. 2A). This result indicates that 3Dpol residues R455 and R456 are not important for the stimulatory activity of 3CDpro in RNA synthesis. The effect of the other mutant 3CDpro protein (3DpolD339A/S341A/D349A) on RNA synthesis was not analyzed.
3CDpro(3CproH40A) has a small stimulatory effect on both minus and plus strand RNA synthesis
To examine the effect of 3CDpro on plus strand RNA synthesis we translated the viral RNA for 4 hr in the absence or presence of extra 3CDpro(3CproH40A). The initiation complexes [34] were isolated by centrifugation and resuspended in reaction mixtures lacking viral RNA but containing [α-32P]CTP. After 1 hr of incubation the RNA products were applied to a nondenaturing agarose gel together with a [α-32P]-labeled full-length poliovirus RNA transcript as a size marker (Fig. 2B, lane 1). The yield of plus strand RNA product obtained from these reactions was equally enhanced by the addition of extra 3CDpro(3CproH40A) or by mutant 3CDpro(3CproH40G, 3DpolR455A/R456A) protein (Fig. 2B, compare lane 4 with lanes 2 and 5). No product was formed in the absence of a viral RNA template (Figs. 2B and 2C, lane 3). When 3CDpro mRNA, containing the R84S/I86A mutations in the RNA binding domain of 3Cpro, was cotranslated with the input viral RNA no stimulation of plus strand RNA synthesis was observed (Fig. 2C, compare lanes 2 and 4). These results indicate that RNA binding by the extra 3CDpro(3CproH40A) is required for the stimulation of plus strand RNA synthesis but mutation R455A/R456A in the 3Dpol domain of the protein is not important for this process.
To compare the stimulatory effect of 3CDpro(3CproH40A) on both minus and plus strand RNA synthesis we used preinintiation replication complexes [2,34], which were collected after 4 hr of incubation of the reactions in the presence of 2 mM guanidine HCl, a potent inhibitor of poliovirus RNA replication. The complexes were resuspended in reactions lacking viral RNA and guanidine and were incubated for an hour with [α-32P]CTP. The RNA products were resolved on a nondenaturing agarose gel. Minus strand RNA synthesis was estimated from the amount of replicative form (RF), in which the minus strand is hybridized to the plus strand template RNA. As shown in Fig. 3B, minus and plus strand RNA synthesis are enhanced about 2-fold and 3-fold, respectively, when the reactions contain extra 3CDpro(3CproH40A). Poliovirus RF and ssRNA obtained from a reaction in which HeLa extracts were replaced by crude replication complexes (CRCs), isolated from PV-infected HeLa cells [35], were used as a size marker for the RF and the plus strand RNA (ssRNA) (Figs. 3B, and 3C, lane 1).
Figure 3 Inhibition of 3CDpro(3CproH40A)-stimulated RNA synthesis by 3Cpro(C147G) in vitro. (A) Inhibition of 3CDpro(3CproH40A)-stimulated total viral RNA synthesis by 3Cpro(C147G). Translation-RNA replication reactions were incubated for the indicated time periods in the presence of [α-35S]CTP (Method II) either in the absence or presence of 3CDpro(CproH40A) (5.5 nM). The total amount of label incorporated into polymer was determined with a filter-binding assay, as described in Materials and Methods. Where indicated 3Cpro(C147G) was added to the reactions at t = 0 either alone or together with 3CDpro(3CproH40A). (B), (C) Inhibition of 3CDpro(3CproH40A)-stimulated minus (B) and plus strand (C) RNA synthesis by 3Cpro(C147G). Translation-RNA replication reactions were carried out in the presence of guanidine HCl for 4 hr and the replication complexes were isolated by centrifugation (Materials and Methods). The pellets were resuspended in translation reactions lacking viral RNA in the presence of [α-32P]CTP and the samples were incubated for 1 hr at 34°C. Following extraction and purification of the RNAs the samples were analyzed on a nondenaturing agarose gel (Materials and Methods). RF: double stranded replicative form RNA; ssRNA: single stranded RNA; CRC: [32P]-labeled RNA products from crude replication complexes (Materials and Methods).
The addition of 3CDpro(3CproH40A) and 3Cpro(C147G) together totally blocks RNA synthesis in translation-RNA replication reactions
We have recently shown that purified 3Cpro(C147G) protein, containing a proteinase active site mutation, when added alone to in vitro translation-RNA replication reactions, has no effect on virus yield. However, when included in reactions along with extra 3CDpro(3CproH40A) the production of virus is reduced about 1 × 104 fold [8]. To determine whether the inhibitory effect of 3Cpro(C147G) is at the level of RNA synthesis, we have examined the time course of RNA synthesis in the presence of both proteins by measuring the amount of [α-35S]UMP incorporated into polymer. As shown in Fig. 3A, the effect of these proteins on RNA synthesis fully parallels their effect on virus synthesis [8]. 3CDpro(3CproH40A) stimulates RNA synthesis up to 3-fold while 3Cpro(C147G) alone exhibits no significant enhancement of the RNA yield. When the two proteins are added together there is essentially no increase in the total amount of RNA produced over a period of 16 hours. Control reactions, lacking a viral RNA template exhibited very little, if any, incorporation of label into a polymeric product (Fig. 3A). All other samples showed some incorporation of label into polymer, over what is measured in the absence of viral RNA (Fig. 3A). This is most likely a result of end labeling of the input viral RNA by newly translated 3Dpol or priming by traces of degraded RNA.
To determine whether 3Cpro(C147G) inhibits plus or minus strand RNA synthesis we labeled with [α-32P]CMP the RNA products formed in preinintiation replication complexes during a 1 hr incubation period, as described above. The samples were analyzed on a nondenaturing agarose gel and as a size marker we used [α-32P]CMP-labeled RNA products made in CRCs (Figs. 3B and 3C, lane 1). Two kinds of products were visible on the gel, the newly made single stranded RNA (ssRNA) and the double stranded replicative intermediate (RF). As shown on Fig. 3B, 3Cpro(C147G) alone has very little, if any, effect on the yield of either of the 2 kinds of RNA products (Fig. 3B and 3C, compare lanes 2 and 3). In the presence of both 3Cpro(C147G) and 3CDpro(3CproH40A), however, the synthesis of both products is completely inhibited (Figs. 3B and 3C, compare lane 4 and lane 5).
3CDpro(3CproH40A) has a small stimulating effect on the early steps of viral particle assembly
The data shown before indicated a modest increase in viral RNA synthesis in the presence of extra 3CDpro(3CproH40A) whereas the production of infectious virus was stimulated about 100 fold. The fact that there is such a large discrepancy between the extent of stimulation of RNA synthesis and virus production by 3CDpro(3CproH40A) suggested to us the possibility that this protein has an additional role at a subsequent step in the viral life cycle, the encapsidation of the progeny viral RNAs. To examine at which step of assembly this might occur, we labeled the viral proteins with [35S]-methionine in the in vitro reactions and analyzed the viral particles produced after 15 hr incubation either in the absence or presence of 3CDpro(3CproH40A). The samples were first loaded on a 5–20% sucrose gradient and sedimented for 15 hr, which resulted in the separation of the 5S protomers and 14S pentamers from the large capsid precursors and mature virions [36]. As a size marker for these small capsid precursors, a parallel gradient was run, onto which a sample of [35S]-labeled PV-infected HeLa cell lysate was applied (designated as control in Figs. 4 and 5). The amount of the 5S and 14S precursors is enhanced less than two fold by the presence of extra 3CDpro(3CproH40A) in the reactions (Figs. 4A and 4B). Similarly, reactions supplemented with mutant 3CDpro proteins, containing mutations either at the RNA binding site of 3Cpro(R84A/I86A) or at interface I in 3Dpol(R455A/R456A), exhibited very little increase in the total amount of 5S and 14S particles, when compared to reactions lacking 3CDpro(3CproH40A) (Figs. 4A and 4B).
Figure 4 Effect of 3CDpro(3CproH40A) on the early stages of poliovirus assembly in vitro. Translation-RNA replication reactions were carried out in the presence of [35S]TransLabel, as described in Materials and Methods. When indicated purified 3CDpro(3CproH40A) protein (5.5 nM) or mRNA (1.4 μg/ml) was added to the reactions at t = 0 hr and the samples were incubated for 16 hr at 34°C. Following RNase treatment and dialysis the samples were loaded on a 5–20% sucrose gradient (Materials and Methods). The samples were centrifuged for 15 hr at 40,000 RPM in a SW41 rotor at 4°C for the separation of 5S protomers and 14S pentamers. The amount of radioactivity at the bottom of the tubes of the gradients was not determined. (A) Comparison of samples obtained in the absence or presence of 3CDpro(3CproH40A) and mutant 3CDpro protein 3Dpol(H40G, R455A/R456A) or mRNA 3Cpro(R84S/I86A). (B) The 14S peak from section (A) is shown enlarged; (C) Western blot analysis with anti VP2 antibodies of samples from the 5S and 14S peaks from the gradient shown on Fig. 4A. The same analysis of the 80S and 155S peaks from the gradient shown on Fig. 5.
Figure 5 Effect of 3CDpro(3CproH40A) on the late stages of poliovirus assembly in vitro. Translation-RNA replication reactions were carried out in the presence of [35S]TransLabel, as described in Materials and Methods. When indicated purified 3CDpro(3CproH40A) protein (5.5 nM) or mRNA (1.4 μg/ml) was added to the reactions at t = 0 hr and the samples were incubated for 16 hr at 34°C. As a control, poliovirus proteins labeled with [35S]TransLabel in vivo in HeLa cells, were used. Following RNase treatment and dialysis the samples were loaded on a 5–20% sucrose gradient (Materials and Methods). The samples were centrifuged for 80 min at 40,000 RPM in a SW41 rotor at 4°C for the separation of 80S empty capsids and 155S virus particles (provirions and virions). (A) Comparison of samples obtained in the absence or presence of 3CDpro(3CproH40A) and mutant 3CDpro protein 3Dpol(H40G, R455A/R456A) or mRNA 3Cpro(R84A/I86A). (B) The 155S peak from section (A) is shown enlarged. (C) Plaque assays of fractions 7–14 in the 155S peak.
To confirm the presence of uncleaved VP0 in the 5S and 14S peak fractions of the gradient derived from reactions supplemented with extra 3CDpro(3CproH40A), we used Western blot analyses with anti VP2 polyclonal antibody (Fig. 4C). As expected, only VP0 and no VP2 could be detected in the 5S and 14S peak fractions containing these small capsid precursors (Fig. 4C).
3CDpro(3CproH40A) has a small stimulatory effect on the late stages of particle assembly
In the next set of experiments we examined the effect of 3CDpro(3CproH40A) on the formation of 80S (empty capsids) and 155S particles (provirion and mature virus). As we discussed before, the role of the 80S particle in viral assembly is unclear. The experimental evidence available at this time favors the hypothesis that empty capsids are dead-end products rather than true intermediates of particle assembly [12,13]. The particle thought to be the direct precursor of the mature virus is the provirion, a structure containing 60 copies of VP0, VP1 and VP3 and the viral RNA [37]. The difference between provirions and mature virus is that in the latter the particle is stabilized by the cleavage of VP0 to VP2 and VP4.
The 80S and 155S viral particles, labeled with [35S]-methionine in vitro, were separated by sedimentation in a 5–20% sucrose gradient for 80 min. Under our experimental conditions the provirions (125S) and mature virus (155S) comigrate [36,37]. As shown in Fig. 5A the yield of 80S particles is stimulated about 2 fold by 3CDpro(3CproH40A) and by 3CDpro(3CproH40G, 3DpolR455A/R456A) but not by 3CDpro(3Cpro R84S/I86A). The formation of 155S particles is enhanced about 3–7 fold by 3CDpro(3CproH40A) but not by the 3CDpro proteins that contain the 3DpolR455A/R456A or 3Cpro R84S/I86A mutations (Figs 5A,5B, 6). To confirm the presence of mature virions in the 155S peak fractions, derived from reactions supplemented with extra 3CDpro(3CproH40A), we used Western blot analysis with anti VP2 polyclonal antibody. As expected, both VP2 and VP0 were observed in the 155S peak but only VP0 was present in the 80S peak fractions of the gradient (Fig. 4C).
Figure 6 3CDpro(3CproH40A) enhances the specific infectivity of virus particles produced in vitro. Translation-RNA replication reactions were carried out in the presence of [35S]TransLabel, as described in Materials and Methods. Where indicated purified 3CDpro(3CproH40A) or 3CDpro(3CproH40G, 3DpolR455A/R456A) protein (5.5 nM) was added to the reactions at t = 0 hr and the samples were incubated for 16 hr at 34°C. Following RNase treatment and dialysis, 0.1% of SDS was added to the samples, as indicated. They were loaded on a 5–20% sucrose gradient (Materials and Methods) and centrifuged for 80 min at 40,000 RPM in a SW41 rotor at 4°C. (A) the 80S peak is shown; (B) the 155S peak is shown.
3CDpro(3CproH40A) strongly enhances the production of mature viral particles
As we discussed above, the extra 3CDpro(3CproH40A) added to translation-RNA replication reactions has a relatively small stimulating effect both on RNA synthesis and on the incorporation of [35S]-methionine into capsid precursors, empty capsids or particles sedimenting at 155S. These results are difficult to reconcile with the 100-fold increase in infectious virus observed in translation RNA-replication reactions that are supplemented with extra 3CDpro(3CproH40A) [8,9]. Taken together these findings suggested the possibility that the presence of extra 3CDpro(3CproH40A) enhances the specific infectivity of the virus particles produced, that is, it enhances the conversion of provirions to virions. To test this hypothesis we measured the yield of infectious virions in the peak fractions sedimenting at 155S in sucrose gradients derived from in vitro reactions incubated with or without extra 3CDpro(3CproH40A). As shown on Fig. 5C, reactions to which extra 3CDpro(3CproH40A) protein was added yielded 155S peaks containing 100 fold higher plaque forming units than reactions that were not supplemented with the protein. Interestingly, neither mutant 3CDpro proteins (3CproR84S/I86A or 3CproH40G, 3DpolR455A/R456A) enhanced the virus yield in the 155S peak of the gradient (Fig. 5), an observation suggesting that both domains of the protein are required for this function. In a parallel experiment we have estimated the total number of viral particles in the 155S peak of the gradient by electron microscopy. We observed about 3-fold increase in viral particles when 3CDpro(3CproH40A) was present in the translation-RNA replication reactions (data not shown).
To obtain further proof that the extra 3CDpro(3CproH40A) enhances the specific infectivity of the virus particles we used SDS treatment of the reaction products prior to sucrose gradient analysis. The incorporation of [35S]-methionine into particles sedimenting at 80S and 155S was determined in reactions treated with SDS. It has been previously demonstrated that only mature virions but not provirions are stable in SDS [37]. As shown on Fig. 6A, there was no increase in 80S particles in SDS-treated samples that contained extra 3CDpro(3CproH40A) (Fig. 6A) suggesting that the sample did not contain significant amounts of provirions. On the other hand, the 80S empty capsid peak, obtained from reactions with no extra 3CDpro(3CproH40A) or with 3CDpro(3CproH40G, 3DpolR455A/R456A) mutant protein, increased by about 4 fold as a result of SDS treatment. Interestingly, most of the extra label that appear in this 80S peak following SDS treatment is not derived from the 155S peak, presumably by the dissociation of provirions into 80S empty capsids and RNA (Fig. 6A). This suggested to us the possibility that in reactions lacking extra 3CDpro(3CproH40A) some of the 80S particles aggregated and pelleted in the gradient. To test this possibility we recovered and analyzed the pellets from the gradients. We observed that the amount of [35S]-label in the pellet, derived from reactions with no extra 3CDpro, was 10-fold higher than in pellets of reactions lacking the extra protein (data not shown). A Western blot analysis of the particles in the pellets indicated the presence of VP0 but no VP2 (data not shown).
As we discussed above, reactions containing extra 3CDpro(3CproH40A) produced 3–7-fold higher amounts of 155S particles than those that lacked the extra protein (Figs. 5A,5B, 6). These particles were stable to SDS treatment (Fig. 6B) suggesting that they are mature virions. In contrast, the small peak of 155S particles obtained from reactions with no extra 3CDpro(3CproH40A) or 3CDpro(3CproH40G, 3DpolR455A/R456A) disappeared upon SDS treatment (Fig. 6B). These results suggest that under these conditions the 155S peaks consists of large amount of provirions that are dissociated into 80S particles and RNA by the SDS treatment. From the amount of [35S]-label resistant to SDS in the 155S peaks (Fig. 6) it can be estimated that the presence of extra 3CDpro(3CproH40A) in translation-RNA replication reactions enhances the yield of mature virus about 40-fold. Western blot analyses with anti VP2 antibodies of gradient samples 8–9 from the 155S peak confirmed the presence of VP0, indicating provirions in reactions lacking extra 3CDpro(3CproH40A) (Fig. 7B) or containing 3CDpro(3CproH40G, 3DpolR455A/R456A) (Fig. 7C). Faster sedimenting particles in fraction 10 of this gradient contained some VP2 characteristic of mature virus. In contrast, reactions that contained extra 3CDpro(3CproH40A) yielded a 155S peak containing predominantly VP2, as judged by the Western analysis (Fig. 7A). Therefore, we conclude that the extra 3CDpro(3CproH40A) enhances the specific infectivity of the viral particles produced.
Figure 7 Comparison of the amount of VP0 and VP2 in 155S particles produced in reactions with and without extra 3CDpro(3CproH40A). Translation RNA-replication reactions were carried out either in the absence or in the presence of extra 3CDpro(3CproH40A) or 3CDpro(3CproH40G, 3DpolR455A/R456A). The reaction products were separated on sucrose gradients, and the peak fractions were run on a SDS-polyacrylamide gel. Western blots were done with a polyclonal antibody to VP2 (Materials and Methods). The amount of VP0 and VP2 in fractions 8–10, in the 155S peak of the gradient shown on Fig. 6, was determined. (A) extra 3CDpro(3CproH40A) added; (B) no extra 3CDpro(3CproH40A) added; (C) 3CDpro(3CproH40G, 3DpolR455A/R456A) added. Lane 1: fraction 8; lane 2, fraction 9; lane 3: fraction 10 of the 155S peak shown in Fig. 6.
3CDpro(3CproH40A) does not stimulate RNA synthesis or virus production in translation RNA-replication reactions programmed with transcript RNA
Transfection of full-length transcript RNAs of poliovirus, made by T7 RNA polymerase, into HeLa cells initiate a complete replication cycle although the yield of virus is only 5% of that obtained in transfections with virion RNA [38]. In the in vitro translation-RNA replication system the yield of virus with transcript RNAs is also significantly reduced to about 1% of what is obtained when the reactions are supplemented with viral RNA [39,40]. This has been attributed to the presence of two extra GMPs at the 5'-end of the transcript RNAs (pppGpGpUpU...), which are removed during replication to yield authentic viral RNA (VPg-pUpU...) [39]. Previous studies have demonstrated that the two GMPs at the 5' end of transcript RNAs do not interfere with minus strand RNA synthesis but greatly reduce the initiation of plus strand RNA synthesis in the in vitro system. Removal of the extra nucleotides with a cis-active hammerhead ribozyme resulted in templates that have regained most of their ability to support efficient plus strand RNA synthesis in the translation-RNA replication system [39].
To determine the effect of 3CDpro(3CproH40A) on virus production, in reactions templated by transcript RNA, we have generated full-length PV transcript RNA from a T7 RNA polymerase promoter and used these to program in vitro translation-RNA replication reactions. In agreement with previous studies, we have observed that the virus yield is 50–100 fold lower in reactions programmed with transcript RNA instead of viral RNA (Fig. 8A, compare lane 1 with lane 3). In contrast, the yield of virus from reactions templated by ribozyme-treated transcript RNAs was essentially the same as what was obtained from viral RNA (Fig. 8A, compare lane 1 with lane 5). Remarkably, the virus yield was not enhanced by 3CDpro(3CproH40A) in either reactions using transcript RNAs with or without ribozyme-treatment (Fig. 8A, compare lanes 5 and 6 and also lanes 3 and 4, respectively).
Figure 8 Extra 3CDpro(3CproH40A) has no effect on virus production and RNA synthesis in reactions programmed with PV transcript RNA. Translation RNA-replication reactions were carried out, as described in Materials and Methods. The viral RNA template was replaced with a PV full-length transcript RNA made from a T7 promoter or with a ribozyme-treated transcript RNA. Where indicated 3CDpro(3CproH40A) (5.5 nM) was added at t = 0 hr. (A) Comparison of virus yields in reactions templated with viral RNA and transcript RNAs. (B) Plus strand RNA synthesis with initiation complexes isolated from reactions programmed with PV transcript RNA made from a T7 promoter (Materials and Methods). Lane 1, CRC: [32P]-labeled RNA products obtained from crude replication complexes (Materials and Methods). (C) Plus strand RNA synthesis with initiation complexes isolated from reactions programmed with ribozyme-treated PV transcript RNA (R+ RNA). Lane 1, CRC: [32P]-labeled RNA products obtained from crude replication complexes (Materials and Methods).
Previous studies have demonstrated that in the in vitro translation-RNA replication system the amount of plus strand RNA product obtained from PV ribozyme-treated transcript RNA or viral RNA is about 100-fold higher than what is produced in reactions with ribozyme-deficient transcript RNAs [40]. To examine whether the lack of enhancement of virus production by 3CDpro(3CproH40A) in our reactions using a ribozyme-deficient transcript RNA is due to a defect in stimulating RNA synthesis we have measured the yield of plus strand RNA. Translation-RNA replication reactions were incubated for 4 hr at 34°C, the initiation complexes were collected by centrifugation and resuspended in reactions lacking transcript RNA. The RNA products were labeled with [α-32P]CTP for 1 hr and the products were applied to a nondenaturing gel. As shown in Fig. 8B, the presence of extra 3CDpro(3CproH40A) in such reactions has no stimulatory effect on plus strand RNA synthesis (compare lane 2 with lane 3). As a size marker for plus strand RNA we have included the [α-32P]-labeled full-length PV ssRNA product made in CRCs (Fig. 8B, lane 1). The same results were obtained when RNA synthesis was measured with ribozyme-treated transcript RNA as template for translation-RNA replication (Fig. 8C, compare lane 2 with lane 3). It should be noted that the addition of extra 3CDpro(3CproH40A) to translation reactions of transcript PV RNA had no effect either on the efficiency of translation or the processing of the polyprotein (data not shown).
The lethal R84S/I86A mutation in the 3Cpro domain of 3CDpro cannot be complemented in vitro by wt 3CDpro
It has been previously demonstrated that in vivo complementation rarely occurs, and if it does, it is very inefficient [7,41]. However, this process is more efficient in the in vitro system because large amounts of complementing proteins are translated from the input RNAs and these are apparently accessible to the replication complex [6]. Our results described in this paper indicate that at least 2 functions of 3CDpro(3CproH40A) are complementable in the in vitro system and both of these functions depend on the RNA binding sequences of the protein. One of these is in RNA synthesis and the other one in virus maturation. To determine whether there are additional functions of 3Cpro/3CDpro that involve RNA binding we have attempted to complement the lethal R84S/I86A mutation in a full length PV transcript RNA either by cotranslation of wt 3CDpro mRNA or by the addition of purified 3CDpro(3CproH40A) to in vitro reactions. As shown in Table 1, the extra wt 3CDpro does not restore the ability of the system to generate infectious virus. It should be noted that the 3CDpro translated both from the mutant PV RNA and the 3CDpro mRNA have full proteolytic activity (data not shown) and therefore these results are not due to a defect in protein processing. We have obtained the same negative results when we cotransfected the R84S/I86A mutant full length PV RNA with wt 3CDpro mRNA into HeLa cells (data not shown). These results can be interpreted to mean that: (1) 3CDpro has one or more additional RNA binding function(s), which is not complementable; (2) that an RNA binding function of 3Cpro cannot be complemented by 3CDpro.
Table 1 Mutation R84S/I86A in the RNA binding domain of 3Cpro cannot be complemented in vitro with wt 3CDpro.a
Sample Infectivity (pfu/μg transcript RNA)
PVM(3CproR84S/I86A) Tr RNA 0
PVM(3CproR84S/I86A) Tr RNA + 1.4 μg/ml 3CDpro mRNA 0
PVM (3CproR84S/I86A) Tr RNA + 400 ng/ml 3CDpro protein 0
a Translation RNA-replication reactions were carried out with a PVM transcript RNA, containing the R84S/I86A mutations in 3Cpro as template. Where indicated the reactions were supplemented with wt 3CDpro mRNA (1.4 μg/ml) or 3CDpro(3CproH40A) purified protein (5.5 nM). The virus yield was measured with a plaque assay (Materials and Methods).
Discussion
We have previously shown that the level of active 3CDpro in in vitro translation-RNA replication reactions, programmed with viral RNA, is suboptimal for efficient virus synthesis and that the addition of extra 3CDpro compensates to some extent for this deficiency [8,9] but the reason for this phenomenon remained unsolved. The results presented in this paper indicate that the stimulatory effect of 3CDpro is both at the level of RNA synthesis and of virus maturation. Since translation, replication, and encapsidation are coupled processes during the growth of poliovirus [13,42,43] one might conclude that the increase in the yield of mature virions simply reflects the stimulation of RNA synthesis. However, although this might be true to some extent, our results indicate that 3CDpro(3CproH40A) exerts its enhancing activity at two distinct stages of the viral growth cycle. This conclusion is supported by three lines of evidence: (1) plus strand RNA synthesis is stimulated by 3CDpro(3CproH40A) about 3-fold but the yield of progeny virus increases 100 fold; (2) although 3CDpro(3CproH40G, 3DpolR455A/R456A), containing mutations at interface I in the 3Dpol domain of the protein, enhance RNA synthesis nearly as efficiently as 3CDpro(3CproH40A) it does not stimulate the yield of mature virus; (3) only those reactions that contain extra 3CDpro(3CproH40A) yield a 155S peak in sucrose gradients with particles resistant to SDS treatment.
Our results with the in vitro translation-RNA replication system do not define the precise role of the extra 3CDpro in stimulating RNA synthesis. The evidence available thus far indicates that in the presence of extra 3CDpro(3CproH40A) (1) minus and plus strand RNA synthesis are stimulated 2- and 3-fold, respectively; (2) the RNA binding sequences (R84/I86) in the 3Cpro domain of the polyprotein are required for the stimulation; (3) the integrity of interface I in the 3Dpol domain of the polyprotein is not important. Whether plus strand RNA synthesis itself is stimulated by the presence of extra 3CDpro or the amount of plus strands increases simply as a result of more minus strands remains to be determined. The fact that the RNA binding domain of the protein in 3Cpro is involved in stimulating RNA synthesis suggests that the extra 3CDpro forms a functional ribonucleoprotein complex (RNP) with an RNA sequence or structure in the viral genome. Poliovirus RNA contains at least 3 different cis-acting elements that are involved in RNA replication. All of these bind 3CDpro, the 5' cloverleaf [17,18,22], the cre(2C) element [20,21] and the 3'NTR [19]. From these 3 structures only the 5' cloverleaf [18,19,22,44] and the cre(2C) stem loop structure [20,21,45] have been shown so far to form a biological relevant RNP complex with 3CDpro. The cloverleaf has been shown to be required for minus strand, and possibly also for plus strand RNA synthesis [17,46]. The RNP complexes of the cloverleaf with 3CDpro, which also include either PCBP2 or 3AB, are also required for both minus and plus strand RNA synthesis [17,19,44,47].
The other important cis-replicating element involved in poliovirus RNA replication, which also binds 3CDpro, is the cre(2C) hairpin [20,21,45]. A conserved AAA sequence in this RNA element serves as template for the synthesis of VPgpU(pU), the primer for RNA synthesis [20,45]. The role of 3CDpro in this reaction is believed to be to enhance the binding of the polymerase/VPg complex to the cre(2C) element [20,21,45]. The question whether the VPgpU(pU) made in this reaction is used exclusively for plus strand RNA synthesis [4] or also for minus strand synthesis remains controversial. The RNA binding sequences (R84/I86) of 3Cpro in 3CDpro but not amino acids R455/R456 at interface I in the 3Dpol domain are essential for the protein's stimulatory activity both in VPg-uridylylation in vitro [20,33] and in the stimulation of RNA synthesis in the translation-RNA replication system. Taken together these results are consistent with a possible role of either the 3CDpro/cloverleaf or the 3CDpro/cre(2C) interactions in the stimulatory activity of the protein in RNA synthesis, which is dependent on the RNA binding activity of the 3Cpro domain.
We have previously reported the interesting observation that the addition of purified protein 3Cpro(C147G) along with 3CDpro(3CproH40A) to translation-RNA replication reactions reduces the virus yield about ten thousand fold [8]. In this paper we show that at least one of the reasons for the nearly total inhibition of virus production under these conditions is that there is a striking inhibition of both minus and plus strand RNA synthesis. One possible explanation of our in vitro results is that the two proteins form a complex, through intermolecular contacts in 3Cpro [48], which is inactive and either cannot bind to the RNA or the RNP complex is nonfunctional. Alternatively, the two proteins interact with the same RNA sequence or structure but only the 3CDpro/RNA complex is functional in RNA synthesis. Of the three cis-replicating elements contained within PV RNA both the cloverleaf and the cre(2C) element have been shown to form RNP complexes with either 3CDpro or 3Cpro [17,21]. In case of the cloverleaf only the 3CDpro/RNP complex is functional in replication but both protein-RNA complexes stimulate VPg-uridylylation on the cre(2C) RNA element [33]. These results suggest that the RNA sequence or structure involved in the stimulatory activity of 3CDpro in RNA synthesis in the in vitro system is the cloverleaf rather than the cre(2C) element.
As we discussed above, the second step in the life cycle of PV where the extra 3CDpro(3CproH40A) appears to exert its stimulatory effect in vitro is during the late stages of particle assembly, and in particular during virus maturation. Although the addition of extra 3CDpro(3CproH40A) leads to a slight increase in the amount of small capsid precursors, the primary effect of the protein is at the step during which provirions are converted to mature viral particles. Although the mechanism of maturation cleavage is not fully understood it has been well established that the process is dependent on the presence of viral RNA [reviewed in [49]]. The exact function of 3CDpro(3CproH40A) in virus maturation is not yet known. Interestingly, both the RNA binding sequences in 3Cpro and the integrity of interface I in the 3Dpol domain of 3CDpro are required for function but the proteolytic activity of the protein is dispensable. The fact that the RNA binding domain of 3Cpro is essential for function indicates that 3CDpro has to interact with a sequence or structure in the viral RNA. The observation that the integrity of interface I in the 3Dpol domain of the protein is also required for this process is more difficult to explain. Although the oligomerization of 3CDpro along interface I in 3Dpol has not yet been directly tested, recent structural studies of the RNA polymerase suggest that oligomerization of the protein along interface I is possible [30]. In addition, recent studies of genetically modified 3CDpro polypeptides in RNA replication strongly support a role of 3CDpro/3CDpro complexes, mediated by 3Dpol domain contacts [50]. Whether the function of interface I in the 3Dpol domain of 3CDpro in virus maturation is related to the RNA binding properties of the protein remains to be determined. Our recent in vitro studies indicate that mutation 3DpolR455A/R456A in the context of 3CDpro alter the RNA binding properties of the protein such that twice as much of the mutant protein is required for optimal binding to a cre(2C) RNA probe than of the wt protein [Pathak and Cameron, unpublished results]. Oligomerization of 3CDpro might also be aided by intermolecular contacts between the 3Cpro domains of two molecules [48]. However, it should be noted that no interaction can be detected between 3Cpro molecules in chemical cross-linking experiments in vitro and only very poor, if any, complex formation can be observed between either 3Cpro/3Cpro or 3CDpro/3CDpro molecules in the yeast two hybrid system [51].
On the basis of these observations we propose 2 possible models for efficient virus maturation in the in vitro translation-RNA replication reactions supplemented with extra 3CDpro(3CproH40A). According to the first model 3CDpro(3CproH40A) interacts with the progeny plus strand RNA, possibly at the cloverleaf, and causes an important conformational change. This step requires the RNA binding activity of the 3Cpro domain of the protein but binding might also be enhanced by the oligomerization of the polypeptide along interface I in the 3Dpol domain. Subsequently the RNA interacts either with the pentamers or the empty capsid and it is encapsidated, yielding a provirion while 3CDpro(3CproH40A) leaves the complex. The correct conformation of the RNA inside the provirions affects the shape of the capsid such that now the cleavage of the VP0s is favored to complete maturation. The second model is similar to the first one except that now 3CDpro(3CproH40A) itself is encapsidated, bound to the progeny RNA. This keeps the RNA in the correct conformation inside the capsid so that the maturation cleavage of VP0 can occur. The second model is supported by previous studies by Newman and Brown who observed that 3CDpro, 3Dpol and 2CATPase proteins were contained within isolated poliovirus and foot-and-mouth disease virus particles [52]. In this context one should note that the scissile bond in VP0 is located on the rim of a trefoil-shaped depression on the capsid's inner surface, which has the potential of binding either RNA or other macromolecules [11]. However, we did not detect any 3CDpro in our 155S peak derived from reactions with extra 3CDpro(3CproH40A) using Western blot analysis with either anti 3Cpro or anti 3Dpol antibodies [data not shown]. In any case, the suboptimal concentration of functional 3CDpro in translation RNA-replication reactions might lead to progeny RNA molecules lacking the proper conformation for encapsidation and efficient virus maturation.
One of the factors that limits the use of the in vitro translation-RNA replication system in studies of RNA replication is the poor function of transcript RNAs as templates in the reaction, lowering the yield of progeny plus strand RNA and of virus to about 1% of what is obtained with virion RNA [39,40]. This has been attributed to the presence of two GMP molecules at the 5' end of RNAs transcribed from a T7 promoter [39]. We hoped that by supplying the inefficient in vitro reactions with an excess of 3CDpro(3CproH40A) the synthesis of plus strands, and consequently the production of mature virus could be enhanced. To our surprise, this does not happen. The simplest explanation of these observations is that the level of endogenous 3CDpro is sufficient for the synthesis of the low level of plus strand RNA that is produced in the system. Therefore supplying the reactions with extra 3CDpro(3CproH40A) would have no stimulatory effect. However, this explanation does not account for the fact that virus synthesis is not stimulated by 3CDpro(3CproH40A) in reactions containing ribozyme-treated transcript RNAs. The yield of virus in such reactions is 50-fold higher than in samples in which ribozyme-deficient transcripts were used as template for translation and RNA replication. The only known difference between viral RNA and ribozyme-treated transcript RNA is the lack of VPg in the latter structure. Therefore our results indicate that the presence of VPg at the 5' end of the input viral RNA [53,54] is an important determinant of the ability of 3CDpro(3CproH40A) to stimulate RNA synthesis and production of viable virions. Interestingly, the addition of extra 3CDpro(3CproH40A) at the beginning of incubation does not stimulate these processes once the newly made VPg-linked viral RNAs are used as templates for replication and packaging. This suggests that at least one of the stimulatory functions of 3CDpro is required at the time RNA synthesis is initiated from the input VPg-linked RNA template. Our results also suggest that either directly or indirectly the presence of VPg on the input RNA template is important for the stimulation by 3CDpro(3CproH40A) of the encapsidation of the newly made viral RNAs. The involvement of VPg in encapsidation has been previously proposed by Reuer et al. [15] who observed that some lethal VPg mutations still permit near normal minus and plus strand RNA synthesis in vivo.
It has been known for some time that complementation between viral proteins is more efficient in the in vitro translation-RNA replication system than in vivo. This is most likely due to relatively large local concentrations of viral proteins that are translated from the input viral RNA template used in the in vitro reactions. The results described in this paper show that at least two functions of 3CDpro are complementable in vitro. One is in RNA synthesis and the other in virus maturation and both of these processes require the RNA binding sequence of the 3Cpro domain. In an attempt to determine whether the RNA binding function of 3CDpro(3CproH40A) is required for additional processes in viral growth we tried to complement the lethal 3CproR84S/I86A mutation in the PV genome in vitro either by the addition of 3CDpro(3CproH40A) protein or wt 3CDpro mRNA. We obtained no virus suggesting that one or more of the RNA binding functions of 3CDpro, distinct from the ones described by us, cannot be complemented in vitro. An alternate explanation of the observation is that 3CDpro cannot substitute for 3Cpro in one or more of its functions.
The results presented in this paper have yielded insights into the steps of the viral life cycle in which the extra 3CDpro(3CproH40A) exerts its stimulatory function in the translation-RNA replication system. Our results also suggest a new role for protein 3CDpro in the life cycle of poliovirus, in virus maturation, which is dependent on the integrity of interface I in the 3Dpol domain of the protein. In addition, we have shown that a VPg-linked PV RNA linked template and the 3Cpro domain of the 3CDpro(3CproH40A) polypeptide are required both for the stimulation of RNA synthesis and for virus maturation. However, the exact mechanism of stimulation by 3CDpro both during RNA synthesis and particle assembly remains to be determined.
Materials and methods
Cells and viruses
HeLa R19 cell monolayers and suspension cultures of HeLa S3 cells were maintained in DMEM supplemented with 5% fetal bovine calf serum. Poliovirus was amplified on HeLa R19 cells as described before. The infectivity of virus stocks was determined by plaque assays on HeLa R19 monolayers, as described before [55].
Preparation of poliovirus RNA
Virus stocks were grown and purified by CsCl gradient centrifugation [55]. Viral RNA was isolated from the purified virus stocks with a 1:1 mixture of phenol and chloroform. The purified RNA was precipitated by the addition of 2 volumes of ethanol.
Preparation of HeLa cytoplamic extracts
HeLa S10 extracts were prepared as previously described [1,56] except for the following modifications: (1) packed cells from 2 liters of HeLa S10 were resuspended in 1.0 volumes (relative to packed cell volume) of hypotonic buffer; (2) the final extracts were not dialyzed.
Translation-RNA replication reactions with HeLa cell-free extracts and plaque assays
Viral RNA was translated at 34°C in the presence of unlabeled methionine, 200 μM each CTP, GTP, UTP, and 1 mM ATP in a total volume of 25 μl [1,5]. After incubation for 12–15 hr the samples were diluted with phosphate-buffered saline and were added to HeLa cell monolayers. Virus titers were determined by plaque assay, as described previously [1,55].
Filter binding assays for measurement of total RNA synthesis
Method I. Translation-RNA replication reactions (125 μl) were incubated at 34°C in the presence of 62.5 μC of [α-35S]CTP (ICN, 600Ci/mmole) but lacking unlabeled CTP. At the indicated times samples were taken and the reactions were stopped by the addition of SDS to a final concentration of 0.5%. The samples were extracted with phenol-chloroform and the RNA was precipitated with ethanol. The pellets were resuspended in 10 mM Tris pH 7.5, 1 mM EDTA and were loaded on a DEAE-81 filter papers (Whatman). The filters were dried and subsequently washed three times with 5% Na2HPO4, once with water and once with 70% ethanol, as described before [57]. Method II. Each translation-RNAreplication reaction was incubated separately at 34°C. At the indicated times (2, 4, 6, 8, and 16 hr) 12.5 μC of [α-35S]CTP was added and incubation was continued for 1 hr. The samples were treated and analyzed as described in Method I.
Preinitiation RNA replication complexes
Preinitiation RNA replication complexes were prepared as described previously [34] except for some minor modifications. Translation-RNA replication reactions, lacking initiation factors, were incubated for 4 hr at 34°C either in the presence or absence of 2 mM guanidine HCl. The complexes were isolated by centrifugation, resuspended in 50 μl HeLa S10 translation/replication reaction mixture without viral RNA, and incubated for 11 hr at 34°C.
Plus and minus strand RNA synthesis
Plus and minus strand RNA synthesis were determined as described previously [2]. Translation RNA replication reactions, programmed with viral RNA, were incubated for 4 hr in the presence of 2 mM guanidine HCl. The preinitiation replication complexes were resuspended in translation-RNA replication reactions lacking viral RNA in the presence of [α-32P]CTP. The reactions were incubated at 34°C for 1 hr, the labeled RNAs were separated by native agarose gel electrophoresis, and the products were visualized by autoradiography. The reaction products were quantitated with a Phosphorimager (Molecular Dynamics Storm 800) by measuring the amount the amount of [α-32P]CMP incorporated into RNA.
Alternatively, plus strand RNA synthesis was measured in translation-RNA replication reactions that were incubated for 4 hr at 34°C, in the absence of guanidine HCl, and the initiation complexes were isolated by centrifugation. They were resuspended in translation-RNA replication reactions lacking viral RNA but supplemented with [α-32P]CTP. The samples were incubated for 1 hr at 34°C and the RNA products were separated on a native agarose gel. The products were visualized by autoradiography.
In vitro transcription and translation
All plasmids were linearized with EcoRI prior to transcription by T7 RNA polymerase. The transcript RNAs were purified by phenol/chloroform extraction and ethanol precipitation. Translation reactions (25 μl) containing 8.8 μC of Trans [35S]Label (ICN Biochemicals) were incubated for 4 hours at 34°C [5]. The samples were analyzed by electrophoresis on sodium deodecyl sulfate-12% polyacrylamide gels, followed by autoradiography.
RNA synthesis with crude replication complexes
Crude replication complexes (CRCs) were prepared by a method similar to what has been described before [35]. HeLa cell monolayers (15 cm) were infected with PVM at a multiplicity of infection of 500. After 6 hr incubation at 37°C the cells were resuspendend in hypotonic buffer [35] and were lysed with a Dounce homogenizer. Cell debris and nuclei were removed by centrifugation for 20 min at 33,000 × g. The pellet was subsequently resuspended in 1 ml of 10 mM Tris-HCl pH 8.0, 10 mM NaCl, and 15% glycerol. Aliquots were stored at -80°C.
RNA synthesis by CRCs was measured as described before [3]. In vitro translation-RNA replication reactions were assembled in which the HeLa extracts were replaced by CRCs (20% by volume). The reaction contained 49% by volume of S10 buffer [2] and 25 μC of [α-32P]CTP.
Sucrose gradient centrifugation of viral particles
HeLa S10 translation-RNA replication reactions (25 μl) were incubated in the presence of 8.8 μC of [35S]TransLabel (ICN Biochemicals) for 12 hr at 34°C. The excess unincorporated label was removed by dialysis. The samples were introduced into a Slide-a-lyzer (Pierce Endogen) dialysis cassette with a M.Wt cut-off of 10 kD and were dialyzed several times against phosphate buffer at 4°C until essentially all the excess label was eliminated. After dialysis the samples were centrifuged at 14,000 × g to remove any precipitated material. The samples were diluted to 500 μl and were centrifuged in a 5–20% sucrose density gradient in phosphate buffered saline containing 0.01% bovine serum albumin in a SW41 rotor at 40,000 rpm at 4°C. To separate 80S empty capsids and 155S virus particles (provirions and virions) the gradients were centrifuged for 80 min [36]. To identify 5S protomers and 14S pentamers the gradients were centrifuged for 15 hr. Fractions (0.5 ml) were collected from the bottom of the gradients and the radioactivity of each sample was determined by scintillation counting. In each sucrose gradient cetrifugation size markers were sedimented in parallel consisting of [35S]-labeled PV-infected HeLa cell extracts.
Western blot analysis
For the identification of the capsid proteins present in sucrose gradient fractions Western blot analysis was used [58]. Samples were loaded on a SDS-polyacrylamide gel (12.5% acrylamide) and after separation the proteins were transferred to a nitrocellulose membrane (Protran; Schleicher&Schuell). The membrane was probed with a rabbit polyclonal antibody to PV capsid protein VP2.
Electron microscopy
Standard electron microscopy processing techniques were used for negative staining. Briefly, formvar coated, 200 mesh nickel grids were prepared. Grids, sample side down were floated on droplets of suspended poliovirus, followed by fixation in a solution of 1% glutaraldehyde in 0.1 M phosphate buffered saline (PBS), pH 7.4. Samples were washed in PBS, then in water followed by phosphotungstic acid. The samples were viewed with a F. E. I. Tecnai 12 BioTwin electron microscope and digital images were captured with an ATM camera system. In each sample the viral particles were counted within a 20 mm2 area.
Proteins
The following PV proteins with a C-terminal his tag were expressed in E. coli and purified by nickel column chromatography (Qiagen): 3CDpro(3CproH40A), a proteinase active site mutant [20]; 3Cpro(3CproC147G), a proteinase active site mutant [33]. The purification of 3CDpro(3CproH40G, 3DpolR455A/R456A) was described previously [33]. This protein contains both a proteinase active site mutation (3CproH40G) and a mutation (3DpolR455A/R456A) at interface I in the 3Dpol domain of the protein.
Plasmids
Poliovirus sequences were derived from plasmid pT7PVM, which contains the full-length (nt 1–7525) plus strand poliovirus cDNA sequence [38]. All constructs were sequenced to ensure their accuracy. The construction of plasmids pLOP315ser and pLOP315(3CproR84S/I86A) was described before [8,9]. Both plasmid DNAs were linearized with EcoRI prior to transcription with T7 RNA polymerase.
Authors' contributions
DF carried out all the experiments and made substantial contributions to the design of the experiments. HP contributed purified mutant enzymes for the study. CEC has contributed to the interpretation of the data and revised the manuscript critically. BR initiated the studies on this subject. EW contributed to the design of the experiments and revised the manuscript critically. AVP planned the experiments and wrote the manuscript. All authors read and approved the final manuscript.
Acknowledgements
We are grateful to D. W. Kim for his help in the preparation of HeLa cell-free extracts and for helpful discussions. We thank R. Andino for the plasmid containing PV1(M) cDNA preceded by a hammer-head ribozyme, prib(+)XPA and S. Van Horn for the electron microscopic analyses. This work was supported by two grants from the National Institute of Allergy and Infectious Diseases (E. Wimmer, R37 AI015122-30; and C. Cameron AI053531).
==== Refs
Molla A Paul AV Wimmer E Cell-free de novo synthesis of poliovirus Science 1991 254 1647 1651 1661029
Barton DJ Flanegan JB Synchronous replication of poliovirus RNA: initiation of negative-strand RNA synthesis requires the guanidine-inhibited activity of protein 2C J Virol 1997 71 8482 8489 9343205
Fogg MH Teterina NL Ehrenfeld E Membrane requirements for uridylylation of the poliovirus VPg protein and viral RNA synthesis in vitro J Virol 2003 77 11408 11416 14557626 10.1128/JVI.77.21.11408-11416.2003
Murray KE Barton DJ Poliovirus CRE-dependent VPg uridylylation is required for positive-strand RNA synthesis but not for negative-strand RNA synthesis J Virol 2003 77 4739 4750 12663781 10.1128/JVI.77.8.4739-4750.2003
Molla A Paul AV Wimmer E Effects of temperature and lipophilic agents on poliovirus formation and RNA synthesis in a cell-free system J Virol 1993 67 5932 5938 8396661
Towner JS Mazanet MM Semler BL Rescue of defective poliovirus RNA replication by 3AB-containing precursor polyproteins J Virol 1998 72 7191 7200 9696813
Wimmer E Hellen CUT Cao X Genetics of poliovirus Annu Rev Genet 1993 27 353 436 8122908 10.1146/annurev.ge.27.120193.002033
Franco D Pathak HB Cameron CE Rombaut B Wimmer E Paul AV Stimulation of poliovirus synthesis in a HeLa cell-free in vitro translation-RNA replication system by viral protein 3CDpro J Virol 2005 79 6358 6367 15858019 10.1128/JVI.79.10.6358-6367.2005
Verlinden Y Cuconati A Wimmer E Rombaut B The viral protein 3CD induces an equilibrium between the viral protein and RNA synthesis in a cell-free system for poliovirus replication Arch Virol 2002 147 731 744 12038684 10.1007/s007050200022
Paul AV Semler BL, Wimmer E Possible unifying mechanism of picornavirus genome replication Molecular biology of picornaviruses 2002 ASM press:Washington, DC 227 246
Rossman MG Semler BL, Wimmer E Picornavirus structure overview Molecular biology of picornaviruses 2002 ASM press:Washington, DC
Verlinden Y Cuconati A Wimmer E Rombaut B Cell-free synthesis of poliovirus: 14S subunits are the key intermediates in the encapsidation of poliovirus RNA J Gen Virol 2000 81 2751 2754 11038388
Nugent CI Kirkegaard K RNA binding properties of poliovirus subviral particles J Virol 1995 69 13 22 7983702
Vance LM Moscufo N Chow M Heinz BA Poliovirus 2C region functions during encapsidation of viral RNA J Virol 1997 71 8759 8765 9343235
Reuer Q Kuhn RJ Wimmer E Characterization of poliovirus clones containing lethal and nonlethal mutations in the genome-linked protein VPg J Virol 1990 64 2967 2975 2159557
Ypma-Wong MF Dewalt PG Johnson VH Lamb JG Semler BL Protein 3CD is the major poliovirus proteinase responsible for cleavage of the P1 capsid precursor Virology 1988 166 265 270 2842953 10.1016/0042-6822(88)90172-9
Andino R Rieckhof E Achacoso PL Baltimore D Poliovirus RNA synthesis utilizes an RNP complex formed around the 5'-end of viral RNA EMBO J 1993 12 3587 3598 8253083
Gamarnik AV Andino R Interactions of viral protein 3CD and poly(rC) binding protein with the 5' nontranslated region of the poliovirus genome J Virol 2000 74 2219 2226 10666252 10.1128/JVI.74.5.2219-2226.2000
Harris K Xiang W Alexander L Lane WS Paul AV Wimmer E Interaction of the poliovirus polypeptide 3CDpro with the 5' and 3' termini of the poliovirus genome: identification of viral and cellular factors cofactors needed for binding J Biol Chem 1994 269 27004 27014 7929441
Paul AV Rieder E Kim DW van Boom JH Wimmer E Identification of an RNA hairpin in poliovirus RNA that serves as the primary template for the uridylylation of VPg in vitro J Virol 2000 74 10359 10370 11044080 10.1128/JVI.74.22.10359-10370.2000
Yin J Paul AV Wimmer E Rieder E Functional dissection of a poliovirus cis-acting replication element [PV-cre(2C)]: analysis of single- and dual-cre viral genomes and proteins that bind specifically to PV-cre RNA J Virol 2003 77 5152 5166 12692218 10.1128/JVI.77.9.5152-5166.2003
Parsley TB Towner JS Blyn LB Ehrenfeld E Semler BL Poly(rC)binding protein 2 forms a ternary complex with the 5'-terminal sequences of poliovirus RNA and the viral 3CD proteinase RNA 1997 3 1124 1134 9326487
Gamarnik AV Andino R Switch from translation to RNA replication in a positive strand RNA virus Genes Dev 1998 12 2293 2304 9694795
Herold J Andino R Poliovirus RNA replication requires genome circularization through a protein bridge Mol Cell 2001 7 581 591 11463383 10.1016/S1097-2765(01)00205-2
Blair WS Parsley TB Bogerd HP Towner JS Semler BL Cullen BR Utilization of a mammalian cell-based RNA binding assay to characterize the RNA binding properties of picornavirus 3C proteinases RNA 1998 4 215 225 9570321
Hammerle T Molla A Wimmer E Mutational analysis of the proposed FG loop of poliovirus proteinase 3C identifies amino acids that are necessary for 3CD cleavage and might be determinants of a function distinct from proteolytic activity J Virol 1992 66 6028 6034 1326654
Cornell CT Semler BL Subdomain specific functions of the RNA polymerase region of poliovirus 3CD polypeptide Virology 2002 298 200 213 12127783 10.1006/viro.2002.1481
Parsley TB Cornell CT Semler BL Modulation of the RNA binding and protein processing activities of poliovirus polypeptide 3CD by the viral RNA polymerase domain J Biol Chem 1999 274 12867 12876 10212275 10.1074/jbc.274.18.12867
Flanegan JB Baltimore D Poliovirus-specific primer-dependent RNA polymerase able to copy poly(A) Proc Natl Acad Sci USA 1977 74 3677 3680 198796
Thompson AA Peersen OB Structural basis for proteolysis-dependent activation of the poliovirus RNA-dependent RNA polymerase EMBO J 2004 23 3462 3471 15306852 10.1038/sj.emboj.7600357
Pata JD Schultz SC Kirkegaard K Functional oligomerization of poliovirus RNA dependent RNA polymerase RNA 1995 1 466 477 7489508
Diamond SE Kirkegaard K Clustered charged-to-alanine mutagenesis of poliovirus RNA-dependent RNA polymerase yields multiple temperature sensitive mutants defective in RNA synthesis J Virol 1994 68 863 876 8289389
Pathak HB Ghosh SKB Roberts AW Sharma SD Yoder JD Arnold JJ Gohara DW Barton DJ Paul AV Cameron CE Structure-function relationships of the RNA-dependent RNA polymerase from poliovirus (3Dpol). A surface of the primary oligomerization domain functions in capsid precursor processing and VPg uridylylation J Biol Chem 2002 277 31551 31562 12077141 10.1074/jbc.M204408200
Barton DJ Black EP Flanegan JB Complete replication of poliovirus in vitro: preinitiation RNA replication complexes require soluble cellular factors for the synthesis of VPg-linked RNA J Virol 1995 69 5516 5527 7636997
Takeda N Kuhn RJ Yang CF Takegami T Wimmer E Initiation of poliovirus plus-strand RNA synthesis in a membrane complex of infected HeLa cells J Virol 1986 60 43 53 3018300
Lee MW Wang W Human rhinovirus type 16: Mutant V1210A requires capsid-binding drug for assembly of pentamers to form virions during morphogenesis J Virol 2003 77 6235 6244 12743280 10.1128/JVI.77.11.6235-6244.2003
Fernandez-Tomas CB Baltimore D Morphogenesis of poliovirus. II. Demonstration of a new intermediate, the provirion J Virol 1973 12 1122 1130 4358164
van der Werf S Bradley J Wimmer E Studier FW Dunn JJ Synthesis of infectious poliovirus RNA by purified T7 RNA polymerase Proc Natl Acad Sci USA 1986 83 2330 2334 3010307
Herold J Andino R Poliovirus requires a precise 5' end for efficient positive-strand RNA synthesis J Virol 2000 74 6394 6400 10864650 10.1128/JVI.74.14.6394-6400.2000
Teterina NL Rinaudo MS Ehrenfeld E Strand-specific RNA synthesis defects in a poliovirus with a mutation in 3A J Virol 2003 77 12679 12691 14610190 10.1128/JVI.77.23.12679-12691.2003
Teterina NL Zhou WD Cho MW Ehrenfeld E Inefficient complementation activity of poliovirus 2C and 3D proteins for rescue of lethal mutations J Virol 1995 69 4245 4254 7769684
Egger D Teterina N Ehrenfeld E Bienz K Formation of the poliovirus replication complex requires coupled viral translation, vesicle production and viral RNA synthesis J Virol 2000 74 6570 6580 10864671 10.1128/JVI.74.14.6570-6580.2000
Novak JE Kirkegaard K Coupling between genome translation and replication in an RNA virus Genes and Dev 1994 8 1726 1737 7958852
Xiang W Harris KS Alexander L Wimmer E Interaction between the 5'-terminal cloverleaf and 3AB/3CDpro of poliovirus is essential for RNA replication J Virol 1995 69 3658 3667 7745714
Rieder E Paul AV Kim DW Van Boom JH Wimmer E Genetic and biochemical studies of poliovirus cis-acting replication element cre in relation to VPg uridylylation J Virol 2000 74 10371 10380 11044081 10.1128/JVI.74.22.10371-10380.2000
Lyons T Murray KE Roberts AW Barton DJ Poliovirus 5'-terminal cloverleaf RNA is required in cis for VPg uridylylation and the initiation of negative-strand RNA synthesis J Virol 2001 75 10696 10708 11602711 10.1128/JVI.75.22.10696-10708.2001
Walter BL Parsley TB Ehrenfeld E Semler BL Distinct poly(rC) binding protein KH domain determinants for poliovirus translation initation and viral RNA replication J Virol 2002 76 12008 12022 12414943 10.1128/JVI.76.23.12008-12022.2002
Mosimann SC Cherney MM Sia S Plotch S James MNG Refined X-ray crystallographic structure of the poliovirus 3C gene product J Mol Biol 1997 273 1032 1047 9367789 10.1006/jmbi.1997.1306
Basavappa R Syed R Flore O Icenogle JP Filman DJ Hogle JM Role and mechanism of the maturation cleavage of VP0 in poliovirus assembly: Structure of the empty capsid assembly intermediate at 2.9 A resolution Protein Science 1994 3 1651 1669 7849583
Cornell CT Brunner JE Semler BL Differential rescue of poliovirus RNA replication functions by genetically modified RNA polymerase precursors J Virol 2004 78 13007 13018 15542652 10.1128/JVI.78.23.13007-13018.2004
Xiang W Cuconati A Hope D Kirkegaard K Wimmer E Complete protein linkage map of poliovirus P3 proteins: interaction of polymerase 3Dpol with VPg and with genetic variants of 3AB J Virol 1998 72 6732 6741 9658121
Newman JFE Brown F Foot-and-Mouth disease virus and poliovirus particles contain proteins of the replication complex J Virol 1997 71 7657 7662 9311848
Ambros V Baltimore D Protein is linked to the 5' end of poliovirus RNA by a phophodiester linkage to tyrosine J Biol Chem 1978 253 5263 5266 209034
Nomoto A Kitamura N Golini F Wimmer E The 5'terminal structures of poliovirion RNA and poliovirus mRNA differ only in the genome-linked protein Proc Natl Acad Sci USA 1977 74 5345 5349 202952
Lu HH Yang CF Murdin AD Klein MH Harber JJ Kew OM Wimmer E Mouse neurovirulence determinants of poliovirus type 1 strain LS-a map to the coding regions of capsid protein VP1 and proteinase 2Apro J Virol 1994 68 7507 7515 7933134
Cuconati A Molla A Wimmer E Brefeldin A inhibits cell-free, de novo synthesis of poliovirus J Virol 1998 72 6456 6464 9658088
Paul AV Rieder E Kim DW van Boom JH Wimmer E Identification of an RNA hairpin in poliovirus RNA that serves as the primary template for the uridylylation of VPg in vitro J Virol 2000 74 10359 10370 11044080 10.1128/JVI.74.22.10359-10370.2000
Pfister T Jones KW Wimmer E A cysteine-rich motif in poliovirus protein 2CATPase is involved in RNA replication and binds zinc in vitro J Virol 2000 74 334 343 10590122
|
16300678
|
PMC1315353
|
CC BY
|
2021-01-04 16:38:57
|
no
|
Virol J. 2005 Nov 21; 2:86
|
utf-8
|
Virol J
| 2,005 |
10.1186/1743-422X-2-86
|
oa_comm
|
==== Front
Theor Biol Med ModelTheoretical Biology & Medical Modelling1742-4682BioMed Central London 1742-4682-2-461628393110.1186/1742-4682-2-46ResearchAnalysis of variation of amplitudes in cell cycle gene expression Liu Delong [email protected] Kevin W [email protected] Russ [email protected] CIIT Ceters for Health Research, 6 Davis Drive, Research Triangle Park, NC 27709, USA2 The SAS Institute Inc., SAS Campus Drive, Cary, NC 27513, USA2005 11 11 2005 2 46 46 31 8 2005 11 11 2005 Copyright © 2005 Liu et al; licensee BioMed Central Ltd.2005Liu 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 gene expression among cells in a population is often considered as noise produced from gene transcription and post-transcription processes and experimental artifacts. Most studies on noise in gene expression have emphasized a few well-characterized genes and proteins. We investigated whether different cell-arresting methods have impacts on the maximum expression levels (amplitudes) of a cell cycle related gene.
Results
By introducing random noise, modeled by a von Mises distribution, to the phase angle in a sinusoidal model in a cell population, we derived a relationship between amplitude and the distribution of noise in maximum transcription time (phase). We applied our analysis to Whitfield's HeLa cell cycle data. Our analysis suggests that among 47 cell cycle related genes common to the 2nd experiment (thymidine-thymidine method) and the 4th experiment (thymidine-nocodazole method): (i) the amplitudes of CDC6 and PCNA, which are expressed during G1/S phase, are smaller in the 2nd experiment than in the 4th, while the amplitude of CDC20, which is expressed during G2/M phase, is smaller in the 4th experiment; and (ii) the two cell-arresting methods had little impact on the amplitudes of the other 43 genes in the 2nd and 4th experiments.
Conclusion
Our analysis suggests that procedures that arrest cells in different stages of the cell cycle differentially affect expression of some cell cycle related genes once the cells are released from arrest. The impact of the cell-arresting method on expression of a cell cycle related gene can be quantitatively estimated from the ratio of two estimated amplitudes in two experiments. The ratio can be used to gauge the variation in the phase/peak expression time distribution involved in stochastic transcription and post-transcriptional processes for the gene. Further investigations are needed using normal, unperturbed and synchronized HeLa cells as a reference to compare how many cell cycle related genes are directly and indirectly affected by various cell-arresting methods.
==== Body
Introduction
Variation in gene expression is often considered as noise or uncertainty arising from experimental artifacts and biological variability. Various studies of noise in gene expression have focused on different scales, ranging from a single gene [1] to a single cell [2,3] to a cell population [4-9]. These studies have greatly helped us understand the effects of stochastic noise in gene expression and gene regulation in various model organisms. In a similar spirit, we were interested in the effects of different cell-arresting methods on the maximum expression levels (amplitudes) of some cell cycle related genes.
Various methods such as chemical induction and temperature shift have been used to arrest cells in genome-wide cell cycle studies [10-13]. Each method may have direct or indirect impacts on the synthesis or degradation of mRNAs from some genes after the interrupted cell cycle resumes. For example Whitfield et al. [11] used thymidine-thymidine (thy-thy) to arrest HeLa cells in G1/S phase and thymidine-nocodazole (thy-noc) to arrest them in G2/M phase. Intuitively, the synthesis or degradation of some mRNAs in G1/S phase and G2/M may be differentially affected by thy-thy and thy-noc arrests, respectively.
Measurements of the intensities of gene expression from microarray experiments are subject to two main sources of variation: (i) technical variability including bioassay preparation, dye-effect and hybridization on chips, (ii) and biological variability including variation in activation of transcription from cell to cell in a population after release from cell cycle arrest. Another implicit feature of microarray data is that gene expression is an average value over a cell population rather than in a single cell. In general, it is difficult to separate these two sources of variation for expression of a gene under given experimental conditions unless multiple repeated measurements are made over time and some prior knowledge of the expression of this gene is available. Periodic expression of some genes may be a good model for examining the effects of various cell-arresting methods on the transcription of known genes during cell cycle experiments.
Some advantages of using cell cycle related gene expression to probe the variation in maximum expression level due to different cell-arresting methods are: (i) cells can be synchronized to some extent so that variation of expression from cell to cell can be reduced; (ii) the expression profiles of some known cell cycle related genes such as PCNA and CDC20 (Figures 1 and 2) have been well characterized as sinusoidal waveforms over multiple cycles in different model organisms [10-13]. This makes it relatively easy to distinguish biological variation from technical variation, which produces random or transient fluctuations around a sinusoidal profile over time.
Figure 1 Log2 expression ratio for PCNA, a known G1/S phase gene, in thymidine-thymidine (exp2) arrest and thymidine-nocodazole arrest (exp4) studies. The solid line ('__') is the fit, which is estimated from the random-periods model (1), to the data ('o') from Whitfield et al. (2002).
Figure 2 Log2 expression ratio for CDC20, a known G2/M phase gene, in thymidine-thymidine (exp2) arrest and thymidine-nocodazole arrest (exp4) studies. The solid line ('__') is the fit, which is estimated from the random-periods model (1), to the data ('o') from Whitfield et al. (2002).
Amplitude, period and phase angle define the dynamics of a sinusoidal profile. In cell cycle or circadian rhythm studies, the phase angle, or time of maximum expression of a cycling gene, has been a primary focus because it reflects the gene's biological role [10-15]. However, the biological implications of amplitudes of cycling genes, referred to as the maximum expression level in one cycle, have not been explored in any previous microarray study of cell cycle or circadian cycle gene expression [10-15]. This might be due to the impression that gene expression from high-throughput data is noisy and therefore not reliable. Alternatively, it may be because no control (reference) mRNA was used across the experiments. When the expression of a cycling gene is measured across multiple time points in cell cycle modeled by a sinusoidal profile, its amplitude can be estimated with reasonable accuracy [16]. When a common reference mRNA is used in cell cycle experiments [11], the estimated amplitudes of the same cycling genes should be comparable across experiments. In addition to phases, changes in amplitude may reveal effects of cell-arrest methods on the expression of some cell cycle related genes.
In a single cell, the amplitude and phase of a cell cycle related gene are considered two independent parameters in a sinusoidal model. Within a cell population, however, variation in amplitude may be dependent on variation in phase angle for some genes of this kind when the cells are stressed at different stages of the cycle. The linking of amplitude to phase variability is similar to Winfree's suggestion about the connection: "Thirty-four years later the situation is beginning to change. It is at least widely recognized now that 'phase' is just one aspect of the circadian clock's 'state,' needing supplementation by at least 'amplitude' (possibly a measure of cell-population phase scatter) before experiments can be designed and interpreted with confidence" [17].
In this paper, we first illustrate how variation in amplitude depends on the distribution of phase angles of a cell cycle related gene in a cell population. We then analyze the effects of two different cell-arresting methods on some known cell cycle related genes expressed in G1/S and G2/M phases, using public cell cycle gene expression datasets.
Methodologies
Three parameters are commonly used for modeling the time-course of expression, yg(t), of a cell cycle related gene g over time t: amplitude, which we denote as Kg; duration of cycle (period), T; and phase angle, φg, which is the time in the cycle when the gene is maximally activated; i.e. yg(t) = f(t; Kg, T, φg). In our previous cell cycle related gene expression studies [16], we introduced a variance parameter σ to yg(t) for modeling attenuation of the amplitude of gene g over time, leading to the following random-periods model (RPM):
where the integral averages the expression level across cells and z is assumed to be distributed as standard Gaussian. The linear terms, ag and bg, give the background gene expression. This model approximated the pattern of cycling, with its attenuation across time, when it was applied to a set of known cell cycle related genes [16].
Here, we introduce random noise, ε, to the phase of gene expression in a cell population into model (1). The expectation, E[ ], of the periodic term, which we call cg(t) in (1) for gene g, is
where ε is von Mises distributed with concentration parameter κ and mean direction 0, and z is, as before, normally distributed with mean 0 and variance 1. Kgmax is the amplitude when ε = 0, i.e. no variation in phase/peak expression time for gene g in a population of perfectly synchronized cells. The expectation of cg(t) in (2), E⌊cg(t)⌋, can be expanded as
If the random variables z and ε are independent, we obtain the simplified expression
Since for the random variable ε with a von Mises distribution, we obtain
Therefore, the amplitude Kg in model (1) is the product of two terms, Kg max and E[cos(ε)] in (3). E[cos(ε)] can be considered a measure of the variability in phase across cells in a given experiment. When the duration of the cell cycle is highly variable, as when σ is large in model (1), one might expect a corresponding attenuation of the amplitude over time.
Since it is difficult to estimate both the amplitude Kg max and the term E[cos(ε)] directly from (3), we propose instead to compare the amplitude parameters in two independent experiments under the same protocol for g gene, by taking the ratio
where, , κg is the concentration parameter of ε with a von Mises distribution [18], and K1g and K2g are the maximum expressions of gene g in experiments 1 and 2, respectively, when the phases or peak expression times for g in a cell population are perfectly synchronized. We have 0 ≤ E(cos(ε)) ≤ 1 as the concentration parameter κg → ∞, the variance goes to 0 and E[cos(ε)] = 1; and as κg = 0, E[cos(ε)] = 0.
Provided that K1g = K2g, we reduce the ratio in (4) to
Equation (5) implies that the ratio between the amplitude parameters of periodic expression of gene g in experiments 1 and 2 can be represented by the ratio of the mean noise variation, which has von Mises distributions in both experiments. When κ1 >κ2, E[c1g(t)]max >E[c2g(t)]max. In biological terms, the concentration parameter, κ, reflects the distribution of phases or peak expression times for a gene within a cell population. Therefore, we can use the ratio of estimated amplitudes from RPM (1) to examine the relative variability in phase/peak expression time for gene g in two cell cycle experiments.
To get a sense of how the ratios of estimated amplitude in (5) vary with κ, we calculated numerical values of E[cos(ε)] for the random variable ε with μ and κ = 1, 2, 3, ..., 20, and plotted κ vs. E[cos(ε)] in Figure 3. For κ = 1, 2, 3, 4, 5, E[cos(ε)] = 0.33, 0.57, 0.71, 0.79, 0.84, respectively. For example, for κ = 2 and 5, the ratio in (5) is 0.57/0.84 = 0.68. Note that E[cos(ε)] increases sharply and monotonically from κ = 1 to κ = 5. Figure 3 suggests that, for a cycling gene in two experiments with relatively large differences in amplitude, the concentration parameters κ in the experiment with small estimated amplitude are relatively small and most likely to be in the range 1 ≤ κ ≤ 5. Although we have no direct knowledge of the true value of κ for a cycling gene in any experiment, we can still use Figure 3 to interpret the variation in transcription of a given gene within a cell population in multiple experiments. For example, within a HeLa cell cycle period of 15 h, phases in the interval (-0.65, 0.65) radians, or peak gene expression times in the interval (-1.5, 1.5) h, are within 95% coverage of the von Mises distribution with concentration parameter κ = 10.
Figure 3 Plot of concentration parameter κ vs. expectation of cos(ε), where ε is von Mises distributed with zero mean direction and concentration κ, i.e., ε ~ VM(κ,0).
In the following two sections, we apply the concepts presented above to the variation in amplitude of a set of cycling genes common to two experiments, using the cell cycle gene expression data of Whitfield et al. [11]. Here, we are primarily interested in assessing the variability of amplitudes of cell cycle related genes commonly expressed in two experiments where cells were arrested by two different methods, and in identifying genes of which the amplitudes Kg do change in two experiments if there is no systematic variation between any pair of experiments.
Testing equality of amplitudes of a set of cycling gene in two experiments
Let and denote the estimated amplitude and the variance of the amplitude for the gth gene in the jth experiment, g = 1, ..., n, where n is the number of genes and j = x, y. is estimated from the random-periods model in (1), and from Wald's sandwich estimator within the random-periods model (1). Prior to testing the equality of amplitude of a cycling gene in two experiments, we need to check whether there is a systematic variation in amplitude, which might be created during sample hybridization. For a set of n genes between two experiments, x and y, we take the difference
and use the Wilcoxon signed rank test to test the null hypothesis: median Δ = 0. If the null hypothesis is rejected, we suspect that there may exist a systematic difference between and in experiments x and y. If we fail to reject the null, there may be no true difference, or the statistical test lacked sufficient power to detect a true difference (which is small compared to the estimated noise in the experiment). In this situation we explore the results further to identify how many and may be equal for g = 1, ..., n by checking whether zero is included in the confidence interval at the level of α, where and are the estimated variances of and . If , transcription of the gene g might not differ between the two experiments.
Example
In our previous work [19], we studied the phase association of 47 cell cycle related genes common to the 2nd, 3rd and 4th experiments of Whitfield et al. [11]. In the present study, we use the same 47 genes commonly expressed in the 2nd and 4th experiments with 26 and 19 time points per gene, respectively. The amplitude, period, geometric standard deviation, phase angle and two parameters describing the linear background, denoted respectively by (), were estimated for each expression time-course experiment using the random-periods model (1) on log2 transformed data. The assumptions underlying the model appear reasonable for these data, although our conclusions are somewhat limited given the small sample size. Owing to the systematically smaller amplitudes of the 47 cell cycle related genes in the 3rd experiment of Whitfield et al. [11], which were identified by the Wilcoxon signed rank test of (6), we excluded the 3rd experiment from our comparison of amplitudes in this study. The estimated amplitudes s, and the variances of the s, g = 1, ..., 47, in the 2nd and 4th experiments are listed in Table 1.
Table 1 Estimated amplitudes , and variances var(), var() of the amplitudes in the 2nd and 4th experiments of Whitfield et al. (2002), respectively.
Assession Gene Symbol K_2 var(K_2) K_4 var(K_4) lower bound upper bound flag
AA088457 0.921 0.026 0.642 0.007 -0.637 0.076 1
AA458994 PMSCL1 0.832 0.018 0.576 0.019 -0.635 0.122 1
AA485454 0.772 0.029 0.743 0.043 -0.554 0.495 1
AA485454 0.772 0.029 0.743 0.043 -0.554 0.495 1
AA282935 MPHOSPH1 0.950 0.030 0.843 0.049 -0.658 0.444 1
N57722 MCM6 0.401 0.013 0.596 0.024 -0.180 0.570 1
AA485454 0.747 0.035 0.743 0.043 -0.551 0.542 1
AA485454 0.747 0.035 0.743 0.043 -0.551 0.542 1
R11407 STK15 1.672 0.049 1.821 0.050 -0.467 0.765 1
T66935 DKFZp762E1312 1.648 0.051 1.319 0.035 -0.903 0.245 1
AA452513 KNSL5 1.162 0.033 1.155 0.062 -0.609 0.595 1
AA157499 MAPK13 1.375 0.045 1.360 0.060 -0.650 0.620 1
AA430092 BUB1 1.083 0.033 1.003 0.085 -0.755 0.593 1
AA053556 MKI67 1.315 0.056 0.790 0.043 -1.144 0.095 1
R96941 C20orf129 1.217 0.076 1.444 0.022 -0.387 0.840 1
AA131908 FLJ10540 0.786 0.016 0.390 0.014 -0.738 -0.053 0
AA279990 TACC3 0.794 0.026 1.023 0.055 -0.329 0.786 1
AA464019 E2-EPF 0.760 0.018 0.987 0.077 -0.378 0.832 1
AA262211 KIAA0008 0.918 0.013 0.688 0.030 -0.635 0.176 1
AI053446 0.964 0.041 0.952 0.050 -0.605 0.581 1
AA620485 ANKT 0.871 0.021 1.150 0.036 -0.192 0.750 1
AA629262 PLK 1.621 0.019 1.510 0.042 -0.597 0.375 1
AA450264 PCNA 0.557 0.008 0.985 0.038 0.007 0.849 0
R06900 RAMP 1.055 0.033 1.322 0.045 -0.280 0.814 1
AA425120 CHAF1B 0.549 0.006 0.552 0.032 -0.378 0.383 1
AA430511 FLJ14642 0.922 0.028 0.859 0.045 -0.592 0.465 1
AA430511 FLJ14642 0.922 0.028 0.786 0.061 -0.721 0.449 1
AA620553 FEN1 0.484 0.010 0.516 0.013 -0.270 0.335 1
AA402431 CENPE 1.468 0.015 1.455 0.082 -0.624 0.599 1
AA608568 CCNA2 1.197 0.016 1.115 0.076 -0.677 0.513 1
W93120 0.584 0.019 1.210 0.091 -0.026 1.278 1
N63744 FLJ10468 1.602 0.021 1.146 0.067 -1.038 0.125 1
R22949 1.055 0.026 0.908 0.045 -0.670 0.377 1
H51719 ORC1L 0.607 0.013 0.469 0.021 -0.500 0.222 1
AA425404 FLJ10156 1.101 0.041 0.841 0.010 -0.702 0.182 1
H59203 CDC6 0.695 0.014 1.060 0.020 0.005 0.724 0
AA292964 CKS2 0.827 0.010 1.516 0.185 -0.177 1.553 1
AA099033 USP1 0.507 0.012 0.750 0.027 -0.145 0.630 1
AA598776 CDC20 1.258 0.017 0.619 0.031 -1.067 -0.212 0
AA676797 CCNF 1.617 0.045 1.141 0.048 -1.072 0.121 1
AA504625 KNSL1 1.222 0.026 0.806 0.033 -0.893 0.062 1
AA235662 FLJ14642 1.041 0.015 0.944 0.041 -0.559 0.365 1
H73329 C20orf1 1.017 0.018 1.255 0.066 -0.330 0.807 1
AA421171 NUF2R 0.982 0.018 1.021 0.049 -0.467 0.546 1
T54121 CCNE1 1.155 0.045 1.144 0.052 -0.623 0.600 1
AA010065 CKS2 0.919 0.006 1.267 0.031 -0.028 0.724 1
Note that the accession numbers and the gene symbols were taken from the dataset of Whitfield et al. (2002). The genes with value 0 in the flag column indicate that the amplitudes are not same in the 2nd and 4th experiments.
Results
The p-value from the Wilcoxon signed rank test on the median Δ = 0 in (6) at the level of α = 0.05 is 0.56, suggesting that the median amplitudes in exp2 and exp4 are similar. Therefore, we can directly compare the estimated amplitudes for each of the 47 genes in the two experiments. The log2 ratios of amplitudes in exp4 over exp2 are plotted in Figure 4. By comparing the amplitudes of the 47 cycling transcripts in these two experiments, we found that the 95% confidence intervals (zα/2 = 1.96, σ = 0.05) for the genes FLJ10540, PCNA, CDC6 and CDC20 did not include zero, suggesting that the estimated amplitudes for these four genes in exp2 and exp4 of Whitfield et al. [11] might be affected by thy-thy arrest in exp2 and thy-noc arrest in exp4. This was not true of the estimated amplitudes of the other 43 genes (Table 1). Note that the amplitudes of CDC6 and PCNA, which are expressed in the G1/S phase, were reduced almost to half in the thy-thy (S phase arrest) experiment relative to thy-noc (M phase arrest) experiment; the amplitude of CDC20, which is expressed in the G2/M phase, was reduced in the thy-noc experiment to half that in the thy-thy experiment.
Figure 4 Plot of ratio of the amplitudes of 47 cell cycle transcripts in exp4 vs. exp2 (Whitfield et al., 2002).
Discussion
In this paper, we have analyzed the effect of the scattering of phase angles of a cell cycle related gene in a cell population on the amplitude of expression of this gene. Our analysis suggests that variation in amplitude for such a gene between two experiments depends on the variation of phase distribution in a population of cells. We illustrated our analysis by comparing the amplitudes of 47 cell cycle related genes in the 2nd and 4th experiments of Whitfield et al. [11], where two different methods were used that resulted in cells being arrested at different stages of the cycle. The amplitudes of 43 of the 47 genes were not significantly affected by the differences in cell-arresting methods. Among the 4 genes that were differentially affected, the amplitudes of the G1/S phase genes CDC 6 and PCNA were smaller in the thy-thy (S phase arrest) experiment 2, while the amplitude of G2/M gene CDC20 was smaller in the thy-noc (M phase arrest) experiment 4 of Whitfield et al. [11]. These results suggest that thy-thy and thy-noc affected the maximum expression levels of some G1/S and G2/M phase genes differentially. It appears plausible that the thy-thy arresting method might completely prevent expression of some G1/S phase genes. Some of these genes could be recovered from the gene list of the 4th experiment using the thy-noc method.
Our results suggest that thy-thy interrupts PCNA and CDC6 mRNA synthesis in S phase arrest, and thy-noc interrupts CDC20 and FLJ10540 mRNA synthesis in G2/M arrest. After the cells are released, synthesis of the mRNAs for some affected genes resumes but with large variation in pace across cells. In other words, the phase distributions of PCNA and CDC6 in the cell population of exp2 are more spread out during the G1/S phase; and the phase distributions of FLJ10540 and CDC20 in the cell population of exp4 are more spread out in the G2/M phase. For example, the ratio between the two amplitudes of CDC20 in exp4 vs. exp2 is about 0.5. According to the ratio defined in (5), we could infer that the upper bound for the concentration parameter of von Mises for CDC20 in exp4 is less than 2.5, provided the for CDC20 in exp2 is very large, e.g. >20. The significant difference between the two distributions with = 2 and 10 is illustrated graphically in Figure A in the Appendix.
Our results show that some cell cycle related genes may be more responsive or sensitive than others to changes in the environment, e.g. cell-arresting chemicals, temperature shift, etc. Raser and O'Shea [8] suggested that noise intrinsic to eukaryotic gene expression is gene-specific, and Fraser et al. [9] suggested that the production of essential and complex-forming proteins involves lower levels of noise than does the production of most other genes. Our findings indicate that the 43 cell cycle related genes with unaltered amplitudes in exp2 and exp4 of Whitfield et al. [11] may be essential to the HeLa cell cycle, and thus less sensitive to perturbation by stress or chemicals. However, CDC6 and CDC20, which are important to the yeast cell cycle [20], were expressed at significantly different amplitudes in the HeLa cell cycle. Further studies are needed to investigate whether some essential cell cycle genes such as CDC6 and CDC20 are cell type specific in response to chemicals.
The amplitude, phase angle and period estimated from (1) for genes from the microarray data are characteristic of cell populations rather than a single cell. Conventionally, amplitude and phase angle are considered independent parameters in a sinusoidal model. However, in microarray studies, where the measured periodic expression for a cell cycle related gene is averaged over a cell population (>106 cells), a phase change in the concentration of von Mises distribution for a gene can contribute to a change in amplitude. Note that our analysis partially addresses Winfree's concern about whether amplitude should be considered as additional information to phase in studies of circadian rhythms [17].
The detection of cell cycle related genes with significantly different amplitudes between exp2 and exp4 of Whitfield et al. [11] depends on: (i) approximation of the true distribution of amplitudes of Kgx and Kgy, g = 1, ..., 47 by a normal distribution; (ii) the design of exp2 and exp4, including number of time points per gene. While these assumptions appear tenable for these data, a more comprehensive analysis of other relevant cell cycle gene expression studies is needed for more definitive conclusions about their validity. The four genes currently identified all have an estimated 1.5 fold change, and with the current sample size, the power to detect such a change is only around 50%. If the number of time points in exp2 and exp4 were larger (e.g. 47 in exp3 of Whitfield et al. [11]), the power for detecting amplitudes with less than 2-fold change would be increased.
One often neglected but important factor in interpreting and analyzing cell cycle related gene expression data is the quality of synchrony of the cell culture. Currently there are no quantitative standards for measuring to what extent cells have been synchronized. The periodic patterns of the 47 genes were measured from stressed or perturbed cells in the 2nd and 4th experiments of Whitfield et al. [11]. Gene expression from normal, un-perturbed and synchronized HeLa cells obtained using the technologies proposed by Helmsteteter et al. [21] may serve as references for comparing the expression of these genes when mRNA synthesis is interrupted by different cell-arresting methods, e.g. temperature shift or chemical induction at various phases of the cell cycle. Good quality control of cell synchrony, as suggested in Cooper et al. [22], will provide a basis for microarray studies of cell cycle related genes. More quantitative measures of cell culture synchrony, and investigation of the impacts of cell culture with various degrees of synchrony on expression of some cell cycle related genes, are needed in future studies.
Conclusion
The amplitudes of some cell cycle related genes were used to measure the effects of two different cell-arresting methods on gene expression. Some genes with periodic expression patterns can be used as models to probe the effects of different cell-arresting methods on expression of these genes, which can be quantitatively described in terms of amplitude and phase. The ratio between the amplitudes estimated in two experiments for a cell cycle related gene can be used to gauge the variation of the phase/peak expression time distribution involved in stochastic transcriptional and post-transcriptional processes for the gene in a cell population. Further investigations are needed using normal, unperturbed and synchronized HeLa cells as a reference for comparing how many cell cycle related genes are directly and indirectly affected by various cell-arresting methods.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
DL conceived of the study, performed the analysis and drafted the manuscript. KWG and RW participated in the draft of the manuscript. All authors read and approved the final manuscript.
Supplementary Material
Additional file
The addition file 'Appendix.doc' is inserted here.
Click here for file
Acknowledgements
The authors thank two anonymous reviewers for constructive comments; we thank Stephen Cooper for his thorough and extensive comments on the manuscript. We also thank the executive editor Dr. Paul Agutter for his help. DL thanks Grace E. Kissling and Mike Whitfield for providing suggestions on an early version of this manuscript. DL thanks Clare Weinberg for stimulating discussion in the early stage of this work, Leping Li for his support, and Shyamal Peddada and David Umback for their encouragement when DL started this work at the NIEHS/NIH. The authors thank Cecilia Tan, Jeffery Schroeter and Elena Kleymenova for their comments on the manuscript.
==== Refs
Ozbudak EM Thattai M Kurtser I Grossman AD von Oudenaarden A Regulation of noise in the expression of a single gene Nat Genet 2002 31 69 73 11967532 10.1038/ng869
Elowitz MB Levine AJ Siggia ED Swain PS Stochastic gene expression in a single cell Science 2002 297 1183 1186 12183631 10.1126/science.1070919
Rosenfeld N Young JW Alon U Swain PS Elowitz MB Gene regulation at the single-cell level Science 2005 307 1962 1965 15790856 10.1126/science.1106914
McAdams HH Arkin A Stochastic mechanisms in gene expression Proc Natl Acad Sci USA 1998 94 814 819 9023339 10.1073/pnas.94.3.814
Thattal M Oudenaarden AV Intrinsic noise in gene regulatory networks Proc Natl Acad Sci USA 2001 98 8614 8619 11438714 10.1073/pnas.151588598
Swain P Elowitz MB Siggia ED Intrinsic and extrinsic contributions to stochasticity in gene expression Proc Natl Acad Sci USA 2002 99 12795 12800 12237400 10.1073/pnas.162041399
Blake WJ Kaern M Cantor CR Collins JJ Noise in eukaryotic gene expression Nature 2003 422 633 637 12687005 10.1038/nature01546
Raser J O'Shea EK Control of stochasticity in eukaryotic gene expression Science 2004 304 1811 1814 15166317 10.1126/science.1098641
Fraser HB Hirsh AE Giaever G Kumm J Eisen MB Noise minimization in eukaryotic gene expression PLoS Biol 2004 2 1 5 10.1371/journal.pbio.0020001
Spellman PT Sherlock G Zhang MQ Iyer VR Anders K Eisen MB Brown PO Botstein D Futcher B Comprehensive identification of cell cycle-regulated genes of the yeast Sacchromyces cerevisiae by microarray hybridization Mol Biol Cell 1998 9 3273 3297 9843569
Whitfield ML Sherlook G Saldanha AJ Murray JI Ball CA Alexander KE Matese JC Perou CM Hurt MM Brown PO Botstein D Identification of genes periodically expressed in the human cell cycle and their expression in tumors Mol Biol Cell 2002 13 1977 2003 12058064 10.1091/mbc.02-02-0030.
Peng X Krishna R Karuturi M Miller LD Lin K Jia Y Kondu P Wang L Wong L-S Liu ET Balasubramanian MK Liu J Identification of cell cycle-regulated genes in fission yeast Mol Biol Cell 2005 16 1026 1042 15616197 10.1091/mbc.E04-04-0299
Rustici G Mata J Kivinen K Lio P Penkett CJ Burns G Hayles J Brazma A Nurse P Bahler J Periodic gene expression program of the fission yeast cell cycle Nature Genet 2004 36 809 817 15195092 10.1038/ng1377
Storch KF Lapan O Leykin I Viswannthan N David FC Wong WH Weitz CJ Extensive and divergent circadian gene expression in liver and heart Nature 2002 417 78 83 11967526 10.1038/nature744
Panda S Antoch MP Miller BH Su AI Schook AB Straume M Schultz PG Kay SA Takahashi JS Hogenesch JB Coordinated transcription of key pathways in the mouse by the circadian clock Cell 2002 109 307 320 12015981 10.1016/S0092-8674(02)00722-5
Liu D Umbach DM Peddada SD Li L Crockett PW Weinberg CR A Random-Periods Model for Expression of Cell-Cycle Genes Proc Natl Acad Sci USA 2004 101 7240 7245 15123814 10.1073/pnas.0402285101
Winfree A The geometry of biological time 2001 2 New York: Springer 228
Mardia KV Jupp PE Directional statistics 2000 New York: John Wiley & Son
Liu D Weinberg C Peddada SD A geometric approach to determine association and coherence of the activation times of cell-cycling genes under differing experimental conditions Bioinformatics 2004 20 2521 2528 15087309 10.1093/bioinformatics/bth274
Murray A Hunt T The cell cycle: an introduction 1993 New York: Oxford University Press
Helmstetter CE Thornton M Romero A Eward KL Synchrony in human, mouse, and bacterial cell cultures: a comparison Cell Cycle 2003 2 42 45 12695687
Cooper S Tenbroek M Ljungman M Bissett P Tarquini M Iyer G Automated, reproducible, membrane-elution for cell-cycle analysis: application to cyclin B1 content during the unperturbed, normal, eukaryotic cell cycle
|
16283931
|
PMC1315354
|
CC BY
|
2021-01-04 16:39:25
|
no
|
Theor Biol Med Model. 2005 Nov 11; 2:46
|
utf-8
|
Theor Biol Med Model
| 2,005 |
10.1186/1742-4682-2-46
|
oa_comm
|
==== Front
Malar JMalaria Journal1475-2875BioMed Central London 1475-2875-4-561632115010.1186/1475-2875-4-56ResearchBone marrow suppression and severe anaemia associated with persistent Plasmodium falciparum infection in African children with microscopically undetectable parasitaemia Helleberg Marie [email protected] Bamenla Q [email protected] Bartholomew D [email protected] George [email protected] Onike [email protected] Jorgen AL [email protected] Centre for Medical Parasitology, Department of Clinical Microbiology 7602, Copenhagen University Hospital, 2100 Copenhagen, Denmark2 Department of Child Health, Korle-Bu Teaching Hospital, Accra, Ghana3 Immunology Unit, Noguchi Memorial Institute for Medical Research, Legon, Ghana2005 1 12 2005 4 56 56 21 7 2005 1 12 2005 Copyright © 2005 Helleberg et al; licensee BioMed Central Ltd.2005Helleberg 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
Severe anaemia can develop in the aftermath of Plasmodium falciparum malaria because of protracted bone marrow suppression, possibly due to residual subpatent parasites.
Materials and methods
Blood was collected from patients with recent malaria and negative malaria microscopy. Detection of the Plasmodium antigens, lactate dehydrogenase (Optimal®), aldolase and histidine rich protein 2 (Now malaria®) were used to differentiate between patients with (1) no malaria, (2) recent cleared malaria, (3) persistent P. falciparum infection. Red cell distribution width (RDW), plasma levels of soluble transferrin receptor (sTfR) and erythropoietin (EPO) were measured as markers of erythropoiesis. Interleukin (IL) 10 and tumour necrosis factor (TNF)α were used as inflammation markers.
Results
EPO was correlated with haemoglobin, irrespective of malaria (R = -0.36, P < 0.001). Persistent P. falciparum infection, but not recent malaria without residual parasites, was associated with bone marrow suppression i.e., low RDW (P < 0.001 vs. P = 0.56) and sTfR (P = 0.02 vs. P = 0.36). TNF-α and IL-10 levels were not associated with bone marrow suppression.
Conclusion
In the treatment of malaria, complete eradication of parasites may prevent subsequent development of anaemia. Severely anaemic children may benefit from antimalarial treatment if antigen tests are positive, even when no parasites can be demonstrated by microscopy.
==== Body
Introduction
Anaemia is a common, life-threatening complication of Plasmodium falciparum malaria in African infants and young children. The anaemia develops when accelerated removal of erythrocytes is not compensated by the bone marrow. Bone marrow suppression is generally present in all malaria patients [1] and has also been described in asymptomatic P. falciparum infection [2]. The fact that some malaria patients develop severe anaemia, whereas others retain normal or near normal haemoglobin (Hb) must thus be explained by the amount of erythrocyte destruction during the period until return of normal bone marrow function. The anaemia may either develop rapidly with severe, acute haemolysis, or take a slow smouldering course, with a relatively slow rate of erythrocyte destruction in the presence of persistent bone marrow suppression [3]. Data on the duration of bone marrow suppression after a malaria attack are conflicting. In some studies there has been evidence of hypoproliferative erythropoiesis and dyserythropoiesis for weeks following treatment [4,5], while other studies have shown that bone marrow suppression is reversed rapidly after treatment [1,6].
In areas where malaria is endemic, a high number of children with severe anaemia but without detectable parasites are hospitalized each year coinciding with the peak of malaria transmission. The history of recent febrile illness and the presence of detectable circulating antigens strongly suggest that these children suffer from the late effects of a recent malaria attack [7]. The question is whether there are signs of bone marrow suppression in these patients, and if so, whether this bone marrow suppression is due to a sustained effect of the malaria attack or to the persistence of parasites that are undetectable by routine microscopy. In the latter case, it is possible that these patients would benefit from a repeated course of antimalarial treatment.
Patients with severe malarial anaemia have an increased ratio between the pro-inflammatory cytokine, tumour necrosis factor (TNF)-α, and the anti-inflammatory cytokine, interleukin (IL)-10 [8], and it has been proposed that inflammatory cytokines may be a causative factor for malarial anaemia [9,10]. If so, protracted bone marrow suppression in the aftermath of a malaria attack might be due to persistence of a dysregulated inflammatory response. However, mild and moderate malarial anaemia is not associated with an inverse ratio between TNF-α and IL-10 [2].
The purpose of the present study was to test the hypothesis that an impaired erythropoietic response to anaemia is associated with a persistent malarial infection in patients without microscopically detectable parasites. In order to differentiate between recent and persistent P. falciparum infection, two rapid tests for soluble malarial antigens, Now® ICT Malaria P.f/P.v (Binax, US) and Optimal® (DiaMed AG, Schweiz) were performed. Binax is based on detection of histidine rich protein 2 (HRP2) and aldolase and can remain positive for weeks following treatment of the infection [11]. Optimal®, on the other hand, detects parasite derived lactate dehydrogenase (pLDH) and test results are positive only in the presence of live parasites [12].
Red cell distribution width (RDW) and plasma levels of erythropoietin (EPO) and soluble transferrin receptor (sTfR) were measured as markers of erythropoiesis. RDW is a measure of size variation in red blood cells and is increased when erythropoiesis is stimulated [13]. sTfR is secreted mainly by erythroblasts, and plasma levels are increased when turnover in the bone marrow is raised as in haemolytic anaemia, whereas levels are normal in the anaemia of chronic disease [14]. Plasma levels of TNF-α and IL-10 were analysed to determine if the bone marrow suppression was correlated to dysregulated inflammation.
Materials and methods
The study took place at the Department of Child Health, Korle Bu Teaching Hospital, Accra, Ghana during the malaria season, July and August, 2003. Two groups of children aged 0.5–12 years were consecutively recruited. Both groups consisted of children with a presumptive or confirmed diagnosis of recent malaria but without detectable parasites in Giemsa-stained blood films, examined by ordinary microscopy.
Group 1
Patients admitted to the emergency room with a presumptive diagnosis of acute malaria but with a negative microscopic test for malaria. The majority of these patients had received treatment for malaria either at home or in various health facilities prior to being referred to hospital. Those who had haemoglobin (Hb) ≤ 5 g/dl or Hb ≥ 8 g/dl were included, unless they had a history of a recent trauma or bleeding. Four hundred and five patients with the same age and sex distribution and with proven malaria by microscopy acted as a positive control group.
Group 2
Children who came for follow-up on day three after initiation of treatment for acute, uncomplicated falciparum malaria at the health clinic attached to the hospital and who had cleared their parasitaemia microscopically. Patients with febrile illness other than malaria, with severe malaria, concomitant infections or known chronic disease were excluded. The children had received amodiaquine, artesunate or a combination of the two as part of an ongoing drug trial.
Children were enrolled after informed consent from parents or guardians. The ethics and protocol review committee at the University of Ghana Medical School had approved the study.
Blood was collected in EDTA-coated tubes and examined using an automated haematology analyser (Sysmex, KX-21, Germany). Giemsa-stained blood films were prepared on day 0 (group 1) and on day 0 (i.e. prior to inclusion), 3 and 7 (group 2). Blood films were considered to be negative if 200 leucocytes had been counted without finding parasites, corresponding to a detection level of approximately 50 parasites/μL. Antigen detection was performed on whole blood from day 0 (group 1) and day 3 (group 2). The rapid diagnostic tests, Now® ICT Malaria P.f/P.v (Binax, US) and Optimal® (DiaMed AG, Schweiz), were used according to the manufacturer's instruction.
The EDTA-plasma was subsequently collected by centrifugation and stored at -20°C until analysis. Plasma concentrations of EPO, sTfR, TNF-α and IL-10 were measured using commercial ELISA kits according to the manufacturer's instructions (EPO, sTfR and IL-10, R&D Systems, MN, USA and TNF-α, BioSource International, CA, USA).
Statistics
All data, except haemoglobin, were logarithmically transformed to achieve normal distribution. Differences between groups were analysed using two-tailed student's t-test, except differences in haemoglobin levels that were analysed by Mann Whitney two sample rank sum test. Associations between parameters were analysed by multiple regression analyses. P-values < 0.05 were considered statistically significant. All calculations were performed using Stata SE 8.0 (Stata Corporation, TX, USA)
Results
Forty eight patients were enrolled into group 1: 24 with severe anaemia and 24 with Hb ≥ 8 g/dl. Two (8.3 %) of the patients with severe anaemia and 15 (62.5 %) of those with Hb ≥ 8 g/dl had no detectable malaria antigens and were, thus, unlikely to suffer from malaria ('No malaria', Table 1). Of the remaining 31 patients, all had a positive HRP-2 test. Sixteen also had a positive pLDH test, suggestive of persistent malaria, whereas 15 had a positive HRP-2 test only, suggestive of recent malaria. Among the patients with a positive antigen test, 25 reported to have taken some antimalarial medication prior to admission, four had not received treatment and information was missing for two patients. Forty seven patients were enroled into group 2. Their median parasite density on day 0 was 25,633 parasites/μL (25th – 75th percentiles: 5096 – 70,755). On the day of inclusion (day 3) all had undetectable parasitaemia by microscopy but had a positive HRP-2 test. Only eight had signs of persistent parasitaemia as indicated by the pLDH test. In both groups, patients with persistent malaria tended to have lower Hb than those with recent malaria; although in group 1, the difference did not reach statistical significance (group1: P = 0.19 and group 2: P = 0.04, Table 1). In order to simplify the result presentation, the two groups are combined, based on the assumption that subpatent P. falciparum infection played a similar role in both groups. However, since the groups had different age distribution and the clinical similarity between the groups could not be established, the data were also analysed separately for the groups. As indicated in the text, this did not affect the conclusions of the study.
Table 1 Patient characteristics
Group 1 Group 2
Persistent malaria Recent malaria No malaria Persistent malaria Recent malaria
Number of patients 16 15 17 8 39
Age1 (years) 2 (1–12) 4 (0.5–10) 2 (0.5–12) 5.8 (1.5–10) 5.8 (1–12)
Temperature on admission1 (°C) 37.0 (36.0–38.5) 38.0 (36.7–38.6) 37.5 (36.6–39.5) 37.2 (36.8–38.8) 38.2 (36.0–40.0)
Sex (boys:girls) 12:4 7:8 10:7 5:3 23:15
Haemoglobin1 (g/dl) 3.4 (2.1–10.2) 4.4 (1.6–11.6) 9.8 (1.5–11.8) 6.6 (4.7–10.5) 9.2 (5.2–12.0)
TNF-α (pg/ml)2 16.2 (6.9–38.0) 13.5 (6.5–27.5) 5.4 (1.2–23.4) 6.2 (0.6–63.1) 8.7 (5.5–13.5)
IL-10 (pg/ml)2 56.2 (15.8–208.2) 29.5 (7.8–112.2) 3.0 (0.9–9.8) 8.9 (3.6–21.9) 8.3 (5.8–11.7)
IL-10/TNF-α2 1.9 (1.0–3.6) 1.4 (0.4–5.6) 1.0 (0.5–2.3) 1.7 (0.1–46.8) 1.0 (0.5–2.0)
1: median (range)
2: mean (95% CI)
Erythropoiesis
The bone marrow response to low Hb was compared between patients with positive pLDH, (persistent malaria, n = 24) and those with negative pLDH, (recent malaria or no malaria, n = 71) in order to detect an effect of persistent P. falciparum infection on erythropoiesis. Both RDW and sTfR increased in response to low Hb as indicated by a negative correlation between RDW and Hb and between sTfR and Hb. However, the levels of RDW and sTfR were lower in patients with persistent infection (filled symbols) than in those with recent malaria or without signs of malaria (open symbols, Figure 1A and 1B). In a multivariate regression analysis with haemoglobin and test results for pLDH and HRP2 as explanatory variables, a positive test for pLDH, but not a positive test for HRP2 alone, was significantly associated with reduced levels of RDW and sTfR (Table 2). For this analysis, all patients in group 1 and group 2 were combined in order to span the range of Hb-levels. However, when the multiple regression analysis was repeated for the two groups separately, the results were similar, although the association between pLDH and sTfR did not reach statistical significance (log RDW: group 1: P < 0.001 and group 2: P < 0.05; log sTfR group 1: P = 0.07 and group 2: P = 0.08). Persistent, microscopy negative malaria reduced the bone marrow response to the same extent as patent malaria. Thus, the slope of the regression line between RDW as the dependent and Hb as the independent variable was reduced to the same extent in microscopy negative patients with persistent malaria as in the 405 patients with a positive microscopy compared to patients without malaria (Figure 2, red slope, purple slope, and green slope, respectively, P < 0.001 for the comparison between no malaria and both other groups).
Figure 1 Effect of P. falciparum on erythropoietic response to anaemia. Associations between haemoglobin and (A) red cell distribution width (RDW), (B) soluble transferrin receptor (sTfR), and C. erythropoietin (EPO) in children with recent malaria (pLDH negative, HRP2 positive), persistent, submicroscopic P. falciparum infection (pLDH and HRP2 positive) or without signs of malaria (pLDH and HRP2 negative).
Table 2 Association between markers of erythropoiesis and Plasmodium antigens in multiple regression analyses with haemoglobin and test results for parasite derived lactate dehydrogenase (pLDH) and histidine rich protein 2 (HRP2) as explanatory variables.
pLDH HRP2 Haemoglobin
β (95% CI) P-value β (95% CI) P-value β (95% CI) P-value
Log RDW1 (%) -0.08 (-0.11–(-0.04)) <0.001 -0.04 (-0.12–0.04) 0.56 0.03 (0.02–0.04) <0.001
Log sTfR2 (nmol/l) -0.18 (-0.32–(-0.03)) 0.02 0.08 (-0.09–0.24) 0.36 0.04 (0.02–0.06) <0.001
Log EPO3 (mIU/ml) 0.13 (-0.16–0.43) 0.39 0.14 (-0.18–0.47) 0.38 0.34 (0.30–0.39) <0.001
1: Red cell distribution width
2: Soluble transferrin receptor
3: Erythropoietin
Figure 2 Comparison of the effect of patent versus subpatent P. falciparum infection on erythropoietic response to anaemia. Association between haemoglobin and red cell distribution width (RDW) in microscopy negative children with persistent malaria, children with microscopically detectable parasites and children without malaria. Regression lines and 95% confidence intervals.
Plasma levels of EPO showed a strong inverse correlation with Hb (Figure 1C, R = -0.36, 95%CI: -0.40 – -0.32, P < 0.001), and neither pLDH nor HRP2 were correlated with EPO, indicating that P. falciparum infection did not affect the secretion of EPO (Table 2).
Forty patients in group 2 were followed up on day 7. In these, bone marrow suppression had receded and there was an inverse linear correlation between log RDW and Hb (R = -0.05, 95%CI: -0.06 – -0.04, P < 0.001, Figure 3). There was no difference between children who had a positive and those who had a negative test for pLDH on day three (P = 0.53), but two children remained severely anaemic.
Figure 3 Erythropoietic response to anaemia after clearance of P. falciparum. Association between haemoglobin and red cell distribution width (RDW) on day 7. Regression line and 95% confidence interval.
Cytokines
The plasma levels of TNF-α and IL-10 were comparable to those in normal Ghanaian children without symptoms of malaria or detectable parasitaemia [2]. There was an inverse linear correlation between log IL10 and Hb (R = -0.08, 95%CI: -0.14 – -0.03, P = 0.005). There was no correlation between log TNF-α and Hb (P = 0.89). The regression line between RDW and Hb was not influenced by TNFα, IL-10 or the IL-10/TNF-α ratio, irrespective of antigen test results (multiple regression analysis, IL-10, P = 0.40, TNF-α, P = 0.53, IL-10/TNF-α, P = 0.40). Analyses with sTfR as a dependent variable gave similar results (IL-10, P = 0.78, TNF-α, P = 0.35, IL-10/TNF-α, P = 0.63). On day 0 (i.e. prior to recruitment), patients in group 2 had elevated plasma IL-10 (mean (95% CI) 450.4 (291.5–609.3) pg/ml) and TNF-α (36.3 (25.7–51.3) pg/ml), but the levels were not correlated with Hb (IL-10, P = 0.31, TNF-α, P = 0.30, IL-10/TNF-α ratio, P = 0.21).
Discussion
Children living in areas with endemic malaria are usually assumed to have malaria if they are hospitalized with severe anaemia that is preceded by an acute febrile illness. Such patients often have undetectable malaria parasites by microscopy of Giemsa-stained blood films, but their leukocytes commonly contain detectable malaria pigment [15]. It has previously been shown that the diagnosis may be aided by the use of rapid antigen detection tests, which can distinguish patients with malarial anaemia from those with anaemia from other causes [7], and this was supported in the present study. Thus, more than 90% of the patients with severe anaemia had detectable parasite antigens compared to only one third of those with normal or near normal Hb. Although microscopy is usually considered the gold standard for the diagnosis of malaria, these results suggest that the antigen detection methods, in particular those based on the detection of pLDH, may be superior to microscopy in the diagnosis of malarial anaemia.
In the present study, patients with detectable pLDH, a sign of persistent parasitaemia, were distinguished from those in whom only HRP2 could be detected as a sign of a cleared, recent infection. Only patients with persistent parasitaemia had signs of bone marrow suppression. This corresponds with previous investigations in which release of immature erythrocytes from the bone marrow coincided with the time of parasite clearance [1,6] and with findings in experimental malaria [16]. In one such study, subpatent P. falciparum infection in vaccinated Aotus monkeys led to severe bone marrow suppression that was rapidly reversed in response to mefloquine induced parasite clearance [17].
The association between P. falciparum infection and suppression of erythropoiesis has been debated. Menendez et al. found that levels of sTfR were increased in infants with malaria [18]. Similarly, Verhoef found increased levels of sTfR in children with malaria and concluded that there was no suppression of erythropoiesis [19]. However, in these studies it was not indicated if levels of sTfR were increased adequately compared to the degree of anaemia. Furthermore, increased levels of sTfR found in malaria patients might be caused by shedding of receptors from proliferating B-lymphocytes [16]. In the present study, however, there was no correlation between sTfR and lymphocyte count (data not shown). Due to the logistics of the study it was not possible to make reticulocyte counts but the finding that persistent, subpatent parasitaemia was associated with reduced levels of both RDW and sTfR strongly points toward parasite-induced suppression of erythropoiesis. This may explain the fact that healthy school children with so-called asymptomatic P. falciparum infection have reduced Hb and signs of bone marrow suppression [2]. Asymptomatic P. falciparum infections are usually accepted as a necessary evil in order to maintain immunity in individuals, who are at great and constant risk of malarial infections [20], and this approach is probably the only realistic option in the near future. However, data from the present study imply that the ultimate goal of malaria control ought to be complete parasite eradication due to the detrimental effects of persistent parasitaemia. This is consistent with a previous clinical trial that linked incomplete haematological recovery with lack of parasitological cure [21]. The Hb levels in infants living with high risk of malaria can be improved by impregnated bednets [22], malaria chemoprophylaxis [23] and presumptive intermittent treatment [24]. It is likely that this effect is mediated partly by a reduction in the risk of bone marrow suppression. From a clinical point of view this study suggests that in malarial endemic areas, antimalarial treatment should not be withheld from severely anaemic patients presenting with signs compatible with ongoing or recent malaria who have not been treated for malaria, even when parasites are undetectable by microscopy. Controlled clinical trials should be performed to determine whether patients who have already received a full course of antimalarial treatment would benefit from a repeat course, and in particular, if it is important to restrict this treatment to children with a positive pLDH test. Paediatric patients that return with repeated episodes of anaemia are common (ref. 7) and the management of these patients puts strain on the health system. Thus, the cost of improved diagnosis and management of these cases have the potential to benefit both the patients and the health system.
Several studies have shown that the suppressed bone marrow response to anaemia, which is a general feature of malaria, is not caused by insufficient secretion of EPO [1,25,26], but this is disputed by other investigations [27-29]. In agreement with the former view, similar levels of EPO in response to low Hb were found in patients with and without malaria. It has been suggested that the bone marrow suppression is a direct effect of TNF-α, which is elevated in malaria [30]. On the other hand, other investigations did not find that inflammatory cytokines to played a role in malarial dyserythropoiesis [31]. In addition, severe malarial anaemia is associated with relatively low TNF-α levels [32] and signs of systemic inflammation [33,34] (Awandare et al. unpublished data) compared with cerebral malaria and uncomplicated malaria. It is thus possible that the effect of cytokine perturbations in malarial anaemia is a lack of parasite control that leads to persistence of the infection and thus indirectly causes bone marrow suppression [32]. It has recently been proposed that phagocytosed haemozoin may play a role in the dyserythropoiesis of malaria through induction of 4-hydroxynonenal [35].
Conclusion
This study has shown that the majority of severe anaemia cases in children living in areas with malaria transmission is due to malaria. This is also the case when no malaria parasites can be detected by microscopy, in which case antigen detection may lead to the diagnosis. These patients have suppressed erythropoiesis, which persists until the parasites have been cleared completely.
Authors' contributions
Marie Helleberg and Jorgen A. L. Kurtzhals designed the study in collaboration with all co-authors. Marie Helleberg, Bamenla Q. Goka, George Obeng-Adjei and Onike Rodriques performed the clinical work; Marie Helleberg and Bartholomew D. Akanmori did the laboratory analysis. Marie Helleberg did the data analysis and drafted the manuscript. All authors contributed significantly to the final version of the manuscript. The authors do not have any commercial or other association that might pose a conflict of interest.
Financial support
The study was supported by grants from the Danish Medical Research Council (SSVF grant no. 22-01-0343 and 22-03-0063) and the Danish International Development Assistance.
Acknowledgements
We are grateful for the collaboration of the patients and the parents, the assistance of the staff at the Department of Child Health, Korle Bu Teaching Hospital and to the staff in the research laboratory.
==== Refs
Kurtzhals JA Rodrigues O Addae M Commey JO Nkrumah FK Hviid L Reversible suppression of bone marrow response to erythropoietin in Plasmodium falciparum malaria Br J Haematol 1997 97 169 174 9136961 10.1046/j.1365-2141.1997.82654.x
Kurtzhals JA Addae MM Akanmori BD Dunyo S Koram KA Appawu MA Nkrumah FK Hviid L Anaemia caused by asymptomatic Plasmodium falciparum infection in semi- immune African schoolchildren Trans R Soc Trop Med Hyg 1999 93 623 627 10717750 10.1016/S0035-9203(99)90073-1
Phillips RE Pasvol G Anaemia of Plasmodium falciparum malaria Baillieres Clin Haematol 1992 5 315 330 1511178
Phillips RE Looareesuwan S Warrell DA Lee SH Karbwang J Warrell MJ White NJ Swasdichai C Weatherall DJ The importance of anaemia in cerebral and uncomplicated falciparum malaria: role of complications, dyserythropoiesis and iron sequestration Q J Med 1986 58 305 323 3526385
Camacho LH Gordeuk VR Wilairatana P Pootrakul P Brittenham GM Looareesuwan S The course of anaemia after the treatment of acute, falciparum malaria Ann Trop Med Parasitol 1998 92 525 537 9797826 10.1080/00034989859221
Abdalla S Weatherall DJ Wickramasinghe SN Hughes M The anaemia of P. falciparum malaria Br J Haematol 1980 46 171 183 7000157
Kurtzhals JA Helleberg M Goka BQ Akanmori BD Severe malaria in west African children Lancet 2003 361 1393 12711504 10.1016/S0140-6736(03)13068-1
Kurtzhals JA Adabayeri V Goka BQ Akanmori BD Oliver-Commey JO Nkrumah FK Behr C Hviid L Low plasma concentrations of interleukin 10 in severe malarial anaemia compared with cerebral and uncomplicated malaria Lancet 1998 351 1768 1772 9635949 10.1016/S0140-6736(97)09439-7
Biemba G Gordeuk VR Thuma PE Mabeza GF Weiss G Prolonged macrophage activation and persistent anaemia in children with complicated malaria Trop Med Int Health 1998 3 60 65 9484971 10.1046/j.1365-3156.1998.00168.x
Wickramasinghe SN Abdalla SH Blood and bone marrow changes in malaria Baillieres Best Pract Res Clin Haematol 2000 13 277 299 10942626
Tjitra E Suprianto S McBroom J Currie BJ Anstey NM Persistent ICT malaria P.f/P.v panmalarial and HRP2 antigen reactivity after treatment of Plasmodium falciparum malaria is associated with gametocytemia and results in false-positive diagnoses of Plasmodium vivax in convalescence J Clin Microbiol 2001 39 1025 1031 11230422 10.1128/JCM.39.3.1025-1031.2001
Piper R Lebras J Wentworth L Hunt-Cooke A Houze S Chiodini P Makler M Immunocapture diagnostic assays for malaria using Plasmodium lactate dehydrogenase (pLDH) Am J Trop Med Hyg 1999 60 109 118 9988333
Roberts GT El Badawi SB Red blood cell distribution width index in some hematologic diseases Am J Clin Pathol 1985 83 222 226 3969961
Beguin Y Clemons GK Pootrakul P Fillet G Quantitative assessment of erythropoiesis and functional classification of anemia based on measurements of serum transferrin receptor and erythropoietin Blood 1993 81 1067 1076 8427988
FK N Severe anaemia associated with malaria in children Ghana Med J 1973 12 17 21
Chang KH Tam M Stevenson MM Inappropriately low reticulocytosis in severe malarial anemia correlates with suppression in the development of late erythroid precursors Blood 2004 103 3727 3735 14739226 10.1182/blood-2003-08-2887
Egan AF Fabucci ME Saul A Kaslow DC Miller LH Aotus New World monkeys: model for studying malaria-induced anemia Blood 2002 99 3863 3866 11986251 10.1182/blood.V99.10.3863
Menendez C Quinto LL Kahigwa E Alvarez L Fernandez R Gimenez N Schellenberg D Aponte JJ Tanner M Alonso PL Effect of malaria on soluble transferrin receptor levels in Tanzanian infants Am J Trop Med Hyg 2001 65 138 142 11508389
Verhoef H West CE Kraaijenhagen R Nzyuko SM King R Mbandi MM van Laatum S Hogervorst R Schep C Kok FJ Malarial anemia leads to adequately increased erythropoiesis in asymptomatic Kenyan children Blood 2002 100 3489 3494 12393621 10.1182/blood-2001-12-0228
Kitua AY Smith TA Alonso PL Urassa H Masanja H Kimario J Tanner M The role of low level Plasmodium falciparum parasitaemia in anaemia among infants living in an area of intense and perennial transmission Trop Med Int Health 1997 2 325 333 9171840
Bloland PB Lackritz EM Kazembe PN Were JB Steketee R Campbell CC Beyond chloroquine: implications of drug resistance for evaluating malaria therapy efficacy and treatment policy in Africa J Infect Dis 1993 167 932 937 8450258
PL A SW L JRM AS K K P G al. SFC A malaria control trial using insecticide-treated bed nets and targeted chemoprophylaxis in a rural area of The Gambia, West Africa. 6. The impact of the interventions on mortality and morbidity from malaria. Trans R Soc Trop Med Hyg 1993 87 S2/37 S2/44
Menendez C Kahigwa E Hirt R Vounatsou P Aponte JJ Font F Acosta CJ Schellenberg DM Galindo CM Kimario J Urassa H Brabin B Smith TA Kitua AY Tanner M Alonso PL Randomised placebo-controlled trial of iron supplementation and malaria chemoprophylaxis for prevention of severe anaemia and malaria in Tanzanian infants Lancet 1997 350 844 850 9310602 10.1016/S0140-6736(97)04229-3
Massaga JJ Kitua AY Lemnge MM Akida JA Malle LN Ronn AM Theander TG Bygbjerg IC Effect of intermittent treatment with amodiaquine on anaemia and malarial fevers in infants in Tanzania: a randomised placebo-controlled trial Lancet 2003 361 1853 1860 12788572 10.1016/S0140-6736(03)13504-0
Burchard GD Radloff P Philipps J Nkeyi M Knobloch J Kremsner PG Increased erythropoietin production in children with severe malarial anemia Am J Trop Med Hyg 1995 53 547 551 7485716
Rencricca NJ Stout JP Coleman RM Erythropoietin production in virulent malaria Infect Immun 1974 10 831 833 4609912
Burgmann H Looareesuwan S Kapiotis S Viravan C Vanijanonta S Hollenstein U Wiesinger E Presterl E Winkler S Graninger W Serum levels of erythropoietin in acute Plasmodium falciparum malaria Am J Trop Med Hyg 1996 54 280 283 8600766
El Hassan AM Saeed AM Fandrey J Jelkmann W Decreased erythropoietin response in Plasmodium falciparum malaria-associated anaemia Eur J Haematol 1997 59 299 304 9414641
Vedovato M De PV Dapporto M Salvatorelli G Defective erythropoietin production in the anaemia of malaria Nephrol Dial Transplant 1999 14 1043 1044 10328513 10.1093/ndt/14.4.1043
Clark IA Chaudhri G Tumour necrosis factor may contribute to the anaemia of malaria by causing dyserythropoiesis and erythrophagocytosis Br J Haematol 1988 70 99 103 3179231
Yap GS Stevenson MM Inhibition of in vitro erythropoiesis by soluble mediators in Plasmodium chabaudi AS malaria: lack of a major role for interleukin 1, tumor necrosis factor alpha, and gamma interferon Infect Immun 1994 62 357 362 8300197
Akanmori BD Kurtzhals JA Goka BQ Adabayeri V Ofori MF Nkrumah FK Behr C Hviid L Distinct patterns of cytokine regulation in discrete clinical forms of Plasmodium falciparum malaria Eur Cytokine Netw 2000 11 113 118 10705308
JA K CM R E T SK D KA K BD A FK N L H Increased eosinophil activity in acute Plasmodium falciparum infection -- association with cerebral malaria Clin Exp Immunol 1998 112 303 307 9649195 10.1046/j.1365-2249.1998.00586.x
Ostrowski SR Ullum H Goka BQ Høyer-Hansen G Obeng-Adjei G Pedersen BK Akanmori BD Kurtzhals JAL The plasma level of the soluble urokinase receptor is increased in malaria patients and associated with a poor clinical or fatal outcome J Infect Dis 2005 191 1331 1341 15776381 10.1086/428854
G G D U E S I R W P P A Hemozoin- and 4-hydroxynonenal-mediated inhibition of erythropoiesis. Possible role in malarial dyserythropoiesis and anemia. Haematologica 2004 89 492 493 15075084
|
16321150
|
PMC1315355
|
CC BY
|
2021-01-04 16:37:31
|
no
|
Malar J. 2005 Dec 1; 4:56
|
utf-8
|
Malar J
| 2,005 |
10.1186/1475-2875-4-56
|
oa_comm
|
==== Front
Reprod Biol EndocrinolReproductive biology and endocrinology : RB&E1477-7827BioMed Central London 1477-7827-3-671631368010.1186/1477-7827-3-67ReviewReactive oxygen species in spermatozoa: methods for monitoring and significance for the origins of genetic disease and infertility Baker Mark A [email protected] R John [email protected] The ARC Centre of Excellence in Biotechnology and Development, Reproductive Science Group, School of Environmental and Life Sciences, University of Newcastle, Callaghan, NSW 2308, Australia2005 29 11 2005 3 67 67 24 8 2005 29 11 2005 Copyright © 2005 Baker and Aitken; licensee BioMed Central Ltd.2005Baker and Aitken; 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.
Human spermatozoa generate low levels of reactive oxygen species in order to stimulate key events, such as tyrosine phosphorylation, associated with sperm capacitation. However, if the generation of these potentially pernicious oxygen metabolites becomes elevated for any reason, spermatozoa possess a limited capacity to protect themselves from oxidative stress. As a consequence, exposure of human spermatozoa to intrinsically- or extrinsically- generated reactive oxygen intermediates can result in a state of oxidative stress characterized by peroxidative damage to the sperm plasma membrane and DNA damage to the mitochondrial and nuclear genomes. Oxidative stress in the male germ line is associated with poor fertilization rates, impaired embryonic development, high levels of abortion and increased morbidity in the offspring, including childhood cancer. In this review, we consider the possible origins of oxidative damage to human spermatozoa and reflect on the important contribution such stress might make to the origins of genetic disease in our species.
==== Body
1. Introduction – origins of genetic disease
The maintenance of genetic integrity in the male germ line has major repercussions for conception, the progress of pregnancy and, ultimately, the health and well-being of the progeny [1]. The human male contributes heavily to germ line mutations [2], and as such, is responsible for most of the dominant genetic diseases observed in our species. Indeed, in some cases, such as multiple endocrine neoplasia or achondroplasia (short-limbed dwarfism), the phenotype is invariably the result of mutations that can be traced back to the paternal germ line [2]. Epidemiological data also suggest that paternally derived genetic damage may contribute significantly to the aetiology of cancer in children and young adults [1,2].
These observations raise important questions about the aetiology of genetic damage in the male germ line and the causal links that exist between the induction of such damage and the inheritance of many childhood diseases. As early as 1912, Wilhelm Weinberg (cited in [2]) reported that children with dominant achondroplasia born to normal parents were among the last-born children in the family. Later work by Penrose [3] suggested that the effect observed by Weinberg was not actually correlated with birth order, nor surprisingly, maternal age. Rather, achondroplasia was a disease associated with paternal age. The implications of these findings were vast. Why is it that a much greater mutation rate apparently exists in the male germ line compared to the female? And why are several X-linked recessive and autosomal-dominant diseases correlated with paternal age?
The current consensus is that replication errors are the probable cause of such mutations as a consequence of the higher number of cell divisions involved in generating a spermatozoon (approximately 840 for a 50 year old male) as opposed to an ovulated egg (approximately 22 divisions regardless of age). If this is the case, then the mutations would have to be generated and retained in mitotically active spermatogonia. In the case of Apert syndrome, for example, there is good reason to believe that the causative mutation (in this case, predominantly a 755C-G transversion in the FGFR-2 gene) arises in spermatogonia and is selectively retained in the germinal epithelium because the mutant germ cells enjoy an unspecified selective advantage [4].
However, the replication-error hypothesis does not hold for all dominant genetic mutations; achondroplasia being a particular case-in-point. By taking sperm DNA from donors of different ages, Tiemann-Boege and colleagues [5] have examined the frequency of nucleotide substitutions in the fibroblast growth factor receptor 3 (FGFR3) gene that are the predominant cause this condition. They concluded that the magnitude of the increased mutation frequency associated with paternal age was insufficient to explain the exponential rise in the incidence of achondroplasia in the offspring [4]. Of the several possible hypotheses that have been advanced to explain this situation, one of the most plausible proposes that replication error is not responsible for the mutations causing for this disease. Rather, age-related premutational lesions may have occurred in these cells that are converted to the authentic mutation (most commonly a glycine to arginine substitution at codon 1,138) following fertilization, as a consequence of aberrant DNA repair in the zygote [5,6].
In order for the premutational lesion hypothesis to account for a mutation that is present in every cell of the body, rather than a mosaic, this putative aberrant repair would have to precede the S phase of the first mitotic division. The oocyte is well endowed with poorly characterized enzymes for effecting DNA repair [7,8], including enzymes that are known to be active prior to the initiation of DNA synthesis during S-phase, at a time when the sperm chromatin is undergoing decondensation [9]. If an oocyte is fertilized with DNA-damaged spermatozoa, these G1-associated DNA repair mechanisms become activated, leading to a dramatic suppression of pronuclear DNA-synthesis via a p53 -dependent mechanism [9].
Of course, aberrant repair of DNA damage in the oocyte could account for a wide variety of genetic aberrations in embryos generated from DNA damaged spermatozoa, not just point mutations. Thus, exposure of spermatozoa to xenobiotics or X-irradiation is known to induce dominant lethal effects (post implantation pregnancy loss) and heritable translocations in the embryos of mated females as a consequence of chromosome mutations (breaks and rearrangements) as well as specific locus mutations [10,11] that could be the result of aberrant repair in the oocyte. If this mechanism is of fundamental importance in the causation of genetic disease, it places emphasis on discovering both the nature and extent of DNA damage in spermatozoa and the fidelity of the repair processes activated in the oocyte. Interestingly, there appear to be profound genetic differences in the capacity of oocytes to repair DNA damage introduced by the fertilizing spermatozoon [12]. However, what could be causing the DNA damage in spermatozoa? A possible clue might be found in the type of mutation most commonly observed in cases of achondroplasia.
As indicated above, the mutation seen in a vast majority of achondroplasia patients is a CG-AT transition in the FGFR3 gene. Since this is the most common base substitution observed following oxidative damage to DNA [13,14], it is plausible that the lesions responsible for the initiation of aberrant DNA repair in the oocyte are oxidative in nature. Such a hypothesis is in keeping with an extensive literature indicating that the functional lesions observed in the spermatozoa of infertile men, are commonly associated with signs of oxidative stress [1].
ROS generation in spermatozoa
Oxidative stress, and its role in the origins of male infertility was first appreciated in 1943, when the Scottish andrologist John MacLeod demonstrated that catalase could support the motility of human spermatozoa incubated under aerobic conditions [15]. His explanation for these findings, that human spermatozoa are vulnerable to oxidative stress created by reactive oxygen species (ROS) such as H2O2, has been confirmed in a number of independent studies [1]. Human spermatozoa are capable of generating ROS and this activity is of physiological significance in promoting the tyrosine phosphorylation events associated with sperm capacitation. The ability of ROS to enhance the tyrosine phosphorylation status of human spermatozoa depends partly on the ability of H2O2 to suppress tyrosine phosphatase activity, and partly on the ability of these molecules to stimulate cAMP generation by the soluble form of adenylyl cyclase (sAC) [16,17]. The cAMP generated in this manner then stimulates tyrosine phosphorylation via a PKA dependent mechanism involving an, as yet, uncharacterised intermediary tyrosine kinase [17]. Redox control of tyrosine phosphorylation during sperm capacitation has been recorded for a large number of species including the rat [18], mouse [19], human [17], bull [16] and stallion [20].
This redox drive to capacitation involves a low, steady state level of ROS production. However, if, for any reason, this physiological rate of ROS generation should increase, or the spermatozoa should become exposed to exogenous ROS generated by, for example, infiltrating leukocytes, then a state of oxidative stress can be readily induced. Spermatozoa are particularly susceptible to such stress as a consequence of their high unsaturated fatty acid content and their limited store of antioxidant enzymes such as superoxide dismutase or glutathione peroxidase [1]. In keeping with this concept, exposure of human spermatozoa to ROS generated by xanthine oxidase disrupts the functional competence of spermatozoa at levels that have little impact on somatic cells and, importantly, this effect can be reversed by the addition of catalase [21,22]. Of interest in the context of this review, is that this same system has been used to demonstrate the damaging effect of ROS on nuclear DNA in spermatozoa [23]. Direct exposure of human spermatozoa to ROS disrupts not only the functional competence of these cells but also their genomic integrity [24]. In relation to the paternal origins of disease, it is significant that endogenous ROS generated by human spermatozoa can, and does, affect sperm function and DNA integrity [6]. The current debate centers on how such oxidative stress is created.
Monitoring ROS production by spermatozoa
Following the initial report from Macleod [15], Aitken and Clarkson [25] went on to interpret the high levels of luminol-dependent chemiluminescence they observed in the spermatozoa of infertile patients, as evidence for ROS generation by such cells. The subpopulations of human spermatozoa responsible for this luminol-dependent activity have subsequently been isolated in the low density region of Percoll gradients [26,27] and linkages established with the aberrant retention of excess residual cytoplasm during spermiogenesis [27,28].
To further understand the biochemical basis of ROS generation in spermatozoa, attention soon focused on the use of other ROS-detecting probes apart from luminol. Owing to its sensitive nature, a popular reagent to use in this context was lucigenin. Lucigenin has been frequently deployed for the detection of ROS and is certainly capable of detecting superoxide anion (O2-•) [29,30]. For example, this probe has been used to detect NADPH oxidase type 2 (NOX2) -dependent O2-• production in phagocytic cells and in cell-free ROS-generating systems, such as xanthine plus xanthine oxidase [31]. In the context of spermatozoa, lucigenin as been used successfully to detect the generation of O2-• in rat sperm suspensions isolated from the cauda epididymides [32]. This signal was shown to be of mitochondrial origin, being inhibited by rotenone and stimulated by lactate, succinate and malate [32]. Although this is very good evidence for the production of mitochondrial ROS generation by rat spermatozoa, it is uncertain as to whether mitochondria are an important source of ROS in the spermatozoa of other species, particularly the human [25].
Notwithstanding the fact that mitochondrial production of ROS in spermatozoa remains largely unexplored, Vernet et al. [32] generated additional evidence for the non-mitochondrial production of ROS by rat spermatozoa. By isolating rat sperm membranes and adding lucigenin together with NADPH as a co-factor, they were able to show definitive chemiluminescent signals in this model system. This signal was inhibited with superoxide dismutase (SOD), DPI (diphenylene iodonium) and zinc [32]. Furthermore, addition of NADPH together with lucigenin to suspensions of human [33], mouse [34] rat [32] wallaby [35] and stallion [36] spermatozoa demonstrated that this lucigenin-dependent redox activity is a ubiquitous feature of mammalian spermatozoa.
In the case of human spermatozoa, NADPH-dependent lucigenin chemiluminescence was of non-mitochondrial in origin, being insensitive to rotenone, antimycin A, carbonyl cyanide m-chlorophenylhydrazone and sodium azide [33]. The initial interpretation of these data was that they represented support for the presence of an NADPH-oxidase in spermatozoa that could be responsible for the high rates of NAD(P)H-induced, lucigenin-dependent chemiluminescence recorded in defective sperm populations [37]. The general concept of such a theoretical oxidase is presented in Fig. 1. In essence, the enzyme serves to transfer electrons from NAD(P)H to ground state oxygen to create O2-• that then dismutates to H2O2 under the influence of intracellular superoxide dismutase (Fig. 1). The interpretation of these NADPH-induced, lucigenin-dependent chemiluminescence signals is complex however [38], because of the tendency of this probe to redox cycle as a result of one-electron reductions conducted by enzymes, such as cytochrome-P450 reductase [39]. This scheme proposes the following steps: (i) a one electron reduction of lucigenin (L) under the influence of cytochrome P450-reductase and NADPH to generate a radical species (LH+•), (ii) a reaction between the latter and ground state oxygen to produce O2-• and recycle the lucigenin back to its native state (L), (iii) and finally a reaction between LH+• and O2-•, generated as a result of the redox cycling of lucigenin, to produce a dioxetane that, in turn, decomposes with the generation of chemiluminescence (Fig. 2).
Figure 1 Schematic representation of NAD(P)H oxidase activity. This enzymes transfers electrons from NAD(P)H to ground state oxygen to create the superoxide anion radical. The latter then dismutates to hydrogen peroxide under the influence of superoxide dismutase. The hydrogen peroxide is predominantly scavenged by glutathione peroxidase, since human spermatozoa possess little catalase activity. Once this peroxidase activity is overwhelmed, a state of oxidative stress may be induced that disrupts the fertilizing capacity of the spermatozoa and the integrity of their DNA.
Figure 2 Schematic representation of the underlying chemistry of lucigenin chemiluminescence. Luc2+ = lucigenin; LH+• = a lucigenin radical created by the one electron reduction of Luc2+. The reaction of LH+• with oxygen generates O2-•. The latter then participates in an oxygenation reaction with LH+• generating a dioxetane that decomposes with the generation of chemiluminescence. Any entity that can bring about the one electron reduction of lucigenin can potentially create a redox cycle in the presence of oxygen that produces high levels of O2-• and chemiluminescence. It is impossible to distinguish the relative contribution of such probe-dependent and cell-dependent chemiluminescence. Hence data obtained with this probe should be interpreted with caution.
As a consequence of this chemistry, O2-•-dependent, SOD-sensitive chemiluminescence can be generated in cellular systems that do not generate ROS themselves but are simply capable of redox-cycling the probe (lucigenin) used to detect this activity. In a similar fashion, NADH-induced lucigenin-dependent chemiluminescence has been shown to be associated with another intracellular reductase, cytochrome b5-reductase (CYB5R) in spermatozoa [40]. Thus, CYB5R was shown to co-elute from an anion exchange column with NADH-induced, lucigenin-dependent chemiluminescence activity, while over-expression of this same enzyme led to a 3-fold increase of this activity in COS7 cells. Although CYB5R is capable of reducing lucigenin in the presence of NADH, a paradox has arisen from these studies in that CYB5R should be inhibited by DPI. However, the NADH-dependent enzyme responsible for lucigenin reduction in spermatozoa is not inhibited with this reagent. Two explanations for this discrepancy are worthy of consideration. First, it is plausible that DPI penetration to the sites of lucigenin action may be limited in intact cells (this would explain why DPI was so much more effective in suppressing the lucigenin chemiluminescence observed in cell-free partially purified CYB5R preparations, compared with transiently transfected intact cells [35]) or alternatively, a second DPI-insensitive enzyme system may exist in whole cells, that is also capable of activating lucigenin.
Although this general redox cycling concept (Fig. 2) seems to explain how SOD-inhibited NAD(P)H-dependent chemiluminescence can be generated in the absence of primary O2-• production, it has also been argued that the reaction between LH+• and O2 is thermodynamically unlikely [41] and that redox cycling of this probe cannot occur in biological systems. In light of such reservations, we cannot be certain of the extent to which the elevated NAD(P)H-induced lucigenin signals detected in defective human spermatozoa [37] reflect primary O2-• production.
NAD(P)H-induced chemiluminescence and oxidative stress
Even if NAD(P)H-induced lucigenin-dependent chemiluminescence does simply reflect the presence of oxidoreductases capable of initiating the redox cycling of the probe, rather than primary O2-• generation, the diagnostic significance of this activity may still reside in its ability to reflect oxidative stress in human sperm populations [37,42,43]. Thus, if the inter-individual differences in lucigenin chemiluminescence reflect the varying availability of certain reductases, this must, in turn, reflect inter-individual differences in the retention of residual cytoplasm during spermiogenesis. Such a conclusion would be in keeping with a large number of studies indicating that defective sperm function is positively correlated with the presence of numerous cytosolic enzymes that are markers of the cytoplasmic space, including lactic acid dehydrogenase, creatine kinase, SOD and glucose-6-phosphate dehydrogenase [6]. These observations, in turn, reflect the fact that defective sperm function is frequently associated with the retention of excess residual cytoplasm as a result of impaired cytoplasmic extrusion during spermiogenesis. Human spermatozoa are unusual in that they have lost the ability shown by most mammalian species to remodel any residual cytoplasm into a cytoplasmic droplet that is ultimately discharged from the cells either during epididymal maturation or at ejaculation. As a consequence, any residual cytoplasm that remains after spermiogenesis has been completed in the human, is retained by the spermatozoa as an amorphous cytoplasmic mass in the neck region of the cell (Fig. 3). Retention of this excess cytoplasm has been associated with the existence of oxidative stress in the germ line in several independent studies [27,28].
Figure 3 Errors during spermiogenesis can lead to the retention of excess residual cytoplasm by human spermatozoa. A) arrows point to cells possessing an irregular cytoplasmic mass in the neck region of the spermatozoon; significantly, human spermatozoa have lost the ability, possessed by most other mammalian species, to create a cytoplasmic droplet which is later discharged from the cell B) The amount of cytoplasm retained by human spermatozoa is highly correlated with the ability of leukocyte-free sperm suspensions to generate a chemiluminescence response to 12-myristate, 13-acetate phorbol ester (PMA), using luminol-peroxidase as the detection system.
An explanation for this association between excess cytoplasmic retention and ROS production has been put forward, based on the correlation between the latter and the cellular content of glucose-6-phosphoate dehydrogenase in human sperm suspensions [28]. This enzyme regulates the rate of glucose flux through the hexose monophosphate shunt and is, in this way, responsible for controlling the intracellular availability of NADPH (Fig. 1). The latter may then serve as a substrate for putative ROS-generating enzymes in the germ line such as NOX 5 [44], and thereby, elicit a state of oxidative stress.
Although this explanation is consistent with the data generated in numerous independent studies, it still has the status of an unproven hypothesis rather than an experimentally proven fact. Definitive evidence for the presence of NOX 5 in human spermatozoa has still not been obtained and we still await independent confirmation of the ability of NOX 5 to generate ROS in spermatozoa. It has even been questioned whether human spermatozoa produce ROS from any source. Using the ability of dihydroethidium to react with O2-• and produce 2OH ethidium we have recently secured incontrovertible evidence that human spermatozoa do in fact generate this free radical species and that the cellular production of O2-• is significantly elevated in populations of defective sperm populations. Furthermore, the data we have generated to date suggests that this is not a consequence of defective mitochondrial function since a variety of mitochondrial inhibitors fail to disrupt the 2OH ethidium signal.
Such results suggest there must be a source of O2-• in human spermatozoa that is extra-mitochondrial. A number of possibilities exist in this context. First of all, the above-mentioned NADPH-oxidase, NOX5, originally identified by Banfi et al. in human testes [44,45], has the potential to generate free radicals in a calcium-dependent manner, as originally described by Aitken and Clarkson [25]. Although the presence of NOX5 in human spermatozoa has not yet been confirmed, a recent publication in the mouse [46] employed Western blot analysis to record the presence of proteins exhibiting cross reactivity with antibodies against various components of the leukocyte NADPH oxidase complex (NOX2) including gp91phox, p67(phox), p47 phox and p40 (phox). The authors claim that this unusual oxidase is regulated by the availability of p40 (phox) and is independent of p22 (phox). They also assert that this oxidase is maximally active in testicular spermatozoa but decreases in concert with sperm maturation. Confirmation of this pattern of NADPH oxidase activity is strategically important since it would add weight to the argument that ROS production by human spermatozoa is an inverse function of their state of functional maturity.
In addition to NADPH oxidase activity, it is also clear that spermatozoa will generate ROS when placed in contact with certain xenobiotics. Examples of such compounds include endocrine disruptors with estrogenic properties that are capable of inducing ROS production by male germ cells and initiating free radical-mediated DNA damage [47]. Similarly, oxidative stress and DNA damage can be induced in spermatozoa via metal-catalysed redox activity involving, for example nickel [48] or iron [49] as well as phthalate esters [50].
Another potential source of ROS in human spermatozoa is a trans-plasma membrane oxidoreductase system that removes electrons from NAD(P)H on the cytoplasmic surface of the cell and transfers them to oxygen on the outer leaf of the plasma membrane via intermediate carriers such as ubiquinone. Evidence to support the existence of such systems in somatic cells, arises from the ability of the latter to reduce artificial membrane-impermeant electron acceptors such as potassium ferricyanide, in concert with the concomitant oxidation of cytosolic NADH [51]. Interestingly, the activity of such plasma membrane oxidase systems has been demonstrated to increase in Rho O cells that lack functional mitochondria [51,52]. In such cases, up-regulation of the plasma membrane redox system may help maintain an adequate pool of NAD+ to fuel the increased glycolysis needed to maintain cell viability in the absence of mitochondrial activity. Human spermatozoa have been shown to possess redox activity typical of such plasma membrane electron transport chains [32,53]. Thus, human spermatozoa possess a capacity to reduce the probe WST-1 in the presence of an intermediate electron acceptor in a similar fashion to the plasma membrane redox system described in somatic cells by Berridge and Tan [54,55]. This transmembrane electron transfer system in spermatozoa shares similarities with the Berridge and Tan activity in being inhibited by SOD, capsaicin, (a potent vanilloid inhibitor) and N-ethyl maleimide (NEM, a membrane permeant alkylating agent). However, the susceptibility of the sperm oxidase activity to the membrane impermeant thiol blocking agent pCMBS (p-chloromercuriphenylsulphonate) as well as retinoic acid, distinguishes the sperm-based activity from that detected in somatic cells [37]. Since mitochondrial function is frequently defective in populations of human spermatozoa [56], it is possible that this trans-plasma membrane redox system is up-regulated in defective human spermatozoa in a similar fashion to the enhanced activity recorded in Rho O cells in order to maintain the redox status of the NAD+/NADH couple. Further characterization of this putative electron transport chain is clearly warranted. However, this task will not be easy given that, by definition, the redox activity ascribed to such systems depends on the close interaction of several independent constituents, not on a single identifiable entity.
Another form of oxidase activity detected by Berridge and Tan [54,55] is a superficial enzyme that removes electrons from exogenously applied sources of reducing equivalents (eg. NAD(P)H) to generate O2-•. Exogenous NAD(P)H will certainly reduce extracellular electron acceptors such as WST-1 in the presence of human spermatozoa [37]. Moreover, the susceptibility of such activity to inhibition with pCMBS corresponds to the activity detected in a variety of cell types, including HeLa and Jurkat cells, by Berridge and Tan [54,55]. However, the activity elicited in the presence of spermatozoa could be distinguished from that generated by somatic cells by virtue of the lack of stimulation observed with NEM [37,54,55].
Nitric oxide (NO) is another oxygen free radical which is apparently generated by defective populations of human spermatozoa. NO is normally generated from L-arginine by three isoforms of nitric oxide synthase (NOS). Recent mouse knock out experiments indicate that the selective deletion of these NOS isoforms has no impact on the ability of the spermatozoa to achieve fertilization. In other words, NO does not appear to have a positive role to play in the generation of functional gametes [57]. However the fact that iNOS deficient spermatozoa exhibited significantly higher in vitro fertilization rates than the wild-type controls, clearly suggests that NO may be involved in the etiology of defective sperm function. NO clearly has a detrimental effect on normal sperm function inhibiting both motility and the competence of these cells for sperm-zona binding [58]. Moreover the NO levels in seminal plasma are negatively correlated with sperm movement in human semen samples [59]. The source of this NO is still an open question. The involvement of NOS is suggested by the negative correlation observed between the pattern of NOS expression on human spermatozoa and percentage motility [60]. However, it has also been pointed out the NO and peroxynitrite formation can be stimulated in mammalian spermatozoa using D-arginine, which cannot be a substrate for NOS. Under these circumstances it is possible that NO is being generated non-enzymatically through an H2O2-mediated attack on arginine [61].
Clearly there are many potential sources of ROS in the male germ line. The task that now confronts us is to determine which of these multifarious sources are responsible for the oxidative stress observed in the spermatozoa of male patients.
Conclusion
Human spermatozoa are redox active cells that are capable of generating O2-• and H2O2. This activity is of fundamental biological importance in regulating the signal transduction pathways that control sperm capacitation. However, excess exposure to ROS can lead to pathological damage to human spermatozoa curtailing their competence for fertilization and disrupting their genetic integrity. DNA damage in these cells appears to be largely oxidative and is associated with a wide variety of adverse outcomes including impaired conception rates, increased incidences of abortion and defects in the offspring, including childhood cancer and dominant genetic diseases such as achondroplasia. It is hypothesized that such effects in the F1 generation involve the aberrant repair of oxidative DNA damage in the newly fertilized zygote. The etiology of oxidative stress in the male germ line is being actively researched at the present time. While spermatozoa certainly generate ROS, the biochemical basis of this activity is uncertain and may be multifactorial. Errors of spermiogenesis associated with the retention of excess residual cytoplasm appear to be associated with oxidative stress as a consequence of enhanced ROS production by uncharacterised 'oxidases', plasma membrane electron transport chains or oxidoreductases capable of activating redox-cycling xenobiotics. Electron leakage from defective sperm mitochondria represent yet another potential source of oxygen radicals. Given the clinical significance of oxidative stress in human spermatozoa, resolving the biochemical basis of this condition is a high priority task for the future.
==== Refs
Aitken RJ The human spermatozoon – a cell in crisis? The Amoroso Lecture J Reprod Fertil 1999 115 1 7 10341716
Crow JF The origins, patterns and implications of human spontaneous mutation Nat Rev Genet 2000 1 40 47 11262873 10.1038/35049558
Penrose LS Parental age and mutation Lancet 1955 269 312 313 13243724 10.1016/S0140-6736(55)92305-9
Goriely A McVean GA Rojmyr M Ingemarsson B Wilkie AO Evidence for selective advantage of pathogenic FGFR2 mutations in the male germ line Science 2003 301 643 646 12893942 10.1126/science.1085710
Tiemann-Boege I Navidi W Grewal R Cohn D Eskenazi B Wyrobek AJ Arnheim N The observed human sperm mutation frequency cannot explain the achondroplasia paternal age effect Proc Natl Acad Sci U S A 2002 99 14952 14957 12397172 10.1073/pnas.232568699
Aitken RJ Human spermatozoa, fruits of creation, seeds of doubt. The Founder's Lecture Reprod Fert Dev 2004 16 655 664
Zheng P Schramm RD Latham KE Developmental regulation and in vitro culture effects on expression of DNA repair and cell cycle checkpoint control genes in rhesus monkey oocytes and embryos Biol Reprod 2005 72 1359 1369 15703371 10.1095/biolreprod.104.039073
Matsuda Y Maemori M Tobari I Relationship between cell cycle stage in the fertilized egg of mice and repair capacity for X-ray-induced damage in the sperm Int J Radiat Biol 1989 56 301 314 2570816
Shimura T Inoue M Taga M Shiraishi K Uematsu N Takei N Yuan ZM Shinohara T Niwa O p53-dependent S-phase damage checkpoint and pronuclear cross talk in mouse zygotes with X-irradiated sperm Mol Cell Biol 2002 22 2220 2228 11884608 10.1128/MCB.22.7.2220-2228.2002
Generoso WM Huff SW Cain KT Relative rates at which dominant-lethal mutations and heritable translocations are induced by alkylating chemicals in postmeiotic male germ cells of mice Genetics 1979 93 163 171 232878
Russell LB Effects of male germ-cell stage on the frequency, nature, and spectrum of induced specific-locus mutations in the mouse Genetica 2004 122 25 36 15619958 10.1007/s10709-004-1443-7
Generoso WM Cain KT Krishna M Huff SW Genetic lesions induced by chemicals in spermatozoa and spermatids of mice are repaired in the egg Proc Natl Acad Sci U S A 1979 76 435 437 284360
Kreutzer DA Essigmann JM Oxidized, deaminated cytosines are a source of C – -> T transitions in vivo Proc Natl Acad Sci U S A 1998 95 3578 3582 9520408 10.1073/pnas.95.7.3578
Crow JF The high spontaneous mutation rate: is it a health risk? Proc Natl Acad Sci USA 1997 94 8380 8386 9237985 10.1073/pnas.94.16.8380
MacLeod J The role of oxygen in the metabolism and motility of human spermatozoa Am J Physiol 1943 138 512 518
Rivlin J Mendel J Rubinstein S Etkovitz N Breitbart H Role of hydrogen peroxide in sperm capacitation and acrosome reaction Biol Reprod 2004 70 518 522 14561655 10.1095/biolreprod.103.020487
Aitken RJ Harkiss D Knox W Paterson M Irvine DS A novel signal transduction cascade in capacitating human spermatozoa characterised by a redox-regulated, cAMP-mediated induction of tyrosine phosphorylation J Cell Sci 1998 111 645 656 9454738
Lewis B Aitken RJ A redox-regulated tyrosine phosphorylation cascade in rat spermatozoa J Androl 2001 22 611 622 11451358
Ecroyd H Jones RC Aitken RJ Endogenous redox activity in mouse spermatozoa and its role in regulating the tyrosine phosphorylation events associated with sperm capacitation Biol Reprod 2003 69 347 354 12672670 10.1095/biolreprod.102.012716
Baumber J Sabeur K Vo A Ball BA Reactive oxygen species promote tyrosine phosphorylation and capacitation in equine spermatozoa Theriogenology 2003 60 1239 1247 14511778 10.1016/S0093-691X(03)00144-4
Aitken RJ Buckingham D Harkiss D Use of a xanthine oxidase oxidant generating system to investigate the cytotoxic effects of reactive oxygen species on human spermatozoa J Reprod Fertil 1993 97 441 450 8388958
Griveau JF Dumont E Renard P Callegari JP Le Lannou D Reactive oxygen species, lipid peroxidation and enzymatic defence systems in human spermatozoa J Reprod Fertil 1995 103 17 26 7707295
Lopes S Jurisicova A Sun JG Casper RF Reactive oxygen species: potential cause for DNA fragmentation in human spermatozoa Human Reprod 1998 13 896 900 10.1093/humrep/13.4.896
Aitken RJ Gordon E Harkiss D Twigg JP Milne P Jennings Z Irvine DS Relative impact of oxidative stress on the functional competence and genomic integrity of human spermatozoa Biol Reprod 1998 59 1037 1046 9780307 10.1095/biolreprod59.5.1037
Aitken RJ Clarkson JS Cellular basis of defective sperm function and its association with the genesis of reactive oxygen species by human spermatozoa J Reprod Fertil 1987 81 459 469 2828610
Gil-Guzman E Ollero M Lopez MC Sharma RK Alvarez JG Thomas AJ JrAgarwal AP Differential production of reactive oxygen species by subsets of human spermatozoa at different stages of maturation Human Reprod 2001 16 1922 1930 10.1093/humrep/16.9.1922
Ollero M Gil-Guzman E Lopez MC Sharma RK Agarwal A Larson K Evenson D Thomas AJ JrAlvarez JG Characterization of subsets of human spermatozoa at different stages of maturation: implications in the diagnosis and treatment of male infertility Human Reprod 2001 16 1912 1921 10.1093/humrep/16.9.1912
Gomez E Buckingham DW Brindle J Lanzafame F Irvine DS Aitken RJ Development of an image analysis system to monitor the retention of residual cytoplasm by human spermatozoa: correlation with biochemical markers of the cytoplasmic space, oxidative stress and sperm function J Androl 1996 17 276 287 8792218
Li Y Stansbury KH Zhu H Trush MA Biochemical characterization of lucigenin (Bis-N-methylacridinium) as a chemiluminescent probe for detecting intramitochondrial superoxide anion radical production Biochem Biophys Res Commun 1999 262 80 87 10448072 10.1006/bbrc.1999.1174
Li Y Zhu H Kuppusamy P Roubaud V Zweier JL Trush MA Validation of lucigenin (bis-N-methylacridinium) as a chemilumigenic probe for detecting superoxide anion radical production by enzymatic and cellular systems J Biol Chem 1998 273 2015 2023 9442038 10.1074/jbc.273.4.2015
Kopprasch S Pietzsch J Graessler J Validation of different chemilumigenic substrates for detecting extracellular generation of reactive oxygen species by phagocytes and endothelial cells Luminescence 2003 18 268 73 14587078 10.1002/bio.737
Vernet P Fulton N Wallace C Aitken RJ Analysis of reactive oxygen species generating systems in rat epididymal spermatozoa Biol Reprod 2001 65 1102 1113 11566731 10.1095/biolreprod65.4.1102
Aitken RJ Fisher H Fulton N Knox W Lewis B Reactive oxygen species generation by human spermatozoa is induced by exogenous NADPH and inhibited by the flavoprotein inhibitors diphenylene iodonium and quinacrine Molec Reprod Dev 1997 47 468 482 9211432 10.1002/(SICI)1098-2795(199708)47:4<468::AID-MRD14>3.0.CO;2-S
Fisher HM Aitken RJ Comparative analysis of the ability of precursor germ cells and epididymal spermatozoa to generate reactive oxygen metabolites J Exp Zool 1997 277 390 400 9127958 10.1002/(SICI)1097-010X(19970401)277:5<390::AID-JEZ5>3.0.CO;2-K
Bennetts L Lin M Aitken RJ Cyclic AMP-dependent tyrosine phosphorylation in tammar wallaby (Macropus eugenii) spermatozoa J Exp Zoo 2004 301 118 130 10.1002/jez.a.20020
Ball BA Vo AT Baumber J Generation of reactive oxygen species by equine spermatozoa Am J Vet Res 2001 62 508 515 11327456
Aitken RJ Ryan AL Curry BJ Baker MA Multiple forms of redox activity in human spermatozoa Molec Hum Reprod 2003 9 645 661 14561808 10.1093/molehr/gag086
de Lamirande E Harakat A Gagnon C Human sperm capacitation induced by biological fluids and progesterone, but not by NADH or NADPH, is associated with the production of superoxide anion J Androl 1998 19 215 225 9570746
Baker MA Krutskikh A Curry BJ McLaughlin EA Aitken RJ Identification of cytochrome P450-reductase as the enzyme responsible for NADPH-dependent lucigenin and tetrazolium salt reduction in rat epididymal sperm preparations Biol Reprod 2004 71 307 318 15031143 10.1095/biolreprod.104.027748
Baker MA Krutskikh A. Curry BJ Hetherington L Aitken RJ Identification of cytochrome b5-reductase as the enzyme responsible for NADH-dependent lucigenin chemiluminescence in human spermatozoa Biol Reprod 2005 73 334 342 15858218 10.1095/biolreprod.104.037960
Afanas'ev IB Ostrachovich EA Korkina LG Lucigenin is a mediator of cytochrome C reduction but not of superoxide anion production Arch Biochem Biophys 1999 366 267 274 10356292 10.1006/abbi.1999.1215
Said TM Agarwal A Sharma RK Mascha E Sikka SC Thomas AJ Jr Human sperm superoxide anion generation and correlation with semen quality in patients with male infertility Fertil Steril 2004 82 871 877 15482762 10.1016/j.fertnstert.2004.02.132
Said TM Agarwal A Sharma RK Thomas AJ Sikka S Impact of sperm morphology on DNA damage caused by oxidative stress induced by β-nicotinamide adenine dinucleotide phosphate Fertil Steril 2005 83 95 103 15652893 10.1016/j.fertnstert.2004.06.056
Banfi B Tirone F Durussel I Knisz J Moskwa P Molnar GZ Krause KH Cox JA Mechanism of Ca2+ activation of the NADPH oxidase 5 (NOX5) J Biol Chem 2004 279 18583 18591 14982937 10.1074/jbc.M310268200
Banfi B Molnar G Maturana A Steger K Hegedus B Demaurex N Krause KH A Ca(2+)-activated NADPH oxidase in testis, spleen, and lymph nodes J Biol Chem 2001 276 37594 37601 11483596 10.1074/jbc.M103034200
Shukla S Jha RK Laloraya M Kumar PG Identification of non-mitochondrial NADPH oxidase and the spatio-temporal organization of its components in mouse spermatozoa Biochem Biophys Res Commun 2005 331 476 483 15850784 10.1016/j.bbrc.2005.03.198
Anderson D Schmid TE Baumgartner A Cemeli-Carratala E Brinkworth MH Wood JM Oestrogenic compounds and oxidative stress (in human sperm and lymphocytes in the Comet assay Mutat Res 2003 544 173 178 14644319 10.1016/j.mrrev.2003.06.016
Doreswamy K Shrilatha B Rajeshkumar T Muralidhara Nickel-induced oxidative stress in testis of mice: evidence of DNA damage and genotoxic effects J Androl 2004 25 996 1003 15477375
Aitken RJ Harkiss D Buckingham DW Analysis of lipid peroxidation mechanisms in human spermatozoa Mol Reprod Dev 1993 35 302 315 8352936 10.1002/mrd.1080350313
Kasahara E Sato EF Miyoshi M Konaka R Hiramoto K Sasaki J Tokuda M Nakano Y Inoue M Role of oxidative stress in germ cell apoptosis induced by di(2-ethylhexyl)phthalate Biochem J 2002 365 849 856 11982482
Larm JA Vaillant F Linnane AW Lawen A Up-regulation of the plasma membrane oxidoreductase as a prerequisite for the viability of human Namalwa rho 0 cells J Biol Chem 1994 269 30097 30100 7982910
Scarlett DJ Herst P Tan A Prata C Berridge M Mitochondrial gene-knockout (rho0) cells: a versatile model for exploring the secrets of trans-plasma membrane electron transport Biofactors 2004 20 199 206 15706057
Aitken RJ Baker MA Oxidative stress and male reproductive biology Reprod Fert Dev 2004 16 581 588 10.1071/RD03089
Berridge MV Tan AS Cell-surface NAD(P)H-oxidase: relationship to trans-plasma membrane NADH-oxidoreductase and a potential source of circulating NADH-oxidase Antioxid Redox Signal 2000 2 277 288 11229532
Berridge MV Tan AS High capacity redox control at the plasma membrane of mammalian cells: trans-membrane, cell surface, and serum NADH-oxidases Antiox Redox Signal 2000 2 231 242
Marchetti C Obert G Deffosez A Formstecher P Marchetti P Study of mitochondrial membrane potential, reactive oxygen species, DNA fragmentation and cell viability by flow cytometry in human sperm Human Reprod 2002 17 1257 1265 10.1093/humrep/17.5.1257
Yang JZ Ajonuma LC Rowlands DK Tsang LL Ho LS Lam SY Chen WY Zhou CX Chung YW Cho CY Tse JY James AE Chan HC The role of inducible nitric oxide synthase in gamete interaction and fertilization: a comparative study on knockout mice of three NOS isoforms Cell Biol Int 2005 29 785 791 16087361 10.1016/j.cellbi.2005.05.005
Wu TP Huang BM Tsai HC Lui MC Liu MY Effects of nitric oxide on human spermatozoa activity, fertilization and mouse embryonic development Arch Androl 2004 50 173 179 15204684
Balercia G Moretti S Vignini A Magagnini M Mantero F Boscaro M Ricciardo-Lamonica G Mazzanti L Role of nitric oxide concentrations on human sperm motility J Androl 2004 25 245 249 14760010
O'Bryan MK Zini A Cheng CY Schlegel PN Human sperm endothelial nitric oxide synthase expression: correlation with sperm motility Fertil Steril 1998 70 1143 1147 9848308 10.1016/S0015-0282(98)00382-3
Aitken RJ Ryan AL Baker MA McLaughlin EA Redox activity associated with the maturation and capacitation of mammalian spermatozoa Free Radic Biol Med 2004 36 994 1010 15059640 10.1016/j.freeradbiomed.2004.01.017
|
16313680
|
PMC1315356
|
CC BY
|
2021-01-04 16:37:13
|
no
|
Reprod Biol Endocrinol. 2005 Nov 29; 3:67
|
utf-8
|
Reprod Biol Endocrinol
| 2,005 |
10.1186/1477-7827-3-67
|
oa_comm
|
==== Front
Cancer Cell IntCancer Cell International1475-2867BioMed Central London 1475-2867-5-321628394810.1186/1475-2867-5-32Primary ResearchA new, fast and semi-automated size determination method (SASDM) for studying multicellular tumor spheroids Monazzam Azita [email protected] Pasha [email protected] Örjan [email protected] Raymond [email protected]ångström Bengt [email protected]öm Mats [email protected] Institute of Oncology, Radiology and Clinical Immunology, Uppsala University2 Uppsala University, Centre for Image Analysis, Lägerhyddsvägen 3, SE-752 37 Uppsala, Sweden3 Uppsala Imanet AB (PET Centre), Uppsala University Hospital, SE-751 85 Uppsala, Sweden4 Department of Pharmaceutical Biosciences, Uppsala Biomedical Centre, SE-751 24 Uppsala, Sweden5 Department of medical Science, Uppsala University Hospital, SE-751 58 Uppsala, Sweden2005 14 11 2005 5 32 32 2 6 2005 14 11 2005 Copyright © 2005 Monazzam et al; licensee BioMed Central Ltd.2005Monazzam 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
Considering the width and importance of using Multicellular Tumor Spheroids (MTS) in oncology research, size determination of MTSs by an accurate and fast method is essential. In the present study an effective, fast and semi-automated method, SASDM, was developed to determinate the size of MTSs. The method was applied and tested in MTSs of three different cell-lines. Frozen section autoradiography and Hemotoxylin Eosin (H&E) staining was used for further confirmation.
Results
SASDM was shown to be effective, user-friendly, and time efficient, and to be more precise than the traditional methods and it was applicable for MTSs of different cell-lines. Furthermore, the results of image analysis showed high correspondence to the results of autoradiography and staining.
Conclusion
The combination of assessment of metabolic condition and image analysis in MTSs provides a good model to evaluate the effect of various anti-cancer treatments.
==== Body
Introduction
The growth of tumor cells such as Multicellular Tumor Cultures (MTSs) has led to important insights in tumor biology [1]. MTSs represent an intermediary level between monolayer growing cells and solid tumors in animals and humans [2]. The cytology and morphology of MTSs are close to experimental tumors in mice and tumors in humans, before neovascularisation occurs [3]. In fact, MTSs represent quite realistically the three-dimensional growth and organization of solid tumors and therefore simulate the cell-cell interactions and micro environmental conditions found in the tumors [4,5]. For example, multicellular aggregates develop a central necrosis, similar to that seen in many tumors in vivo.
Size determination and evaluation of growth pattern are two essential aspects when studying the characteristics and behavior of the MTSs. These are important factors in studies in which e.g. the total uptake of radiotracers is evaluated and in therapeutically oriented investigations, where a drug may induce morphological changes in the tumors.
The most commonly used technique to determine size of MTSs is to measure two diameters of the spheroid, using a calibrated ocular micrometer on an inverted microscope. These values are then used to determine the volume approximately. This measurement is both time consuming and imprecise, especially for irregular MTSs. Moreover, MTSs tend to develop a central necrosis, i.e. the volume of viable cells differs from the total volume of the aggregate. It is therefore necessary to have the possibility to evaluate the volume of viable cells as one entity describing the aggregate and the fraction of the total volume constituted by viable cells as another entity.
In the present study an effective, fast and semi-automated method (SASDM) for more accurate determination of the size of MTSs and its fraction of viable cells was developed and utilized.
Materials and methods
Cell culture
Three standard cell lines were used to investigate the performance of this method:
• MCF-7, a human breast cancer cell line (European Collection of Cell Culture).
• U-343, a human glioma cell line (Westermark et al 1973).
• BON, a human neuroendocrine tumor cell line (a kind gift from Dr. C.M Townsend, University of Texas, Galvestone, USA) derived from a lymph node metastasis of a pancreatic carcinoid.
The MCF-7 cells were grown in MEM-Eagle medium supplemented with 10% FCS, 1 mM sodium pyruvate, 2 mM L-glutamine, 1% non-essential amino acids and 5% Penicillin (Tamro). U-343 cells were cultured in Ham's F-10 medium supplemented with 10% FCS, 1 mM sodium pyruvate and 2 mM L-glutamine (Tamro). BON cells were grown in Ham F-12 K medium (NordCell, Sweden) mixed with DMEM medium supplemented with 10% FCS, 1 mM sodium pyruvate and 2 mM L-glutamine (Tamro).
The medium was changed twice a week and the cells were maintained in exponential growth phase.
Multicellular tumor culture
The tumor cells were trypsinized from the stem monolayer culture, then cell suspensions were seeded in 24-well 1% agarose coated culture plates, with approximately 50,000 cells per well for U-343 and MCF-7 and 15,000 for BON. The cultures were kept at 37°C with 5% CO2, and grown for a period that depended on duplication rate for each cell line [6].
MTS medium for MCF-7 was DMEM supplemented with 10% FCS, 1 mM sodium pyruvate, 2 mM L-glutamine, 1% non-essential amino acids, 5% Penicillin (Tamro), 0.01 mg/ml Insulin and 1 nM β-Estradiol (Sigma Aldrich).
Image Analysis
The aggregates were photographed daily using a Nikon Colorpix 4500 digital camera mounted on a Zeiss Axiovert 135 Microscope . The digital camera was set to use up to 4.0 million effective pixels, with an image resolution of 640 × 480, auto-focus, auto-shooting modes, with applied "Hi" image quality. A 5× magnification objective, 5x/0,12; 44 01 20 on Zeiss Axiovert 135 Microscope was used with adjustment of the strength of background light depending on the color of the medium that was used for growing the desired MTS.
Images were saved as sequences of JPEG files in a 128 MB, Scan Disc Compact Flash card and transferred into the hard disc of a personal computer with installed Windows 2000 for further image and statistical analysis.
A software program in Matlab (The Mathworks, Natick, Massachusetts) with user-friendly interface was developed to perform the image analysis using routines from the "image processing" toolbox. For absolute calibration of the area, a spherical object with a specified diameter of 0.79375 μm, as shown in Fig. 1, was used as reference object.
Figure 1 Reference object used for absolute calibration of area. The units on the axis are number of pixels.
After feeding some input parameters such as name of starting image, number of images to analyze, project name, the software automatically delineated a ROI (Region of Interest) discriminating the necrosis area from the area where the cells were still alive and a ROI around the whole MCT. This was done automatically and sequentially for all images. The total area of the MTS and total area of necrosis were calculated as the total number of pixels inside the ROIs and was converted into μm2 by using the radius of the reference object. To calculate the volume of the MTS the following equation was used.
V=43πr3
MathType@MTEF@5@5@+=feaafiart1ev1aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacH8akY=wiFfYdH8Gipec8Eeeu0xXdbba9frFj0=OqFfea0dXdd9vqai=hGuQ8kuc9pgc9s8qqaq=dirpe0xb9q8qiLsFr0=vr0=vr0dc8meaabaqaciGacaGaaeqabaqabeGadaaakeaacqWGwbGvcqGH9aqpdaWcaaqaaiabisda0aqaaiabiodaZaaacqaHapaCcqWGYbGCdaahaaWcbeqaaiabiodaZaaaaaa@352E@
Where r=area/π
MathType@MTEF@5@5@+=feaafiart1ev1aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacH8akY=wiFfYdH8Gipec8Eeeu0xXdbba9frFj0=OqFfea0dXdd9vqai=hGuQ8kuc9pgc9s8qqaq=dirpe0xb9q8qiLsFr0=vr0=vr0dc8meaabaqaciGacaGaaeqabaqabeGadaaakeaacqWGYbGCcqGH9aqpdaGcaaqaaiabdggaHjabdkhaYjabdwgaLjabdggaHjabc+caViabec8aWbWcbeaaaaa@3735@ was used to calculate the radius of the sphere. The total volume of the necrosis was calculated and subtracted from the total volume of the MTS to obtain the total volume of the living part. Moreover the fraction of total volume constituted by living cells was calculated.
All images with outlined ROIs discriminating the different parts of the MTSs were visualized and automatically saved as TIFF (Tagged Image File Format) for further observations. All desired statistical measurements including the total volume of the whole MTSs and total volume of the necrosis part were saved in an EXCEL file with specified file name for the full group of MTSs.
Validation of method was performed by analyzing pictures from different angles of the same aggregate. The growth curves for the three selected cell lines were generated to illustrate the use of the program. During the first few days of culture after seeding, the aggregates settle, which means they do not follow an exponential growth pattern. The growth curves after this period of 5–9 days were plotted and data were fitted to a monoexponential to indicate the volume doubling time.
Frozen Section Autoradiography and Staining
At a selected time point some of the MTSs were incubated with 18F-labeled Fluoro-Deoxy-Glucose (FDG) to confirm the extent of the necrosis part. After 40 min incubation with 100 MBq/ml FDG and washing for 3*5 min, the MTSs were frozen quickly at -20°C and sectioned for autoradiography [7].
The slices with a thickness of 25 μm were exposed on a phosphor-imaging plate for 20 h. Scanning and imaging were performed using software Image Quant (Molecular Dynamics, Sunnyvale, CA).
For further verification Hematoxylin & Eosin staining of the slices was used. This complementary part was to clarify some histological features in the MTSs.
Results
Multicellular aggregates from all cell lines directly after seeding showed a uniform feature without necrosis, which after a few days changed to the typical feature shown in Figure 2, 4, and 6, with a rim of viable cells surrounding a central necrosis. The shapes differed slightly as did the sizes of individual aggregates. The automatic outlining of the necrosis and the total periphery behaved properly in all cases.
Figure 2 Images of six different MCF-7 cell aggregates. ROIs have been drawn separating the necrosis and living part from the background. The units on the axis are number of pixels.
Figure 4 Images of six different U-343 MTS. ROIs have been drawn separating the necrosis and living part from the background.
Figure 6 Images of BON cell aggregates. ROIs have been drawn separating the necrosis and living part from the background.
The computation time for a set of 25 images was 17 s.
A reproducibility test of the method by repetitive measurements on the same aggregates, after shaking and rotation, showed a variability (Coefficient Of Variance) below 5%.
The MTS volume doubling time was calculated to be 13 days for MCF-7 (Figure 3), 57 days for U-343 (Figure 5) and 5 days for BON (Figure 7).
Figure 3 Volume growth pattern for the MTS of MCF-7 cell line during 17 days of observation. The data represent the mean of 24 MTS.
Figure 5 olume growth pattern for the MTS of U-343 cell line during 15 days of observation. The data represent the mean of 24 MTS.
Figure 7 Volume growth pattern for the MTS of BON cell line during 17 days of observation. The data represent the mean of 24 MTS.
For each cell line, the Coefficient Of Variance within the group was below 15% but considerably higher in-between groups. This further confirms how essential it is to determinate the size of MTS for each experiment set-up.
The result of autoradiography failed to show a clear rim of radioactivity, indicating that the resolution possible with a PET tracer is not sufficient for this task, whereas H&E staining confirmed the necrosis part in the middle of the MTSs (Figure 8, 9).
Figure 8 Frozen section autoradiography of an MCF-7 MTS incubated with 18F-FDG. Serial 25 μm sections are numbered from the top of the MTS to the bottom.
Figure 9 Hemotoxylin and Eosin staining of a BON MTS section. A focus of necrosis is seen in the middle part of the picture.
Discussion
The traditional method to use cultured tumor cells for biology evaluations, including assessment of radiotracer uptake and effects of treatment, uses monolayer culture. This method is easy to apply, allows a good and reproducible source of cells and allows easy calculation of cell number by counting with sophisticated methods. However, there are some clear drawbacks to the method, especially for evaluating radiotracer uptake and monitoring of treatment response. The primary issue is that the growth is severely up-regulated in monolayer as compared to in vivo growth due to lack of adequate contact inhibition. This accelerated growth and lack of direct communication with neighboring cells also questions the physiological relevance, with the expectation that cellular function is aberrant. Furthermore, a monolayer with its rapid growth, quickly changes from a sparse cell culture to a culture with overlay growth, and thereby prohibits long-term follow-up of normal growth pattern compared to growth with pharmacological modulation. To overcome this, it is customary to only study short-term effects or to expose the cells to trypsinization and reseeding, a process that can induce further changes to cellular function.
For radiotracer uptake studies, we have since long been working with multicellular aggregates, grown on agar as a method to have easy access to cells. This method has the additional advantage that it allows easy washing of aggregates followed by a measurement of radiotracer uptake, after which the aggregates can be returned to culture for subsequent evaluations if the radiotracer is a PET tracer with short half-life.
A prerequisite to properly evaluate cell physiology with PET tracers is that the uptake value is related to the amount of living cells, especially in evaluations of treatment effects where changes in the relative proportion of viable cells can be expected. We have therefore developed a routine including quantization of PET tracer uptake that relies on accurately measuring viable cell volume.
The introduced method is effective, user-friendly, and time efficient, and is more accurate than the traditional method. It is known that a tumor cell-line can behave differently under different conditions. One of the advantages of this method is that a growth curve can be generated simply for each cell-line in each laboratory and for each experimental set-up.
Combining SASDM with biological evaluation of the MTS provides a good model to evaluate the effect of various cancer treatments e.g. chemotherapy and radiotherapy.
Conclusion
To conclude, the combination of PET radiotracers and image analysis in MTSs provides a good model to evaluate the relationship between viable volume of tumor and the uptake of metabolic tracer before and after chemotherapy. This feature could be used for a selection of PET biomarker for an early assessment of treatment response.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
Authors AM, PR and MB helped with the design of the study. They created the method for applying SASDM, performed the image and data analysis and drafted the manuscript.
Authors RJ, ÖL and BL helped with some of the practical approaches and the writing of the paper.
Acknowledgements
The authors wish to thank the staff at Uppsala Imanet AB and staff of the chemistry department for the radionuclide production.
==== Refs
Kunz-Schughart LA Kreutz M Knuechel R Multicellular spheroids: a three-dimensional in vitro culture system to study tumor biology Int J Exp Pathol 1998 79 1 23 9614346 10.1046/j.1365-2613.1998.00051.x
Kunz-Schughart LA Santini MT Rainaldi G Hamilton G Mueller-Klieser W Durand RE Multicellular tumor spheroids: intermediates between monolayer culture and in vivo tumor Cell Biol Int 1999 23 157 161 10562436 10.1006/cbir.1999.0384
Santini MT Rainaldi G Three-dimensional spheroid model in tumor biology Pathobiology 1999 67 148 157 10394136 10.1159/000028065
Gati I Bergström M Muhr C Långström B Carlsson J Application of (methyl-11C)-methionine in the multicellular spheroid system J Nucl Med 1991 32 2258 2265 1744712
Mueller-Klieser W Multicellular spheroids. A review on cellular aggregates in cancer research J Cancer Res Clin Oncol 1987 113 101 122 3549738 10.1007/BF00391431
Gati I Bergström M Westerberg G Csoka K Muhr C Carlsson J Effects of prostaglandin and leukotriene inhibitors on the growth of human glioma spheroids Eur J Cancer 1990 26 802 807 2145898
Bergström M Awad R Estrada S Mälman J Lu L Lendvai G Bergström-Pettermann E Langstrom B Autoradiography with positron emitting isotopes in positron emission tomography tracer discovery Mol Imaging Biol 2003 5 390 396 14667493 10.1016/j.mibio.2003.09.004
|
16283948
|
PMC1315357
|
CC BY
|
2021-01-04 16:40:10
|
no
|
Cancer Cell Int. 2005 Nov 14; 5:32
|
utf-8
|
Cancer Cell Int
| 2,005 |
10.1186/1475-2867-5-32
|
oa_comm
|
==== Front
Nutr JNutrition Journal1475-2891BioMed Central London 1475-2891-4-351630768410.1186/1475-2891-4-35ResearchCopper Chaperone for Cu/Zn Superoxide Dismutase is a sensitive biomarker of mild copper deficiency induced by moderately high intakes of zinc Iskandar Monica [email protected] Eleonora [email protected] Keith D [email protected] Bingtuan [email protected]'Abbé Mary R 1mary_l'[email protected] Jesse [email protected] Nutrition Research Division, Food Directorate, Health Products and Food Branch, Health Canada, 2203C Banting Research Centre, Ottawa, ON, K1A 0L2, Canada2005 24 11 2005 4 35 35 23 9 2005 24 11 2005 Copyright © 2005 Iskandar et al; licensee BioMed Central Ltd.2005Iskandar 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
Small increases in zinc (Zn) consumption above recommended amounts have been shown to reduce copper (Cu) status in experimental animals and humans. Recently, we have reported that copper chaperone for Cu/Zn superoxide dismutase (CCS) protein level is increased in tissues of overtly Cu-deficient rats and proposed CCS as a novel biomarker of Cu status.
Methods
Weanling male Wistar rats were fed one of four diets normal in Cu and containing normal (30 mg Zn/kg diet) or moderately high (60, 120 or 240 mg Zn/kg diet) amounts of Zn for 5 weeks. To begin to examine the clinical relevance of CCS, we compared the sensitivity of CCS to mild Cu deficiency, induced by moderately high intakes of Zn, with conventional indices of Cu status.
Results
Liver and erythrocyte CCS expression was significantly (P < 0.05) increased in rats fed the Zn-60 and/or Zn-120 diet compared to rats fed normal levels of Zn (Zn-30). Erythrocyte CCS expression was the most sensitive measure of reduced Cu status and was able to detect a decrease in Cu nutriture in rats fed only twice the recommended amount of Zn. Liver, erythrocyte and white blood cell CCS expression showed a significant (P < 0.05) inverse correlation with plasma and liver Cu concentrations and caeruloplasmin activity. Unexpectedly, rats fed the highest level of Zn (Zn-240) showed overall better Cu status than rats fed a lower level of elevated Zn (Zn-120). Improved Cu status in these rats correlated with increased duodenal mRNA expression of several Zn-trafficking proteins (i.e. MT-1, ZnT-1, ZnT-2 and ZnT-4).
Conclusion
Collectively, these data show that CCS is a sensitive measure of Zn-induced mild Cu deficiency and demonstrate a dose-dependent biphasic response for reduced Cu status by moderately high intakes of Zn.
==== Body
Background
Zinc (Zn) and copper (Cu) play vital roles as structural and catalytic components of metalloenzymes and are essential nutrients required for growth and development [1-3]. It is well recognized that consuming large quantities of Zn for extended periods of time causes severe Cu deficiency and can lead to the development of anaemia and other abnormalities [4-8]. Of potentially greater significance, however, are studies showing that even small increases in Zn consumption above recommended amounts depress Cu status in experimental animals [9,10] and humans [11,12]. This strong antagonism of excess Zn to Cu status was the basis for setting the Tolerable Upper Intake Levels (ULs) for Zn [1]. At present, the mechanism by which Zn impedes Cu absorption is not known.
In humans, overt Cu deficiency is uncommon and usually only seen in specific situations [13]. However, mild Cu deficiency from consuming diets inadequate in Cu [14,15] or high in Zn [16] may be of concern. Notably, studies have shown that a large proportion of young children have Zn intakes exceeding the ULs and much of the Zn consumed by these children is from Zn-fortified foods [16-18]. Although overt Zn toxicity has not been reported in these children, excessive Zn intake may cause small but potentially harmful reductions in Cu balance, stressing the need for sensitive biomarkers able to detect mild Cu deficiency.
Insertion of Cu into Cu/Zn superoxide dismutase (SOD1) requires the copper chaperone for SOD1 (CCS). We have previously shown that CCS protein is up-regulated in liver and erythrocytes of Cu-deficient rats [19] and Cu regulates the degradation of CCS by the 26 S proteosome [20]. Another group has since reported a similar increase in CCS in tissues of Cu-deficient rats and mice [21,22]. CCS is a promising biomarker of reduced Cu status, as CCS protein level is more responsive to Cu deficiency than reduction in SOD1 activity [19], the endpoint used to set the ULs for Zn [1]. Further, CCS is particularly appealing given that commonly used measures of Cu status are affected by various common conditions unrelated to Cu nutriture [14,15,23-25] and thought to be insensitive to marginal deficiency [14,15,26]. The latter is underscored by experiments in rats showing that diets marginally low in Cu induce abnormalities in heart morphology and function, but minimal changes in conventional indices of Cu status [27,28].
Several reports have established increased CCS protein in tissues of overtly Cu-deficient rats and mice [19-22], however, it is not known whether CCS is responsive to small reductions in Cu status. The objectives of this study were two-fold; (1) to examine the effects of graded levels of moderately high dietary Zn on Cu status and (2) to evaluate the ability of CCS to detect small reductions in Cu nutriture, induced by excess Zn, as a first step in determining the usefulness of CCS as a biomarker that can be used in a clinical setting.
Methods
Animals and Test Diets
Weanling (21-day-old) male Wistar rats (n = 12/diet group) (Charles River Canada, St. Constant, Canada) had free access to one of 5 test diets [29] and demineralised drinking water. Food consumption and body weight were measured weekly. After 5 weeks of consuming the diets, rats were killed by exsanguination while anesthetised with 3% isoflurane. Blood was withdrawn from the abdominal aorta. The intestine was extracted and intestinal contents removed by washing with isotonic saline. Extracted intestine, liver and kidneys were frozen until analysis. The Health Products and Food Branch Animal Care Committee of Health Canada approved the experimental protocol. Rats were treated in accordance with the guidelines of the Canadian Council on Animal Care.
Test diets were prepared by adding appropriate amounts of Cu (cupric carbonate) and Zn (zinc carbonate) from cornstarch premixes. Other than differences in Cu and Zn content, composition of the diets were similar to those described previously [19]. Zn and Cu content in samples of each test diet were determined by flame atomic absorption spectrophotometry (AAS) (Perkin-Elmer 5100 PC; Perkin Elmer Cetus Instruments, Norwalk, CT) as described [30].
Blood Fractionation
Blood samples were collected in EDTA tubes and separated into its components by centrifugation. Plasma was frozen in aliquots. White blood cells (WBCs) were carefully removed (trying to avoid contamination with erythrocytes), washed with isotonic saline and centrifuged. The procedure was repeated 3–4 times until a WBC preparation with no visible contamination with erythrocytes was obtained. Erythrocytes were washed 3 times with saline prior to freezing.
Haematological Measurements
Blood samples were collected in vacutainer K3EDTA tubes and shipped to VITA-TECH (Markham, Canada) for analysis of haematological parameters (See additional file 1: Table 1).
Table 1 Total food consumption, body weight and Zn content in tissues of rats fed diets differing in Zn and Cu1,2
Diet Group Test Diets
(mg/kg diet) Total Food
Consumption (g) Body Weight
week 5 (g) Liver Zn
(μg/g dry weight) Kidney Zn
(μg/g dry weight) Mucosal Zn
(μg/g dry weight)
Zn Cu
Zn-30 41.27 ± 0.52a 5.70 ± 0.04a 881.0 ± 20.8a 345.1 ± 6.2a 91.99 ± 1.99a 104.82 ± 1.80a 97.12 ± 2.14a
Zn-60 65.09 ± 3.42b 5.57 ± 0.14a 876.9 ± 28.6a 346.3 ± 7.3a 94.08 ± 1.20a 102.42 ± 1.36a 98.94 ± 1.31a
Zn-120 129.18 ± 0.71c 5.77 ± 0.08a 852.8 ± 26.6a 343.4 ± 6.4a 104.40 ± 1.91b 105.95 ± 1.05a 115.28 ± 2.80b
Zn-240 242.22 ± 0.80d 5.78 ± 0.03a 908.6 ± 28.7a 364.9 ± 9.0b 109.33 ± 2.25b 112.97 ± 2.19b 146.04 ± 6.77c
Cu-D 48.48 ± 2.54e 1.07 ± 0.02b 891.2 ± 28.3a 339.0 ± 8.5a 94.69 ± 1.53a 102.07 ± 0.77a 94.93 ± 1.38a
1 Values in a column without a common letter differ, P < 0.05.
2 Values are means ± SEM, n = 12/diet group.
Western Blotting
Liver protein extracts were prepared by homogenizing in ice-cold 0.5% (v/v) Triton-X-100 buffer containing a protease inhibitor cocktail (Roche, Laval, Canada). WBCs and erythrocytes were lysed in Triton-X-100 buffer or GSH reagent (5 mmol/L KH2PO4/K2HPO4, 2 mmol/L glutathione, pH 7.0), respectively. Extracts (40 μg total protein or haemoglobin) were separated over 8–16% Tris-Glycine gradient gels (Invitrogen, Burlington, Canada) under denaturing and reducing conditions. Gels for each tissue were simultaneously electroblotted onto a single nitrocellulose membrane. The membrane was blocked for 1 h at room temperature (RT) in TBS-Tween [20 mmol/L Tris, 500 mmol/L NaCl, 0.1% Tween 20 (v/v), pH 7.5] supplemented with 5% (wt/v) nonfat dry milk (BioRad, Hercules, CA). Membranes were probed with a CCS antibody (FL-274; Santa Cruz Biotechnology, Santa Cruz, CA) at a final concentration of 0.6 mg/L (for liver and erythrocytes) or 0.2 mg/L (for WBCs) overnight at 4°C. After washing with TBS-Tween, membranes were incubated with an anti-rabbit (0.16 mg/L) HRP-conjugated secondary antibody (BioRad) in blocking solution for 2 h at RT. Antibody-bound proteins were detected by enhanced chemiluminescence and exposure to film. Membranes were stripped with stripping buffer (100 mmol/L 2-mercaptoethanol, 2% (wt/v) SDS, 62.5 mmol/L Tris-HCl, pH 6.8) and re-probed with an antibody against Actin [(I-19)-R, Santa Cruz Biotechnology] or GAPDH (MCA-1D4, Encor Biotech, Alachua, FL) at 0.5 mg/L concentration or at a 1:1000 dilution, respectively. Film was scanned and band intensity determined using Scion Image software (Scion Corporation, Frederick, MD). Band intensities were determined at exposures within the linear response range of the film.
Mineral Analyses in Tissues
Cu and Zn content in liver and kidney samples were determined by flame AAS as described [30]. Plasma Cu concentration was measured by graphite furnace AAS using a SIMAA 6000 (Perkin-Elmer Cetus Instruments) with Zeeman background correction. To determine Zn and Cu levels in the intestinal mucosa, mucosal cells were gently scraped with a glass cover slide from a 10 cm intestinal segment starting 10 cm caudal to the pyloric sphincter. Mucosal scrapings were dried, dissolved in concentrated HNO3 and microwave digested using a CEM (Mars5) microwave (CEM Corporation, Matthews, NC). Zn and Cu in digested samples were determined by flame AAS and graphite furnace AAS, respectively.
Quantitative PCR
A segment (1 cm in length; starting 8 cm caudal to the pyloric sphincter) of frozen intestine was excised and total RNA isolated using QIAGEN's RNeasy mini kit (QIAGEN, Mississauga, Canada). All RNA samples were treated with RNase-free DNase (QIAGEN). cDNA from each sample was generated using an oligo (dT) primer (Ambion, Austin, TX). Sequences for genes (MT-1, ZnT-1, ZnT-2 and ZnT-4) were obtained from GenBank (See additional file 2: Table 2 QPCR primers). To obtain the rat Zip4 sequence, primers specific for the mouse Zip4 gene (Forward: 5'-ACT GGA CGG CCT GTT AAA TAC GCT-3'; Reverse: 5'-TAC TCC GAC TGC TAG AGC CAC GTA-3') were used to amplify the rat Zip4 cDNA from total RNA. The sequence of the PCR product was aligned against the rat genome and the mouse Zip4 gene to confirm amplification of the rat Zip4 cDNA. All primers used for QPCR (See additional file 2: Table 2 QPCR primers) were designed using PrimerQuest (Integrated DNA Technologies, Skokie, IL). QPCR was performed using an Mx4000 Multiplex Quantitative PCR System (Stratagene, La Jolla, CA). Reactions were performed in triplicate using the Brilliant SYBR Green QPCR Core Reagent kit (Stratagene). To ensure amplification of a single homogeneous product, post-amplification dissociation curves were performed. All primer sets produced a single product of expected size (See additional file 3: Figure 1 PCR fragments of genes analysed by QPCR). Changes in gene expression were determined using the standard curve method with β-Actin as the normalizing gene.
Table 2 Tissue Cu concentrations and SOD1 activity 1,2
Diet Group Liver Cu
(μg/g dry weight) Kidney Cu
(μg/g dry weight) Mucosal Cu
(μg/g dry weight) Liver SOD1 Activity
U/mg protein Erythrocyte SOD1 Activity
U/mg Hb
Zn-30 17.50 ± 0.87a 23.08 ± 1.07a 9.66 ± 0.25ab 27.39 ± 1.90a 53.21 ± 2.21a
Zn-60 16.08 ± 0.57ab 21.28 ± 1.04ab 9.21 ± 0.26ab 26.91 ± 1.15a 52.61 ± 1.97a
Zn-120 14.70 ± 1.07b 20.57 ± 0.85b 8.64 ± 0.51a 24.31 ± 1.84a 48.58 ± 1.45a
Zn-240 16.38 ± 0.65ab 21.74 ± 0.82ab 9.78 ± 0.36b 25.25 ± 1.44a 49.06 ± 1.61a
Cu-D 9.45 ± 0.82c 15.84 ± 0.32c 5.50 ± 0.49c 19.40 ± 1.76b 41.81 ± 1.55b
1 Values in a column without a common letter differ, P < 0.05.
2 Values are means ± SEM, n = 12/diet group.
Figure 1 Scatter plots of (A) plasma Cu concentration and (B) Cp activity of rats fed diets differing in Zn and Cu. Each solid circle corresponds to one rat (n = 12/diet group). The horizontal line across all diet groups represents the non-response mean of Zn-30 rats and that of the non-responders from other diet groups (see Methods). The dashed line within each diet group signifies the response mean for that diet group. The non-response mean (diet group Zn-30) and the response means (diet groups Zn-60, Zn-120, Zn-240 and Cu-D) were compared. Diet groups without a common letter differ (P < 0.05).
Enzyme Assays
Liver and erythrocyte SOD1 activity was measured by the cytochrome c reduction assay [31] modified for analysis with a microplate reader. Plasma caeruloplasmin (Cp) activity was measured as described [32].
Statistical Analyses
Data were analysed by one-way ANOVA and differences between means were determined by Fisher's least significant difference test. Data are reported as mean ± SEM. Linear regression analyses were performed to examine the association between conventional indices of Cu status and tissue CCS expression. Pearson's correlation coefficient (r) was calculated to measure the goodness of fit of the data. To test for differences in the response means and the non-response mean for plasma Cu and Cp activity (Figure 1), data were analysed using an ANOVA model with diet group as the main effect. The mean levels were compared using Tukey's Studentized Range Test. Because of the partial response behaviour, these data were fitted using maximum likelihood methods with the diet groups fitted with the same mean as the Zn-30 group for the non-responders and separate means for the responders, each with a common variance and proportion of responders. The response means and the non-response mean were compared using the log-likelihood difference and grouped when not significantly different. Because the maximum likelihood approach does not classify individual observations, a cut-off value of 2 standard deviations from the mean of Zn-30 rats was used to characterise individual rats as responders and non-responders (Table 3). Fisher exact test was used to determine differences in the number of responders between diet groups. Statistical significance was set at P < 0.05. Data were analysed using Statistica 7 software (StatSoft, Tulsa, OK).
Table 3 Effect of dietary Zn and Cu on plasma Cu concentration and Cp activity of rats
Animals Plasma Cu Concentration1,2 Cp Activity3,4
Zn-30 Zn-60 Zn-120 Zn-240 Cu-D Zn-30 Zn-60 Zn-120 Zn-240 Cu-D
No. of non-responders 12 10 5 9 0 11 10 5 9 0
No. of responders5 0a 2ab 7b 3ab 12c 1a 2ab 7b 3ab 12c
Total 12 12 12 12 12 12 12 12 12 12
1 Non-responders had a range of 667.6 – 1291.6 μg/L.
2 Responders (range <0.1 – 572.4 μg/L) were those animals having plasma Cu concentration lower than 2SD below the mean plasma copper concentration of animals in the Zn-30 diet group (Zn-30; mean ± SD = 975.0 ± 159.4 μg/L).
3 Non-responders had a range of 65.3 – 152.8 U/L.
4 Responders (range <0.1 – 57.8 U/L) were those animals having Cp activity lower than 2SD below the mean Cp activity of animals in the Zn-30 diet group (Zn-30; mean ± SD = 106.5 ± 23.9).
5 No. of responders were compared using the Fisher exact test. For each diet group, No. of responders without a common letter differ, P < 0.05.
Results
Test diets were prepared by adding 30, 60, 120 and 240 mg Zn/kg diet, which corresponds to normal amounts of Zn or 2, 4 or 8 times the AIN recommended amount [33], respectively. An additional group of rats was fed a Cu-deficient diet containing normal amounts of Zn and served as a positive control for Cu deficiency. Precise Zn and Cu content of each test diet is shown in Table 1. All test diets contained higher amounts of Zn than what would be expected from amounts added to the diet preparations. Analysis of individual dietary components revealed that casein was the major contributor to the additional Zn in the final diet preparations (data not shown).
Total food consumption of rats from each diet group was similar (Table 1). Rats fed the Zn-120 and Zn-240 diet accumulated larger (P < 0.05) amounts of Zn in the liver compared to control rats fed normal amounts of Zn (Zn-30) (Table 1). Only rats fed the largest amount of Zn (Zn-240) showed a significant (P < 0.05) increase in kidney Zn content. Zn in the intestinal mucosa of rats fed the Zn-120 or the Zn-240 diet was increased (P < 0.05). Zn content in the liver, kidney and intestinal mucosa of rats fed the Cu-deficient diet (Cu-D) was similar to that of rats fed adequate Cu (Zn-30), indicating that the Cu-deficient diet did not affect Zn accumulation in these tissues.
Rats consuming the diet containing the largest amount of Zn (Zn-240) showed a significant (P < 0.05) increase in body weight at week 4 (data not shown) and 5 (Table 1) of the study compared to rats fed normal Zn. Conversely, body weight of rats fed the Zn-60, Zn-120 or Cu-D diet did not differ from that of Zn-30 rats (Table 1). Haemoglobin (Hb) levels and other haematological parameters of rats fed elevated Zn or the Cu-D diet were similar to those of rats fed the Zn-30 diet (See additional file 1: Table 1).
Rats fed 4 times the normal level of Zn (Zn-120) showed a small decrease (P < 0.05) in liver Cu compared to rats fed normal Zn (Table 2). Surprisingly, liver Cu did not decrease further in rats fed the highest amount of Zn (Zn-240). In fact, Cu levels did not differ from those of Zn-30 rats. A similar trend was observed for kidney Cu content. Diets high in Zn did not alter Cu content in the intestinal mucosa when compared to rats fed normal Zn. However, Cu content in the intestinal mucosa was higher (P < 0.05) in Zn-240 compared to Zn-120 rats. Cu levels in the liver, kidney and intestinal mucosa of rats fed the Cu-D diet were markedly decreased (P < 0.05) compared to rats fed the control diet. Liver and erythrocyte SOD1 activity was unaffected (P > 0.05) in rats fed high Zn (Table 2). In contrast, rats fed the Cu-D diet showed a significant (P < 0.05) decrease in SOD1 activity in liver and erythrocytes.
Scatter plots of plasma Cu levels clearly showed that while some rats responded to increased dietary Zn with a reduction in plasma Cu other rats had normal levels (Figure 1A), indicating an all or none response. The number of rats that responded increased from 2 rats when fed the Zn-60 diet to 7 rats when fed the Zn-120 diet (Table 3). Consistent with liver, kidney and intestinal mucosa Cu content indicating improved Cu status in rats fed the Zn-240 diet compared to rats fed the Zn-120 diet, only 3 rats fed the Zn-240 diet were characterised as responders. The response mean for plasma Cu was also higher in Zn-240 compared to Zn-120 rats (Figure 1A). Plasma Cu of Cu-D rats was markedly decreased, consistent with data from a previous study from our laboratory [10]. As expected, given that most plasma Cu is associated with Cp, Cp activity paralleled very closely plasma Cu levels and showed a similar all or none response behaviour (Figure 1B, Table 3). A strong positive correlation (r = 0.984) was found between plasma Cu and Cp activity (data not shown).
Liver CCS was increased (P < 0.05) >1.7-fold in rats fed the Zn-120 diet compared to rats fed normal Zn (Figure 2A). CCS expression was lower (P < 0.05) in Zn-240 rats compared to Zn-120 rats. Erythrocyte CCS was increased (P < 0.05) >1.5-fold in Zn-60 and Zn-120 rats (Figure 2B). The Cu-D diet induced a larger increase in CCS content in liver and erythrocytes (>2.5-fold). WBC CCS expression was not significantly (P > 0.05) increased in rats fed elevated Zn, but was increased (P < 0.05) >3-fold in Cu-D rats (Figure 2C). Rats fed high Zn and characterised as responders for plasma Cu had higher (P < 0.05) liver and WBC CCS expression than non-responders or Zn-30 rats (Table 4). Erythrocyte CCS expression of responders was higher (P < 0.05) than that of Zn-30 rats, but not significantly (P > 0.05) different from non-responders.
Figure 2 (A) Liver, (B) erythrocyte and (C) WBC CCS content of rats fed diets differing in Zn and Cu. CCS expression in liver is expressed relative to that of β-Actin. CCS expression in erythrocytes and WBCs is expressed relative to that of GAPDH. Bars signify the mean ± SEM, n = 12 (for liver and erythrocytes) or 7 (for WBCs)/diet group. Diet groups without a common letter differ (P < 0.05).
Table 4 Comparison of tissue CCS expression between Zn-30 rats and responders and non-responders for plasma Cu1,2
Animals CCS Expression
Liver Erythrocyte WBC
Zn-30 1.00 ± 0.14a
(n = 12) 1.00 ± 0.18a
(n = 12) 1.00 ± 0.20a
(n = 7)
Non-responders3 1.18 ± 0.11a
(n = 24) 1.44 ± 0.16ab
(n = 24) 0.95 ± 0.11a
(n = 17)
Responders4 1.97 ± 0.12b
(n = 12) 1.86 ± 0.20b
(n = 12) 2.02 ± 0.54b
(n = 4)
1 Values in a column without a common letter differ, P < 0.05. Values are means ± SEM.
2 n values are in parentheses.
3 Rats from diet groups Zn-60, Zn-120 and Zn-240 characterised as non-responders for plasma Cu.
4 Rats from diet groups Zn-60, Zn-120 and Zn-240 characterised as responders for plasma Cu.
Pearson linear correlations revealed a strong inverse association between liver CCS and liver (r = -0.746) and plasma (r = -0.816) Cu and Cp activity (r = -0.787) (Table 5). A significant (P < 0.001) association was also observed when rats from the Cu-D diet group were omitted from the analyses. Liver CCS only showed a moderate inverse correlation with liver and erythrocyte SOD1 activity when all rats were used in the analyses. A significant correlation was not found (P > 0.05) when Cu-D rats were omitted from the analyses, consistent with the lack of a significant decrease in SOD1 activity in rats fed elevated Zn (Table 2). Erythrocyte CCS expression was also significantly (P < 0.001) correlated with liver and plasma Cu and Cp activity (Table 5). WBC CCS expression showed a strong inverse correlation with liver (r = -0.768) and plasma (r = -0.813) Cu and Cp activity (r = -0.769).
Table 5 Correlation between CCS content in tissues and liver and plasma Cu concentration, Cp activity and liver and erythrocyte SOD1 activity1,2
Tissue CCS Diet Groups3, n = 60 Diet Groups4, n = 48
Liver Cu Plasma Cu Cp Liver SOD1 Erythrocyte SOD1 Liver Cu Plasma Cu Cp Liver SOD1 Erythrocyte SOD1
Liver -0.746
(<0.001) -0.816
(<0.001) -0.787
(<0.001) -0.418
(<0.005) -0.464
(<0.001) -0.597
(<0.001) -0.649
(<0.001) -0.615
(<0.001) -0.222
(NSD) -0.221
(NSD)
Erythrocyte -0.633
(<0.001) -0.618
(<0.001) -0.590
(<0.001) -0.268
(<0.05) -0.332
(<0.01) -0.425
(<0.005) -0.308
(<0.05) -0.297
(<0.05) -0.001
(NSD) 0.001
(NSD)
WBC -0.7685
(<0.001) -0.8135
(<0.001) -0.7695
(<0.001) -0.4955
(<0.005) -0.5825
(<0.001) -0.4746
(<0.05) -0.5596
(<0.005) -0.5436
(<0.005) -0.0696
(NSD) -0.4296
(<0.05)
1 Pearson's correlation coefficients.
2 P values are in parentheses; NSD = no statistical difference.
3 Diet groups Zn-30, Zn-60, Zn-120, Zn-240, and Cu-D.
4 Diet groups Zn-30, Zn-60, Zn-120, and Zn-240.
5 n = 35.
6 n = 28.
Given that rats fed the highest amount of Zn (Zn-240) showed better Cu status than rats fed a lower level of excess Zn (Zn-120), it prompted us to evaluate changes in expression of duodenal Zn-trafficking proteins in response to these levels of Zn. Messenger RNA for MT-1 and Zn transporters ZnT-1, ZnT-2 and ZnT-4 was significantly (P < 0.05) increased only in Zn-240 rats when compared to Zn-30 rats (Figure 3). MT-1 (~5-fold) and ZnT-2 (~3-fold) showed the largest increases followed by ZnT-1 and ZnT-4. Zip4 expression was unchanged in rats fed elevated Zn compared to rats fed normal Zn. Collectively, these data indicate that mRNA expression of MT-1 and several ZnT transporters in the duodenum of rats is refractory to small increases in dietary Zn, but responds to a higher level of Zn intake.
Figure 3 QPCR analysis of duodenal mRNA expression of Zn-trafficking proteins in rats fed diets differing in Zn and Cu. MT-1, ZnT-1, ZnT-2, ZnT-4 and Zip4 mRNA content is expressed relative to β-Actin expression. Bars represent the mean ± SEM, n = 4/diet group. For each gene, diet groups without a common letter differ (P < 0.05).
Discussion
This is the first study to demonstrate that CCS protein expression responds to mild Cu deficiency induced by moderately high dietary Zn, underscoring the potential usefulness of CCS as a biomarker in a clinical setting. In contrast to rats and mice used in previous studies [19,21,22], rats used in this study had normal haematology and did not show any indications of the development of anaemia, indicating that these rats were not overtly Cu deficient. Further, rats did not show impaired growth, a phenotype associated with severe Cu deficiency in growing rats [19].
Erythrocyte CCS expression was determined to be the most sensitive index of reduced Cu nutriture and was able to detect a decrease in Cu status in rats fed only twice the AIN recommended amount of Zn [33]. This sensitivity may also partly account for the absence of a significant difference in erythrocyte CCS expression between responders and non-responders for plasma Cu, as CCS may have increased in some rats fed high Zn but characterise as non-responders. Notably, erythrocyte CCS expression in non-responders was higher than in Zn-30 rats and this difference was approaching statistical significance (P = 0.09). Liver CCS also responded to elevated dietary Zn and was increased in rats fed 4 times the recommended amount. These are important findings, as these same rats did not show a significant decrease in SOD1 activity, the biomarker of reduced Cu status used to set the ULs for Zn.
Rats fed elevated Zn showed an all or none response for decreased plasma Cu and Cp activity, indicating that these parameters are not ideal for diagnosing small reductions in Cu status and providing an accurate assessment of precise Cu nutriture. These data are consistent with a previous study showing a similar all or none response for Cp activity in rats fed elevated levels of Zn [9]. This response likely reflects the presence of strong homeostatic mechanisms that maintain plasma Cu levels in the normal range until liver stores have been appreciably depleted.
Although a gold standard for assessing reductions in Cu status is lacking, decreased liver Cu content is regarded as one of the most accurate and sensitive measures in experimental animals, while plasma Cu and Cp activity are the most common indicators used in clinical situations. The strong inverse correlation of liver CCS expression with these markers indicates that increased CCS expression was specific for reduced Cu status and is a good predictor of changes in common measures of Cu deficiency. Erythrocyte CCS showed a weaker association with these indicators that may be explained, in part, by the slow turnover of erythrocytes, which may not accurately reflect the current Cu status of the rats. This is in contrast to plasma and liver Cu levels and Cp activity that likely respond more rapidly to changes in Cu availability.
During our experiments we discovered that CCS protein in WBCs is more susceptible to degradation than CCS in erythrocytes, requiring that WBCs be isolated quickly from blood samples. Samples showing degradation of CCS were excluded from our analyses. Of the 7 rats per diet group analysed, we noticed that the majority of rats fed high Zn were characterised as non-responders for decreased plasma Cu and Cp activity. This over sampling of rats showing normal plasma Cu and Cp activity may account for the absence of a significant increase in WBC CCS in rats fed elevated Zn. Nonetheless, WBC CCS expression was increased in rats fed the Cu-deficient diet and was highly correlated with conventional measures of Cu status. Further, rats characterised as responders for plasma Cu (which represents rats with poorer Cu status) had higher WBC CCS levels than non-responders or Zn-30 rats, indicating that WBC CCS responds to Cu deficiency induced by excess zinc. Because of the slow turnover of erythrocytes, WBC CCS may have value as an indicator of early reductions in Cu status. Interestingly, we found that the basal expression level of CCS in WBCs is much higher than in other tissues such as liver, heart and erythrocytes in rats (data not shown).
A major finding of this paper is that rats fed 8 times the recommended amount of Zn (Zn-240) showed overall better Cu status than rats fed 4 times the recommended amount (Zn-120). This was consistently observed with several biomarkers. Remarkably, Cu status indicators of most rats fed the Zn-240 diet were indistinguishable from rats fed a diet normal in Zn. Importantly, liver CCS was significantly lower in Zn-240 compared to Zn-120 rats, further demonstrating the exquisite sensitivity of CCS to changes in Cu status. At this point, we cannot offer a definite explanation as to why other studies from our laboratory [9,34] failed to detect this biphasic response, other than to speculate that differences in the bioavailability of the Zn source, other dietary components or duration of the studies may account for this discrepancy.
Presently, the mechanism by which Zn interferes with Cu absorption is unknown, although it is believed to occur at the site of the intestinal enterocyte. It was thought that increased metallothionein (MT) levels sequestered Cu in the enterocyte leading to Cu deficiency [35,36]. However, MT-null mice fed elevated Zn become Cu deficient indicating that MT induction is not the primary cause of the Cu deficiency [37]. Results presented here are consistent with these data, as rats fed the Zn-60 and Zn-120 diet showed reduced Cu status in the absence of increased duodenal MT-1 transcript. Because MT can bind Cu, however, it would be interesting to determine whether higher MT-1 expression in Zn-240 rats played any role in increasing Cu in the intestinal mucosa and improving Cu status of these rats compared to Zn-120 rats.
Experiments with Caco-2 cells have indicated that elevated levels of Zn affect Cu transport and reduce Cu efflux at the basolateral side [38], suggesting that Zn may block absorption of Cu by affecting the activity of a Cu transporter. Thus, improved Cu status of Zn-240 compared to Zn-120 rats may reflect a secondary effect of the higher Zn diet on the activity of a Cu transporter. It is known that distinct pools of Zn exist within cells and changes in expression of Zn-trafficking proteins can alter the intracellular distribution of Zn [39]. For example, increased expression of ZnT-2 likely promotes the accumulation of Zn within vesicles, decreasing cytoplasmic Zn levels [40]. In addition, increased MT levels result in more Zn bound to MT. Given that transcripts of several Zn-trafficking proteins were increased in the duodenum of rats fed the Zn-240 but not the Zn-120 diet, it is possible that increased expression of one or more of these Zn-trafficking proteins altered the distribution of Zn within absorptive enterocytes of Zn-240 rats in a manner that alleviated the block in activity of a Cu transporter.
Lastly, we report an interesting observation that rats fed the highest level of Zn had increased body weight. Although it is well recognized that Zn deficiency can result in impaired growth [41], it is not established that consuming larger quantities of Zn, above recommend amounts, can enhance growth. Given that lower doses of Zn above the AIN recommended amount clearly depressed Cu status of rats, the benefit of consuming this higher level of Zn is presently unclear.
Conclusion
In this study we report two major discoveries. Firstly, we show that CCS is a sensitive biomarker of Zn-induced mild Cu deficiency in rats. Given this finding, further studies evaluating CCS as a biomarker in humans are necessary, as CCS may prove to be a better measure of reduced Cu status than conventional indicators and impact studies aimed at setting more accurate Dietary Reference Intakes for Zn and Cu. Secondly, we demonstrate a dose-dependent biphasic response for the reduction of Cu status by moderately high intakes of Zn. Although it is not known whether humans respond to excess Zn in a similar manner to rats, this finding offers a caution when interpreting data from studies reporting no reduction in Cu status that have used a single dose of supplemental Zn and emphasizes the importance of using a dose-response approach.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
MI performed the QPCR and statistical analyses. MI and ES carried out the tissue mineral analyses and the determination of tissue CCS expression. ES performed the enzyme assays. KT prepared the test diets and participated in the animal phase of the study. BW isolated the blood components. ML participated in planning the study and was involved in the interpretation of the results. JB wrote the manuscript and was involved in planning and coordinating the study and interpreting the data. All authors read and approved the final manuscript.
Supplementary Material
Additional file 1
White blood cells (WBC), erythrocytes (ERCS), haemoglobin (Hb), haematocrit (HCT), mean corpuscular volume (MCV), mean corpuscular haemoglobin (MCH), mean corpuscular haemoglobin concentration (MCHC), reticulocytes (RTC), platelet count (PLT) and red cell distribution width (RDW) of weanling male Wistar rats fed diets with differing amounts of Zn and Cu for 5 weeks (Table 1).
Click here for file
Additional file 2
QPCR primers (Table 2).
Click here for file
Additional file 3
PCR fragments of genes analysed by QPCR (Figure 1).
Click here for file
Acknowledgements
We thank the Animal Resource Division of Health Canada for assistance in care of the rats. We would also like to acknowledge Stephen Hayward for assistance with the statistical analyses. Bingtuan Wang is a National Institute of Nutrition (NIN) Post-doctoral Fellow. This study was funded by the Bureau of Nutritional Sciences, Health Canada. This is publication no. 607 of the Bureau of Nutritional Sciences.
==== Refs
Institute of Medicine Dietary reference intakes for vitamin A, vitamin K, arsenic, boron, chromium, copper, iodine, iron, manganese, molybdenum, nickel, silicon, vanadium, and zinc Washington, DC 2001
Schumann K Classen HG Dieter HH Konig J Multhaup G Rukgauer M Summer KH Bernhardt J Biesalski HK Hohenheim consensus workshop: copper Eur J Clin Nutr 2002 56 469 83 12032645 10.1038/sj.ejcn.1601315
McCall KA Huang C Fierke CA Function and mechanism of zinc metalloenzymes J Nutr 2000 130 1437S 46S 10801957
Botash AS Nasca J Dubowy R Weinberger HL Oliphant M Zinc-induced copper deficiency in an infant Am J Dis Child 1992 146 709 11 1595625
Porea TJ Belmont JW Mahoney DH Jr Zinc-induced anemia and neutropenia in an adolescent J Pediatr 2000 136 688 90 10802505 10.1067/mpd.2000.103355
Broun ER Greist A Tricot G Hoffman R Excessive zinc ingestion. A reversible cause of sideroblastic anemia and bone marrow depression JAMA 1990 264 1441 3 2094240 10.1001/jama.264.11.1441
Hein MS Copper deficiency anemia and nephrosis in zinc-toxicity: a case report S D J Med 2003 56 143 7 12728841
Willis MS Monaghan SA Miller ML McKenna RW Perkins WD Levinson BS Bhushan V Kroft SH Zinc-induced copper deficiency: a report of three cases initially recognized on bone marrow examination Am J Clin Pathol 2005 123 125 31 15762288 10.1309/V6GV-YW2Q-TYD5-C5PJ
L'Abbe MR Fischer PW The effects of dietary zinc on the activity of copper-requiring metalloenzymes in the rat J Nutr 1984 114 823 8 6327958
L'Abbe MR Fischer PW The effects of high dietary zinc and copper deficiency on the activity of copper-requiring metalloenzymes in the growing rat J Nutr 1984 114 813 22 6327957
Fischer PW Giroux A L'Abbe MR Effect of zinc supplementation on copper status in adult man Am J Clin Nutr 1984 40 743 6 6486080
Yadrick MK Kenney MA Winterfeldt EA Iron, copper, and zinc status: response to supplementation with zinc or zinc and iron in adult females Am J Clin Nutr 1989 49 145 50 2912000
Williams DM Copper deficiency in humans Semin Hematol 1983 20 118 28 6410510
Milne DB Assessment of copper nutritional status Clin Chem 1994 40 1479 84 8044985
Milne DB Copper intake and assessment of copper status Am J Clin Nutr 1998 67 1041S 1045S 9587149
Arsenault JE Brown KH Zinc intake of US preschool children exceeds new dietary reference intakes Am J Clin Nutr 2003 78 1011 7 14594789
Bialostosky K Wright JD Kennedy-Stephenson J McDowell MA Johnson CL Dietary intakes of macronutrients, micronutrients, and other dietary constituents: United States, 1988–94 Vital Health Stat 11 2002 245 1 168 15787426
Hunt CD Meacham SL Aluminum, boron, calcium, copper, iron, magnesium, manganese, molybdenum, phosphorus, potassium, sodium, and zinc: concentrations in common western foods and estimated daily intakes by infants; toddlers; and male and female adolescents, adults, and seniors in the United States J Am Diet Assoc 2001 101 1058 60 11573761 10.1016/S0002-8223(01)00260-7
Bertinato J Iskandar M L'Abbe MR Copper deficiency induces the upregulation of the copper chaperone for Cu/Zn superoxide dismutase in weanling male rats J Nutr 2003 133 28 31 12514262
Bertinato J L'Abbe MR Copper modulates the degradation of copper chaperone for Cu,Zn superoxide dismutase by the 26 S proteosome J Biol Chem 2003 278 35071 8 12832419 10.1074/jbc.M302242200
West EC Prohaska JR Cu,Zn-superoxide dismutase is lower and copper chaperone CCS is higher in erythrocytes of copper-deficient rats and mice Exp Biol Med (Maywood) 2004 229 756 64 15337829
Prohaska JR Broderius M Brokate B Metallochaperone for Cu,Zn-superoxide dismutase (CCS) protein but not mRNA is higher in organs from copper-deficient mice and rats Arch Biochem Biophys 2003 417 227 34 12941305 10.1016/S0003-9861(03)00364-3
Fischer PW L'Abbe MR Giroux A Effects of smoking, drinking, exercise and estrogen use on indices of copper status in healthy adults Nutr Res 1990 10 1081 90
Lukaski HC Hoverson BS Gallagher SK Bolonchuk WW Physical training and copper, iron, and zinc status of swimmers Am J Clin Nutr 1990 51 1093 9 2349924
Del Villano DC Miller SI Schacter LP Tischfield JA Elevated superoxide dismutase in black alcoholics Science 1980 207 991 3 7352301
Danks DM Copper deficiency in humans Annu Rev Nutr 1988 8 235 57 3060166 10.1146/annurev.nu.08.070188.001315
Wildman RE Hopkins R Failla ML Medeiros DM Marginal copper-restricted diets produce altered cardiac ultrastructure in the rat Proc Soc Exp Biol Med 1995 210 43 9 7675797
Li Y Wang L Schuschke DA Zhou Z Saari JT Kang YJ Marginal dietary copper restriction induces cardiomyopathy in rats J Nutr 2005 135 2130 6 16140888
Reeves PG Nielsen FH Fahey GC Jr AIN-93 purified diets for laboratory rodents: final report of the American Institute of Nutrition ad hoc writing committee on the reformulation of the AIN-76A rodent diet J Nutr 1993 123 1939 51 8229312
Cockell KA Fischer PW Belonje B Elemental composition of anatomically distinct regions of rat liver Biol Trace Elem Res 1999 70 251 63 10610064
L'Abbe MR Fischer PW Automated assay of superoxide dismutase in blood Methods Enzymol 1990 186 232 7 2233295
Schosinsky KH Lehmann HP Beeler MF Measurement of ceruloplasmin from its oxidase activity in serum by use of o-dianisidine dihydrochloride Clin Chem 1974 20 1556 63 4214636
Report of the American Institute of Nutrition ad hoc Committee on Standards for Nutritional Studies J Nutr 1977 107 1340 8 874577
Fischer PW Giroux A L'Abbe MR The effect of dietary zinc on intestinal copper absorption Am J Clin Nutr 1981 34 1670 5 7282591
Hall AC Young BW Bremner I Intestinal metallothionein and the mutual antagonism between copper and zinc in the rat J Inorg Biochem 1979 11 57 66 479878 10.1016/S0162-0134(00)80054-9
Fischer PW Giroux A L'Abbe MR Effects of zinc on mucosal copper binding and on the kinetics of copper absorption J Nutr 1983 113 462 9 6822916
Reeves PG Copper metabolism in metallothionein-null mice fed a high zinc diet J of Nutr Biochem 1998 9 598 601 10.1016/S0955-2863(98)00055-2
Reeves PG Briske-Anderson M Johnson L Physiologic concentrations of zinc affect the kinetics of copper uptake and transport in the human intestinal cell model, Caco-2 J Nutr 1998 128 1794 801 9772151
Liuzzi JP Cousins RJ Mammalian zinc transporters Annu Rev Nutr 2004 24 151 72 15189117 10.1146/annurev.nutr.24.012003.132402
Palmiter RD Cole TB Findley SD ZnT-2, a mammalian protein that confers resistance to zinc by facilitating vesicular sequestration EMBO J 1996 15 1784 91 8617223
Brown KH Peerson JM Allen LH Effect of zinc supplementation on children's growth: a meta-analysis of intervention trials Bibl Nutr Dieta 1998 54 76 83 9597173
|
16307684
|
PMC1315358
|
CC BY
|
2021-01-04 16:39:30
|
no
|
Nutr J. 2005 Nov 24; 4:35
|
utf-8
|
Nutr J
| 2,005 |
10.1186/1475-2891-4-35
|
oa_comm
|
==== Front
BMC Dev BiolBMC Developmental Biology1471-213XBioMed Central London 1471-213X-5-271632422010.1186/1471-213X-5-27Research ArticleSelection of reference genes for quantitative real-time PCR in bovine preimplantation embryos Goossens Karen [email protected] Poucke Mario [email protected] Soom Ann [email protected] Jo [email protected] Zeveren Alex [email protected] Luc J [email protected] Department of Animal Nutrition, Genetics, Breeding and Ethology, Faculty of Veterinary Medicine, Ghent University, Heidestraat 19, B-9820 Merelbeke, Belgium2 Department of Reproduction, Obstetrics and Herd Health, Faculty of Veterinary Medicine, Ghent University, Salisburylaan 133, B-9820 Merelbeke, Belgium3 Center for Medical Genetics Ghent, Ghent University Hospital, Medical Research Building, De Pintelaan 185, B-9000 Ghent, Belgium2005 3 12 2005 5 27 27 9 6 2005 3 12 2005 Copyright © 2005 Goossens et al; licensee BioMed Central Ltd.2005Goossens 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
Real-time quantitative PCR is a sensitive and very efficient technique to examine gene transcription patterns in preimplantation embryos, in order to gain information about embryo development and to optimize assisted reproductive technologies. Critical to the succesful application of real-time PCR is careful assay design, reaction optimization and validation to maximize sensitivity and accuracy. In most of the studies published GAPD, ACTB or 18S rRNA have been used as a single reference gene without prior verification of their expression stability. Normalization of the data using unstable controls can result in erroneous conclusions, especially when only one reference gene is used.
Results
In this study the transcription levels of 8 commonly used reference genes (ACTB, GAPD, Histone H2A, TBP, HPRT1, SDHA, YWHAZ and 18S rRNA) were determined at different preimplantation stages (2-cell, 8-cell, blastocyst and hatched blastocyst) in order to select the most stable genes to normalize quantitative data within different preimplantation embryo stages.
Conclusion
Using the geNorm application YWHAZ, GAPD and SDHA were found to be the most stable genes across the examined embryonic stages, while the commonly used ACTB was shown to be highly regulated. We recommend the use of the geometric mean of those 3 reference genes as an accurate normalization factor, which allows small expression differences to be reliably measured.
==== Body
Background
Preimplantation bovine embryo development is characterized by distinct biological steps, including first cleavage division, activation of the embryonic genome, compaction and blastocyst formation with the derivation of two different cell lines, the inner cell mass and the trophectodermal cells. These processes are regulated by differential expression of developmentally important genes, mostly expressed in a stage- and time-dependent manner following the common maternal and/or embryonic expression pattern [1]. The acquisition of knowledge about the physiological timetable of gene expression during preimplantation development is crucial for a better understanding of mammalian embryo development and is useful for further refinement of assisted reproductive technology in mammals [2].
Until recently, most studies of oocyte and embryo physiology were based on microscopic observations, but it is commonly agreed that the evaluation of embryo morphology alone does not answer most of the questions [3]. New insights into preimplantation development were gained through the measurement of differential mRNA levels in oocytes and preimplantation embryos by Reverse Transcription (RT-) PCR methods [4-7] that replace less sensitive and more laborious methods like Northern blot analysis and RNase protection assay. However due to differential reaction efficiencies and kinetics in RT-PCR, the amount of final product after amplification may not accurately reflect the initial sample mRNA concentration [8].
Real-time RT-PCR assays in which data are accurately normalized, are significantly less variable than commonly used conventional RT-PCR procedures. Real-time quantification at the exponential phase is not affected by any reaction components becomming limited in the plateau phase. Although small differences in transcript levels can be measured by endpoint RT-PCR, a lot of optimizations and post-PCR manipulations are required. These optimizations have to be performed for every individual sample, because the RNA expression can vary a lot between individual samples. Therefore, real-time RT-PCR has been recognised as the method of choice for accurate and sensitive quantification of mRNA transcripts [9,10]. Many studies have now been published where RNA quantification has been assessed in early domestic animal embryos. This technique has the advantage of speed, high throughput and accuracy over a large dynamic range of quantification and is especially suitable when only a small number of cells are available [11]. Several authors [12-14] have demonstrated the reproducibility of the 2 step SYBR Green I real-time RT-PCR reaction by determination of the intra- and interassay variation.
However, a lack of standards, variation in assay design, diversity of protocols, instruments and analysis methods make that real-time qRT-PCR results should be treated with caution and that agreed standards and operating procedures are required [15].
Several variables need to be controlled for gene-transcription analysis, such as the amount of starting material, enzymatic efficiencies, and differences between tissues or cells in overall transcriptional activity. Many methods are used to control some of these these variables, for example normalization against the total cell number, against the mass of the input material or against the RNA mass quantity. Exogenously added mRNA or spikes can be used for standardization when the spike is added before the RNA extraction [16].
Normalization against internal control genes is most frequently used because it can control all variables. Those internal control genes, also known as reference genes, are often referred to as housekeeping genes, assuming that those genes are expressed at a constant level in certain tissues, at all stages of development and are unaffected by the experimental treatment. To date most of the standardizations are done to reference genes such as β-actin (ACTB), glyceraldehyde 3-phosphate dehydrogenase (GAPD) or 18S rRNA. However a number of studies have provided solid evidence that their transcription levels are not constant between different developmental stages and different experimental conditions [17-19]. Normalization of the data using these types of apparent controls can result in false conclusions being made regarding transcription levels. Therefore, validation of candidate reference genes is critical for accurate analysis of gene expression [20].
Most experiments include only a single reference gene. Vandesompele et al. [21] demonstrated that the conventional use of a single gene for normalization leads to relative large errors and they validated the geometric mean of multiple carefully selected reference genes as an accurate normalization factor.
In this study, we determined the mRNA expression levels of 8 commonly used reference genes at different preimplantation stages and calculated a normalization factor based on multiple control genes for more accurate and reliable normalization of gene-expression data in bovine preimplantation embryos.
Results
Sample quality
For each assay, embryos with good morphological characteristics [22] were selected from 3 different in vitro embryo production (IVP) experiments. The mean percentage of obtained embryos from all cultivated oocytes at the different developmental stages were 65 ± 6% for the 2-cell stage, 44 ± 5% for the 8-cell stage, 25 ± 4% for the blastocyst stage and 16 ± 3% for the hatched blastocyst stage. Because IVP is time consuming and only a restricted amount of fertilized oocytes develop to the desired embryonic stages, the number of assays was restricted to 3.
Total RNA was isolated from pools of 20 embryos per assay and for each examined developmental stage. Because of the very small cell numbers used for RNA extraction, the RNA quantity could not be measured by the BioPhotometer (Eppendorf, Leuven) or the Nanodrop ND 1000 spectrophotometer.
A minus RT control demonstrated the presence of a considerable amount of contaminating genomic DNA. This illustrated the necessity of a DNase treatment, which removed all the contaminating genomic DNA from the RNA samples.
Although the RNA quality and quantity could not be determined, a real-time PCR with the reference gene GAPD gave cycle threshold (Ct) values in the range of 23 to 27 and a single band on agarose gel. This first-strand cDNA was 2.5 times diluted with 10 mM Tris HCl pH 8 and used for further real-time applications.
Transcription profiling of the reference genes
An initial screening of the transcription profiles of the selected reference genes by RT-PCR showed that all of those genes were expressed across the preimplantation embryo stages of interest. None of the eight selected genes was excluded from the study.
Gene-specific amplification was confirmed by a single peak in melt-curve analysis and a single band with the expected size in agarose gel electrophoresis. No primer-dimer formation was detected and the identities of the PCR products were confirmed by sequencing (Table 1).
Table 1 Information on the primers used for real-time PCR
Gene Genbank Accession number species Sequence Product size (bp) Ta (°C) % homology
ACTB[23] AY141970 Cow 5'-CCTCACGGAACGTGGTTACA-3' 5'-TCCTTGATGTCACGCACAATTT-3' 87 58 100 %
GAPD XM_618013 Cow 5'-TTCAACGGCACAGTCAAGG-3' 5'-ACATACTCAGCACCAGCATCAC-3' 119 62 100 %
Histone H2A[15] U62674 Mouse 5'-GTCGTGGCAAGCAAGGAG-3' 5'-GATCTCGGCCGTTAGGTACTC-3' 182 60 82 %
TBP[33] NM_003194 Human 5'-CCTAAAGACCATTGCACTTCG-3' 5'-CTTCACTCTTGGCTCCTGTG-3' 146 57 94 %
HPRT1 AF176419 Cow 5'-TGCTGAGGATTTGGAGAAGG-3' 5'-CAACAGGTCGGCAAAGAACT-3' 154 58 100 %
SDHA NM_174178 Cow 5'-GCAGAACCTGATGCTTTGTG-3' 5'-CGTAGGAGAGCGTGTGCTT-3' 185 60 100 %
YWHAZ BM446307 Cow 5'-GCATCCCACAGACTATTTCC-3' 5'-GCAAAGACAATGACAGACCA-3' 120 60 97 %
18S rRNA AF176811 Cow 5'-AGAAACGGCTACCACATCCA-3' 5'-CACCAGACTTGCCCTCCA-3' 169 62 100 %
For all genes studied, standard curves derived from 10-fold serial dilutions of pooled cDNA gave correlation coefficients greater than 0.97 and efficiencies greater than 90%. The reaction efficiencies were used to transform the Ct-values into raw data for analysis with the geNorm software [21].
Three identical real-time qPCR assays were performed. In each assay the transcription levels of the selected reference genes were measured in duplicate, at 4 different stages of preimplantation development.
To compare the RNA transcription levels across the stages of embryonic development, the Ct values were compared. The Ct-value is defined as the number of cycles needed for the fluorescence signal to reach a specific threshold level of detection and is inversely correlated with the amount of template nucleic acid present in the reaction [23]. Most of the genes had Ct values in the range of 25 to 33 but 18S rRNA was more abundant (Ct levels <15).
GeNorm analysis
Analysis of the gene expression stability over the different embryonic stages was done using the geNorm software. The ranking of the 8 control genes according to their M value was equivalent between the 3 assays. GAPD, YWHAZ, SDHA and 18S rRNA were the 4 most stable genes in each of the 3 assays, only the order of the genes was different. ACTB was the least stable gene in the 3 assays. The results are listed in Table 2.
Table 2 Ranking of the reference genes in order of their expression stability per assay, decreasing from top to bottom. The reference genes chosen to calculate the normalization factor are printed in bold.
Assay 1 Assay 2 Assay 3 Combined*
SDHA SDHA GAPD GAPD
18S rRNA GAPD YWHAZ YWHAZ
GAPD 18S rRNA 18S rRNA 18S rRNA
YWHAZ YWHAZ SDHA SDHA
HPRT1 Histone H2A Histone H2A Histone H2A
Histone H2A HPRT1 TBP HPRT1
TBP TBP HPRT1 TBP
ACTB ACTB ACTB ACTB
*After correction for run-to-run variation
To ensure comparability between the 3 assays, we compared the Ct values and efficiencies of the relative standard curves, derived from the same pooled cDNA stock, between the three independent assays and made a correction for the plate-to-plate variation according to the qBase algorithm (Hellemans et al., in preparation) [24]. This correction factor for inter-assay variation was necessary for the determination of the most stable reference gene over the 3 assays together. The results of the geNorm analysis of the combined assay are shown in Figure 1A. The ranking of the genes in this combined assay is in agreement with the ranking of the 3 individual assays.
Figure 1 (A-B): Gene expression stability of the candidate reference genes analyzed by the geNorm program. (A) Average expression stability values (M) of the control genes over the 3 assays together, plotted from least stable (left) to most stable (right). (B) Pairwise variation analysis over the 3 assays together between the normalization factors NFn and NFn+1, to determine the optimal number of control genes for normalization. The higher V4/5 and V7/8 values are due to the inclusion of a relative unstable gene and are in accordance with the average expression stability M.
To determine how many reference genes should be used, normalization factors (NFn), based on the geometric mean of the expression levels of the n best reference genes, were calculated by stepwise inclusion of an extra, less stable reference gene according to Vandesompele et al. [21]. Figure 1B shows the pairwise variation Vn/Vn+1 between 2 sequential normalization factors NFn and NFn+1. A large variation means that the added gene has a significant effect and should probably be included for calculation of the normalization factor. In this case, the inclusion of a 4th gene has no significanf effect (low V3/4 value) on the NF. The 3 member set GAPD, SDHA and YWHAZ is an excellent choice for the calculation of the NF.
Discussion
Analysis of expression patterns of genes essential in early embryo development, provides a useful tool to assess the normality of the embryos and a tool to optimize assisted reproduction technologies [25]. New insights into early embryo development of mammals are commonly gained through the measurement of different mRNA levels by real-time qPCR. This technique has revolutionized the quantification of mRNA but requires careful assay design and reaction optimization to maximize sensitivity, accuracy and precision [26].
The problem of measuring transcript levels throughout preimplantation development is confounded by the fact that cell numbers and cell sizes are constantly changing during this developmental interval. Untill the maternal-zygotic transition, the mRNA is mainly of maternal origin. Once the genome is activated, the cell number will influence the amount of mRNA available [3]. To allow ontogenic analysis, the embryos were compared as a single unit and the reference genes will correct for the differences between the embryos.
Using in vitro culture, another variable that must be taken into account is the proportion of normal embryos in the sample. Incompetent embryos may be mixed with competent embryos during the analysis [3]. The gene expression in incompetent embryos may be different from those in competent embryos and could introduce a bias [27-29]. As such, there might be an influence of abnormal embryos on the choice of reference genes when using single embryo samples. Therefore we used groups of 20 pooled embryos with good morphological characteristics to minimize the influence of the quality of the individual embryo.
RNA quality and quantity are critical for succesful gene expression analysis. Due to the limited amount, RNA analysis was not possible, but an RNA extraction method optimized for small sample quantities, DNase treatment and appropriate control methods resulted in as reliable as possible results.
Critical to the successful application of real-time qPCR is the prevention of amplification of contaminating genomic DNA, resulting in an overestimation of the amount of RNA present. But even more importantly, yielding unreliable data especially for low abundant single exon genes or genes with retropseudogenes in the genome. Minus RT-controls before and after the DNAse treatment demonstrated the necessity and efficacy of the DNase treatment.
When intercalating dyes such as SYBR green I are used, attention should be paid to the formation of primer-dimers. Melting curve analysis and agarose gel electrophoresis confirmed that the fluorescent signal was specifically from the desired amplicons, not from artefacts.
Accurate normalization is required to correct real-time data for differences in cellular input, RNA quality and enzymatic efficiency between the samples. Under controlled conditions of reproducable extraction of good-quality RNA, the gene transcript number is ideally standardized to the number of cells [21]. During the bovine preimplantation period the cell numbers and cell size are constantly changing. Comparing mRNA levels within the same developmental stage is feasible but doing ontogenic analyses, which are essential to understand transitions in gene expression, is more problematic [3].
Another way of normalization is using the mass of the input material. However, in our case it was impossible to quantify those parameters because only minimal amounts of RNA were available and the total amount of RNA present throughout the preimplantation period is not constant. So normalization to total RNA requires a reliable RNA quantification method, and fails to take into account the variability of the RT-reaction. Probably the strongest argument against the use of total RNA mass for normalization is the fact that it predominantly consists of rRNA molecules and is not always representative of the mRNA fraction [21].
The addition of exogenously added mRNA can only be used when the exact amount of cells or starting material is known [16]. Spikes only correct for differences in enzymatic efficiencies but do not account for the quality and quantity of the input sample. Therefore, the use of spikes is not assumption free.
It is now generally accepted that transcription levels should be normalized to an invariable internal control gene. Those reference genes are often adapted from the literature and used against a variety of experimental conditions. An ideal reference gene should be expressed at a constant level among different experimental conditions and at all stages of development. An ontogenic study of several commonly used reference genes showed that their mRNA levels are not stable throughout preimplantation development [19]. If the used reference gene fluctuates between the samples, the subsequent normalization will cause erroneous results [30]. As the biological function of many genes is still unknown, it is difficult to predict how experimental conditions will affect the expression of the putative reference genes. Thus a safer approach is to use the geometric average expression of several genes that show small variance. Vandesompele et al. [21] postulated that gene pairs that have stable expression patterns relative to each other are proper control genes. YWHAZ, GAPD, SDHA and 18S rRNA were found to be the best endogenous control genes in preimplantation embryo samples as represented by their low M values, the marker of gene stability. ACTB was the worst scoring reference gene, in the set of 8 tested reference genes (Figure 1A). This is a remarkable result given the fact that in several publications on gene expression analysis in embryos, ACTB was the only reference gene used. However, the differential mRNA expression of ACTB is in accordance with previous reports that prove the upregulation of ACTB during preimplantation embryo development and predict a role for ACTB during blastocyst formation [31-33]. In previous studies Histone H2A was determined as the most stable reference gene during preimplantation embryo development [8,19], but those authors only considered Ct values and did not correct for the amount of input material. To validate the presumed stable expression of a given control gene, prior knowledge of a reliable measure to normalize this gene in order to remove any nonspecific variation is required. To address this circular problem, Vandesompele et al. [21] developed a gene-stability measure to determine the expression stability of control genes on the basis of non-normalized expression levels. By using this approach, the changing RNA content during the developmental stages was taken into account [34]. Besides geNorm, other programs and strategies are described in literature to select the best reference genes. BestKeeper [35] is an Excel-application also based on pairwise correlation. Normfinder [13] is a model based approach and enables estimation not only of the overall variation of the candidate normalization genes, but also of the variation between sample subgroups of the sample set.
A normalization factor (NF) based on the geometric mean of the best performing reference genes was calculated. The number of genes used to calculate this NF is a trade-off between practical considerations and accuracy. In this case, the 3 most stable reference genes were used to calculate the normalization factor (NF3). Figure 1B demonstrates that the inclusion of a 4th reference gene has no significant contribution to the newly calculated normalization factor NF4.
Ultimately, our choice for the normalizing set is the geometric mean of the transcription levels of GAPD, SDHA and YWHAZ. 18S rRNA was excluded because rRNA genes have general disadvantages when used as reference genes. Their transcription is carried out by RNA polymerase I, therefore the regulation of rRNA synthesis is independent from mRNA synthesis, which is carried out by RNA polymerase II [36]. Besides, rRNA genes are highly abundant compared to the target mRNA transcripts, this imbalance makes it difficult to accurately subtract the baseline values in real-time qPCR analysis [21]. 18S rRNA was evaluated in this study because it is a commonly used reference gene.
Conclusion
In conclusion, a method for gDNA free RNA extraction from embryos was optimized and a reference gene assay for reliable normalization of real-time PCR data, obtained from bovine preimplantation embryo samples was designed. Transcription profiling of 8 different reference genes showed that the use of a single reference gene is not reliable and will result in erroneous conclusions. Instead GAPD, SDHA and YWHAZ should be used.
Methods
In vitro production of bovine embryos
Bovine embryos were produced by routine in vitro methods as described by Yuan and colleagues [37]. Briefly, bovine oocytes were obtained from ovaries collected at a local slaughterhouse. Immature cumulus-oocyte complexes were selected from follicular fluid, washed three times in HEPES-TALP and matured for 22 to 26 h in groups of 100 in 500 μl maturation medium at 39°C in a humified 5% CO2 incubator. After maturation the oocytes were inseminated with frozen-thawed sperm of a dairy bull (1 × 106 spermatozoa/ml). The cumulus cells and spermatozoa were mechanically removed from the presumptive zygotes, which were placed in groups of 25 in 50 μl droplets of synthetic oviduct fluid supplemented with 5% fetal calf serum and cultured up to the desired stages. The embryos were collected at the indicative time period after fertilization: 2-cell (24–36 h), 8-cell (48–64 h), blastocyst (day 7) and hatched blastocyst (day 8). All embryos were washed three times in PBS, collected in pools of 20 and frozen at -80°C until RNA extraction.
RNA extraction and cDNA synthesis
Total RNA was isolated from 20 pooled embryos using the PicoPure RNA Isolation Kit (Arcturus, Mountain View, CA) according to the manufacturer's instructions. This kit is engineered to recover high-quality total RNA from pico-scale samples.
For genomic DNA removal an in-solution DNase digestion was carried out by treating the total RNA with 2 units of RQ1 DNase (Promega, Leiden) followed by a spin-column purification (Microcon YM-100, Millipore, Brussels). A minus RT control was performed with primers for GAPD to check the removal of all the contaminating genomic DNA.
First-strand cDNA was synthesized from the total amount of RNA using the iScript cDNA synthesis kit (Bio-Rad, Nazareth), following the manufacturer's instructions. The iScript Reverse Transcriptase is a modified MMLV-derived reverse transcriptase and the iScript Reaction Mix contains both oligo(dT) and random primers. After the RT reaction and RT control with primers for GAPD, the cDNA was 2.5 times diluted in 10 mM Tris HCl pH 8.0.
Reference gene selection and primer design
Eight reference genes were selected (ACTB, GAPD, Histone H2A, TBP, HPRT1, SDHA, YWHAZ and 18S rRNA) that belong to different functional classes to reduce the chance that the genes might be co-regulated (Table 3).
Table 3 Functions of the selected reference genes
Symbol Gene name Function
ACTB β-actin Cytoskeletal structural protein
GAPD Glyceraldehyde-3-phosphate dehydrogenase Glycolytic enzyme
Histone H2A Histone 2 alpha Nucleosome structure
TBP TATA box binding protein General RNA polymerase II transcription factor
HPRT1 Hypoxanthine phosphoribosyl-transferase I Purine synthesis in salvage pathway
SDHA Succinate dehydrogenase flavoprotein subunit A Electron transporter in the TCA cycle and respiratory chain
YWHAZ Tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein, zeta polypeptide Signal transduction by binding to phosphoserine-containing proteins
18S rRNA 18S ribosomal RNA Ribosome unit
Primers for Histone H2A were taken from Robert and colleagues [19], primers for TBP were taken from Vigneault and colleagues [38] and primers for ACTB were taken from Fair and colleagues [27]. The other primers were designed by the Primer 3 software [39] and were based on RNA or DNA sequences found in Genbank. The reported bovine sequences were preferentially used and the specificity of the primers was tested using a BLAST analysis against the genomic NCBI database. PCR amplicons were characterized using Mfold [40] in order to predict the nature of any secondary structures which might influence the PCR efficiency. The PCR products were cloned (pCR 2.1 vector, Invitrogen, Merelbeke) and sequenced for verification (Thermo Sequenase Primer Cycle Sequencing Kit, Amersham Bioscience, Roosendaal) with a ALF Express sequencer (Amersham Bioscience, Roosendaal) [GenBank: DQ066891, DQ066892, DQ066893, DQ066894, DQ066895, DQ066896, DQ066897 and DQ066898]. Primer and amplicon information are listed in Table 1.
Real-Time quantitative PCR
Three replicates of 20 pooled embryos were used for each developmental stage (2-cell, 8-cell, blastocyst and hatched blastocyst) as described by Robert et al. [19].
All PCR reactions were performed in a 15 μl reaction volume on the iCycler iQ Real-Time PCR Detection System (Bio-Rad, Nazareth) using the iQ SYBR Green Supermix (Bio-Rad, Nazareth), 200 nM of each specific primer and 2.5 μl of diluted cDNA or one embryo equivalent per reaction.
The PCR program consisted of an initial denaturation step at 95°C for 3 minutes to activate the Taq DNA polymerase, followed by 45 cycles of denaturation at 95°C for 20 seconds and a combined primer annealing/extension at the specific annealing temperature for 40 seconds during which fluorescence was measured. A melt curve was produced to confirm a single gene-specific peak and to detect primer/dimer formation by heating the samples from 70 to 95°C in 0.5°C increments with a dwell time at each temperature of 10 seconds while continuously monitoring the fluorescence. PCR efficiencies were calculated using a relative standard curve derived from a pooled cDNA mixture (a ten-fold dilution series with four measuring points). This pooled cDNA was obtained from bovine heart, kidney, liver, muscle, lung and placenta tissue, using Total RNA Isolation Reagent (TRIR, ABgene, Epsom) for the RNA isolation and the iScript cDNA synthesis kit (Bio-Rad, Nazareth) for the RT-reaction.
Each reaction was run in duplicate, whereby a no-template control was included.
Determination of reference gene expression stability
To determine the stability of the selected reference genes, the geNorm Visual Basic application for Microsoft Excel was used as described by Vandesompele et al. [21].
This approach relies on the principle that the expression ratio of two perfect reference genes should be identical in all samples, regardless of the experimental condition or cell type. Increasing variation in this ratio corresponds to decreasing expression stability. The program calculates the gene stability measure M by determining the average pair-wise variation between a particular reference gene and all other control genes. Genes with higher M values have greater variation in RNA expression. By stepwise exclusion of the least stable gene and recalculation of the M values, the most stable reference genes are identified. Finally, a normalisation factor (NF) was calculated based on the geometric mean of the expression levels of the best-performing reference genes.
Authors' contributions
KG performed all the experimental procedures and was the primary author of the manuscript. MVP participated in the study design and provided real-time support. AVS contributed to the IVF experiments. JV provided expert input in data analysis. AVZ and LJP participated in the design of the project, helped to draft the manuscript and supervised the study. All authors read and approved the final manuscript.
Acknowledgements
The authors like to thank Machteld Baetens, Bieke Scharlaken, Johanna Mestagh and Griet Spaepen for their excellent technical assistance.
==== Refs
Niemann H Wrenzycki C Alterations of expression of developmentally important genes in preimplantation bovine embryos by in vitro culture conditions: implications for subsequent development Theriogenology 2000 53 21 34 10735059 10.1016/S0093-691X(99)00237-X
Steuerwald N Cohen J Herrera RJ Brenner CA Analysis of gene expression in single oocytes and embryos by real-time rapid cycle fluorescence monitored RT-PCR Mol Hum Reprod 1999 11 1034 1039 10541565 10.1093/molehr/5.11.1034
Sirard MA Dufort I Vallée M Massicotte L Gravel C Reghenas H Watson AJ King WA Robert C Potential and limitations of bovine-specific arrays for the analysis of mRNA levels in early development: premilimary analysis using a bovine embryonic array Reprod Fertil Dev 2005 17 47 57 15745631 10.1071/RD04113
Zhang X Kidder GM Zhang C Khamsi F Armstrong DT Expression of plasminogen activator genes and enzymatic activities in rat preimplantation embryos J Reprod Fertil 1994 101 235 240 8064688
Gutierrez-Adan A Behboodi E Murray JD Anderson GB Early transcription of the SRY gene by bovine preimplantation embryos Mol Reprod Dev 1997 48 246 250 9291474 10.1002/(SICI)1098-2795(199710)48:2<246::AID-MRD12>3.0.CO;2-W
Wrenzycki C Hermann D Carnwath JW Niemann H Expression of RNA from developmentally important genes in preimplantation bovine embryos produced in TCM supplemented with BSA J Reprod Fertil 1998 112 387 398 9640278
Wrenzycki C Herrmann D Niemann H Timing of blastocyst expansion affects spatial messenger RNA expression patterns of genes in bovine blastocysts produced in vitro Biol Reprod 2003 68 2073 2080 12606328 10.1095/biolreprod.102.012104
Jeong YJ Choi HW Shin HS Cui XS Kim NH Gerton GL Jun JH Optimization of real-time RT-PCR methods for the analysis of gene expression in mouse eggs and preimplantation embryos Mol Reprod Dev 2005 71 284 289 15806558 10.1002/mrd.20269
Bustin SA Absolute quantification of mRNA using real-time reverse transcription polymerase chain reaction assays J Mol Endocrinol 2000 25 169 193 11013345 10.1677/jme.0.0250169
Bustin SA Quantification of mRNA using real-time reverse transcription PCR (RT-PCR): trends and problems J Mol Endocrinol 2002 29 23 39 12200227 10.1677/jme.0.0290023
Fink L Seeger W Ermert L Hanze J Stahl U Grimminger F Kummer W Bohle RM Real-Time quantitative RT-PCR after laser-assisted cell picking Nat Med 1998 4 1329 1333 9809560 10.1038/3327
Vandesompele J De Paepe A Speleman F Elimination of primer-dimer artifacts and genomic coamplification using a two-step SYBR Green I Real-Time RT-PCR Anal Biochem 2002 303 95 98 11906156 10.1006/abio.2001.5564
Andersen CL Jensen JL Ørntoft TF Normalization of real-time quantitative reverse transcription-PCR data: a model-based variance estimation approach to identify genes suited for normalization, applied to bladder and colon cancer data sets Cancer Res 2004 64 5245 5250 15289330 10.1158/0008-5472.CAN-04-0496
Lai JP Douglas SD Wang YJ Ho WZ Real-time reverse transcription-PCR quantification of substance P receptor (NK-1R) mRNA Clin Diagn Lab Immunol 2005 12 537 541 15817763 10.1128/CDLI.12.4.537-541.2005
Bustin SA Nolan T Pitfalls of quantitative real-time reverse-transcription Polymerase Chain Reaction J Biomol Tech 2004 15 155 166 15331581
Donnison M Pfeffer PL Isolation of genes associated with developmentally competent bovine oocytes and quantitation of their expression levels during development Biol Reprod 2004 71 1813 1321 15286031 10.1095/biolreprod.104.032367
Thellin O Zorzi W Lakaye B De Borman B Coumans B Hennen G Gristar T Igout A Heinen E Housekeeping genes as internal standards: use and limits J Biotechnol 1999 75 291 295 10617337 10.1016/S0168-1656(99)00163-7
Suzuki T Higgins PJ Crawford DR Control selection for RNA quantitation Biotechniques 2000 29 332 337 10948434
Robert C McGraw S Massicotte L Pravetoni M Gandolfi F Sirard MA Quantification of housekeeping transcript levels during the development of bovine preimplantation embryos Biol Reprod 2002 67 1465 1472 12390877 10.1095/biolreprod.102.006320
Zhang X Ding L Sandford AJ Selection of reference genes for gene expression studies in human neutrophils by real-time PCR BMC Mol Biol 2005 6 4 15720708 10.1186/1471-2199-6-4
Vandesompele J De Preter K Pattyn F Poppe B Van Roy N De Paepe A Speleman F Accurate normalization of Real-Time quantitative RT-PCR by geometric averaging of multiple internal control genes Genome Biol 2002 3 34 10.1186/gb-2002-3-7-research0034
Van Soom A Mateusen B Leroy J De Kruif A Assessment of mammalian embryo quality: what can we learn from embryo morphology? Reprod Biomed Online 2003 7 664 670 14748965
Walker NJ Tech Sight. A technique whose time has come Science 2002 296 557 559 11964485 10.1126/science.296.5567.557
Hellemans qBase Management and automated analysis of real-time quantitative PCR data
Wrenzycki C Herrmann D Lucas-Hahn A Korsawe K Lemme E Niemann H Messenger RNA expression patterns in bovine embryos derived from in vitro procedures and their implications for development Reprod Fertil Dev 2005 17 23 35 15745629 10.1071/RD04109
Peters IR Helps CR Hall EJ Day MJ Real-time RT-PCR: considerations for efficient and sensitive assay design J Immunol Methods 2004 286 203 217 15087233 10.1016/j.jim.2004.01.003
Fair T Murphy M Rizos D Moss C Martin F Boland MP Lonergan P Analysis of differential maternal mRNA expression in developmentally competent and incompetent bovine two-cell embryos Mol Reprod Dev 2004 67 136 144 14694428 10.1002/mrd.10385
Donnison M Pfeffer PL Isolation of genes associated with developmentally competent bovine oocytes and quantitation of their levels during development Biol Reprod 2004 71 1813 1821 15286031 10.1095/biolreprod.104.032367
Gardner DK Lane M Ex vivo embryo development and effects on gene expression and imprinting Reprod Fertil Dev 2005 17 361 370 15745644 10.1071/RD04103
Dheda K Huggett JF Bustin SA Johnson MA Rook G Zumla A Validation of housekeeping genes for normalizing RNA expression in real-time PCR Biotechniques 2004 37 112 119 15283208
Tanaka TS Ko MS A global view of gene expression in the preimplantation mouse embryo: morula versus blastocyst Eur J Obstet Gynecol Reprod Biol 2004 115 85 91 15223171 10.1016/j.ejogrb.2004.01.026
Bilodeau-Goeseels S Schultz GA Changes in the relative abundance of various housekeeping gene transcripts in in vitro-produced early bovine embryos Mol Reprod Dev 1997 47 413 420 9211425 10.1002/(SICI)1098-2795(199708)47:4<413::AID-MRD7>3.0.CO;2-N
Giebelhaus DH Weitlauf HM Schultz GA Actin mRNA content in normal and delayed implanting mouse embryos Dev Biol 1985 107 407 413 3972162 10.1016/0012-1606(85)90322-7
Bilodeau-Goeseels S Schultz GA Changes in ribosomal nucleic acid content within in vitro-produced bovine embryos Biol Reprod 1997 56 1323 1329 9160734 10.1095/biolreprod56.5.1323
Pfaffl MW Tichopad A Prgomet C Neuvians TP Determination of stable housekeeping genes, differentially regulated target genes and sample integrity: BestKeeper – Excel-based tool using pair-wise correlations Biotechnol Lett 2004 26 509 515 15127793 10.1023/B:BILE.0000019559.84305.47
Radonic A Thulke S Bae HG Müller MA Siegert W Nitsche A Reference gene selection for quantitative real-time PCR analysis in virus infected cells: SARS corona virus, Yellow fever virus, Human Herpesvirus-6 and Cytomegalovirus infections Virol J 2005 2 7 15705200 10.1186/1743-422X-2-7
Yuan YQ Van Soom A Leroy JLMR Dewulf J Van Zeveren A de Kruif A Peelman LJ Apoptosis in cumulus cells, but not in oocytes, may influence bovine embryonic developmental competence Theriogenology 2005 63 2147 2163 15826680 10.1016/j.theriogenology.2004.09.054
Vigneault C McGraw S Massicotte L Sirard MA Transcription factor expression patterns in bovine in vitro-derived embryos prior to maternal-zygotic transition Biol Reprod 2004 70 1701 1709 14960490 10.1095/biolreprod.103.022970
Rozen S Skaletsky H Primer3 on the WWW for general users and for biologist programmers Methods Mol Biol 2000 132 365 86 10547847
Zuker M Mfold web server for nucleic acid folding and hybridization prediction Nucleic Acids Res 2003 31 3406 3415 12824337 10.1093/nar/gkg595
|
16324220
|
PMC1315359
|
CC BY
|
2021-01-04 16:30:31
|
no
|
BMC Dev Biol. 2005 Dec 3; 5:27
|
utf-8
|
BMC Dev Biol
| 2,005 |
10.1186/1471-213X-5-27
|
oa_comm
|
==== Front
BMC PediatrBMC Pediatrics1471-2431BioMed Central London 1471-2431-5-401628008010.1186/1471-2431-5-40Research ArticleActuarial survival of a large Canadian cohort of preterm infants Jones Huw P [email protected] Stella [email protected] Catherine MG [email protected] Arne [email protected] Abraham [email protected] Anne [email protected] Shoo K [email protected] Canadian Neonatal Network [email protected] Department of Pediatrics, St Mary's Hospital, Portsmouth, UK2 Canadian Neonatal Network Coordinating Centre, Edmonton, AB, Canada3 Department of Pediatrics, University of Manitoba, Winnipeg, MB, Canada4 Department of Pediatrics, University of Toronto, Toronto, ON, Canada5 Department of Pediatrics, University of Alberta, Edmonton, AB, Canada6 Department of Pediatrics, University of British Columbia, Vancouver, BC, Canada2005 9 11 2005 5 40 40 4 7 2005 9 11 2005 Copyright © 2005 Jones et al; licensee BioMed Central Ltd.2005Jones 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 increased survival of preterm and very low birth weight infants in recent years has been well documented but continued surveillance is required in order to monitor the effects of new therapeutic interventions. Gestation and birth weight specific survival rates most accurately reflect the outcome of perinatal care. Our aims were to determine survival to discharge for a large Canadian cohort of preterm infants admitted to the neonatal intensive care unit (NICU), and to examine the effect of gender on survival and the effect of increasing postnatal age on predicted survival.
Methods
Outcomes for all 19,507 infants admitted to 17 NICUs throughout Canada between January 1996 and October 1997 were collected prospectively. Babies with congenital anomalies were excluded from the study population. Gestation and birth weight specific survival for all infants with birth weight <1,500 g (n = 3419) or gestation ≤30 weeks (n = 3119) were recorded. Actuarial survival curves were constructed to show changes in expected survival with increasing postnatal age.
Results
Survival to discharge at 24 weeks gestation was 54%, compared to 82% at 26 weeks and 95% at 30 weeks. In infants with birth weights 600–699, survival to discharge was 62%, compared to 79% at 700–799 g and 96% at 1,000–1,099 g. In infants born at 24 weeks gestational age, survival was higher in females but there were no significant gender differences above 24 weeks gestation. Actuarial analysis showed that risk of death was highest in the first 5 days. For infants born at 24 weeks gestation, estimated survival probability to 48 hours, 7 days and 4 weeks were 88 (CI 84,92)%, 70 (CI 64, 76)% and 60 (CI 53,66)% respectively. For smaller birth weights, female survival probabilities were higher than males for the first 40 days of life.
Conclusion
Actuarial analysis provides useful information when counseling parents and highlights the importance of frequently revising the prediction for long term survival particularly after the first few days of life.
==== Body
Background
The improvement in survival rates for preterm and very low birth weight infants has been well documented during the last 20 years [1-3]. More recently, a marked reduction in mortality rates has been reported following the introduction and widespread use of antenatal steroids and exogenous surfactant [4-6]. Despite initial concerns, studies of later neurodevelopmental outcome have not shown an increase in major disability rates as a result of improved survival [7-9]. As our understanding of disease processes increases and new therapies continue to be developed, continued surveillance of up to date outcome data is essential in order to monitor the effectiveness of current practice.
Early reports of survival rates for high risk infants were based on birth weight alone, as assessment of gestational age was relatively imprecise prior to the introduction of routine first trimester ultrasound scans. Subsequently, survival data based on gestational age has become more widely reported [10,11]. This information is more useful for obstetric decision-making in the prenatal period and there is increasing evidence that mortality and later morbidity in high risk infants relates more closely to gestation than to birth weight [10,12]. Indeed a combination of both variables may give an even more accurate prediction of outcome once the infant is born [13].
Of further concern is that survival data in many published studies is derived from single tertiary units or collaborations of such centres which are not geographically based [1,14]. This increases the likelihood of sample bias and may not give a true reflection of survival expectations for the population as a whole. The effects of gender and multiple births on survival have also been of interest in previous studies, many showing higher survival rates for females and/or singletons [11,15-19]. This additional information may enhance our prediction of survival based on gestation or birth weight alone. More recently, attention has focused on the timing of infant death, specifically utilising actuarial survival analysis to predict future life expectancy from a given age [20-23]. This provides useful additional information when making decisions regarding ongoing management in the neonatal intensive care unit (NICU).
Our aims were to determine survival to discharge for a large Canadian cohort of high-risk infants (<1500 g or ≤30 weeks) representing admissions to 17 NICUs with 75% of tertiary level NICU beds in Canada, and to examine the effects of gestation, birth weight, gender and multiple birth on survival. Utilizing actuarial survival analysis, the effect of increasing postnatal age on expected survival to discharge was investigated.
Methods
Study population
The Canadian Neonatal Network comprised 17 tertiary NICUs across Canada in 1996 [24]. It was funded by the Medical Research Council of Canada and other institutions (see acknowledgements) in 1996 to facilitate neonatal research by creating a national neonatal-perinatal database. The Canadian population in 1996 was about 30 million with approximately 357,000 births annually [25]. This study included all infants with a birth weight <1500 grams or gestational age ≤30 weeks who were admitted to the Canadian Neonatal Network during a 22-month period between January 1996 and October 1997. Infants with congenital anomalies were excluded because they have different mortality and morbidity risks [26].
Data collection
Prospective data were collected locally by trained research assistants and transmitted electronically to the Canadian Neonatal Network Coordinating Centre for verification and analysis. Collected data included demographic variables, obstetric information, neonatal illness severity (Score for Neonatal Acute Physiology, Version II [SNAP-II]) [27], therapeutic intensity (Neonatal Therapeutic Intensity Scoring System [NTISS]) [28] and selected outcomes and resource use.
Definition of study variables
Study variables were defined according to the Canadian Neonatal Network SNAP Project Abstractor Manual [29]. An admission was defined as a stay in the NICU for at least 24 hours or death/transfer to another unit within 24 hours. Gestational age was defined as the best obstetric estimate based on early prenatal ultrasound, obstetric examination and obstetric history, unless the post-natal pediatric estimate of gestation differed from the obstetric estimate by more than two weeks. In that case, the pediatric estimate of gestational age was used instead. An infant was defined as small-for-gestational age (SGA) if the birth weight was less than the 10th percentile for gestational age according to the growth charts established by Arbuckle [30] in 1989 for the Canadian population. SNAP-II [27] is a neonatal illness severity score calculated from 6 empirically weighted physiologic measurements made during the first 12 hours of admission to the NICU. NTISS [28] is a score of neonatal therapeutic intensity calculated from a checklist of 63 NICU therapies used in a 24 hour period, weighted according to invasiveness and cost. Chronic lung disease was defined as oxygen dependency at 36 weeks corrected GA for an infant who was born at ≤32 weeks gestation [31]. Intraventricular hemorrhage (IVH) was defined according to the criteria of Papile [32] from head ultrasound performed before 14 days of life. Necrotizing enterocolitis (NEC) was defined according to Bell's criteria (stage 2 or higher) [33] and was classified as medical (clinical symptoms and signs plus evidence of pneumatosis on abdominal X'ray) or surgical (histological evidence of NEC on surgical specimen of intestine). Retinopathy of prematurity (ROP) was defined according to the International Classification for Retinopathy of Prematurity [34] and the Reese Classification of cicatrical disease [35]. Nosocomial infection was defined using blood and cerebrospinal fluid culture results according to Freeman's criteria [36]. Patent ductus arteriosus was defined as clinical diagnosis plus treatment with indomethacin or surgical ligation or both. Seizures were defined as clinically significant episodes witnessed by a nurse or physician and for which anti-convulsant treatment was given. Congenital anomalies were classified according to the WHO International Classification of Diseases, 9 th Revision (ICD-9) [37].
Data analysis
Data analysis was carried out on two populations – preterm babies with gestation age ≤30 weeks and Very Low Birth Weight (VLBW) babies with birth weight <1,500 grams. A two sample t-test was used to separately test the effect of gender and multiple births on population demographics, adverse outcome and resource use. Similarly a two sample t-test was used to test the effect of gender on survival proportions for each week of gestational age and each 100 grams birth weight range. For the actuarial analysis, the event of interest was death and the time-to-event was the number of days survived. Babies that survived to discharge had their length of stay included as right censored observations. The actuarial survival estimator which estimates the probability of survival to and beyond a set number of days was calculated in daily intervals for the first 60 days of life. The effect of gender, gestation age and birth weight on survival were studied by obtaining actuarial survival estimators for each category. The probability of survival to discharge, for an infant who has survived to a given day in the NICU, was calculated and graphed by gestational age and birth weight category.
Results
Between January 8, 1996 and October 31, 1997, 19,507 infants were admitted to participating NICUs. Excluding those with congenital anomalies, there were 3,119 infants ≤30 weeks gestational age and 3,409 infants with birth weight <1,500 grams. Characteristics of the two groups are shown in Table 1. Of note, for infants <1,500 grams birth weight, 52% were born by caesarean section, 21% were outborn and 68% received antenatal steroids.
Table 1 Characteristics of preterm and very low birth weight (VLBW) infants in the study cohort
Preterm (≤30 weeks) VLBW (<1500 g)
n = 3119 n = 3409
Males (%) 56 53
Multiple births (%) 27 29
Cesarean section (%) 46 51
Antenatal steroids (%) 72 70
Outborn (%) 22 21
Small for gestational age 3 rd percentile (%) 14.7 26.1
Mean birth weight (grams) 1071 1041
Mean gestational age (weeks) 27 28
Survival to NICU discharge
Survival to discharge increased from 45% for infants weighing <600 g at birth to over 95% for those with birth weight >1200 g (Fig 1). There was a similar increase for gestational age groups with survival increasing from about 14% at 22 weeks to over 93% at 28 weeks and above (Fig 2).
Figure 1 Birth weight specific survival.
Figure 2 Gestational age specific survival.
Effect of gender on survival and morbidity
When analysed by gestational age groups, survival for male and female infants was not significantly different except for those born at 24 weeks gestation in whom survival for females was 17 (3, 29)% higher (Fig 3). When analysed according to birth weight groups, survival for females in the 600–699 g group was 7 (6,28)% significantly higher than for males (Fig 4). However, within many birth weight groups, females were significantly more mature than males and would therefore be expected to have higher survival as a consequence of gestational age (Tables 2 and 3). In contrast to survival, female infants had significantly (p < 0.05) lower incidence of chronic lung disease and severe intraventricular hemorrhage than male infants (Table 4). The high incidence of SGA among infants born at <1,500 g reflects the inherent tendency to select SGA infants when birth weight criteria is used to categorise infants instead of gestational age.
Table 2 Comparison of mean gestational age (GA) between male and female infants in each birth weight group (VLBW)
Mean GA (weeks)
Birth Weight (g) Male Female p-value
n
<600 203 24.2 24.4 ns
600–699 303 24.7 25.5 <0.05
700–799 318 25.5 26.0 <0.05
800–899 344 26.3 27.1 <0.05
900–999 297 27.2 27.4 ns
1000–1099 375 28.2 28.6 ns
1100–1199 368 28.8 29.3 <0.05
1200–1299 381 29.4 29.9 <0.05
1300–1399 372 29.7 30.9 <0.05
1400–1499 448 30.7 31.0 ns
Table 3 Comparison of mean birth weight between male and female infants in each GA group (Preterm)
Weight (grams)
GA (weeks) Male Female p-value
n
23 105 625 576 <0.05
34 234 699 679 ns
25 329 765 755 ns
26 374 897 813 <0.05
27 388 1005 943 <0.05
28 466 1155 1102 ns
29 579 1311 1236 <0.05
30 612 1485 1382 <0.05
Table 4 Male/Female characteristics of preterm (≤30 weeks gestation) and very low birth weight Infant (< 1,500 g) groups
Preterm (≤30 weeks gestation) Very low birth weight (< 1,500 g)
Male n = 1741 (56%) Female n = 1376 (44%) Male n = 1797 (53%) Female n = 1601 (47%)
Demographics
Mean birth weight (grams) 1102 1032* 1050 1031*
Mean gestational age (weeks) 27.4 27.4 27.9 28.3*
Small for gestational age (%) 13.6 16.1* 24.4 28.2*
5-minute Apgar score (mean) 7.2 7.3 7.2 7.4*
Outborn (%) 21.4 21.7 21.4 19.9
Antenatal steroids (%) 69.3 71.3 66.2 69.4*
SNAP-II (mean) 26.6 27.2 26.5 24.8*
Outcomes
Survival (%) 85.2 86.9 85.6 88.9*
Necrotizing enterocolitis (%) 6.5 7.0 6.6 6.3
Patent ductus arteriosus (%) 28.0 32.3* 27.3 28.8
Seizures (%) 4.6 4.9 4.5 4.2
Chronic lung disease (%) 25.1 21.7* 25.7 18.9*
Primary infection (%) 1.8 2.0 1.6 1.6
Nosocomial infection (%) 21.2 22.1 21.9 20.9
Intraventricular hemorrhage (≥grade 3) (%) 9.6 7.5* 9.1 6.4*
Retinopathy of prematurity (≥stage 3) (%) 10.9 10.8 10.5 9.7
Resource use
NTISS (mean) 17.5 17.1 17.1 16.0*
Assisted ventilation (%) 88.7 86.7 83.9 79.6*
Number of ventilated days (mean) 18.0 17.1 17.5 14.9*
Surgery (%) 21.2 14.5* 21.6 12.9*
Length of NICU stay (mean) (days) 48.1 48.9 47.6 45.7
* significant (p < 0.05) difference between male and female infants
Figure 3 Male/Female survival by gestational age.
Figure 4 Male/Female survival by birth weight.
Multiple births and antenatal steroids
Of the 3119 infants born ≤30 weeks, 2,277 (73%) were singleton deliveries and 841 (27%) were the products of multiple gestation pregnancies (Table 5). There was no significant difference in survival, illness severity or resource use between the two groups. Antenatal steroid use was 7(4,11)% higher in multiple gestation for VLBW babies, and delivery by cesarean section was 9(5,13)% more likely in multiple gestation pregnancies for VLBW. Antenatal steroid use was 7(3,11)% higher in multiple gestation for preterm babies, and delivery by cesarean section was 9(6,13)% more likely in multiple gestation pregnancies for preterm. A full or partial course of antenatal steroids was given to 70% of infants ≤30 weeks gestation and to 68% of infants between <1,500 g birth weight. The use of antenatal steroids was lower in the most preterm infants ≤24 weeks gestation.
Table 5 Characteristics of singleton and multiple birth preterm infants
Preterm (≤30 weeks gestation) Very low birth weight (< 1,500 g)
Singleton n = 2277 (73%) Multiple n = 841 (27%) Singleton n = 2424 (71%) Multiple n = 975 (29%)
Demographics
Mean birth weight (grams) 1071 1072 1031 1066
Mean gestational age (weeks) 27.3 27.6* 27.9 28.4*
Small for gestational age (%) 15.0 13.8 26.3 25.7
5-minute Apgar score (mean) 7.2 7.4* 7.2 7.5*
Outborn (%) 22.5 18.9* 21.9 17.6*
Antenatal Steroids (%) 68.2 75.4* 65.7 73.2*
SNAP-II (mean) 27.1 26.2 26.3 24.1*
Outcomes
Survival (%) 86.1 85.7 87.1 87.4
Necrotizing enterocolitis (%) 6.9 6.3 6.8 5.6
Patent ductus arteriosus (%) 29.6 30.9 28.1 28.0
Seizures (%) 5.3 3.2* 4.8 3.2*
Chronic lung disease (%) 24.4 21.7 23.4 20.2
Primary infection (%) 2.2 0.8* 1.9 0.9*
Nosocomial infection (%) 22.1 21.6 22.0 20.0
Intraventricular hemorrhage (≥grade 3) (%) 9.2 7.3 8.4 6.5
Retinopathy of prematurity (≥stage 3) (%) 12.1 7.4* 11.3 7.2*
Resource use
Caesarean section (%) 43.0 52.4* 48.7 57.7*
NTISS (mean) 17.3 17.3 16.8 16.1*
Assisted ventilation (%) 87.6 88.6 82.8 79.6*
Number of ventilated days (mean) 18.2 15.9* 17.1 14.1*
Surgery (%) 18.8 17.1 18.4 15.3*
Length of NICU stay (mean) (days) 48.8 47.6 47.9 43.9*
* denotes p < 0.05 level of significance
Actuarial survival
Actuarial survival curves are given in Figures 5, 6, 7, 8, 9 for the first 60 days. Points on the curves give S(t); the estimated probability of a baby surviving to at least t days. Babies that were discharged from the NICU can not afford to be ignored as these were most likely to be the longer lived observations. Babies that were discharged from the NICU were therefore included as censored observations in the analysis. By comparing the slope of the curve (which gives the instantaneous risk of death), the highest risk of death is within the first 6 days. Risk of death is higher for smaller babies and babies with lower GA. Risk of death also decreases with time. Figure 9 shows that actuarial survival is higher among females than males during the first 40 days, and narrows after that. Figures 10 and 11 give the probability of survival to discharge for an infant who has survived to a given day in the NICU, stratified by gestational age and birth weight respectively.
Figure 5 Actuarial survival curves stratified by gestation (weeks), by gender.
Figure 6 Actuarial survival curves by gestational age, with 95% CI.
Figure 7 Actuarial survival curves stratified by birth weight category; by Gender.
Figure 8 Actuarial survival curves by birth weight category.
Figure 9 Actuarial survival curves by sex.
Figure 10 Probability of survival to discharge (y-axis), for male and female infants surviving to a given day in the NICU (x-axis), stratified by gestation age.
Figure 11 Probability of survival to discharge (y-axis), for male and female infants surviving to a given day in the NICU (x-axis), stratified by birth weight (g).
Discussion
Since our study is based on a large, geographically defined Canadian cohort, it has greater relevance than those from single units and provides a more realistic picture of neonatal outcomes. In addition, our study cohort was derived from large regional centres with comparable levels of care conforming to current North American standards for neonatal intensive care. Our results are similar to those from previous publications on NICU outcomes, such as that of the NICHD Neonatal Network [1-4], the Vermont-Oxford Trials Network [38,39] and the Australia-New Zealand Neonatal Network [40] (although some networks examined all live born infants instead of NICU admissions). Although improvement in survival rates were previously reported among the most preterm groups [1-4,24,38-40], more recent publications report no further improvement in outcomes [41,42], and demonstrate the importance of reporting trends over time.
An advantage in survival for female infants has been reported by previous studies with the greatest difference apparent when stratified by birth weight groups [41-44]. In contrast, when analysed by gestational age, we found a gender difference in survival only for infants born at 24 weeks GA or less. This may reflect the fact that at any given birth weight, female infants tend to be more mature. It has also been previously suggested that the advantage in survival for female infants is related to a more favourable hormonal milieu in the female fetus leading to accelerated lung maturation compared to the male [45-48]. In the present era, it is also possible that the increased use of antenatal steroids and exogenous surfactant has improved overall survival but more notably in male compared to female infants. However, not all gender differences were eliminated and female infants continued to have lower incidences of chronic lung disease and severe intraventricular hemorrhage than male infants. Actuarial survival analysis also showed that female infants had lower mortality rates than male infants during the first 40 days of life, indicating a difference in time of death.
Actuarial analysis adds a further dimension to standard survival data and highlights several points of interest. For all gestational age and birth weight groups the risk of death is greatest during the first few days of life with relatively few deaths occurring after day 7. This emphasises the importance of revising the chances of survival at regular intervals especially during the first few days of life (as in figures 10 and 11). Such updated survival information is useful when informing parents and is necessary for ongoing management decisions. Furthermore, actuarial survival curves are helpful when studying the economic aspects of neonatal intensive care. The majority of deaths occur during the first few days of life and therefore the proportion of health care expenditure on eventual non-survivors is relatively small. With reference to research and health promotion, actuarial data aid the development of treatment strategies. Our data highlights where interventions would have the greatest benefit e.g. early deaths within the first few days and late deaths after the first 28 days. The effects of new interventions on survival patterns could then be monitored. Finally, this report extends the available information on various aspects of low birth weight infants previously reported by the Canadian Neonatal Network [24,26,27,49-58].
One limitation of our survival data is that they only include infants admitted for neonatal intensive care. They do not take into account stillbirths after the onset of preterm labour nor delivery room deaths of live born infants. Therefore, they provide an overestimate of survival chances if used to counsel parents during preterm labour or as a guide to obstetric management. However, as most high risk pregnancies are currently managed in large perinatal centres, with skilled personnel certified in neonatal resuscitation, the majority of infants born at 24 weeks and above are successfully resuscitated and admitted for intensive care, and our results would be highly applicable to them. Discharge policies may affect survival rates but since infants are usually discharged only when they are sufficiently well, it is unlikely that discharge policies will significantly affect our results. For the actuarial survival analysis, survival to 60 days of life was considered. However, deaths may occur even after this time during the post-discharge period and these were not considered in the study.
Conclusion
Up to date survival rates are essential when evaluating perinatal services. Previously reported effect of gender on overall survival are no longer apparent among infants >24 weeks gestation but there is a gender difference in the time of death. Actuarial survival analysis emphasises the importance of frequently revising predictions for survival in high-risk infants, particularly during the first week of life.
Abbreviations
BW Birth weight
GA Gestational age
NICU Neonatal intensive care unit
NTISS Neonatal therapeutic intensity scoring system
SNAP-II Score for Neonatal Acute Physiology, Version II
VLBW Very low birth weight
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
Huw Jones interpreted data and drafted the manuscript. Stella Karuri performed statistical analysis and interpretation. Shoo K Lee was the principal investigator and interpreted the data and drafted the manuscript. Catherine Cronin, Arne Ohlson and Abraham Peliowski and Anne Synnes were site investigators. All these individuals read and approved the final manuscript. The CNN represents all site investigators, and was responsible for organization and administration of the SNAP study, and subsequent data flow.
Pre-publication history
The pre-publication history for this paper can be accessed here:
Acknowledgements
This study was supported by Grant 40503 and Grant 00152 from the Medical Research Council of Canada. Additional funding was provided by the B.C.'s Children's Hospital Foundation; Calgary Regional Health Authority; Dalhousie University Neonatal-Perinatal Research Fund; Division of Neonatology, Children's Hospital of Eastern Ontario; Child Health Program, Health Care Corporation of St John's; The Neonatology Program, Hospital for Sick Children; Lawson Research Institute; Midland Walwyn Capital Inc; Division of Neonatology, Hamilton Health Sciences Corporation; Mount Sinai Hospital; North York General Hospital Foundation; Saint Joseph's Health Centre; University of Saskatchewan Neonatal Research Fund; University of Western Ontario; Women's College Hospital.
Members of the Canadian Neonatal Network: Shoo K. Lee (Coordinator, Canadian Neonatal Network); Wayne Andrews (Charles A. Janeway Child Health Centre, St John's, NF); Ranjit Baboolal (North York Hospital, N. York, ON); Jill Boulton (St Joseph's Health Centre, London, ON; previously at Mt Sinai Hospital, Toronto, ON); David Brabyn (Royal Columbian Hospital, New Westminster, BC); David S.C. Lee (St Joseph's Health Centre; London, ON); Derek Matthew (Victoria General Hospital (Victoria, BC); Douglas D. McMillan (Foothill's Hospital, Calgary, AB); Christine Newman (Hospital for Sick Children; Toronto, ON); Arne Ohlsson (Mt Sinai Hospital, Toronto, ON; formerly at Women's College Hospital, Toronto, ON); Abraham Peliowski (Royal Alexandra Hospital, Edmonton, AB); Margaret Pendray (Children's & Women's Health Centre of British Columbia (Vancouver, BC); Koravangattu Sankaran, (Royal University Hospital, Saskatoon, SK); Barbara Schmidt (Hamilton Health Sciences Corporation, Hamilton, ON); Mary Seshia (Health Sciences Centre, Winnipeg, MB); Anne Synnes (Children's and Women's Health Centre of British Columbia, Vancouver, BC; formerly at Montreal Children's Hospital, Montreal, PQ); Paul Thiessen (Children's & Women's Health Centre of British Columbia (Vancouver, BC); Robin Walker (Children's Hospital of Eastern Ontario and Ottawa General Hospital, Ottawa, ON); Robin Whyte (IWK-Grace Health Centre for Women, Children and Families, Halifax, NS); Catherine M. G. Cronin (Winnipeg Regional Health Authority).
Coordinating centre staff (Vancouver, BC): Aireen Wingert; Stella Karuri; Sarka Lisonkova.
==== Refs
Hack M Wright LL Shankaran S Tyson JE Horbar JD Bauer CR Younes N Very low birth weight outcomes of the National Institute of Child Health and Human Development Neonatal Network, November 1989 to October 1990 Am J Obstet Gynecol 1995 172 457 464 7856670 10.1016/0002-9378(95)90557-X
Fanaroff AA Wright LL Stevenson DK Shankaran S Donovan EP Ehrenkranz RA Younes N Korones SB Stoll BJ Tyson JE Bauer CR Oh W Lemons JA Papile LA Verter J Very low birthweight outcomes of the National Institute of Child Health and Human Development Neonatal Research Network, May 1991 through December 1992 Am J Obstet Gynecol 1995 173 1423 1431 7503180 10.1016/0002-9378(95)90628-2
Stevenson DK Wright LL Lemons JA Oh W Korones SB Papile LA Bauer CR Stoll BJ Tyson JE Shankaran S Fanaroff AA Donovan EF Ehrenkranz RA Verter J Very low birth weight outcomes of the National Institute of Child Health and Human Development Neonatal Research Network, January 1993 through December 1994 Am J Obstet Gynecol 1998 179 1632 1639 9855609 10.1016/S0002-9378(98)70037-7
Ferrara TB Hoekstra RE Couser RJ Gaziano EP Calvin SE Payne NR Fangman JJ Survival and follow-up of infants born at 23–26 weeks of gestational age: effects of surfactant therapy J Pediatr 1994 124 119 124 8283360
Jobe AH Mitchell BR Gunkel JH Beneficial effects of the combined use of prenatal corticosteroids and postnatal surfactant on preterm infants Am J Obstet Gynecol 1993 168 508 513 8438919
Corbet A Bucciarelli R Goldman S Mammel M Wold D Long W Decreased mortality rate among small premature infants treated at birth with a single dose of synthetic surfactant: a multicenter controlled trial. American Exosurf Pediatric Study Group 1 J Pediatr 1991 118 277 284 1993961
O'Shea TM Preisser JS Klinepeter KL Dillard RG Trends in mortality and cerebral palsy in a geographically based cohort of very low birth weight neonates born between 1982 to 1994 Pediatrics 1998 101 642 647 9521949 10.1542/peds.101.4.642
O'Shea TM Klinepeter KL Goldstein DJ Jackson BW Dillard RG Survival and developmental disability in infants with birth weights of 501 to 800 grams, born between 1979 and 1994 Pediatrics 1997 100 982 986 9374569 10.1542/peds.100.6.982
Cooke RWI Trends in incidence of cranial ultrasound lesions and cerebral palsy in very low birth weight infants 1982–93 Arch Dis Child Fetal Neonatal Ed 1999 80 F115 F117 10325787
Wariyar U Richmond S Hey E Pregnancy outcome at 24–31 weeks' gestation: neonatal survivors Arch Dis Child 1989 64 678 686 2471464
Synnes AR Ling EW Whitfield MF Mackinnon M Lopes L Wong G Effer SB Perinatal outcomes of a large cohort of extremely low gestational age infants (twenty-three to twenty-eight completed weeks of gestation) J Pediatr 1994 125 952 60 7996370
Phelps DL Brown DR Tung B Cassady G McClead RE Purohit DM Palmer EA 28-Day Survival Rates of 6676 Neonates with Birth Weights of 1250 Grams or Less Pediatrics 1991 87 7 17 1984621
Draper ES Manktelow B Field DJ James D Prediction of survival for preterm births by weight and gestational age: retrospective population based study BMJ 1999 319 1093 1097 10531097
Hack M Horbar JD Malloy MH Tyson JE Wright E Wright L Very low birth weight outcomes of the National Institute of Child Health and Human Development Neonatal Network Pediatrics 1991 87 587 597 2020502
Tyson JE Younes N Verter J Wright LL Viability, morbidity, and resource use among newborns of 501 to 800 g birth weight. National Institute of Child Health and Human Development Neonatal Research Network JAMA 1996 276 1645 1651 8922450 10.1001/jama.276.20.1645
Brothwood M Wolke D Gamsu H Benson J Cooper D Prognosis of the very low birth weight baby in relation to gender Arch Dis Child 1986 61 559 564 2425752
Resnick MB Carter RL Ariet M Bucciarelli RL Evans JH Furlough RR Ausbon WW Curran JS Effect of birth weight, race, and sex on survival of low-birth-weight infants in neonatal intensive care Am J Obstet Gynecol 1989 161 184 7 2750802
Hoffman EL Bennett FC Birth Weight Less than 800 Grams: Changing Outcomes and Influences of Gender and Gestation Number Pediatrics 1990 86 27 34 2359681
Khoury MJ Marks JS McCarthy BJ Zaro SM Factors affecting the sex differential in neonatal mortality: The role of respiratory distress syndrome Am J Obstet Gynecol 1985 151 777 782 3976790
Cartlidge PHT Stewart JH Survival of very low birth weight and very preterm infants in a geographically defined population Acta Paediatrica 1997 86 105 110 9116412
Cooper TR Berseth CL Adams JM Weisman LE Actuarial survival in the Premature Infant Less than 30 Weeks' Gestation Pediatrics 1998 101 975 978 9606222 10.1542/peds.101.6.975
Meadow W Reimshisel T Lantos J Birth weight specific mortality for extremely low birth weight infants vanishes by four days of life: epidemiology and ethics in the neonatal intensive care unit Pediatrics 1996 97 636 643 8628599
Gould JB Benitz WE Liu H Mortality and time to death in very low birth weight infants: California, 1987 and 1993 Pediatrics 2000 105 e37 10699139 10.1542/peds.105.3.e37
Lee SK McMillan DD Ohlsson A Pendray M Synnes A Whyte R Chien LY Sale J Variations in practice and outcomes of the Canadian NICU Network 1996–7 Pediatrics 2000 106 1070 9 11061777 10.1542/peds.106.5.1070
Synnes A Berry M Jones H Pendray M Stewart SD Lee SK for The Canadian Neonatal Network Infants with congenital anomalies admitted to neonatal intensive care units Am J Perinatology 2004 21 199 208 10.1055/s-2004-828604
Richardson DK Corcoran JD Escobar GJ Lee SK for the Canadian NICU Network the Kaiser Permanente Neonatal Minimum Sata Set Wide Area Network and the SNAP-II Study Group SNAP-II and SNAPPE-II: Simplified newborn illness severity and mortality risk scores J Pediatr 2001 138 92 100 11148519 10.1067/mpd.2001.109608
Gray JE Richardson DK McCormick MC Workman-Daniels K Goldman DA Neonatal therapeutic intervention scoring system (NTISS): a therapy-based severity of illness assessment tool Pediatrics 1992 90 561 567 1408510
Canadian Neonatal Network Abstractor Manual, Vancouver, BC 1995
Arbuckle TE Sherman GJ An analysis of birth weight by gestational age in Canada CMAJ 1989 140 157 165 2597238
Shennan AT Dunn MS Ohlsson A Lennox K Hoskins EM Abnormal pulmonary outcomes in preterm infants: prediction from oxygen requirement in the neonatal period Pediatrics 1988 82 527 532 3174313
Papile LA Munsick-Bruno G Schaefer A Relationship of cerebral intra-ventricular hemorrhage and early childhood handicaps J Pediatr 1983 103 272 277
Bell MJ Ternberg JL Feigin RD Keating JP Marshall R Barton L Brotherton T Neonatal necrotizing enterocolitis. Therapeutic decisions based upon clinical staging Ann Surg 1978 187 1 7 413500
Committee members An international classification of retinopathy of prematurity Pediatrics 1984 74 127 133 6547526
Reese AB King MJ Owens WC a classification of retrolental fibroplasia Am J Opthalmol 1953 36 1333
Freeman J Epstein MF Smith NE Platt R Sidebottom DG Goldman DA Extra hospital stay and antibiotic usage with nocosomial coagulase negative staphylococcal bacteremia in two neonatal intensive care unit populations Am J Dis Child 1990 144 324 329 2305739
WHO International Classification of Diseases, 9 th Revision WHO, Geneva 1975
Horbar JD Badger J Lewit EM Rogowski J Shiono PH Hospital and patient characteristics associated with variation in 28-day mortality rates for very low birth weight infants Pediatrics 1997 99 149 156 9024438 10.1542/peds.99.2.149
The investigators of the Vermont-Oxford Trials Network Very low birth weight outcomes for 1990 Pediatrics 1993 91 540 545 8441556
Donoghue D Australia & New Zealand Neonatal Network Neonatal Network Series Number 2 AIHW National Perinatal Statistics Unit, Sydney 1997 ISSN 1326-012X, AIHW Catalogue no. PER 5
Tucker J McGuire W Epidemiology of preterm birth BMJ 2004 329 675 678 15374920 10.1136/bmj.329.7467.675
Battin M Ling EW Whitfield MF Mackinnon M Effer SB Has the outcome for extremely low gestational age (ELGA) infants improved following recent advances in neonatal intensive care? Am J Perinatol 1998 15 469 477 9788645
Hack M Fanaroff AA Outcomes of children of extremely low birthweight and gestational age in the 1990's Early Hum Devel 1999 53 193 218 10088988 10.1016/S0378-3782(98)00052-8
Hack M Friedman H Fanaroff AA Outcomes of extremely low birth weight infants Pediatrics 1996 98 931 937 8909488
Cummings JJ D'Eugenio DB Gross SJ A controlled trial of dexamethasone in preterm infants at high risk for bronchopulmonary dysplasia N Engl J Med 1989 320 1505 10 2657423
Khosla SS Brehier A Eisenfeld AJ Ingleson LD Parks PA Rooney SA Effects of estrogen on fetal rabbit lung maturation: morphological and biochemical studies Pediatr Res 1981 15 1274 1281 7290778
Zachman RD Morison JC Curet LB Gustafson N the Collaborative Group on Antenatal Steroid Therapy Lecithin:sphingomyelin ratio in the amniotic fluid of male and female fetuses J Reprod Med 1989 34 203 206 2724233
Nielson HC Testosterone regulation of sex differences in fetal lung development Proc Soc Exp Biol Med 1992 199 446 452 1549623
Chien LY Ohlsson A Seshia M Boulton J Sankaran K Lee SK The Canadian Neonatal Network Variation in antenatal corticosteroid treatment – a persistent problem 30 years later Obstetr & Gynecol 2002 99 401 408 10.1016/S0029-7844(01)01732-X
Chan KJ Ohlsson A Synnes A Lee DSC Chien LY Lee SK The Canadian NICU Network Survival, morbidity and resource use of <25 weeks gestation infants Am J Obstet Gynecol 2001 185 220 226 11483932 10.1067/mob.2001.115280
Synnes A Chien LY Peliowski A Baboolal R Lee SK The Canadian Neonatal Network Variations in Intraventricular Hemorrhage Incidence Rates among Canadian Neonatal Intensive Care Units J Pediatr 2001 138 525 531 11295716 10.1067/mpd.2001.111822
Chien LY Whyte R Thiessen P Matthew D Aziz K Lee SK The Canadian Neonatal Network Improved outcome of preterm infants when delivered in tertiary care centers Obstet & Gynecol 2001 98 247 252 11506840 10.1016/S0029-7844(01)01438-7
Lee SK McMillan DD Ohlsson A Boulton J Lee DSC Ting S Liston R The Canadian Neonatal Network Benefit of preterm birth at tertiary care centers is related to gestational age Am J Obstet Gynecol 2003 188 617 622 12634630 10.1067/mob.2003.139
Lee SK Lee DCS Andrews WL Baboolal R Pendray M Stewart SD The Canadian Neonatal Network Higher mortality among infants admitted to neonatal intensive care units at nights J Pediatr 2003 143 592 597 14615728 10.1067/S0022-3476(03)00367-6
Sankaran K Puckett B Lee DSC Seshia M Boulton J Qiu ZG Lee SK Canadian Neonatal Network Variations in incidence of necrotizing enterocolitis in Canadian neonatal intensive care units J Pediatr Gastro Nutr 2004 39 366 372 10.1097/00005176-200410000-00012
Hayter M Anderson L Claydon J Magee LA Little R Liston RM Lee SK Dadelszen PV Canadian Neonatal Network Variations in early and intermediate neonatal outcomes for inborn babies admitted to a Canadian NICU and born of hypertensive pregnancies JOGC 2005 27 25 32 15937579
Shah P Shah V Qiu ZG Ohlsson A Lee SK Canadian Neonatal Network Improved outcomes of outborn preterm infants if admitted to perinatal centers vs free-standing pediatric hospitals J Pediatr 2005 146 626 631 15870665 10.1016/j.jpeds.2005.01.030
Lau J Magee F Qiu ZG Hoube J von Dadelszen P Lee SK Maternal-fetal chorioamnionitis is associated with higher neonatal mortality, morbidity and resource use than maternal chorioamnionitis Am J Obstet Gynecol 2005 193 708 713 16150264 10.1016/j.ajog.2005.01.017
|
16280080
|
PMC1315360
|
CC BY
|
2021-01-04 16:31:06
|
no
|
BMC Pediatr. 2005 Nov 9; 5:40
|
utf-8
|
BMC Pediatr
| 2,005 |
10.1186/1471-2431-5-40
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1635410310.1371/journal.pmed.0030001Policy ForumInfectious DiseasesHealth PolicyHealth PolicyInternational healthPublic HealthInfectious DiseasesThe New International Health Regulations and the Federalism Dilemma Policy ForumWilson Kumanan *McDougall Christopher Upshur Ross Kumanan Wilson is in the Department of Medicine, University of Toronto, Toronto, Canada, and the Institute of Intergovernmental Relations, Queen's University, Kingston, Canada. Christopher McDougall is in the Primary Care Research Unit, Sunnybrook and Women's College Health Sciences Centre, Toronto, Canada. Ross Upshur is in the Departments of Family and Community Medicine and Public Health Sciences, University of Toronto, and the Primary Care Research Unit, Sunnybrook and Women's College Health Sciences Centre.
*To whom correspondence should be addressed. E-mail: [email protected]
Competing Interests: The authors declare that they have no competing interests.
1 2006 20 12 2005 3 1 e1Copyright: © 2006 Wilson et al.2006This 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 recent revision of the International Health Regulations, say Wilson and colleagues, is both long overdue and eminently necessary to face the challenges of an increasingly globalized world.
==== Body
In the aftermath of severe acute respiratory syndrome (SARS) (Figure 1) and in anticipation of avian flu, the international health community has recognized that pandemic planning and response is an inherently multigovernmental concern. The ability of pathogens to cross borders and rapidly spread around the globe requires highly coordinated public health responses that involve the cooperation of local, regional, national, and supranational governments (Figure 2). The understanding of this reality has informed the current International Health Regulations (IHR) revision process.
Figure 1 “8 Steps Towards SARS Prevention”—Public Information Poster Issued by the Chinese Government, 2003
Figure 2 Processing SARS Specimens at the Special Pathogens Branch of the CDC
The Centers for Disease Control and Prevention (CDC) worked closely with the WHO and other partners, as part of a global collaboration to address the SARS outbreak.
(Photo: James Gathany/CDC Public Health Image Laboratory)
Approved in May 2005, the revised IHR have increased the disease surveillance requirements of “states parties” in an effort to better inform the pandemic response process and to protect the health of the global community [1]. Furthermore, the revisions have also outlined recommendations that the World Health Organization (WHO) could issue if an outbreak originating in one country is perceived to be a threat to other countries. The revision of the IHR is both long overdue and eminently necessary to face the challenges of an increasingly globalized world [2]. The practical implementation of these proposals, however, may encounter obstacles. This is particularly true for those WHO member nations that have federal systems of government (federations), and could ultimately threaten their ongoing support of the new regulations.
Federalism and Public Health Response
Federalism is a type of political system in which the advantages of shared rule are combined with those of regional government [3]. Countries with federal governments make up about 40% of the world's population, and include the second most populous country (India) and the world's largest economy (United States) [4]. Federal systems of government offer many advantages, including allowing for the distinctiveness of the regions within a nation to be recognized and for region-specific policy approaches to be developed. However, one of the limitations of federations is that the division of powers can create an obstacle to the development of centralized approaches to national challenges. Such scenarios can arise when the country's constitution distributes the key powers in question to the regions. This characteristic of federal systems poses a dilemma when international treaties are signed by a federal government, but the cooperation of regional governments (states, provinces, etc.) is necessary for compliance with the treaty.
Canada's experience with SARS outlines the challenges of such constitutional division of powers when it comes to managing public health crises. During SARS, the Canadian federal government's ability to obtain data from the Province of Ontario was dependent on voluntary transfer, since the management of infectious disease outbreaks falls under provincial jurisdiction [5,6]. Reviews of the response to SARS showed that the transfer of data from the provincial government to the federal government was a key obstacle to the management of the crisis and, in particular, limited the federal government's ability to effectively communicate the status of the outbreak to the WHO (Problem 7 of [7]).
The United States encountered similar jurisdictional problems when developing strategies to address the threat of bioterrorism following the 2001 anthrax attacks, because public health is primarily within the jurisdiction of the states [8,9]. Concerns about the inadequacy of some state public health legislation at the time provided the impetus for the development of a Model State Emergency Health Powers Act [10].
In Australia, powers over emergency response to public health crises also primarily reside at the state level, with the federal government having limited authority except for quarantine. The development of a coordinated Australian approach to managing new infectious threats has thus been a challenge, and there is potential for confusion over who has authority in the event of a public health emergency that crosses state borders [11].
Adding to the challenges of developing effective intergovernmental approaches to disease outbreaks in federations is their multidimensional nature. The response to an infectious outbreak could involve issues of national security, emergency response, environmental protection, and food and water safety. Powers over these areas may be differentially allocated across the various orders of government. Such a scenario could produce conflict or confusion when attempting to determine which order of government has the ultimate authority over the management of the outbreak. This in turn may contribute to a failure to adequately manage an outbreak and to the spread of the outbreak across borders within a country, and potentially into other countries.
The New International Health Regulations
The revisions of the IHR create important new challenges to all countries, and in particular to those with federal systems of government (Table 1) [12]. The new IHR require all states parties to designate representatives to implement the surveillance, response, and notification requirements of the regulations. These requirements cover all jurisdictions from the community level to the national level. While the previous version only applied to three infectious diseases, the new IHR apply to “all events that may constitute a public health emergency of international concern,” and a decision-making instrument to assist in the identification of such events is included in the regulations. Notifiable events within a state are to be reported by a “national IHR Focal Point” to the WHO within 24 hours. States are also responsible for strengthening their surveillance system and are required to complete both a capacity assessment within two years of the approval of the revised Regulations, and the development of public health infrastructure that ensures full compliance within five years of the entry into force (in June 2007) of the regulations.
Table 1 Some Changes to the IHR That Are Relevant to Federations
The new IHR also explicitly outline new WHO powers, which include an information-gathering prerogative that is not limited solely to official state notifications or consultations, but which covers all “the available scientific evidence and other relevant information”. The WHO is also empowered to share information with other states parties if an affected state “does not accept the offer of collaboration” and “when justified by the magnitude of the public health risk”. The revisions also formally empower the WHO to issue temporary and standing recommendations if an outbreak is classified as a public health emergency of international concern. These could include recommendations to issue travel restrictions for persons from affected areas. The recommendations would ideally be made with the consent of the affected country, although provisions exist for such action in the absence of the target member state's consent.
Federalism and the New International Health Regulations
The realities of federations, and the fragmentation of powers within them, could become particularly problematic when attempting to operationalize the new IHR. This is particularly true of the surveillance and reporting requirements, and it is of particular concern given the implications of new WHO powers. There is a real concern that federations may not be able to comply with the IHR, which could result in the issuance of temporary recommendations that would penalize federations for political and administrative features that they perceive to be beyond their control.
While a primarily unitary state may have sufficient centralized powers to ensure that the surveillance and reporting requirements embodied in the new IHR are met, the allocation of powers within federations may not permit this. For example, the IHR revisions require all member nations to notify the WHO “within 24 hours of assessment of public health information, of all events which may constitute a public health emergency of international concern within its territory…as well as any health measure implemented in response to those events”. However, federal governments may not have the authority to collect the data necessary for reporting to the WHO and the transfer of data from the affected regions may not occur voluntarily. While, according to the new IHR, the federal government can designate regional representatives to carry out some of the measures, the government is ultimately dependent on cooperation from these regional authorities that may not necessarily be forthcoming.
There are several reasons why regional governments, for example, may not want to provide complete information on the nature of an outbreak. These include concern about the impact of disclosure on their economy, aversion to federal scrutiny, concerns about stigmatization, and a fundamental belief that the issue being addressed is within their jurisdiction. Failure to report could have significant consequences, most importantly the delay of national and international responses to prevent the spread of the disease. The resistance of regional governments to sharing of information could leave a country susceptible to measures introduced by the WHO, particularly since the WHO now has authority to conduct surveillance and to utilize information gathered from nongovernmental sources through the Global Public Health Intelligence Network [13,14].
The issuance of travel advisories, as occurred during SARS, or of temporary recommendations to member nations not to accept travelers from an affected region, could have damaging effects on the economy of that region. The perception in Canada was that the SARS travel advisory may not have been warranted, and had a serious negative impact on the economy of Toronto [15,16]. In developing countries where tourism is essential to the national economy, such an advisory could be catastrophic. Federations could view the use of such measures by the WHO as unnecessarily punitive and an invasion of national sovereignty. Ultimately, this could lead to a lack of support for the WHO, the IHR and other global strategies for disease control [12].
Federalism and International Agreements
The United States, recognizing the challenges the IHR could pose for federations, had requested the insertion of a clause that would acknowledge the unique governance structures of federations. The decision not to include such a clause in the revised IHR prompted the United States to notify that they intend to submit a statement of reservation, specifically commenting that they will “implement the IHR in a manner consistent with (their) federal system of government” [17].
The challenges posed by the IHR to federations are not unique, however, as federations are often confronted with difficulties in implementing international agreements. For example, the
GATT/WTO agreement included a federal clause which states that “[e]ach contracting party shall take such reasonable measures as may be available to it to ensure observance of the provisions of this Agreement by the regional and local governments and authorities within its territory”. However, there has been variability in the interpretation of this clause, with certain countries claiming that this requires the use of any constitutional power available to adhere to the agreement, and others arguing that this clause should not allow internal federal structures to be compromised [18,19]. The Kyoto Protocol to the United Nations Framework Convention on Climate Change has generated similar challenges to implementation in federal states, and it is notable that two key federations, the US and Australia, have not ratified the treaty [20].
The problems federations have with compliance with international treaties can occur at several levels. First, as already mentioned, the necessary powers to ensure compliance with a treaty may not fall within the jurisdiction of the federal level of government. While in many federations, the federal power to sign treaties may permit them to override the jurisdiction of regional governments, this is not necessarily always the case. And in scenarios in which the federal government does have the necessary constitutional authority, they may choose not to exercise it because of political concerns about creating conflict with regional governments. Even in scenarios in which the federal government has the necessary legislative power and does choose to exercise it, regional governments may not be able to cope with the financial and/or practical burdens of compliance [21]. This latter issue, of unfunded mandates, is not unique to federal states and constitutes a challenge that all WHO member nations will need to address when determining how to implement the surveillance requirements of the new IHR.
Guidance to Federations
It is apparent that this “federalism dilemma” will need to be addressed, both by member nations and by the WHO, if the revised IHR are to be implemented successfully. It is primarily incumbent upon federations, as responsible members of the international community, to take every measure available to ensure that they can comply with the new IHR. These measures would include using what constitutional means may be available to centralize necessary public health powers concerning surveillance and outbreak response, making efforts to establish effective collaborative intergovernmental arrangements, and developing appropriate public health capacity at the local level. The intergovernmental acrimony to which federations are susceptible would clearly not be acceptable if, at a time of crisis, it produced a dysfunctional response that resulted in the international spread of disease.
The WHO, in turn, must make efforts to assist federations in this regard. These efforts could include the provision of guidelines on strategies federal governments can use to address some of the challenges we have described. For example, the constitutions of some federations may have unexploited powers that federal governments could utilize to gain the necessary authority. Constitutions are often interpreted in a flexible manner by courts, in recognition of the realities of a changing world. The threat of pandemic infections could be taken strongly into consideration when courts are interpreting the use of federal powers. Of course, there are important limitations to this strategy, including the fact that any constitutional interpretation will need to strike a balance between new public health powers and respect for fundamental human rights and traditional allocations of government power, both of which may constrain the expansion of federal authority. Furthermore, heavy-handed, top-down approaches to managing disease outbreaks are not ideal, given the critical importance of local and regional public health activities. Ideally, responses would be a collaborative venture between orders of government.
To facilitate such collaboration, federal governments could enter into agreements with regional governments to ensure cooperation on matters such as the timely and adequate transfer of data. The likelihood of adherence to such agreements would be greatly enhanced if conditional funding were provided by the federal government to assist in developing the necessary surveillance and response infrastructure at the local level. This option is particularly important to consider because adequate federal powers will have no effect in the absence of adequate surveillance infrastructure or public health capacity. Governance strategies must go hand in hand with effective surveillance and the development of necessary public health capacity [22].
Conclusion
Given the importance of the IHR revision process, every effort should be made to ensure that member nations can comply with the new requirements. The size and power of several federations in the World Health Organization require that the particular nuances of their governance structure be acknowledged and respected. Failure to do so could threaten the long-term support of the IHR by key federations, such as the United States, India, and Russia, which would undermine their fundamental objective of protecting the global community. Ultimately, however, it is the responsibility of federations to make the appropriate adjustments in their approach to public health governance so that they can effectively identify, respond to, and communicate information on disease outbreaks.
This article emerged from a Canadian Institutes of Health Research funded project (grant SAR67798). KW is a Canadian Institutes of Health Research New Investigator. RU is supported by a New Investigator Award from the Canadian Institutes of Health Research and a Research Scholar Award from the Department of Family and Community Medicine, University of Toronto.
The following members of the Joint Centre for Bioethics SARS Global Health Ethics Research Group also contributed to this article: Abdallah Daar, McLaughlin Centre for Molecular Medicine, Joint Centre for Bioethics, University of Toronto; Peter A. Singer, Department of Medicine, University of Toronto and Joint Centre for Bioethics, University of Toronto; and Shawn Tracy, Primary Care Research Unit, Sunnybrook and Women's College Health Sciences Centre.
Citation: Wilson K, McDougall C, Upshur R, the Joint Centre for Bioethics SARS Global Health Ethics Research Group (2006) The new international health regulations and the federalism dilemma. PLoS Med 3(1): e1.
Abbreviations
IHRInternational Health Regulations
SARSsevere acute respiratory syndrome
WHOWorld Health Organization
==== Refs
References
World Health Organization Third report of committee A 2005 Fifty-eighth world health assembly. Agenda item 13.1. Available: http://www.who.int/gb/ebwha/pdf_files/WHA58/A58_55-en.pdf . Accessed 1 November 2005
Nicoll A Jones J Aavitsland P Giesecke J Proposed new International Health Regulations BMJ 2005 330 321 322 15705671
Watts R Watts R Definition of terms and principles of federalism Comparing federal systems in the 1990s 1996 Kingston (Canada) Institute of Intergovernmental Relations, Queen's University 6 14
Griffiths A Handbook of federal countries, 2002 2002 Montreal (Canada) McGill-Queen's University Press 528
Wilson K The complexities of multilevel governance in public health Can J Public Health 2004 95 409 412 15622787
Wilson K McCrea-Logie J Lazar H Understanding the impact of intergovernmental relations on public health: Lessons form reform initiatives in the blood system and health surveillance Can Public Policy 2004 30 177 194
Campbell A The SARS commission interim report: SARS and public health in Ontario 2004 April 15 Available: http://www.sarscommission.ca/report/Interim_Report.pdf . Accessed 1 November 2005
Fidler DP Legal issues surrounding public health emergencies Public Health Rep 2001 116 Suppl 2 79 86 11880678
Parmet WE After September 11: Rethinking public health federalism J Law Med Ethics 2002 30 201 211 12066598
Gostin LO Sapsin JW Teret SP The Model State Emergency Health Powers Act: Planning for and response to bioterrorism and naturally occurring infectious diseases JAMA 2002 288 622 628 12150674
Howse G Managing emerging infectious diseases: Is a federal system an impediment to effective laws? Aust New Zealand Health Policy 2004 1 7 15679928
Gostin LO International infectious disease law: Revision of the World Health Organization's International Health Regulations JAMA 2004 291 2623 2627 15173154
Fidler DP Emerging trends in international law concerning global infectious disease control Emerg Infect Dis 2003 9 285 290 12643821
Fidler DP SARS: Political pathology of the first post-Westphalian pathogen J Law Med Ethics 2003 31 485 505 14968652
Svoboda T Henry B Shulman L Kennedy E Rea E Public health measures to control the spread of the severe acute respiratory syndrome during the outbreak in Toronto N Engl J Med 2004 350 2352 2361 15175437
Kondro W Canadians still stung by WHO's SARS travel advisory Lancet 2003 361 1624 12747887
Bureau of International Organization Affairs. US Department of State U.S. statement for the record concerning the World Health Organization's revised International Health Regulations 2005 May 23 Available: http://www.state.gov/p/io/rls/rm/46714.htm . Accessed 3 November 2005
Hayes ET Changing notions of sovereignty and federalism in the international economic system: A reassessment of WTO regulation of federal states and the regional and local governments within their territories Northwest J Int Law Bus 2004 25 1 36
Kukucha CJ From Kyoto to the WTO: Evaluating the constitutional legitimacy of the provinces in Canadian foreign trade and environmental policy Can J Polit Sci 2005 38 129 152
Rabe BG North American federalism and climate change policy: American state and Canadian provincial policy development Widener Law J 2005 14 121 172
Eastman J Re-entering the arena: Restoring a judicial role for enforcing limitations on federal mandates Harv J Law Public Policy 2002 25 931 952
[Anonymous] Public-health preparedness requires more than surveillance Lancet 2004 364 1639 1640 15530602
|
16354103
|
PMC1315361
|
CC BY
|
2021-01-05 10:40:32
|
no
|
PLoS Med. 2006 Jan 20; 3(1):e1
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0030001
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1635410410.1371/journal.pmed.0030009Policy ForumMental HealthPediatricsPsychiatryChild PsychiatryAdolescent PsychiatryMood Disorders (Including Depression)PsychiatryPediatricsPerspectives on the Efficacy of Antidepressants for Child and Adolescent Depression Policy ForumRaz Amir Amir Raz is in the Department of Psychiatry, Division of Child and Adolescent Psychiatry, Columbia University College of Physicians and Surgeons, New York, New York, United States of America, and at the New York State Psychiatric Institute, New York, New York, United States of America. E-mail: [email protected]
Competing Interests: The author declares that he has no competing interests
1 2006 20 12 2005 3 1 e9Copyright: © 2006 Amir Raz.2006This 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.Raz looks at the ongoing controversies surrounding the use of SSRI antidepressants in children.
==== Body
Practitioners of pediatric medicine may still be undecided as to whether the newer generation of antidepressant drugs is effective for child and adolescent depression (CAD) [1]. Since 1989, when selective serotonin reuptake inhibitors (SSRIs) were introduced in the United States, they have become the top-selling drug category; as many as one in eight adult Americans having tried at least one SSRI in the past ten years. Despite their popularity in treating adult depression, the efficacy of SSRIs for CAD remains in dispute. In this article, I examine some of the core problems in medical research that have led to this disagreement.
Specificity, Safety, and Efficacy
Advances in molecular biology and neuroscience have fostered increasingly specific drugs. However, the pharmaceutical industry promotes an idea of drug specificity that may extend beyond the existing data. For example, SSRIs may selectively block the reuptake of serotonin, as claimed by many SSRI manufacturers, but they also influence numerous postsynaptic serotonin receptor systems, instigating multiple neurochemical effects. Furthermore, certain neurotransmitter systems are so tightly entwined that affecting one inevitably influences others (e.g., selective norepinepherine reuptake inhibitors also influence the serotonergic system). Hence, drugs often have effects that seem unrelated to the presumed therapeutic outcome (e.g., tricyclic antidepressants [TCAs] and SSRIs have significant effects on fast sodium channels and platelet function, respectively). And one drug can treat a variety of syndromes. For example, SSRIs are effective for symptoms ranging from obsessive-compulsive disorder to panic and anxiety. Thus, specificity, as defined by the pharmaceutical industry, is perhaps an overextended notion.
Antidepressant medications have become central to managing CAD [2]. Because double-blind trials of TCAs have failed to show greater efficacy than placebo for treating CAD [3,4], and concerns have been raised about the side effects of TCAs, SSRIs have been seen as the viable option for treating CAD [5]. Indeed, the 21st century ushered in major clinical guidelines endorsing SSRIs as first-line pharmacotherapy for CAD in both North America and the United Kingdom [6,7]. Most rigorous studies that tested the safety and efficacy of these medications in depressed adolescents began after these drugs were deemed “first line” by the professional community of child and adolescent psychiatrists.
Yet the recent history of SSRIs is replete with inconsistent verdicts about their safety. For example, in January 2003, the US Food and Drug Administration (FDA) approved fluoxetine for children and adolescents. However, about five months later, concerns arose among psychiatrists about whether the drug was associated with suicidal thinking and behavior in children and adolescents. Nevertheless, in December 2003, the UK Medicines and Healthcare Products Regulatory Agency (MHRA) supported the use of fluoxetine in children and adolescents [8,9]. It stated that for three other SSRIs (sertraline, citalopram, and escitalopram) the risks outweighed the benefits, while the balance of risks and benefits was “unassessable” for a fourth SSRI, fluvoxamine [8]. In September 2004, based on a review of 24 trials of nine different antidepressant drugs that were used to treat CAD, obsessive compulsive disorder, or “other psychiatric disorders,” the FDA also supported the use of fluoxetine in treating CAD [10]. Prior to this, on March 22, 2004, the FDA had issued a “black box” warning label on all antidepressants, cautioning that these medications may “increase the risk of suicidal thinking and behavior (suicidality) in children and adolescents with major depressive disorder (MDD) and other psychiatric disorders” [11].
In its September 2004 review, the FDA endorsed SSRI safety as well as an arbitrary improvement criterion—some decrease, relative to placebo, on a reputable scale (e.g., a two-point drop on the Hamilton depression scale)—without addressing actual clinical efficacy. However, the resulting assumption among some practitioners was that SSRIs in general are effective, as well as safe, for CAD. This conclusion is consistent with recent reviews [12,13], drawing on both published and unpublished data, reporting that at least fluoxetine is seen as a safe and efficacious treatment for CAD. The tendency to embrace SSRIs for CAD demonstrates a trend in pediatric mental health practice toward taking efficacy for granted and focusing on safety.
Statistical Significance versus Clinical Significance
Paying too much attention to significance tests and too little attention to the analysis methods [14] or other aspects of the data (e.g., the estimates of the magnitude of the effects [15]) may blur the difference between statistical significance and clinical importance [16,17]. Furthermore, statistical significance itself can be clinically meaningless.
For example, a hypothetical study with a large sample size might show that an average heart rate of 69 on placebo compared to 71 on a drug is statistically significant, but this effect is likely to be clinically meaningless. In fact, the only standard for determining drug efficacy for “quality-of-life illnesses,” including CAD, is a placebo-controlled trial; a comparator (i.e., a “horse-race,” or “drug A versus drug B”) trial typically boosts the drug effect (P. Roose, personal communication and [18]). (This is partly due to the fact that when people know that they are being treated with either a more or less potent medicine, drug response tends to be more vigorous than when they know that they may be on either an actual drug or a placebo). Finally, scientific experiments rarely control for variables known to influence drug response (e.g., expectation, suggestion, motivation, site location, and trial length) (P. Roose, unpublished data).
Clinical significance—a meaningful change in the symptomatic state or functioning of an individual patient—requires independent replication of results [19]. Different fields have different criteria for clinical significance. Additionally, the statistical method used to calculate clinical significance affects the estimates of meaningful change [14]. Indeed, some professional associations (e.g., the American Psychological Association) have deemed effect sizes and confidence intervals to be more meaningful measures than significance testing [20].
To determine clinical significance through “risk-benefit” analysis [21], one must weigh the potential benefit of improved symptomatology, accompanied by adverse side effects, against the risk of leaving the disease untreated. Trying to address the issue of clinical significance, researchers have considered such parameters as the number needed to treat (NNT), the number needed to harm, and the number needed to prevent [22]. In the case of depression and CAD, FDA approval of fluoxetine implies that the FDA considered the number needed to harm to be reasonable. But what is a good value for NNT? Ideally, it would be as close as possible to one (i.e., we need to treat only one person in order to see a desired effect in one person), but actually the NNT tends to be much higher than one [23], and it is unclear what range of values permits clinicians to conclude that a favorable benefit-to-risk ratio exists.
In addition, the NNT must be interpreted by using a comparison group. For example, in a placebo controlled trial, NNT = 3 means that on average one of three patients will derive specific benefit from the treatment above and beyond placebo, which is rarely used clinically. Thus, NNT may be more clinically meaningful as an active comparator than in relation to a placebo. The absence of clear criteria for clinical efficacy is probably partially responsible for interpretation of the same data as being both for [24] and against [25] the efficacy of a specific drug.
Lacking clear criteria for clinical significance, statistical significance is perhaps the most convenient substitute [26]. But researchers may pay too much attention to the results of significance tests, thereby overlooking clinical significance. Statistical significance may not be a sufficient criterion for recommending a drug [27]. In the case of adult antidepressants, FDA approval requires that two randomized clinical trials (RCTs) show that drug performance is at least two Hamilton-scale points better than a placebo; however, this arbitrary criterion does not signify clinical efficacy. In fact, it is unclear what criteria should be used to assess clinical significance. For example, how many studies are needed to convince a clinician that a drug is efficacious? Clinical significance relies on replication and probably requires a meta-analysis. Recent [21,50] as well as future meta-analyses, including one currently under preparation by A. Drews, I. Kirsch, and D.O. Antonuccio entitled “A Meta-Analysis of Antidepressants Trials for Depressed Children: Small Benefits, Large Stakes”, may further illuminate the efficacy of antidepressants for CAD.
Belief Systems
While strong opinions on either side of any controversy may appear extreme, it is important not to disregard these beliefs immediately. For example, on February 2, 2004, Irving Kirsch and David Antonuccio offered their testimony to the FDA on the efficacy of antidepressants for treating children with depression. At that time, only a dozen RCTs had examined the efficacy of antidepressants in CAD (four assessed SSRIs, seven assessed TCAs, and one assessed both SSRIs and TCAs) [28–38] (see sidebar). Eight of these RCTs failed to find any significant benefit of medication over placebo. While no TCA-placebo comparisons showed significant differences, four of the five SSRI-placebo RCTs (plus a fifth that included SSRIs and TCAs) claimed significant differences between drug and placebo, but only on clinician-rated, not patient-rated, measures.
Incentives for Conducting Pediatric Clinical Trials
Giving pharmaceutical companies an incentive to conduct pediatric tests, Congress passed the FDA Modernization Act in November 1997. Section 111 of this “pediatric exclusivity” partnership act offered drug sponsors six months (sometimes up to a year) of additional market exclusivity if they conducted pediatric studies on drugs still under some exclusivity provision. Under the FDA Modernization Act, the pharmaceutical company could continue to set the market price, keeping generic forms of the drug off the market. The original act has since been revised and extended through 2007 under the Best Pharmaceuticals for Children Act, but concerns still linger regarding disproportionate profits accrued by drug companies from the six-month extension compared to the cost of clinical trials, and the absence of commitment by the companies to publish or make readily available the safety and efficacy results of these trials [92]. While it is easy to see why pharmaceutical companies are eager to conduct pediatric research (six-month worldwide exclusivity for fluoxetine, for example, is estimated to be worth about a billion dollars), clinicians and patients rarely have access to these findings [93].
Since either means or standard deviations were missing in 25% of these RCTs, only nine were amenable to meta-analytic scrutiny. When Kirsch and Antonuccio combined data from these nine studies for analysis, the placebo response was 87% of the drug response, 75% of the SSRI response, and 97% of the TCA response. These results seem to indicate that TCAs have no significant pharmacological effect on CAD. They concluded that the effect of SSRIs may be statistically significant, but possibly not clinically significant.
While some psychopharmacologists dismiss investigators such as Kirsch and Antonuccio as “outliers” or inherently biased against the drug industry, such “outlying” accounts should nevertheless be examined. Opinions on either side of this issue should be considered, especially in the absence of definitive data and facts.
Market Forces
In the US, the pharmaceutical industry, Congress, and advocacy groups are known to both lobby and contribute generously to the FDA and may partially inspire its decisions and policies (P. Roose, personal communication). The influence of the pharmaceutical industry permeates science [39], and evidence points to the increasing commercial impact of biomedical research on scientific reporting [40,41]. Indeed, industry funding for research tends to yield favorable reports concerning the tested drug [42]. For example, among the authors of original research papers, reviews, and letters to the editor that were supportive of the use of specific drugs, 96% had financial relationships with the drugs' manufacturers, whereas for publications deemed neutral or critical, the figures were only 60% and 37%, respectively [43,44].
Further, since negative results are often discounted or not published [45,46,47], the message conveyed to the popular press and the public is often positively skewed [48], emphasizing benefits over risks and predicting improbable breakthroughs [49]. This trend may create unrealistic expectations about scientific advances or products and may lead to inappropriate and expensive utilization patterns. However, commercial pressure is not the only source of ambitious interpretations; another is researchers who are eager to promote their latest findings. Thus, conflicts of interest in SSRI trials for treating CAD may potentially cloud results.
Different Reviews Have Had Different Results
Although most previous reviews have been partisan, a few have presented a balanced account [1,50,51]. The first, though least comprehensive, review of RCTs conducted on newer antidepressants for CAD examined six published studies, including studies on venlafaxine and three SSRIs: fluoxetine (three studies), sertraline (one study), and paroxetine (one study) [52]. In their limited meta-analysis, the authors used a random effects model to pool averaged, selected outcomes across the five SSRI studies. They found a small effect size of 0.26 (95% confidence interval [CI], 0.13–0.40), which they described as equivalent to a three-to-four-point improvement on the revised children's depression rating scale (which ranges from 17 to 113). They concluded that a large benefit from newer antidepressant drugs is unlikely. Reviewing the same studies, another report judged the efficacy data to be inconclusive [53].
Since the above reviews [52,53] included in their analysis a negative study of fluoxetine which involved a small (n = 30), clinically heterogeneous (mixed inpatients and outpatients) participant group [39], some later reviews opted to exclude this negative study from their analyses. For example, one study analyzed five of the six published papers addressed in the above-mentioned reviews [52,53], as well as data from six unpublished studies accessed through collaboration with the UK's MHRA [50]. These unpublished studies included two investigations of venlafaxine, paroxetine, and citalopram, respectively (one of the citalopram studies was subsequently published [54]). After extracting raw data for outcome measures, including remission, response to treatment, and depressive symptoms scores, the authors reanalyzed, and—where possible—meta-analyzed published and unpublished studies of each drug, using fixed-effects and random-effects models. The authors concluded that only fluoxetine had evidence of efficacy that was robust: across two published trials, fluoxetine was more likely than placebo to bring about remission (number needed to benefit 6; 95% CI, 4–15), or a clinically meaningful response (number needed to benefit 5; 95% CI, 4–13). The study did not draw on unpublished fluoxetine efficacy data.
The most recent systematic review of newer antidepressants for CAD to date [55] included all of the data previously reviewed [50,52,53] as well as unpublished data regarding two RCTs of nefazodone, and the recently published Treatment for Adolescents with Depression Study (TADS) [23], which compared fluoxetine, placebo, and cognitive behavioral therapy (CBT) alone and in combination (see Table 1). After examining efficacy outcomes and the influence of methodology (site selection, study population, study design, and outcome measures) on those outcomes, the authors found that the more methodologically sound SSRI studies tended to have better outcomes. For example, in the TADS, 60.6% (95% CI, 51%–70%) of adolescents responded to fluoxetine alone, as opposed to a 34.8 (95% CI, 26%– 44%) response rate for placebo. Based on these findings and no evidence of differences among these drugs in adult populations, the authors concluded that most newer antidepressants are likely effective, and that different results have been largely due to methodological differences in studies of CAD. They concluded, therefore, that at least fluoxetine is clinically effective for CAD.
Table 1 A List of Published and Unpublished Sources Commonly Cited in Recent Reviews Reporting on SSRIs for Child and Adolescent Depression
aNot included meta-analysis by [52].
b Two separate trials analyzed and published together.
cReviewed prior to publication (draft differed slightly from published paper).
This table is an updated version of Table 1 in [13]. Only the more recent studies account for unpublished data, which are important to the formulation of a comprehensive and informed opinion regarding drug efficacy and safety.
Methodologic Oversights in Published Studies
The TADS concluded that “medical management of MDD with fluoxetine, including careful monitoring for adverse events, should be made widely available” [23]. However, some researchers disagree with this conclusion [25].
The TADS explores four experimental arms: placebo, fluoxetine, CBT, and CBT + fluoxetine. Some of these treatment groups were “blind” (i.e., participants were unaware of what treatment they were receiving, as in the case for placebo) while others were informed (as in the case for CBT). However, Table 2 shows the difficulty of interpreting a study that compares blinded and unblinded treatment groups (e.g., differences between treatment groups might be due to the varying influence of expectation). Thus, in the interpretation of the TADS findings, subtle methodological caveats go unrecognized.
Table 2 A Subtle Flaw in the TADS: Condition I versus Condition II
In the TADS [23], one of the better studies in the field, half the participants knew what treatment they were getting. The TADS treatment comparisons report differences between placebo (blind) versus fluoxetine (blind); CBT (not blind) versus fluoxetine (blind); and CBT plus fluoxetine (both not blind) versus fluoxetine (blind). It is difficult to interpret findings comparing a blind arm of an experiment with one that is not blind. Given that it is difficult to blind CBT [96], critical comparisons such as that of CBT plus fluoxetine versus fluoxetine (all not blind) were not included in the TADS study, although perhaps they should have been.
Relative to other studies, CBT does not fare well compared to both placebo and fluoxetine in the TADS. However, other studies comparing antidepressants and CBT showed that both were moderately effective in relieving depression in adults [56]. Adult neuroimaging findings suggest that the two methods work by improving the functioning of different brain circuits: CBT operates on cortical areas related to attention and comprehension, including the anterior cingulate cortex, whereas antidepressants operate on subcortical areas. Applying the adult data to children, these exploratory imaging results may provide both a context for testing drugs against nonpharmacological therapy and a basis for considering how to treat or even prevent depression in those who are most susceptible [57].
Despite being regarded as efficacious in adults, meta-analysis of published RCTs indicates that 75% of antidepressant response in adults is duplicated by placebo [58]. This initial meta-analysis was amply critiqued by Klein for multiple limitations [59]. However, follow-up analyses using a different data set taken from the FDA, to which Klein's objections do not apply, again reported that about 80% of the response to antidepressants in adults was duplicated in placebo control groups [60,61]. Together with the notion that antidepressant medication effects are typically weaker in children than adults [58,60], these conclusions accord with earlier reviews that challenge the effects of antidepressants in CAD [4,62–67]. Merely labeling a pill an antidepressant does not make it so. In fact, the existing data suggest that antidepressants are probably more effective in treating anxiety than depression [68].
Another limitation draws on the implications of using antidepressants in early life [69]. Serotonin acts as a brain and glial growth factor in early development. Some serotonin receptors act during development to establish normal anxiety-like behavior later in life, while others play a role in synapse formation [70,71]. Some exploratory findings suggest that artificial perturbation of serotonin function in early life may alter the normal development of brain systems related to stress, motor development, and motor control [71,72]. Furthermore, early exposure to fluoxetine produced abnormal emotional behaviors in adult mice [69]. The critical role of serotonin in the maturation of brain systems that modulate emotional function in the adult suggests that, in concert with genetic makeup, low serotonin levels during early development may increase vulnerability to psychiatric disorders [69,73]. Early exposure to SSRIs, therefore, can potentially exact a heavy price in later life [74,75].
These caveats suggest that, in addition to the potential implications of using antidepressants in early life, there are few compelling data sets, free of funding from drug companies, concerning the efficacy of antidepressant medications over and above their placebo value for CAD. These caveats also show the difficulty of assessing the clinical significance of the unique effects attributable to antidepressant medications. Because antidepressants work just slightly better than placebos, even according to data endorsed by the pharmaceutical industry, the image of antidepressants as more effective may be overreaching and perhaps a consequence of methodological artifacts [76].
Public Health versus Individual Decisions
The US Surgeon General makes decisions based on the greater good of a vast population: a mere two-point improvement on the Hamilton depression scale may constitute a meaningful public-health benefit (D. Shaffer, unpublished data). However, parents decide whether their depressed adolescent child should receive CBT, fluoxetine, or start a vigorous exercise regime, based on an individually tailored risk-benefit analysis.
Most clinicians recommend psychotherapy for mild to moderate CAD and reserve SSRIs for severe CAD or when therapy is not effective [21]. Yet given the large numbers of people suffering from depressive disorders (i.e., an estimated 1.5 million adolescents (12–18 years of age) with MDD in the US alone [77]), it is easy to see why offering therapy would be difficult (e.g., number of therapists and insurance considerations), thereby making the drug option more popular.
Despite early suggestions in the literature [78–80], accounts of the association between adult suicidality and the use of SSRIs have been inconclusive [81–83]. One early meta-analysis showed that SSRIs potentially decreased suicidal ideation as measured by a single question on the Hamilton depression score [83], but a more recent study reported a non-significant increase in suicide rates between patients assigned to SSRIs and those assigned to placebo or other antidepressants [82]. Despite the FDA's black-box label, some accounts suggest that since the introduction of SSRIs the number of successful suicides has steadily declined [84,85]. One recent meta-analysis reported that SSRIs increased the risk of suicide attempts, but not completions, across all indications [86]. Another, a meta-analysis of drug company data that were submitted to the MHRA's safety review, reported that SSRIs did not appear to increase the risk of suicide attempts or thoughts [87].
Data regarding SSRIs and youth suicide are sparser, but no less controversial [12,13,88,89]. One study reported an inverse relationship between regional change in use of antidepressants and suicide [90]. Nonetheless, the highest possible standard should be applied to scientific data involving drug treatment of children because they are essentially involuntary patients: when a medication is prescribed for a young child, the adult caregiver ensures that the child takes the medication, regardless of the child's own desires. Yet, studies of adolescent compliance with medication treatment report notoriously low compliance outside of the controlled settings of clinical trials.
The sparse RCT findings suggest that improvement may not always be clinically significant. When evaluating a medication with side effects, potential clinical implications for later life, and questionable effectiveness, it should be compared to interventions such as exercise and CBT, which have shown some therapeutic effects on depression without medical side effects and risks [91]. Originally celebrated but recently disparaged by modern psychiatry, therapeutic rapport may prove more clinically significant than drug specificity.
Conclusions
Given all of these limitations, patients and physicians should demand stronger evidence for the efficacy of antidepressants for CAD.
Some advocates assert that rather than using medication with side effects and low effectiveness, children should be offered interventions that produce therapeutic effects on depression without the medical side effects and associated risks [91]. However, clinicians and laypeople must apply comparable standards for evaluating the efficacy of drug and psychotherapy data. Whereas medical drug research occurs in a formally regulated, albeit imperfect, environment, safety and efficacy in psychotherapy research are largely unregulated. Moreover, unlike drug assays, psychotherapy studies do not typically report adverse events, their meta-analyses are sparse, and their experimental design lacks a placebo condition (see Table 2).
Finally, although antidepressants undoubtedly affect brain biochemistry, interpreting these neural changes is controversial, and a risk-benefit analysis of side effects and long-term health risks may cast a long shadow on the current preference for antidepressants as first-line treatment for CAD. Only more studies, and the passage of enough time to examine the putative long-term effects, will determine the efficacy of antidepressants in CAD. Clinicians, patients, families, and the public should be cognizant of these issues and exercise critical judgment as they make informed decisions.
Citation: Raz A (2006) Perspectives on the efficacy of antidepressants for child and adolescent depression. PLoS Med 3(1): e9.
Abbreviations
CADchild and adolescent depression
CBTcognitive behavioral therapy
CIconfidence interval
FDAFood and Drug Administration
MDDmajor depressive disorder
MHRAMedicines and Healthcare Products Regulatory Agency
RCTrandomized clinical trial
SSRIselective serotonin reuptake inhibitor
TADSTreatment for Adolescents with Depression Study
TCAtricyclic antidepressant
==== Refs
References
Vitiello B Swedo S Antidepressant medications in children N Engl J Med 2004 350 1489 1491 15071123
Wong IC Besag FM Santosh PJ Murray ML Use of selective serotonin reuptake inhibitors in children and adolescents Drug Saf 2004 27 991 1000 15471506
Geller B Reising D Leonard HL Riddle MA Walsh BT Critical review of tricyclic antidepressant use in children and adolescents J Am Acad Child Adolesc Psychiatry 1999 38 513 516 10230182
Hazell P O'Connell D Heathcote D Robertson J Henry D Efficacy of tricyclic drugs in treating child and adolescent depression: A meta-analysis BMJ 1995 310 897 901 7719178
Emslie GJ Walkup JT Pliszka SR Ernst M Nontricyclic antidepressants: Current trends in children and adolescents J Am Acad Child Adolesc Psychiatry 1999 38 517 528 10230183
Park RJ Goodyer IM Clinical guidelines for depressive disorders in childhood and adolescence Eur Child Adolesc Psychiatry 2000 9 147 161 11095037
[Anonymous] Summary of the practice parameters for the assessment and treatment of children and adolescents with depressive disorders J Am Acad Child Adolesc Psychiatry 1998 37 1234 1238 10075518
United Kingdom Committee on Safety of Medicines Medicines and Healthcare Products Regulatory Agency Use of selective serotonin reuptake inhibitors (SSRIs) in children and adolescents with major depressive disorder (MDD): Only fluoxetine (Prozac) shown to have a favourable balance of risks and benefits for the treatment of MDD in the under 18s 2003 London United Kingdom Committee on Safety of Medicines, Medicines and Healthcare Products Regulatory Agency Available: http://www.focusproject.org.uk/pooled/articles/BF_NEWSART/view.asp?Q=BF_NEWSART_83376 . Accessed 9 November 2005
National Institute for Health and Clinical Excellence Depression in children: Identification and management of depression in children and young people in primary, community and secondary care 2005 London National Institute for Health and Clinical Excellence Available: http://www.nice.org.uk/pdf/Depn_child_2ndcons_Fullguideline.pdf . Accessed 9 November 2005
US Food and Drug Administration FDA launches a multi-pronged strategy to strengthen safeguards for children treated with antidepressant medications 2004 October 15 Rockville (Maryland) Food and Drug Administration News Available: http://www.fda.gov/bbs/topics/news/2004/NEW01124.html . Accessed 9 November 2005
Center for Drug Evaluation and Research USFDA Worsening depression and suicidality in patients being treated with antidepressant medications 2004 March 22 Rockville (Maryland) Food and Drug Administration Available: http://www.fda.gov/cder/drug/antidepressants/AntidepressanstPHA.htm . Accessed 9 November 2005
Whittington CJ Kendall T Pilling S Are the SSRIs and atypical antidepressants safe and effective for children and adolescents? Curr Opin Psychiatry 2005 18 21 25 16639179
Cheung A Emslie GJ Taryn ML Efficacy and safety of antidepressants in youth depression J Can Acad Adolesc Psychiatry 2004 13 98 104
Bauer S Lambert MJ Nielsen SL Clinical significance methods: A comparison of statistical techniques J Pers Assess 2004 82 60 70 14979835
Rutledge T Loh C Effect sizes and statistical testing in the determination of clinical significance in behavioral medicine research Ann Behav Med 2004 27 138 145 15026298
Hojat M Xu G A visitor's guide to effect sizes: Statistical significance versus practical (clinical) importance of research findings Adv Health Sci Educ Theory Pract 2004 9 241 249 15316274
Glaros AG Statistical and clinical significance: Alternative methods for understanding the importance of research findings J Ir Dent Assoc 2004 50 128 131 15481525
Roose SP Sackeim HA Clinical trials in late-life depression: Revisited Am J Geriatr Psychiatry 2002 10 503 505 12213683
Bhardwaj SS Camacho F Derrow A Fleischer AB Feldman SR Statistical significance and clinical relevance: The importance of power in clinical trials in dermatology Arch Dermatol 2004 140 1520 1523 15611433
Cohen J The earth is round (p < .05) Am Psychol 1990 49 997 1003
Haby MM Tonge B Littlefield L Carter R Vos T Cost-effectiveness of cognitive behavioural therapy and selective serotonin reuptake inhibitors for major depression in children and adolescents Aust N Z J Psychiatry 2004 38 579 591 15298580
Laupacis A Sackett DL Roberts RS An assessment of clinically useful measures of the consequences of treatment N Engl J Med 1988 318 1728 1733 3374545
The Treatment for Adolescents with Depression Study (TADS): Demographic and clinical characteristics J Am Acad Child Adolesc Psychiatry 2005 44 28 40 15608541
Carroll BJ Adolescents with depression JAMA 2004 292 2578
Antonuccio D Burns D Adolescents with depression JAMA 2004 292 2577
Gigerenzer G Keren G Lewis C The superego, the ego, and the id in statistical reasoning A handbook for data analysis in the behavioral sciences: Methodological issues 1993 Hillsdale (New Jersey) Erlbaum 311 339
Killeen PR An alternative to null-hypothesis significance tests Psychol Sci 2005 16 345 353 15869691
Boulos C Kutcher S Marton P Simeon J Ferguson B Response to desipramine treatment in adolescent major depression Psychopharmacol Bull 1991 27 59 65 1862207
Emslie GJ Rush AJ Weinberg WA Kowatch RA Hughes CW A double-blind, randomized, placebo-controlled trial of fluoxetine in children and adolescents with depression Arch Gen Psychiatry 1997 54 1031 1037 9366660
Emslie GJ Heiligenstein JH Wagner KD Hoog SL Ernest DE Fluoxetine for acute treatment of depression in children and adolescents: A placebo-controlled, randomized clinical trial J Am Acad Child Adolesc Psychiatry 2002 41 1205 1215 12364842
Geller B Cooper TB Graham DL Marsteller FA Bryant DM Double-blind placebo-controlled study of nortriptyline in depressed adolescents using a “fixed plasma level” design Psychopharmacol Bull 1990 26 85 90 2196631
Geller B Cooper TB Graham DL Fetner HH Marsteller FA Pharmacokinetically designed double-blind placebo-controlled study of nortriptyline in 6- to 12-year-olds with major depressive disorder J Am Acad Child Adolesc Psychiatry 1992 31 34 44 1537779
Keller MB Ryan ND Strober M Klein RG Kutcher SP Efficacy of paroxetine in the treatment of adolescent major depression: A randomized, controlled trial J Am Acad Child Adolesc Psychiatry 2001 40 762 772 11437014
Kramer AD Feiguine RJ Clinical effects of amitriptyline in adolescent depression: a pilot study J Am Acad Child Psychiatry 1981 20 636 644 7031113
Kutcher S Boulos C Ward B Marton P Simeon J Response to desipramine treatment in adolescent depression: A fixed-dose, placebo-controlled trial J Am Acad Child Adolesc Psychiatry 1994 33 686 694 8056732
Preskorn SH Weller EB Hughes CW Weller RA Bolte K Depression in prepubertal children: Dexamethasone nonsuppression predicts differential response to imipramine vs. placebo Psychopharmacol Bull 1987 23 128 133 3602308
Puig-Antich J Perel JM Lupatkin W Chambers WJ Tabrizi MA Imipramine in prepubertal major depressive disorders Arch Gen Psychiatry 1987 44 81 89 3541830
Simeon JG Dinicola VF Ferguson HB Copping W Adolescent depression: A placebo-controlled fluoxetine treatment study and follow-up Prog Neuropsychopharmacol Biol Psychiatry 1990 14 791 795 2293257
Ferner RE The influence of big pharma BMJ 2005 330 855 856 15831847
Caulfield T The commercialisation of medical and scientific reporting PLoS Medicine 2004 1 e38 10.1371/journal.pmed.0010038 15630461
Antonuccio DO Danton WG McClanahan TM Psychology in the prescription era: Building a firewall between marketing and science Am Psychol 2003 58 1028 1043 14664691
Bhandari M Busse JW Jackowski D Montori VM Schunemann H Association between industry funding and statistically significant pro-industry findings in medical and surgical randomized trials CMAJ 2004 170 477 480 14970094
Stelfox HT Chua G O'Rourke K Detsky AS Conflict of interest in the debate over calcium-channel antagonists N Engl J Med 1998 338 101 106 9420342
van Kolfschooten F Conflicts of interest: Can you believe what you read? Nature 2002 416 360 363 11919595
Cassels A Hughes MA Cole C Mintzes B Lexchin J Drugs in the news: An analysis of Canadian newspaper coverage of new prescription drugs CMAJ 2003 168 1133 1137 12719316
Koren G Klein N Bias against negative studies in newspaper reports of medical research JAMA 1991 266 1824 1826 1890712
Lenzer J Drug secrets: What the FDA isn't telling 2005 Available: http://www.slate.com/id/2126918 . Accessed 1 November 2005
Bubela TM Caulfield TA Do the print media “hype” genetic research? A comparison of newspaper stories and peer-reviewed research papers CMAJ 2004 170 1399 1407 15111473
Nelkin D Beyond risk. Reporting about genetics in the post-Asilomar press Perspect Biol Med 2001 44 199 207 11370154
Whittington CJ Kendall T Fonagy P Cottrell D Cotgrove A Selective serotonin reuptake inhibitors in childhood depression: Systematic review of published versus unpublished data Lancet 2004 363 1341 1345 15110490
National Institute for Health and Clinical Excellence Depression in children: Identification and management of depression in children and young people in primary, community and secondary care. Second draft for consultation 2005 London National Institute for Health and Clinical Excellence Available: http://www.nice.org.uk/pdf/Scope_Depression_Child.pdf ; http://www.nice.org.uk/pdf/Depn_child_2ndcons_%20App_P.pdf . Accessed 15 November 2005
Jureidini JN Doecke CJ Mansfield PR Haby MM Menkes DB Efficacy and safety of antidepressants for children and adolescents BMJ 2004 328 879 883 15073072
Courtney DB Selective serotonin reuptake inhibitor and venlafaxine use in children and adolescents with major depressive disorder: A systematic review of published randomized controlled trials Can J Psychiatry 2004 49 557 563 15453105
Wagner KD Robb AS Findling RL Jin J Gutierrez MM A randomized, placebo-controlled trial of citalopram for the treatment of major depression in children and adolescents Am J Psychiatry 2004 161 1079 1083 15169696
Cheung AH Emslie GJ Mayes TL Review of the efficacy and safety of antidepressants in youth depression J Child Psychol Psychiatry 2005 46 735 754 15972068
Mayberg HS Modulating dysfunctional limbic-cortical circuits in depression: Towards development of brain-based algorithms for diagnosis and optimised treatment Br Med Bull 2003 65 193 207 12697626
Posner MI Rothbart MK Influencing brain networks: Implications for education Trends Cogn Sci 2005 9 99 103 15737817
Kirsch I Sapirstein G Listening to Prozac but hearing placebo: A meta analysis of antidepressant medication Prev Treat 1 1998 Available: http://www.journals.apa.org/prevention/volume1/pre0010002a.html . Accessed 9 November 2005
Klein DF Listening to meta-analysis but hearing bias Prev Treat 1 1998 Available: http://www.journals.apa.org/prevention/volume1/pre0010006c.html . Accessed 9 November 2005
Kirsch I Moore TJ Scoboria A Nicholls SS The emperor's new drugs: An analysis of antidepressant medication data submitted to the U.S. Food and Drug Administration Prev Treat 5 2002 Available: http://www.journals.apa.org/prevention/volume5/pre0050023a.html . Accessed 9 November 2005
Moncrieff J Kirsch I Efficacy of antidepressants in adults BMJ 2005 331 155 157 16020858
Dujovne VF Barnard MU Rapoff MA Pharmacological and cognitive-behavioral approaches in the treatment of childhood depression: A review and critique Clin Psychol Rev 1995 15 589
Sommers-Flanagan J Sommers-Flanagan R Efficacy of antidepressant medication with depressed youth: What psychologists should know Prof Psychol Res Pr 1996 27 145 153
Michael KD Crowley SL How effective are treatments for child and adolescent depression? A meta-analytic review Clin Psychol Rev 2002 22 247 11806021
Kastelic EA Labellarte MJ Riddle MA Selective serotonin reuptake inhibitors for children and adolescents Curr Psychiatry Rep 2000 2 117 123 11122943
Ambrosini PJ Bianchi MD Rabinovich H Elia J Antidepressant treatments in children and adolescents. I. Affective disorders J Am Acad Child Adolesc Psychiatry 1993 32 1 6 8428861
Fisher RL Fisher S Antidepressants for children. Is scientific support necessary? J Nerv Ment Dis 1996 184 99 102 8596118
Schmitt R Gazalle FK Lima MS Cunha A Souza J The efficacy of antidepressants for generalized anxiety disorder: A systematic review and meta-analysis Rev Bras Psiquiatr 2005 27 18 24 15867979
Ansorge MS Zhou M Lira A Hen R Gingrich JA Early-life blockade of the 5-HT transporter alters emotional behavior in adult mice Science 2004 306 879 881 15514160
Hariri AR Mattay VS Tessitore A Kolachana B Fera F Serotonin transporter genetic variation and the response of the human amygdala Science 2002 297 400 403 12130784
Hariri AR Drabant EM Munoz KE Kolachana BS Mattay VS A susceptibility gene for affective disorders and the response of the human amygdala Arch Gen Psychiatry 2005 62 146 152 15699291
Casper RC Fleisher BE Lee-Ancajas JC Gilles A Gaylor E Follow-up of children of depressed mothers exposed or not exposed to antidepressant drugs during pregnancy J Pediatr 2003 142 402 408 12712058
Caspi A Sugden K Moffitt TE Taylor A Craig IW Influence of life stress on depression: Moderation by a polymorphism in the 5-HTT gene Science 2003 301 386 389 12869766
Coyle JT Popper C Biochemical development of the brain: Neurotransmitters and child psychiatry Psychiatric pharmacosciences of children and adolescents 1997 Washington (District of Columbia) American Psychiatric Press 3 25
Coyle JT Psychotropic drug use in very young children JAMA 2000 283 1059 1060 10697069
Moncrieff J The antidepressant debate Br J Psychiatry 2002 180 193 194 11872507
Rifkin A Rifkin W Adolescents with depression JAMA 2004 292 2577 2578
Masand P Gupta S Dewan M Suicidal ideation related to fluoxetine treatment N Engl J Med 1991 324 420
Rothschild AJ Locke CA Reexposure to fluoxetine after serious suicide attempts by three patients: The role of akathisia J Clin Psychiatry 1991 52 491 493 1752848
Teicher MH Glod C Cole JO Emergence of intense suicidal preoccupation during fluoxetine treatment Am J Psychiatry 1990 147 207 210 2301661
Jick H Kaye JA Jick SS Antidepressants and the risk of suicidal behaviors JAMA 2004 292 338 343 15265848
Khan A Khan S Kolts R Brown WA Suicide rates in clinical trials of SSRIs, other antidepressants, and placebo: Analysis of FDA reports Am J Psychiatry 2003 160 790 792 12668373
Baldwin D Bullock T Montgomery D Montgomery S 5-HT reuptake inhibitors, tricyclic antidepressants and suicidal behaviour Int Clin Psychopharmacol 1991 6 Suppl 3 49 55 1839632
Grunebaum MF Ellis SP Li S Oquendo MA Mann JJ Antidepressants and suicide risk in the United States, 1985–1999 J Clin Psychiatry 2004 65 1456 1462 15554756
Healy D Whitaker C Antidepressants and suicide: Risk-benefit conundrums J Psychiatry Neurosci 2003 28 331 337 14517576
Fergusson D Doucette S Glass KC Shapiro S Healy D Association between suicide attempts and selective serotonin reuptake inhibitors: Systematic review of randomised controlled trials BMJ 2005 330 396 15718539
Gunnell D Saperia J Ashby D Selective serotonin reuptake inhibitors (SSRIs) and suicide in adults: Meta-analysis of drug company data from placebo controlled, randomised controlled trials submitted to the MHRA's safety review BMJ 2005 330 385 15718537
Caballero J Nahata MC Selective serotonin-reuptake inhibitors and suicidal ideation and behavior in children Am J Health Syst Pharm 2005 62 864 867 15821279
Licinio J Wong ML Depression, antidepressants and suicidality: A critical appraisal Nat Rev Drug Discov 2005 4 165 171 15688079
Olfson M Shaffer D Marcus SC Greenberg T Relationship between antidepressant medication treatment and suicide in adolescents Arch Gen Psychiatry 2003 60 978 982 14557142
Clarke GN Rohde P Lewinsohn PM Hops H Seeley JR Cognitive-behavioral treatment of adolescent depression: Efficacy of acute group treatment and booster sessions J Am Acad Child Adolesc Psychiatry 1999 38 272 279 10087688
[Anonymous] Waxman wants release of study summaries Washington Drug Letter 2004 July 26 Available: http://www.fdanews.com/wdl/36_29/capitolhill/27380-1.html . Accessed 2 November 2005
[Anonymous] Waxman wants withheld data on pediatric depressants Drug Industry Daily Online 2004 July 21 Available: http://www.fdanews.com/did/3_141/capitolhill/27248-1.html . Accessed 2 November 2005
Mandoki MW Tapia MR Tapia MA Sumner GS Parker JL Venlafaxine in the treatment of children and adolescents with major depression Psychopharmacol Bull 1997 33 149 154 9133767
Wagner KD Ambrosini P Rynn M Wohlberg C Yang R Efficacy of sertraline in the treatment of children and adolescents with major depressive disorder: Two randomized controlled trials JAMA 2003 290 1033 1041 12941675
Kirsch I Placebo psychotherapy: Synonym or oxymoron? J Clin Psychol 2005 61 791 803 15827992
|
16354104
|
PMC1315362
|
CC BY
|
2021-01-05 10:40:32
|
no
|
PLoS Med. 2006 Jan 20; 3(1):e9
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0030009
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1635410510.1371/journal.pmed.0030022Research ArticleGenetics/Genomics/Gene TherapyPhysiologyCardiology/Cardiac SurgeryDiabetes/Endocrinology/MetabolismStatisticsBiochemistryCardiovascular MedicineDiabetesEndocrinologyGeneticsNutrition and MetabolismEndothelial Lipase Concentrations Are Increased in Metabolic Syndrome and Associated with Coronary Atherosclerosis EL and Metabolic SyndromeBadellino Karen O
1
*Wolfe Megan L
1
Reilly Muredach P
1
Rader Daniel J
1
1Schools of Nursing and Medicine, Institute for Translational Medicine and Therapeutics, University of Pennsylvania, Philadelphia, Pennsylvania, United States of AmericaQuertermous Tom Academic EditorStanford UniversityUnited States of America* To whom correspondence should be addressed. E-mail: [email protected]
Competing Interests: The authors have declared that no competing interests exist.
Author Contributions: KOB and DJR designed the study. KOB and DJR performed the experiments. KOB, MLW, and MPR analyzed the data. MLW and DJR enrolled patients. KOB, MPR, and DJR contributed to writing the paper.
2 2006 20 12 2005 3 2 e2214 4 2005 26 10 2005 Copyright: © 2006 Badellino et al.2006This 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.
Endothelial Lipase: A New Risk Factor for Atherosclerosis?
Background
Endothelial lipase (EL), a new member of the lipase family, has been shown to modulate high-density lipoprotein (HDL-C) metabolism and atherosclerosis in mouse models. We hypothesized that EL concentrations would be associated with decreased HDL-C and increased atherosclerosis in humans.
Methods and Findings
Healthy individuals with a family history of premature coronary heart disease (n = 858) were recruited as part of the Study of the Inherited Risk of Atherosclerosis. Blood was drawn in the fasting state before and, in a subgroup (n = 510), after administration of a single dose of intravenous heparin. Plasma lipids were measured enzymatically, lipoprotein subclasses were assessed by nuclear magnetic resonance, and coronary artery calcification (CAC) was quantified by electron beam computed tomography. Plasma EL mass was measured using a newly developed enzyme-linked immunosorbent assay. Median EL mass in pre-heparin plasma was 442 (interquartile range = 324–617) ng/ml. Median post-heparin mass was approximately 3-fold higher, 1,313 (888–1,927) ng/ml. The correlation between pre-heparin EL mass and post-heparin EL mass was 0.46 (p < 0.001). EL mass concentrations in both pre- and post-heparin plasma significantly correlated with all NCEP ATPIII-defined metabolic syndrome factors: waist circumference (r = 0.28 and 0.22, respectively, p < 0.001 for each), blood pressure (r = 0.18 and 0.24, p < 0.001 for each), triglycerides (r = 0.22, p < 0.001; and 0.13, p = 0.004), HDL cholesterol (r = –0.11, p = 0.002; and –0.18, p < 0.001), and fasting glucose (r = 0.11 and 0.16, p = 0.001 for both). EL mass in both routine (odds ratio [OR] = 1.67, p = 0.01) and post-heparin (OR = 2.42, p = 0.003) plasma was associated with CAC as determined by ordinal regression after adjustment for age, gender, waist circumference, vasoactive medications, hormone replacement therapy (women), and established cardiovascular risk factors.
Conclusions
We report, to our knowledge for the first time, that human plasma EL concentrations, in both post-heparin and routine pre-heparin plasma, are significantly associated with metabolic syndrome features and with subclinical atherosclerosis. EL may be a pro-atherogenic factor in humans, especially in overweight individuals and those with metabolic syndrome.
Studies in rodents had suggested that endothelial lipase might play a role in the development of atherosclerosis. This study finds an association between EL levels and early-stage atherosclerosis in humans.
==== Body
Introduction
Two members of the lipase family, lipoprotein lipase (LPL) and hepatic lipase (HL) are known to influence plasma lipoprotein metabolism and risk of atherosclerosis in humans [1–5]. Endothelial lipase (EL) is a more recently discovered member of this family of lipases [6,7]. In contrast to LPL and HL, it is expressed in endothelial cells and, in the majority of reports, has relatively more phospholipase activity than the other two enzymes. Similar to HL and LPL, it is secreted and binds to the endothelial surface. In vitro, EL effectively hydrolyzes high-density lipoprotein (HDL) phospholipids. In mice, EL has a major influence on HDL metabolism. Adenovirally mediated overexpression of human EL in mice dramatically reduced HDL cholesterol (HDL-C) concentrations [6], shown to be due to rapid catabolism of HDL [8]. Transgenic overexpression of human EL was shown to reduce HDL-C concentrations as well [9]. Conversely, antibody inhibition of EL in mice significantly increased HDL-C concentrations [10], and EL knockout mice have a significant increase in HDL-C concentrations [9,11]. Furthermore, EL knockout mice crossed onto the apolipoprotein E knockout background have decreased atherosclerosis [12]. Based on these studies in mice, it has been suggested that EL may be an important risk factor and may affect the development of atherosclerosis in humans.
The relationship of EL to variation in lipoprotein concentrations and atherosclerosis in humans has not been reported. Several genetic association studies have suggested that rare variants and common polymorphisms in the human EL gene might be associated with variation in HDL-C concentrations [11,13,14], but they have not been conclusive and were not accompanied by direct measurement of EL concentrations. We hypothesized that concentrations of EL are inversely associated with HDL-C concentrations and directly associated with the metabolic syndrome and atherosclerosis in humans. To test this, we developed an immunoassay for human EL and used it to measure EL mass concentrations in both pre-heparin and post-heparin plasma samples in individuals enrolled in the Study of the Inherited Risk of Atherosclerosis (SIRCA). We determined the association of EL concentrations with lipoprotein concentrations, other cardiovascular risk factors and coronary artery calcification (CAC), a noninvasive measure of coronary atherosclerosis [15–17].
Methods
Study Protocol
SIRCA is a cross-sectional study of asymptomatic individuals and their families designed to investigate novel biomarkers and genetic factors associated with coronary atherosclerosis. The study design and initial findings have previously been published [18,19]. Briefly, persons were eligible for SIRCA if they had a family history of premature coronary artery disease (CAD), were free of clinical CAD, and were men, 20–75 years of age or women, 30–75 years. Exclusion criteria included other major CAD risk factors: known diabetes, total cholesterol higher than 300 mg/dl, cigarette smoking of one pack or more per day, or blood pressure higher than 160/100 mm Hg. The University of Pennsylvania Institutional Review Board approved the study protocol. Informed consent was obtained from each participant. This report focuses on 858 random unrelated individuals from the SIRCA in whom plasma EL concentrations were measured.
Evaluated Parameters
Participants were assessed at the General Clinical Research Center in the University of Pennsylvania Medical Center after a 12-h overnight fast. A questionnaire was administered; height, weight, waist circumference, and blood pressure were obtained; and 30 ml of whole blood was drawn (“pre-heparin” plasma) into EDTA-containing tubes. In a subset of individuals (n = 510), blood was drawn 10 min after intravenous administration of heparin (60 units per kilogram body weight). After the blood was centrifuged, the plasma was removed and stored at –80 °C until use (“post-heparin” plasma). Plasma total cholesterol, HDL cholesterol, and triglyceride concentrations were measured enzymatically on a Cobas Fara II (Roche Diagnostic Systems, Indianapolis, Indiana, United States) using Sigma reagents (Sigma, St. Louis, Missouri, United States) in a Centers for Disease Control-standardized lipid laboratory. LDL cholesterol was calculated using the Friedewald formula.
Post-heparin plasma samples in EDTA were analyzed for lipoprotein subclasses by nuclear magnetic resonance (NMR) [20–22] (LipoSciences, Raleigh, North Carolina, United States). The NMR method uses the characteristic proton signals broadcast by lipoprotein subclasses to determine both the size and the quantity of each size of lipoprotein. A summary of the reporting format can be found at http://www.liposcience.com. The ten subclasses (and their size ranges) routinely used are: large very low-density lipoprotein (VLDL) (60–200 nm), intermediate VLDL (35–60 nm), small VLDL (27–35 nm), IDL (23–27 nm), large low-density lipoprotein (LDL) (21.3–23 nm), intermediate LDL (19.8–21.2 nm), small LDL (18.3–19.7 nm), large HDL (8.8–13.0 nm), intermediate HDL (8.2–8.8 nm), and small HDL (7.3–8.2 nm).
Electron beam computed tomography was performed on an Imatron C-150 XP/LXP Evolution scanner (Imatron, San Francisco, California, United States) connected to a MagicView workstation for volumetric image reconstruction. Using ECG triggering with the individual in a supine position, 40 contiguous, 3 mm-thick axial slices were acquired just below the carina through the apex of the heart. Global CAC scores were determined according to the method of Agatston [23].
Sandwich Enzyme-Linked Immunosorbent Assay for Human EL
Detailed information on the development and quality control of the sandwich ELISA can be found in Protocol S1. The wells of a 96-well microtiter plate were coated with rabbit anti-human EL antibody. Various concentrations of purified recombinant human EL in phosphate-buffered saline containing 1% BSA were added to the wells as a standard. Plasma samples were diluted 1:10 in phosphate-buffered saline and applied to the wells. Specifically bound protein was incubated with biotin-conjugated rabbit anti-human EL antibody, followed by streptavidin-horseradish peroxidase conjugate, and detection with o-phenylenediamine. The reaction was stopped with 2.5 M sulfuric acid and the plate read at 490 nm. A standard curve of 490 nm absorbance versus the known concentrations of EL was constructed. The concentration of the plasma samples was determined by comparison to the standard curve multiplied by the dilution factor. Using this ELISA, EL mass concentrations were quantified in plasma collected in the routine manner (“pre-heparin” plasma) in the full sample (n = 858) as well as in the post-heparin plasma samples described above (n = 510).
Statistical Analysis
Data are reported as median and interquartile range (IQR) or mean ± standard deviation for continuous variables and as proportions for categorical variables. The distributions of both pre-heparin and post-heparin EL mass concentrations were highly skewed rightward, so analyses were performed on log-transformed data. Variables were determined to be normally distributed using the Shapiro-Wilk test. Spearman correlations of EL mass with cardiovascular risk factors and with concentrations of lipoprotein particles, measured by NMR, are presented. The association of EL mass with categorical variables was examined using Kruskal-Wallis rank test. Data were analyzed in the total group and in men and women separately, because the distributions of CAC, HDL, EL, and many risk factors vary with gender [24,25]. Differences in log-transformed EL mass by gender, and cut-points in waist circumference and other risk factors were measured by t-test. Median CAC scores were compared across quartiles of plasma EL (pre-heparin: 6–268, 268.5–422, 422.5–641, and 641.5–2,043; post-heparin: 102–888, 888.5–1,313, 1,313.5–1,927, and 1,927.5–7,505) using the Wilcoxon test for trend. Ordinal logistic regression is a method appropriate for the analysis of CAC data that has a markedly non-normal distribution and a significant proportion of participants with no detectable CAC [26]. CAC scores were divided into ordered outcome categories (0, 1–10, 11–100, 101–400, and >400) using published criteria [27] as described [26,28]. The association of EL mass with CAC was assessed in multivariable models that also included: (1) gender and age (age and age2); (2) established risk factors, gender, and age; and (3) waist circumference, medications (including hormone replacement therapy in women), established risk factors, gender, and age. Established risk factors included total (or LDL) and HDL cholesterol, triglycerides, systolic blood pressure, smoking (current versus never and ex-smokers), race, exercise (none versus any), fasting glucose, and alcohol intake (drinks per week). The results of ordinal logistic regression are presented as the odds ratio (OR) of being in a higher CAC category comparing the highest quartile of EL mass to the lowest quartile. The proportional odds assumption of ordinal regression was satisfied for all models [29]. Data analysis was performed by the authors (KOB, MR, and MW) using Stata 8.1 (Stata, College Station, Texas, United States).
Results
Characteristics of Study Participants
The characteristics of the study participants are summarized in Table 1. Compared to the 2002 CDC report on the body weight status of US adults [30], there were fewer individuals of healthy weight and more obese individuals in our cohort. The lipid profiles of our participants were similar to those reported for 20- to 59-year-old participants in the National Health and Nutrition Examination Survey III [31].
Table 1 Clinical and Biochemical Characteristics of the Cohort
Distribution of EL Mass Concentrations in Pre-Heparin and Post-Heparin Plasma
Our previous studies indicated that EL, like HL and LPL [32], is an avid heparin-binding protein (unpublished data), and we suspected that it would be released by injection of heparin in humans. Indeed, a preliminary study in 60 individuals with normal lipid profiles indicated an approximately 3- to 4-fold increase in EL mass concentrations from pre-heparin to post-heparin plasma (unpublished data). Interestingly, EL mass was abundant in pre-heparin plasma of the entire SIRCA cohort (n = 858), median = 442 (range 324–617) ng/ml. After administration of heparin, median plasma EL concentrations were approximately 3-fold higher than in pre-heparin samples, 1,313 (888–1,927) ng/ml (n = 510). There were no significant differences in pre-heparin EL mass concentrations between men and women. While median post-heparin EL concentrations were 16% lower in women, 1,165 (703–1,705) ng/ml, than men, 1,396 (944–2,031) ng/ml, this difference could be attributed to greater weight in men.
The correlation between pre-heparin and post-heparin EL mass was 0.46, p < 0.001. Using linear regression, pre-heparin EL mass accounted for 15% of the variation in post-heparin EL mass. While including age and gender added less than 1% each, waist circumference accounted for 5% of the variation in post-heparin EL mass.
Association of EL Mass Concentrations with Nonlipid Cardiovascular Risk Factors
Differences in EL mass between groups in the presence and absence of other risk factors were compared by Kruskal-Wallis χ2 test. There were significant positive associations between hypertension and both pre-heparin (p = 0.004) and post-heparin EL mass (p = 0.003). There were significant negative associations between exercise and EL mass in both pre-heparin (p < 0.001) and post-heparin (p = 0.015) plasma. Smoking was associated with higher post-heparin EL mass (p = 0.005).
Correlations of plasma EL mass with age, body mass index (BMI), waist circumference, blood pressure, fasting glucose, and the homeostasis model assessment (HOMA) index, a measure of insulin resistance [33], are shown in Table 2. There was a significant but modest positive correlation between pre-heparin EL mass and age in the entire cohort (r = 0.14, p = 0.004) and in women (r = 0.13, p = 0.013) (Table 2). There were significant positive correlations between both pre-heparin (r = 0.28, p < 0.001) and post-heparin (r = 0.22, p < 0.001) EL concentrations and both BMI and waist circumference, and these associations remained significant when each gender was examined separately (Table 2). In addition, EL concentrations were greater in obese (BMI > 30) compared to lean men and women in pre-heparin plasma, 575 ± 364 ng/ml versus 456 ± 328 ng/ml, p < 0.001, and in post-heparin plasma, 1,820 ± 1,253 ng/ml versus 1,417 ± 836 ng/ml, p < 0.001. Both pre- and post-heparin EL mass were correlated with an increased HOMA-IR score: r = 0.27, p < 0.001 and r = 0.24, p < 0.001, respectively.
Table 2 Association of EL Mass with Cardiovascular Risk Factor
Association of EL Concentrations with Plasma Lipid and Lipoprotein Concentrations
Both pre-heparin and post-heparin EL mass concentrations were positively associated with aspects of an atherogenic lipoprotein profile (Table 2). In pre- and post-heparin plasma, EL mass concentrations were significantly positively correlated with fasting plasma triglyceride and apolipoprotein B (apoB) concentrations in men and women and with LDL-C in women but not men.
To further explore the relationship between plasma EL mass and apoB-containing lipoproteins, we examined the correlations between plasma EL mass and lipoprotein particle size as determined by NMR lipoprotein analysis (Table 3). A significant positive correlation was found between post-heparin EL mass and large VLDL concentrations in men and women, consistent with the positive association of EL with triglycerides. There was a very modest association of EL mass with concentrations of intermediate-size LDL particles in the entire group, which was not significant when assessed separately in men or women. There were no other significant associations of EL mass concentrations with apoB-containing lipoprotein subclasses.
Table 3 Correlations between Post-Heparin EL Mass and Lipoprotein Subclasses Assessed by NMR
There was a modest, but highly statistically significant, negative association between HDL-C concentrations and both pre- and post-heparin EL concentrations (see Table 2). EL mass concentrations in the lowest and the highest HDL quartile were significantly different, with the difference more pronounced in post-heparin plasma, 1,766 ± 1,231 ng/ml in the lowest HDL quartile versus 1,342 ± 917 ng/ml in the highest (p = 0.001). Interestingly, the NMR analysis revealed a negative association of post-heparin EL concentrations with large HDL particles, and a positive association of EL mass with small HDL particles (Table 3). There was no correlation between either pre-heparin or post-heparin EL mass and apoA-I concentrations.
High triglycerides and low HDL-C are two of the criteria used to diagnose metabolic syndrome. We examined the correlation between EL mass and HDL-C concentrations in participants with central obesity as defined by NCEP ATPIII metabolic syndrome waist circumference criteria (greater than 35 inches in women and greater than 40 inches in men). There were stronger negative correlations between post-heparin EL mass in both men and women, r = –0.25, p = 0.03 and r = –0.31, p = 0.03, respectively.
Since there were significant correlations between EL mass and each of the metabolic syndrome parameters, we examined differences in median post-heparin EL mass with increasing numbers of metabolic syndrome factors present (Figure 1). There was an additive effect of increasing numbers of metabolic syndrome factors, with median EL mass increasing from 328 ng/ml in individuals with no factors to 642 ng/ml in participants with all five factors present.
Figure 1 Pre-Heparin EL Mass Concentrations Were Determined in Subgroups According to the Presence of Increasing Numbers of Metabolic Syndrome Factors
There was an increase in median pre-heparin EL mass as the number of metabolic syndrome factors increased. Median values and IQR are listed above each box.
Association of EL Concentrations with Coronary Artery Calcification
CAC scores increased across quartiles of both pre- and post-heparin EL concentrations in men (for trend, p < 0.001 for both) and women (for trend, p < 0.001 for both) (Figure 2). Using multivariable ordinal regression analyses of the entire cohort, plasma concentrations of both pre- and post-heparin EL mass were significantly associated with CAC after controlling for age, gender, and established risk factors, and after further adjustment for plasma lipids, waist circumference, and vasoactive medications (Table 4). In stratified analyses, the fourth EL quartile compared to the first was associated with higher CAC scores for both pre-heparin EL mass, OR = 1.82 (confidence interval [CI] 1.02–3.24, p = 0.043) and post-heparin EL mass, OR = 7.63 (CI 2.44–23.9, p < 0.001) in women, while in men pre-heparin EL mass, OR = 1.7 (CI 1.0–2.9, p = 0.05) but not post-heparin EL mass, OR = 1.46 (CI 0.7–3.08, p = 0.31) was significantly associated with higher CAC scores.
Figure 2 Coronary Artery Calcification Scores Were Compared across Quartiles of Pre-Heparin EL Mass in Both Men and Women
The box represents the 25th to 75th percentile, and the line represents the median CAC score within a given pre-heparin EL mass quartile. The whiskers represent the fifth and 95th percentiles. The diamonds are outliers above the 95th percentile. There was an increase in CAC scores as EL mass increased.
Table 4 Multivariable Association of Plasma EL Concentrations with Coronary Artery Calcification
Discussion
We report, to our knowledge for the first time, measurement of plasma concentrations of EL mass in routine, pre-heparin, and post-heparin plasma of human individuals. Using this newly developed sandwich ELISA, we demonstrated that EL mass concentrations in pre-heparin plasma were higher than those of LPL [34–37], and injection of heparin resulted in a 3-fold increase in plasma EL mass to concentrations similar to those reported for HL [34,38]. Importantly, within individuals, the pre-heparin and post-heparin EL concentrations are significantly correlated, suggesting that pre-heparin EL mass may be a surrogate marker for total vascular EL expression. Indeed, most of the associations between post-heparin EL mass and metabolic syndrome factors, cardiovascular risk factors, lipids, and CAC were also found for pre-heparin concentrations. This suggests that future studies of the association of EL mass concentrations with cardiovascular outcomes may be able to be performed using routine, pre-heparin plasma samples.
We previously reported that overexpression of EL in hyperlipidemic mouse models reduced concentrations of apoB-containing lipoproteins [39]. However, in this human study, we found that concentrations of EL are positively associated with plasma concentrations of LDL-C, triglycerides, and apoB, but did not correlate with the number of particles in LDL subfractions. This is in contrast to HL, which hydrolyzes remnant lipoproteins and is negatively associated with remnant particles and positively associated with concentrations of small, dense LDL particles [40]. The positive association of EL with apoB-containing lipoproteins is most likely the reflection of the concomitant increase in VLDL and triglycerides [41] and EL mass found in obese individuals and those with metabolic syndrome.
This is the first study to demonstrate that, in humans, EL plasma concentrations are negatively associated with HDL-C concentrations. While the negative correlations found between plasma EL mass concentrations and HDL-C concentrations are decidedly modest, they are highly statistically significant. The correlation between plasma EL mass and HDL-C may, in fact, underestimate the association of EL activity with HDL-C, given that the EL mass assay may only partially reflect the biological activity of EL. This finding is consistent with published reports in mice that overexpression of EL dramatically reduces HDL-C concentrations [6,9], and either inhibition of EL or genetic deletion causes an increase in HDL-C [9,10,11]. Furthermore, we show that EL concentrations are negatively associated with large HDL but positively associated with small HDL, consistent with the model proven in mice [8], whereby EL hydrolyzes surface phospholipids on large HDL, creating smaller, phospholipid-depleted HDL particles. This pattern of negative correlation with large HDL and positive correlation with small HDL has also been reported for HL [42]. Of significance was the lack of correlation between EL mass concentrations and apoA-I. This finding is consistent with the report by Jahangiri [43] that EL remodeling of HDL to smaller particles does not mediate apoA-I dissociation.
The apolipoprotein content of HDL has been shown to influence the activity of HL and EL toward HDL. Several reports by Patsch et al. [44] and Mowri et al. [45] have suggested that an increase in the apoA-II content increases HL hydrolysis of the triglyceride in the larger HDL2 particle. Recent studies by Hedrick et al. [46], using apoA-II transgenic mice, suggest that apoA-II inhibits HL activity, but the addition of apoA-I partially reverses this inhibition. Boucher et al. [47] found similar results in in vitro studies, in which the addition of apoA-II to apoA-I–containing HDL inhibited substrate hydrolysis by increasing the affinity of HL for the HDL particle. The authors reported that the addition of apoA-II to particles increased HDL size and induced a conformational change in apoA-I. In contrast, Caiazza et al. [48] used spherical HDL of identical size, containing cholesteryl ester as the sole lipid core with phosphatidylcholine and apoA-I, apoA-II, or both to determine the influence of apolipoproteins on EL activity. They found that EL hydrolysis of phospholipids in apoA-II–containing HDL was negligible, measurable in apoA-I–containing HDL, and greatest in spherical HDL containing both apoA-I and apoA-II. While the absence of triglycerides in these spherical particles makes application to native HDL difficult, it does suggest that there are differences in the subpopulations of HDL to which HL and EL bind.
Both Jin et al. [49] and Hirata et al. [50] reported that stimulation of cultured endothelial cells with tumor necrosis factor-α and interleukin 1β caused an increase in EL mRNA expression and protein secretion. Elevated plasma concentrations of tumor necrosis factor-α are found in individuals with metabolic syndrome [51]. The positive correlations between plasma EL mass, obesity, and the other metabolic syndrome factors provide indirect evidence of a cytokine-mediated increase in endothelial EL protein secretion in the setting of obesity and metabolic syndrome. The significant difference in EL mass between lean and obese participants also suggests that this may occur in vivo.
An important finding of this study in healthy asymptomatic individuals is that EL mass concentrations are positively associated with CAC, a measure of subclinical atherosclerosis in humans, even after controlling for cardiovascular risk factors, plasma lipids, and vasoactive medications. The observation that EL tends to be a stronger predictor of CAC in women than in men could reflect the fact that HL activity is lower in women [52]. In the presence of less HL activity, EL might have a greater relative influence on HDL metabolism and atherosclerosis.
Importantly, our analysis was based on measurement of EL mass, not EL activity. LPL and HL activity are generally measured in post-heparin plasma by assaying their triglyceride lipase activity. However, EL has much less triglyceride lipase activity compared with these enzymes. Therefore, EL activity in human plasma will need to be assessed based on its phospholipase activity and will require differentiation of EL-specific activity from other sources of phospholipase activity, such as HL. The development of a validated assay will allow determination of the association between EL mass and activity and EL activity with plasma lipids and atherosclerosis.
In summary, EL mass is present in measurable amounts in pre-heparin plasma, and increases 3-fold after administration of heparin. EL concentrations are inversely associated with total HDL-C concentrations and positively associated with obesity, triglycerides, fasting glucose, and hypertension, factors composing the metabolic syndrome. Finally, plasma EL concentrations are associated with subclinical atherosclerosis independent of all established risk factors, including plasma lipids. These data support the concept that plasma EL concentrations may modulate metabolic dyslipidemia and atherosclerosis. Prospective studies will be needed to determine if EL is, indeed, a risk factor.
Supporting Information
Protocol S1 Detailed Information on the Development and Quality Control of the Sandwich ELISA
(40 KB DOC).
Click here for additional data file.
Patient Summary
Background
Atherosclerosis (clogging of the arteries), characterized by progressive injury to blood vessels in critical organs such as the heart and the brain, can lead to heart attacks and strokes. It is the most common cause of death and loss of quality of life in Western societies and is on the rise worldwide. In addition to high blood pressure, diabetes, obesity, and smoking, cholesterol levels are a major risk factor for atherosclerosis. Twenty years ago, we thought that high cholesterol was bad, but now we know that there several types of cholesterol, including “good cholesterol” (called HDL-C) and “bad cholesterol” (LDL-C), and that healthy people have high levels of HDL-C and low levels of LDL-C.
Why Was This Study Done?
The researchers who did this study and others had previously found a link between a molecule call endothelial lipase (EL) and atherosclerosis in mice. In mice, EL seems to decrease the levels of HDL-C, the “good cholesterol,” and to make the mice more prone to atherosclerosis. Overall, mice with lower levels of EL seemed to be better off. The question was whether EL levels influenced cholesterol levels and atherosclerosis in humans as well. Human studies until now had searched for a connection between different variants of the EL gene and atherosclerosis, but had not yielded clear answers.
What Did the Researchers Do and Find?
In this study, the researchers directly measured levels of EL in the blood of over 800 human participants and looked for links with cholesterol levels and other atherosclerosis risk factors. They found that, just as in mice, higher blood concentrations of EL are linked with lower levels of HDL-C (the effect was clear but not very large). They also found that higher concentrations of EL were linked with other risk factors for atherosclerosis such as obesity and hypertension. Finally, they found that raised EL concentrations were linked with signs of early atherosclerosis in the participants, and this link held true even when they corrected for all other risk factors, i.e., when they compared people who had the same weight, blood pressure, and cholesterol levels.
What Does This Mean?
The study found a link in humans between high EL concentrations in the blood, low HDL-C levels, and early stages of atherosclerosis. This suggests (but does not prove) that EL concentrations influence the risk of atherosclerosis and could be used to predict an individual's risk. Proving that EL levels predict risk can only be done in so-called “prospective studies” (studies that look forward over time, rather than go backward). Such studies would measure EL blood concentrations in a large number of people now, subdivide them into groups that differed in their EL levels but not in any other factors known to influence atherosclerosis, and then check years later whether there are differences in atherosclerosis, heart attacks, and strokes between the groups. Such studies will take time. While they are ongoing, it seems worth exploring the possibility that treatments to reduce EL levels might help prevent atherosclerosis.
Where Can I Find More Information Online?
Patient information pages from the National Lipid Association: http://www.lipid.org/clinical/patients/
Medline Plus pages on vascular diseases: http://www.nlm.nih.gov/medlineplus/vasculardiseases.html
A collection of atherosclerosis resources for patients and families from the Guam Medical Library:
http://guam-dl.slis.ua.edu/patientinfo/cardiology/cardiovascular/atherosclerosis.html
Atherosclerosis page at the University of Maryland Web site: http://www.umm.edu/cardiac/athero.htm
American Heart Association Web site:
http://www.americanheart.org
KOB was supported by NIH training grant T32 HL-07443–25 and is currently supported by K23 HL74967-01A1. This work was also supported by R01 HL55323 (DJR) from the National Heart Lung and Blood Institute, a Burroughs Wellcome Fund Clinical Scientist Award in Translational Research (DJR), and the General Clinical Research Center of the University of Pennsylvania (M01-RR00040). DJR is also a recipient of a Doris Duke Distinguished Clinical Investigator Award. We extend our appreciation to Jennifer Dykhouse and Kimberly McMahon for their excellent technical assistance. The funding agencies had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Citation: Badellino KO, Wolfe ML, Reilly MP, Rader DJ (2006) Endothelial lipase concentrations are increased in metabolic syndrome and associated with coronary atherosclerosis. PLoS Med 3(2): e22.
Abbreviations
BMIbody mass index
CACcoronary artery calcification
CADcoronary artery disease
CIconfidence interval
ELendothelial lipase
HDL-Chigh-density lipoprotein cholesterol
HLhepatic lipase
HOMAhomeostasis model assessment
IQRinterquartile range
LDLlow-density lipoprotein
LPLlipoprotein lipase
NMRnuclear magnetic resonance
SIRCAStudy of the Inherited Risk of Atherosclerosis
VLDLvery low-density lipoprotein
==== Refs
References
Hokanson JE Kamboh MI Scarboro S Eckel RH Hamman RF Effects of the hepatic lipase gene and physical activity on coronary heart disease risk Am J Epidemiol 2003 158 836 843 14585761
Dugi KA Brandauer K Schmidt N Nau B Schneider JG Low hepatic lipase activity is a novel risk factor for coronary artery disease Circulation 2001 104 3057 3062 11748100
Dugi KA Feuerstein IM Hill S Shih J Santamarina-Fojo S Lipoprotein lipase correlates positively and hepatic lipase inversely with calcific atherosclerosis in homozygous familial hypercholesterolemia Arterioscler Thromb Vasc Biol 1997 17 354 364 9081692
Arca M Campagna F Montali A Barilla F Mangieri E The common mutations in the lipoprotein lipase gene in Italy: Effects on plasma lipids and angiographically assessed coronary atherosclerosis Clin Genet 2000 58 369 374 11140837
Stein Y Stein O Lipoprotein lipase and atherosclerosis Atherosclerosis 2003 170 1 9 12957676
Jaye M Lynch KJ Krawiec J Marchadier D Maugeais C A novel endothelial-derived lipase that modulates HDL metabolism Nat Genet 1999 21 424 428 10192396
Hirata K Dichek HL Cioffi JA Choi SY Leeper NJ Cloning of a unique lipase from endothelial cells extends the lipase gene family J Biol Chem 1999 274 14170 14175 10318835
Maugeais C Tietge UJ Broedl UC Marchadier D Cain W Dose-dependent acceleration of high-density lipoprotein catabolism by endothelial lipase Circulation 2003 108 2121 2126 14517167
Ishida T Choi S Kundu RK Hirata K Rubin EM Endothelial lipase is a major determinant of HDL level J Clin Invest 2003 111 347 355 12569160
Jin W Millar JS Broedl U Glick JM Rader DJ Inhibition of endothelial lipase causes increased HDL cholesterol concentrations in vivo J Clin Invest 2003 111 357 362 12569161
Ma K Cilingiroglu M Otvos JD Ballantyne CM Marian AJ Endothelial lipase is a major genetic determinant for high-density lipoprotein concentration, structure, and metabolism Proc Natl Acad Sci U S A 2003 100 2748 2753 12601178
Ishida T Choi SY Kundu RK Spin J Yamashita T Endothelial lipase modulates susceptibility to atherosclerosis in apolipoprotein-E-deficient mice J Biol Chem 2004 279 45085 45092 15304490
deLemos AS, Wolfe ML, Long CJ, Sivapackianathan R, Rader DJ Identification of genetic variants in endothelial lipase in persons with elevated high-density lipoprotein cholesterol Circulation 2002 106 1321 1326 12221047
Yamakawa-Kobayashi K Yanagi H Endo K Arinami T Hamaguchi H Relationship between serum HDL-C concentrations and common genetic variants of the endothelial lipase gene in Japanese school-aged children Hum Genet 2003 113 311 315 12884003
Rumberger JA Sheedy PF Breen JF Fitzpatrick LA Schwartz RS Electron beam computed tomography and coronary artery disease: Scanning for coronary artery calcification Mayo Clin Proc 1996 71 369 377 8637260
Devries S Wolfkiel C Fusman B Bakdash H Ahmed A Influence of age and gender on the presence of coronary calcium detected by ultrafast computed tomography J Am Coll Cardiol 1995 25 76 82 7798530
Nallamothu BK Saint S Bielak LF Sonnad SS Peyser PA Electron-beam computed tomography in the diagnosis of coronary artery disease: A meta-analysis Arch Intern Med 2001 161 833 838 11268225
Valdes AM Wolfe ML Tate HC Gefter W Rut A Association of traditional risk factors with coronary calcification in persons with a family history of premature coronary heart disease: The study of the inherited risk of coronary atherosclerosis J Investig Med 2001 49 353 361
Reilly MP Wolfe ML Localio AR Rader DJ C-reactive protein and coronary artery calcification: The Study of Inherited Risk of Coronary Atherosclerosis (SIRCA) Arterioscler Thromb Vasc Biol 2003 23 1851 1856 12933535
Otvos J Measurement of triglyceride-rich lipoproteins by nuclear magnetic resonance spectroscopy Clin Cardiol 1999 22 II21 27 10376193
Kuller L Arnold A Tracy R Otvos J Burke G Nuclear magnetic resonance spectroscopy of lipoproteins and risk of coronary heart disease in the cardiovascular health study Arterioscler Thromb Vasc Biol 2002 22 1175 1180 12117734
Mackey RH Kuller LH Sutton-Tyrrell K Evans RW Holubkov R Hormone therapy, lipoprotein subclasses and coronary calcification Arch Intern Med 2005 165 510 515 15767525
Agatston AS Janowitz WR Hildner FJ Zusmer NR Viamonte M Quantification of coronary artery calcium using ultrafast computed tomography J Am Coll Cardiol 1990 15 827 832 2407762
Bello N Mosca L Epidemiology of coronary heart disease in women Prog Cardiovasc Dis 2004 46 287 295 14961452
Chrysohoou C Panagiotakos DB Pitsavos C Kokkinos P Marinakis N Gender differences on the risk evaluation of acute coronary syndromes: The CARDIO2000 study Prev Cardiol 2003 6 71 77 12732792
Reilly MP Wolfe ML Localio AR Rader DJ Coronary artery calcification and cardiovascular risk factors: Impact of the analytic approach Atherosclerosis 2004 173 69 78 15177125
Rumberger JA Brundage BH Rader DJ Kondos G Electron beam computed tomographic coronary calcium scanning: A review and guidelines for use in asymptomatic persons Mayo Clin Proc 1999 74 243 252 10089993
Reilly MP Wolfe ML Rhodes T Girman C Mehta N Measures of insulin resistance add incremental value to the clinical diagnosis of metabolic syndrome in association with coronary atherosclerosis Circulation 2004 110 803 809 15289378
Fu,V Estimating generalized ordered logit models STATA Tech Bulletin 1998 44 27 30
Schoenborn CA Adams PF Barnes PM Body weight status of adults: United States, 1997–98 Adv Data 2002 2002 1 15
Tande DL Hotchkiss L Cotugna N The associations between blood lipids and the Food Guide Pyramid: Findings from the Third National Health and Nutrition Examination Survey Prev Med 2004 38 452 457 15020178
Ehnholm C Shaw W Greten H Brown WV Purification from human plasma of a heparin-released lipase with activity against triglyceride and phospholipids J Biol Chem 1975 250 6756 6761 1158882
Bonora E Targher G Alberiche M Bonadonna RC Saggiani F Homeostasis model assessment closely mirrors the glucose clamp technique in the assessment of insulin sensitivity Diabetes Care 2000 23 57 63 10857969
Ikeda Y Takagi A Ohkaru Y Nogi K Iwanaga T A sandwich-enzyme immunoassay for the quantification of lipoprotein lipase and hepatic triglyceride lipase in human postheparin plasma using monoclonal antibodies to the corresponding enzymes J Lipid Res 1990 31 1911 1924 2079612
Kobayashi J Hashimoto H Fukamachi I Tashiro J Shirai K Lipoprotein lipase mass and activity in severe hypertriglyceridemia Clin Chim Acta 1993 216 113 123 8222262
Kobayashi J Pre-heparin lipoprotein lipase mass J Atheroscler Thromb 2004 11 1 5 15067192
Kimura H Ohkaru Y Katoh K Ishii H Sunahara N Development and evaluation of a direct sandwich enzyme-linked immunosorbent assay for the quantification of lipoprotein lipase mass in human plasma Clin Biochem 1999 32 15 23 10074887
Nishimura M Ohkaru Y Ishii H Sunahara N Takagi A Development and evaluation of a direct sandwich-enzyme-linked immunosorbent assay for the quantification of human hepatic triglyceride lipase mass in human plasma J Immunol Methods 2000 235 41 51 10675756
Broedl UC Maugeais C Millar JS Jin W Moore RE Endothelial lipase promotes the catabolism of ApoB-containing lipoproteins Circ Res 2004 94 1554 1561 15117821
Murdoch SJ Carr MC Kennedy H Brunzell JD Albers JJ Selective and independent associations of phospholipid transfer protein and hepatic lipase with the LDL subfraction distribution J Lipid Res 2002 43 1256 1263 12177169
van Wijk JP Halkes CJ Erkelens DW Castro Cabezas M Fasting and daylong triglycerides in obesity with and without type 2 diabetes Metabolism 2003 52 1043 1049 12898471
Barrans A Collet X Barbaras R Jaspard B Manent J Hepatic lipase induces the formation of pre-beta 1 high density lipoprotein (HDL) from triacylglycerol-rich HDL2. A study comparing liver perfusion to in vitro incubation with lipases J Biol Chem 1994 269 11572 11577 8157689
Jahangiri A Rader DJ Marchadier D Curtiss LK Bonnet DJ Evidence that endothelial lipase remodels high density lipoproteins without mediating the dissociation of apolipoprotein A-I J Lipid Res 2005 46 896 903 15687350
Patsch JR Prasad S Gotto AM Patsch W High density lipoprotein 2. Relationship of the plasma levels of this lipoprotein species to its composition, to the magnitude of postprandial lipemia, and to the activities of lipoprotein lipase and hepatic lipase J Clin Invest 1987 80 341 347 3611351
Mowri HO Patsch JR Gotto AM Patsch W Apolipoprotein A-II influences the substrate properties of human HDL2 and HDL3 for hepatic lipase Arterioscler Thromb Vasc Biol 1996 16 755 762 8640403
Hedrick CC Castellani LW Wong H Lusis AJ In vivo interactions of apolipoprotein A-II, apolipoprotein A-I, and hepatic lipase contributing to HDL structure and antiatherogenic functions J Lipid Res 2001 42 563 570 11290828
Boucher J Ramsamy TA Biaschi S Sahoo D Neville TA Apolipoprotein A-II regulates HDL stability and affects hepatic lipase association and activity J Lipid Res 2004 45 849 858 14967812
Caiazza D Jahangiri A Rader DJ Marchadier D Rye KA Apolipoproteins regulate kinetics of endothelial lipase-mediated hydrolysis of phospholipids in reconstituted high-density lipoproteins Biochemistry 2004 43 11898 11905 15362876
Jin W Sun GS Marchadier D Octtaviani E Glick JM Endothelial cells secrete triglyceride lipase and phospholipase activities in response to cytokines as a result of endothelial lipase Circ Res 2003 92 644 650 12609972
Hirata K Ishida T Matsushita H Tsao PS Quertermous T Regulated expression of endothelial cell-derived lipase Biochem Biophys Res Commun 2000 272 90 93 10872808
Nilsson J Jovinge S Niemann A Reneland R Lithell H Relationship between plasma tumor necrosis factor-α and insulin sensitivity in elderly men with non-insulin-dependent diabetes mellitus Arterioscler Thromb Vasc Biol 1998 18 1199 1202 9714125
Carr MC Hokanson JE Zambon A Deeb SS Barrett PH The contribution of intraabdominal fat to gender differences in hepatic lipase activity and low/high density lipoprotein heterogeneity J Clin Endocrinol Metab 2001 86 2831 2837 11397895
DeGeest BR Van Lithout SA Collen D Humoral immune response in mice against a circulating antigen induced by adenoviral transfer is strictly dependent on expression in antigen-presenting cells Blood 2003 101 2551 2556 12446451
Ruel IL Couture P Cohn JS Bensadoun A Marcil M Evidence that hepatic lipase deficiency in humans is not associated with proatherogenic changes in HDL composition and metabolism J Lipid Res 2004 45 1528 1537 15175359
|
16354105
|
PMC1316064
|
CC BY
|
2021-01-05 11:14:05
|
no
|
PLoS Med. 2006 Feb 20; 3(2):e22
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0030022
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1635410710.1371/journal.pmed.0030017Research ArticleCell BiologyImmunologyAllergy/ImmunologyDiabetes/Endocrinology/MetabolismPathologyBiochemistryEndocrinologyDiabetesImmunology and AllergyAutoimmune DiseasesInterleukin-1 Stimulates β-Cell Necrosis and Release of the Immunological Adjuvant HMGB1 β-Cell Necrosis and HMGB1 ReleaseSteer Sarah A
1
Scarim Anna L
1
Chambers Kari T
1
Corbett John A
1
*1The Edward A. Doisy Department of Biochemistry and Molecular Biology, St. Louis University School of Medicine, St. Louis, Missouri, United States of AmericaHabener Joel Academic EditorHarvard Medical SchoolUnited States of America.*To whom correspondence should be addressed. E-mail: [email protected]
Competing Interests: The authors have declared that no competing interests exist.
Author Contributions: SAS, ALS, KTC, and JAC designed the study, analyzed the data, and contributed to writing the paper.
2 2006 20 12 2005 3 2 e173 5 2005 10 10 2005 Copyright: © 2006 Steer et al.2006This 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.
Death by Necrosis: The Early Stages of Type 1 Diabetes
Background
There are at least two phases of β-cell death during the development of autoimmune diabetes: an initiation event that results in the release of β-cell-specific antigens, and a second, antigen-driven event in which β-cell death is mediated by the actions of T lymphocytes. In this report, the mechanisms by which the macrophage-derived cytokine interleukin (IL)-1 induces β-cell death are examined. IL-1, known to inhibit glucose-induced insulin secretion by stimulating inducible nitric oxide synthase expression and increased production of nitric oxide by β-cells, also induces β-cell death.
Methods and Findings
To ascertain the mechanisms of cell death, the effects of IL-1 and known activators of apoptosis on β-cell viability were examined. While IL-1 stimulates β-cell DNA damage, this cytokine fails to activate caspase-3 or to induce phosphatidylserine (PS) externalization; however, apoptosis inducers activate caspase-3 and the externalization of PS on β-cells. In contrast, IL-1 stimulates the release of the immunological adjuvant high mobility group box 1 protein (HMGB1; a biochemical maker of necrosis) in a nitric oxide-dependent manner, while apoptosis inducers fail to stimulate HMGB1 release. The release of HMGB1 by β-cells treated with IL-1 is not sensitive to caspase-3 inhibition, while inhibition of this caspase attenuates β-cell death in response to known inducers of apoptosis.
Patient Summary
Background
Type 1 diabetes (also called autoimmune diabetes or juvenile diabetes) is an autoimmune disease. For unknown reasons, at some point in childhood or adolescence, the body's own immune system starts attacking and destroying the insulin-producing islet cells in the pancreas. Once the majority of islet cells are destroyed, patients can no longer produce insulin to regulate their blood sugar and must depend on strict diets and insulin injections. Scientists are trying to understand the early events during the development of the disease. There are two fundamentally different kinds of cell death in cells of higher animals and humans, called apoptosis and necrosis. Apoptosis (also called programmed cell death) is an organized, clean way in which cells die without spilling their contents and without causing an inflammatory immune response. They are simply gobbled up by other cells that serve as the body's garbage collectors. Necrosis, on the other hand, is a more messy process and one that does activate the immune system and cause local inflammation.
Why Was This Study Done?
The scientists who did this study are interested in the early stages of islet cell death. Specifically, they wanted to know whether islet cells during the early events of autoimmune diabetes die via apoptosis or necrosis. Earlier experiments to address this question had yielded no clear-cut results.
What Did the Researchers Do and Find?
All the experiments for this study were done in cultured cells in the laboratory. For the most part, the researchers used rodent islet cells, and then they confirmed the crucial finding in human islet cells. They grew the cells under conditions that resembled, to the best of their knowledge, the early stages of diabetes, which caused some of the cells to die. They then did a variety of tests to see whether that cell death was through apoptosis or necrosis, and the results showed that the latter was the case. They also identified some of the key factors involved in promoting and executing the necrosis process.
What Does This Mean?
One must always be careful to extrapolate from laboratory results like these. With this caveat, the results suggest that early in the development of diabetes cells die by necrosis, and they point to some of the key factors involved. These are important results that will inform future studies toward the goal of understanding autoimmune diabetes well enough to prevent or stop its development.
Where Can I Find More Information Online?
The following Web sites provide information on autoimmune diabetes.
MedlinePlus pages on type 1 diabetes:
http://www.nlm.nih.gov/medlineplus/ency/article/000305.htm
Web site of the Juvenile Diabetes Research Foundation:
http://www.jdrf.org/index.cfm?page_id=101982
Pages on type 1 diabetes from the Canadian Diabetes Association:
http://www.diabetes.ca/Section_About/type1.asp
Type 1 diabetes pages from the UK National Institute for Health and Clinical Excellence:
http://www.nice.org.uk/page.aspx?o=213575
UK National Diabetes Information Clearinghouse:
http://diabetes.niddk.nih.gov/index.htm
American Diabetes Association Web site:
http://www.diabetes.org
Conclusions
These findings indicate that IL-1 induces β-cell necrosis and support the hypothesis that macrophage-derived cytokines may participate in the initial stages of diabetes development by inducing β-cell death by a mechanism that promotes antigen release (necrosis) and islet inflammation (HMGB1 release).
Results from rodent and human cells suggest that necrosis of β-cells plays a role in the early stages of type 1 diabetes.
==== Body
Introduction
Insulin-dependent diabetes is an autoimmune disease characterized by selective destruction of insulin-producing β-cells found in pancreatic islets of Langerhans [1]. The development of diabetes is characterized initially by insulitis, in which leukocytes migrate to and invade islets. This is followed by overt diabetes, distinguished by insulin insufficiency due to the destruction of a majority of β-cells [2,3]. Cytokines produced by inflammatory leukocytes and resident and inflammatory macrophages are believed to contribute to the loss of β-cell function and viability during the development of autoimmune diabetes [4,5]. Interleukin (IL)-1 was identified as the active component of conditioned medium derived from activated human mononuclear cells that causes an inhibition of insulin secretion and the loss of islet cell viability, provided the initial evidence supporting a role for cytokines in the pathogenesis of autoimmune diabetes [6,7].
The mechanisms by which cytokines impair β-cell function have been described in detail. Alone or in combination with interferon (IFN)-γ and tumor necrosis factor (TNF), IL-1 has been shown to stimulate β-cell expression of inducible nitric oxide synthase (iNOS), and the resulting production of nitric oxide impairs β-cell oxidation of glucose to CO2, reduces the activity of the Krebs cycle enzyme aconitase, and reduces islet ATP levels by more than 4-fold [8,9]. Importantly, glucose-induced insulin secretion is dependent on β-cell depolarization and Ca2+ entry, events that are mediated by the inhibition of ATP-sensitive K+ channels due to the accumulation of ATP that is produced by mitochondrial oxidation of glucose to CO2 [10,11]. The targeted disruption of mitochondrial oxidation resulting in the reduction of cellular levels of ATP is one mechanism by which nitric oxide impairs glucose-stimulated insulin secretion. The damaging actions of cytokines on β-cell function appear to be selective for β-cells, as IL-1 fails to impair the ability of α-cells to oxidize glucose to CO2 [12]. Furthermore, β-cells are the primary islet-cell source of iNOS in response to cytokine treatment [13,14]. The macrophage-derived cytokine IL-1 is the major regulator of iNOS expression by β-cells, IL-1 alone is sufficient to stimulate iNOS expression by rat β-cells, and combinations of IL-1 and IFN-γ stimulate β-cell expression of iNOS in most mouse strains and in human islets [4,9,15]. Biochemical and genetic evidence supports a primary role for nitric oxide as the major mediator of cytokine-induced β-cell damage. The NOS inhibitors NG-monomethyl-L-arginine (NMMA) and aminoguanidine (AG) prevent the inhibitory actions of cytokines on insulin secretion, glucose oxidation, and aconitase activity by rat, mouse, and human islets [4,8,9]. In addition, islets isolated from iNOS-deficient mice are resistant to the inhibitory actions of cytokines on insulin secretion [16], and expression of iNOS under control of the rat insulin promoter results in the development of diabetes in a nitric oxide-dependent manner that occurs in the absence of insulitis [17].
While autoimmune diabetes is characterized by selective destruction of β-cells, the biochemical mechanisms by which β-cells are killed are poorly defined. In vitro studies using isolated islets and primary β-cells have shown that nitric oxide is a primary mediator of impaired β-cell function in response to cytokine treatment; however, the role of this free radical as a mediator of β-cell death and the mechanism of cell death has been debated. Much of this debate has centered on whether cytokine-induced islet cell death is apoptotic or necrotic and if nitric oxide is a mediator [18–20]. Early studies by Kolb and coworkers demonstrated that nitric oxide is responsible for IL-1-stimulated DNA damage in β-cells [21]. More recently, cytokines have been reported to induce β-cell apoptosis by a process that requires prolonged exposures of 7–9 d and that appears to occur by nitric oxide-independent mechanisms [19,22]. Similar to the inhibitory actions of cytokines on β-cell function, the macrophage-derived cytokine IL-1 appears to be the primary regulator of cytokine-induced β-cell death, be it by necrosis or apoptosis [18,19,23].
In classical terms, apoptosis is defined as programmed cell death by a pathway that removes unwanted cells in the absence of an inflammatory response. It is a highly organized, energy-dependent process that is characterized by caspase activation, the ordered cellular degradation of proteins and organelles, and maintenance of plasma membrane integrity. The final result is noninflammatory phagocytosis of the dying cell, which minimizes leakage of cellular contents and inflammation [24,25]. In contrast, necrosis is an inflammatory process that is characterized by the loss of membrane integrity and the leakage of cellular contents into the extracellular space. It is an unregulated form of cell death that does not require energy or caspase activation [26]. When considering the role of macrophage-derived cytokines such as IL-1 in the context of the initiation of autoimmune diabetes, one would anticipate that this cytokine would kill β-cells by a mechanism that would promote inflammation. In this report, we compare the effects of IL-1 with known inducers of apoptosis on β-cell death.
Methods
Materials and Animals
Sprague Dawley rats were purchased from Harlan (Indianapolis, Indiana, United States). RINm5F (rat insulinoma) cells were obtained from Washington University Tissue Culture Support Center (St. Louis, Missouri, United States). RPMI 1640 containing L-glutamine, CMRL-1066 tissue culture medium, L-glutamine, penicillin, and streptomycin were obtained from GIBCO-BRL (Grand Island, New York, United States). FCS was obtained from Hyclone Laboratories (Logan, Utah, United States). Human recombinant IL-1β was obtained from Cistron Biotechnology (Pine Brook, New Jersey, United States). Sodium (Z)-1(N,N-diethylamino) diazen-1-ium-1,2-diolate (DEANO) was obtained from Alexis (San Diego, California, United States). Enhanced chemiluminescence reagent was purchased from Amersham Pharmacia Biotech (Piscataway, New Jersey, United States). MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide), neutral red, DAPI (4′,6-diamidino-2-phenylindole dihydrochloride), camptothecin, and staurosporine were obtained from Sigma (St. Louis, Missouri, United States). In Situ Cell Death Detection kit was obtained from Roche (Indianapolis, Indiana, United States). Caspase-3 Fluorometric Assay was obtained from R&D Systems (Minneapolis, Minnesota, United States). Annexin V-FITC Apoptosis Detection kit and the caspase-3 inhibitor Z-DQMD-CHO were obtained from Calbiochem (San Diego, California, United States). Guinea pig anti-human insulin antibody was obtained from Linco Research (St. Louis, Missouri, United States). CY3-conjugated donkey anti-guinea pig antibody and HRP-conjugated donkey anti-rabbit and donkey anti-mouse antisera were obtained from Jackson Immunoresearch Laboratories (West Grove, Pennsylvania, United States). Rabbit anti-mouse HMGB1 antibody was obtained from BD Biosciences Pharmingen (San Diego, California, United States). All other reagents were from commercially available sources.
Islet Isolation and Cell Culture
Islets were isolated from 250- to 300-g male Sprague Dawley rats by collagenase digestion as previously described [27]. Islets were cultured overnight in complete CMRL-1066 (containing 2 mM L-glutamine, 10% heat-inactivated fetal calf serum, 100 U/ml penicillin, 100 μg/ml streptomycin, and 5.5 mM glucose) at 37 °C under an atmosphere of 95% air and 5% CO2 prior to experimentation. For annexin V and terminal nick-end labeling (TUNEL) assays of cell death, rat islets were treated with IL-1 for the indicated times followed by dispersion into individual cells by trypsin digestion as previously described [27]. RINm5F cells were grown in RPMI 1640 tissue culture medium to a density of ∼75% prior to experimentation. The cells were then removed from growth flasks by treatment with 0.05% trypsin and 0.02% EDTA for 5 min at 37 °C, washed two times with complete CMRL-1066 media, and plated in CMRL-1066 at the indicated density.
Human islets were obtained from the Islet Cell Resource Center at Washington University School of Medicine (St. Louis, Missouri, United States). Before experiments, the human islets were cultured for 48 h at 37 °C in complete CMRL-1066 tissue culture medium in an atmosphere of 95% air and 5% CO2.
Cell Death Assays
Cell viability was examined using the MTT assay and neutral red uptake. The MTT assay is based on the ability of viable cells to reduce MTT to insoluble formazan crystals [28]. For this assay, RINm5F cells or rat islet cells (1 × 105 cells in 200 μl of complete CMRL-1066) were washed with tissue culture medium following treatments, and MTT was added at a final concentration of 0.5 mg/ml. Following an incubation for 3 h at 37 °C, the supernatants were removed and discarded, and the cells were lysed with acid-isopropanol (100 μl of 0.04 N HCl in isopropanol) and mixed thoroughly to dissolve the crystals for 5 min at room temperature. The optical density of this product was determined at a wavelength of 570 nm.
Cell viability was also evaluated by examining the accumulation of neutral red dye in viable cells [29]. For this assay, RINm5F cells or dispersed rat islet cells (1 × 105 cells in 200 μl of complete CMRL-1066) were cultured for 24 or 48 h with the indicated concentrations of IL-1, camptothecin, or staurosporine; the culture supernatants were removed and discarded; and the cells were incubated in fresh medium containing 50 μg/ml neutral red at 37 °C. Following a 2-h incubation, the supernatant was removed, the cells were washed with a 1% formaldehyde-1% CaCl2 solution, and the neutral red dye was extracted in 100 μl of a 50% ethanol-1% acetic acid lysing solution. The accumulation of neutral red dye in the lysing solution was measured at a wavelength of 570 nm. Cell viability using both the MTT and neutral red assays is expressed as the percent of dead cells.
Annexin V-FITC Staining
RINm5F cells or rat islet cells (1 × 105 cells in 200 μl of complete CMRL-1066) were treated as indicated in the figure legends. Cells were removed from plates by cell scraping (RINm5F) or washing (islet cells). Externalization of phosphatidylserine (PS) was evaluated by annexin V staining according to manufacturer's instructions (Calbiochem) and quantified by flow cytometry [30].
TUNEL Staining/Immunohistochemistry
RINm5F cells or rat islet cells (4 × 105 cells in 400 μl of complete CMRL-1066) were treated as indicated in the figure legends. The cells were removed from plates and centrifuged onto glass slides. DNA damage was quantified by TUNEL staining according to the manufacturer's instructions using the In Situ Cell Death Detection Kit, Fluorescein (Roche). Cells containing DNA damage were colocalized with DAPI by incubating cells with DAPI (5 μg/ml) for 15 min followed by three washes with PBS (pH 7.4). DNA damage in rat islet cells was colocalized with insulin-containing cells as outlined previously [31].
Caspase-3 Activity
RINm5F cells or rat islet cells (4 × 105 cells in 400 μl of complete CMRL-1066) were treated as indicated in the figure legends. Cells were isolated and lysed, and capase-3 activity in the cell lysate was determined according to the manufacturer's instructions (Caspase-3 Fluorometric Assay Kit, R&D Systems). Caspase-3 activity was normalized to protein content and is presented as the fold-increase over the level of activity measured in the untreated control.
Western Blot Analysis
Protein lysates prepared from RINm5F cells and rat islets were separated by SDS-PAGE and then transferred to nitrocellulose membranes (Amersham Pharmacia Biotech) under semidry transfer conditions as previously described [32]. Antigen was detected by enhanced chemiluminescence according to the manufacturer's specifications (Amersham Pharmacia Biotech). Antibody dilutions were: rabbit anti-mouse HMGB1, 1:1,000; and HRP-conjugated donkey anti-rabbit, 1:7,000.
Nitrite Determination
Nitrite production was determined by mixing 50 μl of cell culture medium with 50 μl of Griess reagent [33]. The absorbance at 540 nm was measured, and nitrite concentrations were calculated from a sodium nitrite standard curve.
Statistics
Statistical comparisons were made between groups using one-way ANOVA. Significant differences between groups (p < 0.05) were determined by Newman-Keuls post-hoc analysis.
Results
Nitric Oxide Mediates Cytokine-Induced Islet Cell Death
While it is clear that the inhibitory actions of cytokines such as IL-1 on glucose-stimulated insulin secretion are mediated by β-cell production of nitric oxide, the role of nitric oxide in islet cell death has been debated [8,18,19]. Short exposures (24–48 h) of islets and insulinoma cell lines to IL-1 alone, or in combination with IFN-γ, and TNF cause islet cell DNA damage and reduced oxidative capacity in a nitric oxide-dependent fashion [34–36]; however, prolonged exposures (7–9 d) of islets or purified β-cells with IL-1 or cytokine combinations have been reported to induce β-cell apoptosis in a nitric oxide-independent manner [19,22]. To investigate the mechanisms of cytokine-induced β-cell death, we initially examined whether IL-1 reduces β-cell viability using the MTT assay. Following a 24-h incubation with IL-1, death of RINm5F cells increased ∼20% (Figure 1A), and this cell death correlated with a ∼10-fold increase in production of nitrite, an oxidative metabolite of nitric oxide (Figure 1B). Following a 48-h incubation with IL-1, RINm5F cell death increased to ∼40% (Figure 1A), and cell death correlated with a ∼20-fold increase in nitrite production (Figure 1B). The NOS inhibitor NMMA prevented nitrite production (Figure 1B) and significantly attenuated IL-1-induced RINm5F cell death (Figure 1A). Similar to its effect in RINm5F cells, IL-1 induced the death of ∼25% of islet cells following 24- and 48-h incubations (Figure 1C). IL-1-induced islet cell death correlated with an increase in nitric oxide production (Figure 1D), and NMMA attenuated both islet cell death and nitric oxide production. These findings suggest that IL-1 induces islet and RINm5F cell death, that cell death correlates with the production of nitric oxide, and that cell death is attenuated by the inhibition of NOS.
Figure 1 Nitric Oxide Mediates IL-1-Induced β-Cell Death as Determined by the MTT Assay
RINm5F cells (A and B) or rat islets (C and D) were treated with IL-1 (10 units/ml) and NMMA (2 mM) for either 24 or 48 h, as indicated. The cells were isolated and viability determined using the MTT assay (A and C). The culture supernatants were isolated and nitrite production was determined using the Griess assay (B and D).
(E and F) The effects of exogenously produced nitric oxide supplied by the donor compound DEANO (500 μM) and the apoptosis inducer camptothecin (25 μM) on RINm5F cell viability (E) and the apoptosis inducer staurosporine on rat islet cell viability (F) were determined by MTT assay.
Cell viability data is expressed as percent death. Results for cell viability and nitrite production are the average ± standard error of the mean (SEM) of three independent experiments. *p < 0.05, significantly different from untreated controls. **p < 0.05, significantly different from IL-1-treated condition.
To examine whether nitric oxide can directly modulate cell viability, the effects of exogenously supplied nitric oxide on RINm5F cell viability were examined. RINm5F cells were chosen for these studies because they represent a homogenous population of β-cells, and the effects of cytokines on these cells have been characterized in detail [9]. Treatment of RINm5F cells for 24 h with the nitric oxide donor compound DEANO resulted in a 25% increase in cell death (Figure 1E). Following a 48-h exposure to DEANO, the level of death was not further increased, a finding that is likely due to the rapid decomposition of DEANO (which has a half-life of ∼2 min under these conditions). The effects of DEANO were compared to known inducers of apoptosis. The topoisomerase inhibitor camptothecin is a classical inducer of apoptosis in rapidly proliferating cell lines [37] and, as shown in Figure 1E, induces the death of ∼40% and 75% of RINm5F cells following 24- and 48-h incubations, respectively. Camptothecin is not an effective inducer of islet cell death (most likely due to the quiescent nature of these cells [unpublished data]); however, the kinase inhibitor staurosporine, another known inducer of apoptosis [38], stimulated the death of ∼40% of islet cells following a 24-h incubation (Figure 1F).
A second assay of cell viability, neutral red uptake, was used to confirm the observations that cytokines induce β-cell death (Figure 2). This assay is based on the ability of viable cells to accumulate neutral red dye, while nonviable cells fail to accumulate this dye [29]. Similar to the results presented for the MTT assay in Figure 1, IL-1 reduced RINm5F cell viability by ∼20%–25% following 24- or 48-h incubations, and cell death in response to this cytokine was attenuated by NMMA (Figure 2A). The apoptosis inducer camptothecin also reduced RINm5F cell viability by ∼60%–80% following 24- and 48-h incubations, as determined by neutral red uptake (Figure 2B). These findings provide evidence, in two different assays of cell death, that IL-1 reduces β-cell viability and that the loss of viability is mediated, in part, by the production of nitric oxide.
Figure 2 Effects of IL-1 on β-Cell Viability as Determined by Neutral Red Uptake
RINm5F cells were treated for 24 h or 48 h with IL-1 (10 units/ml), NMMA (2 mM) (A), or camptothecin (25 μM) (B), and the cells were isolated and percentage of death determined using the neutral red uptake assay. Results are average ± SEM of three independent experiments. *p < 0.05, significantly different from untreated controls. **p < 0.05, significantly different from IL-1-treated condition.
Morphological Analysis of IL-1-Induced β-Cell Death
While performing the cell viability experiments, clear morphological differences were observed for RINm5F cells treated with IL-1 as compared to the apoptosis inducer camptothecin. Control, untreated RINm5F cells grew as a flat monolayer of cells with distinct cell borders (Figure 3A). Treatment with IL-1 resulted in morphological changes that included cell swelling and the appearance of holes in the cytoplasm (Figure 3B), changes that are consistent with a necrotic form of cell death. Nitric oxide appears to be responsible for the morphological changes induced by IL-1, as NMMA attenuated RINm5F cell swelling and the formation of cytoplasmic holes (Figure 3C). The morphological changes in RINm5F cells undergoing apoptosis were strikingly different from the alterations induced by IL-1. RINm5F cells that underwent apoptosis following a 24-h treatment with camptothecin were severely shrunken and contained compacted nuclei, consistent with an apoptotic morphology (Figure 3D). This morphological analysis of the effects of IL-1 on RINm5F cell viability suggests that cytokine-induced β-cell death is not caused by classical apoptotic mechanisms and may occur by necrotic mechanisms.
Figure 3 Morphological Analysis of IL-1-Induced RINm5F Cell Death
RINm5F cells were treated for 24 h with IL-1 (10 units/ml) (B and C), NMMA (2 mM) (C), or camptothecin (25 μM) (D), and cellular morphology was examined by phase-contrast microscopy. Treatment with IL-1 resulted in cell swelling and detachment (B), events that are prevented by NMMA (C). In contrast to the effects of IL-1, induction of apoptosis using camptothecin (D) resulted in cell shrinkage and condensation of the nucleus. Untreated control cells are shown in (A). Results are representative of three independent experiments.
IL-1 Stimulates β-cell DNA Damage in a Nitric Oxide-Dependent Manner
The TUNEL assay is commonly used to detect DNA damage in cells, and the results have frequently been used as evidence to support an apoptotic pathway of cell death. Using the TUNEL assay, the effects of IL-1 on the integrity of RINm5F and rat islet cell DNA were examined. In this assay, DNA strand breaks were detected by the incorporation of fluorescein dUTP (green fluorescence, Figure 4), and nuclei were visualized by co-staining with DAPI (blue fluorescence, Figure 4) or colocalized with insulin-containing islet cells (using a CY3-conjugated secondary antibody; red fluorescence, Figure 5). Treatment of RINm5F cells for 24 h with IL-1 resulted in DNA damage in ∼50% of the cells (see Figure 4B), and the level of DNA damage increased to ∼65% of the cells following a 48-h incubation (see Figure 4C). Importantly, IL-1-induced DNA damage was attenuated by NMMA to levels similar to those observed in untreated control cells (∼5%–8% TUNEL-positive; see Figure 4E and 4F). Representative fluorescent micrographs indicate that the morphology of TUNEL-positive RINm5F cells following treatment with IL-1 is significantly different from the morphology of cells treated with the apoptosis inducer camptothecin. While TUNEL staining in most RINm5F cells treated with IL-1 appeared uniform throughout the nuclei and colocalized with DAPI staining, there were a number of RINm5F cells in which the TUNEL staining appears to spill out of the nucleus (see Figure 4B and 4C, white arrows), indicating a loss of nuclear membrane integrity. Camptothecin stimulated DNA damage in ∼60% of the RINm5F cells, as shown in Figure 4D (white arrow), and the pattern of TUNEL and DAPI staining was distinctly different from that observed following IL-1 treatment. Camptothecin-induced TUNEL and DAPI staining appears more intense and punctate, indicating that chromatin condensation has occurred.
Figure 4 IL-1 Stimulates RINm5F Cell DNA Damage in a Nitric Oxide-Dependent Manner
Treatment of RINm5F cells for 24 h (B) or 48 h (C) with IL-1 (10 units/ml) resulted in DNA damage, as determined by TUNEL staining (green fluorescence). DNA damage was attenuated by NMMA (E and F). The apoptosis inducer camptothecin (25 μM) stimulated DNA damage (D), and this DNA damage is morphologically distinct from that induced by IL-1. Nuclei are stained with DAPI (blue). The morphological differences in TUNEL staining, the loss of nuclear membrane integrity in response to IL-1 (B and C) and the condensation of DNA in response to camptothecin (D) are highlighted using white arrows. The levels of DNA damage in control cells (A) were less than 5%. Results are representative of three independent experiments.
Figure 5 IL-1 Stimulates Islet Cell DNA Damage in a Nitric Oxide-Dependent Manner
Treatment of rat islet cells for 24 h with IL-1 resulted in DNA damage as determined by TUNEL staining, and ∼95% of this damage occurred in insulin-containing cells (red fluorescence) (B). Nitric oxide appears to mediate IL-1-induced DNA damage, as it was attenuated by the NOS inhibitor NMMA (2 mM) (D). The apoptosis inducer staurosporine (1 μM) also stimulated DNA damage in both insulin containing and non-insulin containing cells (C). Less than 10% of control, untreated islet cells were TUNEL-positive (A). Results are representative of three independent experiments.
Similar to RINm5F cells, treatment of rat islets for 48 h with IL-1 resulted in DNA damage in ∼50% of islet cells, and over 95% of this DNA damage was found in β-cells, as determined by immunolocalization of TUNEL-positive cells with insulin-containing cells (Figure 5B). IL-1-induced DNA damage was prevented by NMMA (Figure 5D), indicating that nitric oxide is primarily responsible for the DNA damage observed in rat islet cells treated with IL-1. IL-1-induced DNA damage was compared with damage caused by the apoptosis inducer staurosporine. Treatment of rat islets for 48 h with staurosporine resulted in TUNEL-positive staining in ∼50% of islet cells, and this DNA damage was observed in both insulin-containing and non-insulin-containing cells (Figure 5C). Similar to RINm5F cells (see Figure 4), the morphology of TUNEL staining in the IL-1-treated islet cells appears uniform throughout the nucleus (Figure 5B, white arrows). In contrast, staurosporine-induced TUNEL staining in islet cells was more intense and punctate (Figure 5C, white arrows). Taken together, these findings suggest that IL-1 stimulates DNA damage in β-cells, and that nitric oxide is a primary mediator of this damage. In addition, these findings indicate that DNA damage induced by IL-1 results in morphological changes in β-cells that are markedly different from the DNA damage observed in β-cells undergoing staurosporine-induced apoptosis.
IL-1 Fails to Activate Caspase-3 in RINm5F Cells or Rat Islets
To directly examine whether IL-1 induces classical apoptosis in rat islets or RINm5F cells, the effects of this cytokine on the activation of caspase-3 were examined. We decided to examine caspase-3 because it is an executioner caspase that is a downstream target of caspases involved in both the intrinsic and extrinsic pathways of apoptosis [39,40]. IL-1 failed to stimulate caspase-3 activation in RINm5F cells (Figure 6A) or rat islets (Figure 6B) treated for either 24 or 48 h. Since nitric oxide is a known inhibitor of caspase activity [41–43], the effects of IL-1 alone and in combination with NMMA on caspase-3 activity were examined. In the presence of NMMA, IL-1 still failed to stimulate caspase-3 activation in RINm5F cells or rat islets. In contrast to the lack of caspase-3 activation in response to IL-1, camptothecin induced a ∼9-fold increase in caspase-3 activity in RINm5F cells (Figure 6A), and staurosporine stimulated a ∼5-fold increase in caspase-3 activity in rat islets (Figure 6B).
Figure 6 IL-1 Fails to Activate Caspase-3 in RINm5F Cells or Rat Islets
(A) IL-1 (10 units/ml), alone or in combination with NMMA (2 mM) and the nitric oxide donor DEANO (500 μM), failed to activate caspase-3 in RINm5F cells following a 24- or 48-h exposure. As a positive control, the apoptosis inducer camptothecin (25 μM) stimulated a 9-fold increase in RINm5F cell caspase-3 activity.
(B) Similar to RINm5F cells, in rat islets IL-1 (10 units/ml) alone or in combination with NMMA (2 mM) failed to activate caspase-3 following a 24- or 48-h incubation. The apoptosis inducer staurosporine (1 μM) stimulated a 5-fold increase in rat islet caspase-3 activity. Results are the average ± SEM of three independent experiments. *p < 0.05, significantly different from untreated control.
Effects of IL-1 on Annexin V Staining in RINm5F Cells and Rat Islets
The externalization of PS on the outer membrane surface of cells is a hallmark of cells undergoing early or intermediate stages of apoptosis that can be quantified by flow cytometry using the calcium-dependent phospholipid-binding protein annexin V [30]. Using this assay, we show that IL-1 fails to stimulate PS externalization on RINm5F cells (Figure 7A) or rat islet cells (Figure 7B). In contrast, ∼55% of RINm5F cells stained positive for annexin V following a 24 or 48-h incubation with camptothecin, and ∼15% of rat islet cells stained positive for annexin V following a 24- or 48-h incubation with staurosporine. The absence of annexin V staining on rat islet cells and RINm5F cells treated with IL-1 is consistent with the lack of caspase-3 activation, and suggests that, while IL-1 stimulates β-cell death, it does not appear to occur by classical apoptotic mechanisms.
Figure 7 Effects of IL-1 on Annexin V Staining in RINm5F Cells and Rat Islets
Treatment of RINm5F cells (A) or rat islet cells (B) with 10 units/ml IL-1 did not stimulate PS externalization, as determined by annexin V staining and quantified by flow cytometry. In contrast, treatment of RINm5F Cells with camptothecin (25 μM) (A) or rat islet cells with staurosporine (1 μM) (B) stimulated the externalization of PS. Results are presented as percent increase in annexin V-positive cells above the basal levels (∼3%–7%) detected on untreated control RINm5F or rat islet cells and are the average ± SEM of three independent experiments. *p < 0.05, significantly different from untreated control.
Biochemical Evidence that IL-1 Stimulates β-Cell Necrosis
HMGB1 is a chromatin-binding protein that is selectively released by cells undergoing necrosis [26,44–46]. It is not released from cells undergoing apoptosis, as it is sequestered by chromatin [45]. To determine whether IL-1 stimulates β-cell necrosis, RINm5F cells were treated for 24 or 48 h in the presence or absence of NMMA, and the supernatants were isolated and examined by Western blot analysis for the presence of HMGB1. As shown in Figure 8A, IL-1 stimulates the release of low levels of HMGB1 by RINm5F cells, which is first apparent following a 24-h incubation, with higher levels released following a 48-h incubation. In a similar manner, IL-1 stimulated the release of HMGB1 by rat islets (Figure 8B). NMMA attenuates IL-1-induced HMGB1 release by both rat islets and RINm5F cells, indicating that nitric oxide is a mediator of β-cell necrosis. To confirm that nitric oxide induces β-cell necrosis, we show that a 24-h incubation of RINm5F cells (Figure 8C) or rat islets (Figure 8D) with the nitric oxide donor DEANO results in HMGB1 release. In contrast, barely detectable levels of HMGB1 were released by RINm5F cells treated with camptothecin or islets incubated with staurosporine for 24 h (Figures 8C and 8D). As a control, inactivated DEANO failed to stimulate HMGB1 release by either rat islets or RINm5F cells (unpublished data). The release of HMGB1 by rat islets and RINm5F cells provides biochemical evidence that IL-1 stimulates β-cell death by necrotic mechanisms.
Figure 8 IL-1 Stimulates HMGB1 Release in RINm5F Cells and Rat Islets
(A and B) RINm5F cells (A) or rat islets (B) were treated for 24 or 48 h with 10 units/ml of IL-1 and 2 mM NMMA as indicated. The levels of HMGB1 released into the supernatant (s) and contained in the cells (p) were determined by Western blot analysis. Results show that IL-1 stimulates the release of HMGB1 from RINm5F cells and rat islets, and that this release is attenuated by the NOS inhibitor NMMA.
(C) Treatment of RINm5F cells for 24 h with DEANO (500 μM) stimulated HMGB1 release, while the apoptosis inducer camptothecin (25 μM) failed to stimulate HMGB1 release from RINm5F cells.
(D) Nitric oxide supplied exogenously via treatment with DEANO (500 μM) also stimulated HMGB1 release by rat islets, while HMGB1 was not released under conditions of β-cell apoptosis in response to treatment with staurosporine. Results are representative of three independent experiments.
Effect of Caspase-3 Inhibition on Cytokine-Induced β-Cell Death
To further explore the mechanisms of β-cell death, the effects of caspase-3 inhibition on β-cell death in response to apoptosis inducers and IL-1 were examined. As expected, the caspase-3 inhibitor Z-DQMD-CHO attenuated camptothecin-induced RINm5F cell death, as determined by MTT assay (Figure 9A), annexin V staining (Figure 9B), and caspase-3 activation (unpublished data). In contrast, inhibition of caspase-3 failed to attenuate IL-1-induced RINm5F cell death following a 24- or 48-h incubation (Figure 9C). Furthermore, caspase-3 inhibition failed to prevent RINm5F cell HMGB1 release induced by the exogenous addition of nitric oxide by the donor DEANO (Figure 9D), or following a 48-h incubation with IL-1 (Figure 9E). These findings demonstrate that β-cell death in response to IL-1 treatment does not require the activation of caspase-3, an executioner caspase, and that caspase-3 inhibition does not prevent β-cell necrosis in response to IL-1 treatment or exposure to exogenous nitric oxide, as assessed by the release of HMGB1.
Figure 9 Caspase-3 Inhibition Prevents β-Cell Apoptosis but Does Not Prevent IL-1-Induced Necrosis
The caspase-3 inhibitor Z-DQMD-CHO (100 μM) attenuated RINm5F cell death induced by a 24-h incubation with camptothecin as determined by MTT assay (A) or annexin V staining (B). This caspase-3 inhibitor failed to prevent RINm5F cell death in response to a 48-h incubation with 10 units/ml IL-1 as determined by the MTT assay (C). Z-DQMD-CHO also did not attenuate HMGB1 release by RINm5F cells treated for 24 h with 500 μM DEANO (D) or for 48 h with 10 units/ml IL-1 (E). The levels of HMGB1 released into the supernatant (s) and contained in the cells (p) were determined by Western blot analysis. Results for MTT assay (A and B) and annexin V staining (C) are the average ± SEM of three independent experiments. *p < 0.05, significantly different from untreated control. **p < 0.05, significantly different from camptothecin-treated condition. Results for HMGB1 release are representative of three independent experiments.
Cytokines Stimulate HMGB1 Release from Human Islets in a Nitric Oxide-Dependent Manner
In contrast to the effects of IL-1 on rat islets, a combination of cytokines (IL-1, TNF, and IFN-γ) is required to stimulate nitric oxide synthase expression and nitric oxide production from human islets. Previous studies have shown that, individually, these cytokines fail to stimulate iNOS expression or nitric oxide production, or to induce human islet cell death [47–49]. To examine whether nitric oxide modulates human islet viability, we examined the effects of a 48-h incubation of human islets with IL-1, IFN-γ, and TNF on the release of HMGB1. Treatment of human islets with this cytokine mixture stimulated a ∼7 fold increase in nitrite production (Figure 10A), and nitric oxide production correlated with high levels of iNOS expression (unpublished observation). Importantly, under conditions in which human islets produce nitric oxide, HMGB1 accumulated in the supernatant of these islet cultures (Figure 10B). HMGB1 release appears to be dependent on the production of nitric oxide, as the iNOS selective inhibitor AG attenuated the production of nitric oxide and the release of HMGB1 by human islets. The apoptosis inducer staurosporine failed to stimulate the production of nitric oxide or release of HMGB1. Alone, neither IL-1, TNF, nor IFN-γ stimulated nitric oxide production, iNOS expression, or the release of HMGB1 from human islets following a 24- or 48-h incubation (unpublished data), providing evidence that the increased production of nitric oxide by human islets in response to the combination of cytokines is required for the release of HMGB1 by human islets. Overall, these findings are consistent with the effects of IL-1 on rat islets and RINm5F cells, in that they provide direct evidence that nitric oxide mediates the release of HMGB1, a marker of necrotic cell death, by human islets in response to cytokine treatment.
Figure 10 Nitric Oxide-Dependent Release of HMGB1 from Cytokine-Treated Human Islets
Human islets (200 islets in 400 μl of complete CMRL-1066) were treated with a combination of IL-1 (50 units/ml), IFN-γ (500 units/ml), and TNF (50 nM) in the presence or absence of the NOS inhibitor AG (2 mM) or treated with staurosporine (1 μM) alone.
(A) Following 48h -incubation the supernatant was removed and nitrite production was determined using the Griess assay. The levels of HMGB1 released into the supernatant (s) and contained in the islets (p) were determined by Western blot analysis.
(B) Results indicate that cytokine-stimulated HMGB1 release correlates with nitric oxide production, and that inhibition of nitric oxide production using AG attenuates HMGB1 release from human islets. Similar results have been obtained in three of five independent preparations of human islets.
Discussion
Autoimmune diabetes is characterized by an inability to regulate blood glucose levels due to the autoimmune destruction of insulin-producing pancreatic β-cells. The development of autoimmune diabetes is a T cell-dependent process, as they represent a major cellular constituent found in insulitis, and it is possible to prevent diabetes by T cell depletion [2,3,50]. While T cells are responsible for killing the majority of β-cells during the development of diabetes, relatively little is known concerning the events that initiate this autoimmune response directed against β-cells. Mathis and coworkers [3,51] have recently proposed that β-cell death during normal physiological islet remodeling, which peaks at 14–17 d after birth, may supply β-cell antigens that can then be engulfed by dendritic cells. These antigen-presenting cells (APCs) then ferry the antigen to pancreatic lymph nodes, where they may activate naïve diabetogenic T cells [3,51]. Although the mechanisms of β-cell death during islet remodeling are likely apoptotic, it is possible to mount a T cell response by antigens produced by cells undergoing apoptotic cell death [52].
While experimental evidence supports this model for initiation of diabetes in the NOD mouse [51], this model does not take into account the low concordance rate and variable time of onset of diabetes observed in identical twins [53,54]. Specifically, if β-cell death during normal physiological islet remodeling supplies antigens that activate naïve diabetogenic T cells, one would have to assume that the extent/efficiency of β-cell death and/or antigen processing and presentation to naïve T cells may be highly variable in individuals with an identical genetic makeup. An alternative hypothesis to explain the initiation of autoimmune diabetes, proposed by Lacy [55,56], suggests that intraislet macrophage activation and local release of macrophage-derived inflammatory mediators initiates β-cell death, antigen release, and autoimmunity directed against β-cells. The initiating event leading to intraislet macrophage activation may be viral, bacterial, or chemical in origin. It is well known that viral or bacterial infection stimulates macrophage production of IL-1 and TNF, and these cytokines have been shown to impair β-cell function and induce β-cell death [4,5,57]. In support of this model, previous studies have shown that feeding rodents a diet deficient in essential free fatty acids, known to deplete resident macrophages [58,59], attenuates diabetes induced by the chemical toxin streptozotocin [60] and the natural progression of diabetes in BB rats [61]. In addition, a number of studies have shown that the local release of IL-1 by activated intraislet macrophages in rodent and human islets leads to β-cell dysfunction and damage [13,14,62].
In this report we have evaluated the cellular mechanisms by which the macrophage-derived cytokine IL-1 kills β-cells. Using three different methods to evaluate cell viability, MTT oxidation, neutral red uptake, and TUNEL staining (DNA damage), we show that IL-1 induces β-cell death (see Figures 1–3). Nitric oxide appears to be one mediator of β-cell death, as NOS inhibition attenuated cell death in response to IL-1, and nitric oxide, supplied exogenously using donor compounds, reduced β-cell viability. The mechanism by which IL-1 stimulates β-cell death is both morphologically and biochemically different from known inducers of apoptosis. Morphologically, IL-1-induced β-cell death is characterized by cell swelling and diffuse TUNEL staining–changes indicative of the loss of membrane integrity, or necrosis. In response to inducers of apoptosis, the morphology of dying β-cells differs drastically from the effects of IL-1. Apoptosis induction results in cell shrinkage and tight, punctate TUNEL staining, consistent with the form of cell death that we have observed with known inducers of apoptosis. To investigate biochemical mechanisms responsible for β-cell death, we evaluated the effects of IL-1 on caspase-3 activation and PS externalization (annexin staining). Caspase-3 activity was chosen because it the executioner caspase activated by initiator caspases involved in both extrinsic and intrinsic apoptotic pathways [39,40]. While many reports have suggested that IL-1 induces β-cell death by apoptosis, we show that IL-1 fails to activate caspase-3 or to stimulate annexin V staining in islets or RINm5F cells. In contrast, known apoptosis inducers stimulated 6- to 9-fold increases in caspase-3 activity and significantly increased annexin V staining on RINm5F cells and rat islet cells. The lack of RINm5F cell and islet caspase activity following treatment with IL-1 is not surprising, as IL-1 stimulates β-cell production of nitric oxide, and nitric oxide has been shown to inhibit caspase activity by S-nitrosation of the active site cysteine [41,43].
Since these findings suggest that the macrophage-derived cytokine IL-1 does not induce apoptosis, we sought to identify whether IL-1 kills β-cells by necrosis. The morphological changes (see Figure 3) and the diffuse TUNEL staining in RINm5F cells and rat islet cells (see Figures 4 and 5) induced by IL-1 treatment are consistent with a necrotic type of cell death. To provide biochemical evidence in support of IL-1-induced β-cell necrosis, the effects of this cytokine on the release of HMGB1 from β-cells was examined. HMGB1 is a chromatin-binding protein that functions to distort the double helix, allowing proper physical interactions between transcription factors and chromatin [63]. Recent studies have shown that HMGB1 is released from cells dying by necrosis, but it remains tightly bound to chromatin in cells undergoing apoptosis [26,44–46]. In this study, we show that IL-1 stimulates HMGB1 release from RINm5F cells and rat islets in a nitric oxide-dependent manner, and that HMGB1 release is not affected by the inhibition of caspase activity. These findings suggest that IL-1-induced HMGB1 release is mediated, at least in part, by nitric oxide, a conclusion further supported by the ability of nitric oxide to directly stimulate HMGB1 release by RINm5F cells and rat islets. Importantly, under conditions in which RINm5F cells or islets have been forced to undergo apoptosis, HMGB1 was not released. Overall, this morphological and biochemical evidence supports β-cell necrosis as the primary mechanism by which IL-1 induces β-cell death, and that this death pathway is mediated, in part, by nitric oxide.
While IL-1 is sufficient to inhibit the function and to induce the death of rat β-cells, alone, the exogenous addition of IL-1 does not inhibit the function or induce death of human β-cells [47,48]. A combination of the cytokines IL-1, TNF, and IFN-γ is required to impair human β-cell function and induce islet cell death [47,48]. Treatment of human islets for 24- or 48-h with IL-1 + IFN-γ + TNF results in inhibition of glucose-stimulated insulin secretion and mitochondrial aconitase activity, effects attenuated by the iNOS inhibitors NMMA and AG [47,64]. In addition, the local release of IL-1 by intraislet macrophages in response to TNF + IFN-γ + LPS treatment results in inhibition of glucose-stimulated insulin secretion that is attenuated by inhibitors of iNOS [14]. These findings support a role for nitric oxide as a mediator of cytokine-induced β-cell damage in human following exposures of 24 and 48 h. In contrast, extended incubation (for 6–9 d) of human islets with IL-1 + TNF + IFN-γ has been reported to inhibit insulin secretion and induce β-cell apoptosis in a nitric oxide-independent manner [19,48]. The latter findings have led to speculation that human islets are less sensitive to nitric oxide than are rodent islets [65]. Comparisons of cellular responses to cytokines across species are complicated by the genetic background of each species. This is especially true when using human islets in which the genetic background is different for each individual human islet preparation. In contrast, studies with rodent islets are usually performed with a specific strain of rat or mouse. Importantly, Nerup and coworkers have shown that the ability of IL-1 to inhibit β-cell function and to induce iNOS expression varies in three different rat strains [66]. These findings suggest that the genetic background, even within a given species of animal, contributes to the islet response to cytokines. Taking this into account, we have evaluated the effects of IL-1, TNF, and IFN-γ on nitric oxide production and HMGB1 release by human islets. In three of five preparations, the combination of cytokines stimulated HMGB1 release by human islets by a mechanism that was sensitive to inhibition by the iNOS inhibitors AG (Figure 10) and NMMA (unpublished data). Importantly, in two human islet preparations the combination of cytokines failed to stimulate HMGB1 release and also failed to stimulate nitric oxide production or iNOS expression. The reason for this lack of response to cytokines in two of the human islet preparations is unknown; however, it has been reported that elevated levels of heat-shock protein 70 attenuate the ability of β-cells to express iNOS and produce nitric oxide, and the strain differences in response of rat islets to IL-1 have been reported to be associated with differences in levels of heat-shock protein 70 expression [67–69]. Taking this into account, our findings suggest that when human islets produce sufficient levels of nitric oxide in response to cytokines, one consequence is the release of HMGB1. In contrast, when human islets fail to respond to cytokines by producing nitric oxide, they also do not release HMGB1.
HMGB1 is not only a marker of cells undergoing necrosis, but it is also immunologically active. It has been shown to bind to Toll2 and Toll4 receptors and to activate nuclear factor κB-dependent gene expression [70–72], and antagonists of HMGB1 have been shown to attenuate sepsis in animal models [73]. It is well established that necrotic cells release immunological adjuvants that activate APCs and that amplify and sustain T cell-dependent immune responses [74–76]. Recently, Rovere-Querini et al. showed that necrotic HMGB1-deficient cells have a reduced capacity to activate APCs, and that HMGB1 neutralization attenuates the ability of supernatants derived from necrotic wild-type cells to activate APCs [77]. HMGB1 also appears to enhance primary antibody responses to soluble antigens and to enhance cytotoxic lymphocyte responses [77]. While caution should be exercised when extrapolating in vitro findings to mechanisms of disease pathogenesis, it is tempting to speculate on the potential role of intraislet macrophage activation as a precipitating event in the development of autoimmune diabetes. In previous studies, we have shown that the 10–15 resident macrophages found in islets, when activated, produce IL-1 to levels sufficient to inhibit β-cell function in a manner similar to that occurring with the exogenous addition of this cytokine to islets [13,14,78]. We now show that, at these same concentrations, IL-1 kills β-cells by necrotic mechanisms that are associated with the release of the immunological adjuvant HMGB1–a molecule known to augment APC activation and cytolytic responses of T cells.
We would like to thank Colleen Bratcher for expert technical assistance. National Institutes of Health Grant DK 52194 and DK 68839 (JAC) and predoctoral fellowship awards from the American Heart Association (SAS and KTC) supported this work. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Citation: Steer SA, Scarim AL, Chambers KT, Corbett JA (2006) Interleukin-1 stimulates β-cell necrosis and release of the immunological adjuvant HMGB1. PLoS Med 3(2): e17.
Abbreviations
AGaminoguanidine
APCantigen-presenting cell
DAPI4′,6-diamidino-2-phenylindole dihydrochloride
DEANOsodium (Z)-1(N,N-diethylamino) diazen-1-ium-1,2-diolate
HMGB1high-mobility group box 1 protein
IFNinterferon
ILinterleukin
iNOSinducible nitric oxide synthase
MTT3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide
NMMANG-monomethyl-L-arginine
PSphosphatidylserine
SEMstandard error of the mean
TNFtumor necrosis factor
TUNELterminal nick-end labeling
==== Refs
References
Gepts W Pathologic anatomy of the pancreas in juvenile diabetes mellitus Diabetes 1965 14 619 633 5318831
Bach JF Insulin-dependent diabetes mellitus as an autoimmune disease Endocr Rev 1994 15 516 542 7988484
Mathis D Vence L Benoist C Beta-cell death during progression to diabetes Nature 2001 414 792 798 11742411
Mandrup-Poulsen T The role of interleukin-1 in the pathogenesis of IDDM Diabetologia 1996 39 1005 1029 8877284
Rabinovitch A Suarez-Pinzon WL Cytokines and their roles in pancreatic islet beta-cell destruction and insulin-dependent diabetes mellitus Biochem Pharmacol 1998 55 1139 1149 9719467
Bendtzen K Barington T Mandrup-Poulsen T Pedersen JG Svenson M Measurement of human IL-1 by LIF induction, pancreatic islet-cell cytotoxicity, and bone resorption Lymphokine Res 1986 5 Suppl 1 S93 98 3537550
Mandrup-Poulsen T Bendtzen K Nielsen JH Bendixen G Nerup J Cytokines cause functional and structural damage to isolated islets of Langerhans Allergy 1985 40 424 429 3901813
Eizirik DL Flodstrom M Karlsen AE Welsh N The harmony of the spheres: Inducible nitric oxide synthase and related genes in pancreatic beta cells Diabetologia 1996 39 875 890 8858209
Heitmeier MR Corbett JA Ignarro LG Cytotoxic role of nitric oxide in diabetes Nitric oxide biology and pathobiology 2000 San Diego Academic Press 785 810
Koster JC Marshall BA Ensor N Corbett JA Nichols CG Targeted overactivity of beta cell K(ATP) channels induces profound neonatal diabetes Cell 2000 100 645 654 10761930
Misler S Barnett DW Gillis KD Pressel DM Electrophysiology of stimulus-secretion coupling in human beta-cells Diabetes 1992 41 1221 1228 1397696
Corbett JA Wang JL Sweetland MA Lancaster JR McDaniel ML Interleukin 1 beta induces the formation of nitric oxide by beta-cells purified from rodent islets of Langerhans. Evidence for the beta-cell as a source and site of action of nitric oxide J Clin Invest 1992 90 2384 2391 1334975
Corbett JA McDaniel ML Intraislet release of interleukin 1 inhibits beta cell function by inducing beta cell expression of inducible nitric oxide synthase J Exp Med 1995 181 559 568 7530759
Arnush M Heitmeier MR Scarim AL Marino MH Manning PT IL-1 produced and released endogenously within human islets inhibits beta cell function J Clin Invest 1998 102 516 526 9691088
Thomas HE Darwiche R Corbett JA Kay TWH Interleukin-1 plus gamma-interferon-induced pancreatic beta-cell dysfunction is mediated by beta-cell nitric oxide production Diabetes 2002 51 311 316 11812737
Andersson AK Flodstrom M Sandler S Cytokine-induced inhibition of insulin release from mouse pancreatic beta-cells deficient in inducible nitric oxide synthase Biochem Biophys Res Commun 2001 281 396 403 11181061
Takamura T Kato I Kimura N Nakazawa T Yonekura H Transgenic mice overexpressing type 2 nitric-oxide synthase in pancreatic beta cells develop insulin-dependent diabetes without insulitis J Biol Chem 1998 273 2493 2496 9446547
Mandrup-Poulsen T Beta-cell apoptosis: Stimuli and signaling Diabetes 2001 50 Suppl 1 S58 S63 11272204
Eizirik DL Mandrup-Poulsen T Darville MI Ling ZD Van de Casteele M A choice of death–The signal-transduction of immune-mediated beta-cell apoptosis Diabetologia 2001 44 2115 2133 11793013
Eizirik DL Darville MI Beta-cell apoptosis and defense mechanisms: Lessons from type 1 diabetes Diabetes 2001 50 Suppl 1 64 69
Heller B Wang ZQ Wagner EF Radons J Burkle A Inactivation of the poly(ADP-ribose) polymerase gene affects oxygen radical and nitric oxide toxicity in islet cells J Biol Chem 1995 270 11176 11180 7744749
Liu D Pavlovic D Chen MC Flodstrom M Sandler S Cytokines induce apoptosis in beta-cells isolated from mice lacking the inducible isoform of nitric oxide synthase (iNOS–/– ) Diabetes 2000 49 1116 1122 10909967
Kay TW Darwiche R Irawaty W Chong MM Pennington HL The role of cytokines as effectors of tissue destruction in autoimmunity Adv Exp Med Biol 2003 520 73 86 12613573
Thornberry NA Lazebnik Y Caspases: Enemies within Science 1998 281 1312 1316 9721091
Zimmermann KC Bonzon C Green DR The machinery of programmed cell death Pharmacol Ther 2001 92 57 70 11750036
Zong WX Ditsworth D Bauer DE Wang ZQ Thompson CB Alkylating DNA damage stimulates a regulated form of necrotic cell death Genes Dev 2004 18 1272 1282 15145826
Kelly CB Blair LA Corbett JA Scarim AL Isolation of islets of Langerhans from rodent pancreas Methods Mol Med 2003 83 3 14 12619711
Mosmann T Rapid colorimetric assay for cellular growth and survival: Application to proliferation and cytotoxicity assays J Immunol Methods 1983 65 55 63 6606682
Borenfreund E Puerner JA Toxicity determined in vitro by morphological alterations and neutral red absorption Toxicol Lett 1985 24 119 124 3983963
Fadok VA Voelker DR Campbell PA Cohen JJ Bratton DL Exposure of phosphatidylserine on the surface of apoptotic lymphocytes triggers specific recognition and removal by macrophages J Immunol 1992 148 2207 2216 1545126
Scarim AL Arnush M Blair LA Concepcion J Heitmeier MR Mechanisms of beta-cell death in response to double-stranded (ds) RNA and interferon-gamma: dsRNA-dependent protein kinase apoptosis and nitric oxide-dependent necrosis Am J Pathol 2001 159 273 283 11438474
Heitmeier MR Scarim AL Corbett JA Interferon-gamma increases the sensitivity of islets of Langerhans for inducible nitric-oxide synthase expression induced by interleukin 1 J Biol Chem 1997 272 13697 13704 9153221
Green LC Wagner DA Glogowski J Skipper PL Wishnok JS Analysis of nitrate, nitrite, and [15N]nitrate in biological fluids Anal Biochem 1982 126 131 138 7181105
Saldeen J Lee JC Welsh N Role of p38 mitogen-activated protein kinase (p38 MAPK) in cytokine-induced rat islet cell apoptosis Biochem Pharmacol 2001 61 1561 1569 11377386
Mandrup-Poulsen T Apoptotic signal transduction pathways in diabetes Biochem Pharmacol 2003 66 1433 1440 14555218
Chen G Hohmeier HE Gasa R Tran VV Newgard CB Selection of insulinoma cell lines with resistance to interleukin-1beta- and gamma-interferon-induced cytotoxicity Diabetes 2000 49 562 570 10871193
Morris EJ Geller HM Induction of neuronal apoptosis by camptothecin, an inhibitor of DNA topoisomerase-I: Evidence for cell cycle-independent toxicity J Cell Biol 1996 134 757 770 8707853
Bertrand R Solary E O'Connor P Kohn KW Pommier Y Induction of a common pathway of apoptosis by staurosporine Exp Cell Res 1994 211 314 321 8143779
Wilson MR Apoptotic signal transduction: Emerging pathways Biochem Cell Biol 1998 76 573 582 10099777
Boatright KM Salvesen GS Mechanisms of caspase activation Curr Opin Cell Biol 2003 15 725 731 14644197
Rossig L Fichtlscherer B Breitschopf K Haendeler J Zeiher AM Nitric oxide inhibits caspase-3 by S-nitrosation in vivo J Biol Chem 1999 274 6823 6826 10066732
Kim YM Talanian RV Billiar TR Nitric oxide inhibits apoptosis by preventing increases in caspase-3-like activity via two distinct mechanisms J Biol Chem 1997 272 31138 31148 9388267
Li J Billiar TR Talanian RV Kim YM Nitric oxide reversibly inhibits seven members of the caspase family via S-nitrosylation Biochem Biophys Res Commun 1997 240 419 424 9388494
Degryse B Bonaldi T Scaffidi P Muller S Resnati M The high mobility group (HMG) boxes of the nuclear protein HMG1 induce chemotaxis and cytoskeleton reorganization in rat smooth muscle cells J Cell Biol 2001 152 1197 1206 11257120
Muller S Scaffidi P Degryse B Bonaldi T Ronfani L New EMBO members' review: Rhe double life of HMGB1 chromatin protein: Architectural factor and extracellular signal EMBO J 2001 20 4337 4340 11500360
Scaffidi P Misteli T Bianchi ME Release of chromatin protein HMGB1 by necrotic cells triggers inflammation Nature 2002 418 191 195 12110890
Corbett JA Sweetland MA Wang JL Lancaster JR McDaniel ML Nitric oxide mediates cytokine-induced inhibition of insulin secretion by human islets of Langerhans Proc Natl Acad Sci U S A 1993 90 1731 1735 8383325
Eizirik DL Sandler S Welsh N Cetkovic-Cvrlje M Nieman A Cytokines suppress human islet function irrespective of their effects on nitric oxide generation J Clin Invest 1994 93 1968 1974 7514190
Kawahara DJ Kenney JS Species differences in human and rat islet sensitivity to human cytokines. Monoclonal anti-interleukin-1 (IL-1) influences on direct and indirect IL-1-mediated islet effects Cytokine 1991 3 117 124 1888882
Kay TW Chaplin HL Parker JL Stephens LA Thomas HE CD4+ and CD8+ T lymphocytes: Clarification of their pathogenic roles in diabetes in the NOD mouse Res Immunol 1997 148 320 327 9352596
Turley S Poirot L Hattori M Benoist C Mathis D Physiological beta cell death triggers priming of self-reactive T cells by dendritic cells in a type-1 diabetes model J Exp Med 2003 198 1527 1537 14623908
Steinman RM Turley S Mellman I Inaba K The induction of tolerance by dendritic cells that have captured apoptotic cells J Exp Med 2000 191 411 416 10662786
Kumar D Gemayel NS Deapen D Kapadia D Yamashita PH North-American twins with IDDM. Genetic, etiological, and clinical significance of disease concordance according to age, zygosity, and the interval after diagnosis in first twin Diabetes 1993 42 1351 1363 8349046
Barnett AH Eff C Leslie RD Pyke DA Diabetes in identical twins. A study of 200 pairs Diabetologia 1981 20 87 93 7193616
Lacy PE The intraislet macrophage and type I diabetes Mt Sinai J Med 1994 61 170 174 8022429
Lacy PE Finke EH Activation of intraislet lymphoid cells causes destruction of islet cells Am J Pathol 1991 138 1183 1190 1902627
McDaniel ML Kwon G Hill JR Marshall CA Corbett JA Cytokines and nitric oxide in islet inflammation and diabetes Proc Soc Exp Biol Med 1996 211 24 32 8594615
Schreiner GF Flye W Brunt E Korber K Lefkowith JB Essential fatty acid depletion of renal allografts and prevention of rejection Science 1988 240 1032 1033 3285468
Lefkowith JB Schreiner G Essential fatty acid deficiency depletes rat glomeruli of resident macrophages and inhibits angiotensin II-induced eicosanoid synthesis J Clin Invest 1987 80 947 956 3116045
Wright JR Lefkowith JB Schreiner G Lacy PE Essential fatty acid deficiency prevents multiple low-dose streptozotocin-induced diabetes in CD-1 mice Proc Natl Acad Sci U S A 1988 85 6137 6141 3045812
Lefkowith J Schreiner G Cormier J Handler ES Driscoll HK Prevention of diabetes in the BB rat by essential fatty acid deficiency. Relationship between physiological and biochemical changes J Exp Med 1990 171 729 743 2307932
Heitmeier MR Arnush M Scarim AL Corbett JA Pancreatic beta-cell damage mediated by beta-cell production of interleukin-1. A novel mechanism for virus-induced diabetes J Biol Chem 2001 276 11151 11158 11108714
Bianchi ME Manfredi A Chromatin and cell death Biochim Biophys Acta 2004 1677 181 186 15020058
Scarim AL Heitmeier MR Corbett JA Irreversible inhibition of metabolic function and islet destruction after a 36-hour exposure to interleukin-1beta Endocrinology 1997 138 5301 5307 9389514
Eizirik D Pipeleers D Ling Z Welsh N Hellerstrom C Major species differences between humans and rodents in the susceptibility to pancreatic beta-cell injury Proc Natl Acad Sci U S A 1994 91 9253 9256 7937750
Reimers J Andersen H Mauricio D Pociot F Karlsen A Strain-dependent differences in sensitivity of rat beta-cells to interleukin 1 beta in vitro and in vivo: Association with islet nitric oxide synthesis Diabetes 1996 45 771 778 8635652
Bellmann K Jaattela M Wissing D Burkart V Kolb H Heat shock protein hsp70 overexpression confers resistance against nitric oxide FEBS Lett 1996 391 185 188 8706913
Scarim AL Heitmeier MR Corbett JA Heat shock inhibits cytokine-induced nitric oxide synthase expression by rat and human islets Endocrinology 1998 139 5050 5057 9832444
Johannesen J Pociot F Karlsen A Mandrup-Poulsen T Nerup J Strain-dependent difference in inducible nitric oxide synthesis (iNOS) expression in rat pancreatic islets correlates with interferon regulating factor 1 (IRF-1) and heat shock protein 70 (HSP70) expression Eur Cytokine Netw 2001 12 501 509 11566631
Park JS Svetkauskaite D He Q Kim JY Strassheim D Involvement of toll-like receptors 2 and 4 in cellular activation by high mobility group box 1 protein J Biol Chem 2004 279 7370 7377 14660645
Andersson AC Seppala U Rudin A Activation of human neonatal monocyte-derived dendritic cells by lipopolysaccharide down-regulates birch allergen-induced Th2 differentiation Eur J Immunol 2004 34 3516 3524 15517608
Kokkola R Andersson A Mullins G Ostberg T Treutiger CJ RAGE is the major receptor for the proinflammatory activity of HMGB1 in rodent macrophages Scand J Immunol 2005 61 1 9 15644117
Yang H Ochani M Li J Qiang X Tanovic M Reversing established sepsis with antagonists of endogenous high-mobility group box 1
Proc Natl Acad Sci U S A 2004 101 296 301 14695889
Kurts C Miller JF Subramaniam RM Carbone FR Heath WR Major histocompatibility complex class I-restricted cross-presentation is biased towards high dose antigens and those released during cellular destruction J Exp Med 1998 188 409 414 9670054
Basu S Binder RJ Suto R Anderson KM Srivastava PK Necrotic but not apoptotic cell death releases heat shock proteins, which deliver a partial maturation signal to dendritic cells and activate the NF-kappa B pathway Int Immunol 2000 12 1539 1546 11058573
Millar DG Garza KM Odermatt B Elford AR Ono N Hsp70 promotes antigen-presenting cell function and converts T-cell tolerance to autoimmunity in vivo Nat Med 2003 9 1469 1476 14625545
Rovere-Querini P Capobianco A Scaffidi P Valentinis B Catalanotti F HMGB1 is an endogenous immune adjuvant released by necrotic cells EMBO Rep 2004 5 825 830 15272298
Arnush M Scarim AL Heitmeier MR Kelly CB Corbett JA Potential role of resident islet macrophage activation in the initiation of autoimmune diabetes J Immunol 1998 160 2684 2691 9510167
|
16354107
|
PMC1316065
|
CC BY
|
2021-01-05 11:14:05
|
no
|
PLoS Med. 2006 Feb 20; 3(2):e17
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0030017
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1635410810.1371/journal.pmed.0030018Research ArticleInfectious DiseasesEpidemiology/Public HealthStatisticsEpidemiologyPublic HealthCohort studiesProspective Study of One Million Deaths in India: Rationale, Design, and Validation Results Million Death Study in IndiaJha Prabhat
1
*Gajalakshmi Vendhan
2
Gupta Prakash C
3
Kumar Rajesh
4
Mony Prem
5
Dhingra Neeraj
1
Peto Richard
6
RGI-CGHR Prospective Study Collaborators 1Centre for Global Health Research, Public Health Sciences, St Michael's Hospital, and McLaughlin Centre for Molecular Medicine, University of Toronto, Toronto, Canada2Epidemiological Research Centre, Chennai, India3Healis-Seskarhia Institute of Public Health, Navi Mumbai, India4School of Public Health, Post Graduate Institute of Medical Education and Research, Chandigarh, India5Institute of Population Health and Clinical Research, St. John's Medical College, Bangalore, India6Clinical Trial and Epidemiological Studies Unit, University of Oxford, Oxford, United KingdomAvila Mauricio Hernandez Academic EditorNational Institute of Public HealthMexico*To whom correspondence should be addressed. E-mail: [email protected]
Competing Interests: The authors have declared that no competing interests exist.
Author Contributions: All authors contributed to the design of the study, to analyses of the data, and to the writing of the paper.
2 2006 20 12 2005 3 2 e1823 5 2005 18 10 2005 Copyright: © 2006 Jha et al.2006This 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.
Measuring Mortality in Developing Countries
Background
Over 75% of the annual estimated 9.5 million deaths in India occur in the home, and the large majority of these do not have a certified cause. India and other developing countries urgently need reliable quantification of the causes of death. They also need better epidemiological evidence about the relevance of physical (such as blood pressure and obesity), behavioral (such as smoking, alcohol, HIV-1 risk taking, and immunization history), and biological (such as blood lipids and gene polymorphisms) measurements to the development of disease in individuals or disease rates in populations. We report here on the rationale, design, and implementation of the world's largest prospective study of the causes and correlates of mortality.
Methods and Findings
We will monitor nearly 14 million people in 2.4 million nationally representative Indian households (6.3 million people in 1.1 million households in the 1998–2003 sample frame and 7.6 million people in 1.3 million households in the 2004–2014 sample frame) for vital status and, if dead, the causes of death through a well-validated verbal autopsy (VA) instrument. About 300,000 deaths from 1998–2003 and some 700,000 deaths from 2004–2014 are expected; of these about 850,000 will be coded by two physicians to provide causes of death by gender, age, socioeconomic status, and geographical region. Pilot studies will evaluate the addition of physical and biological measurements, specifically dried blood spots.
Preliminary results from over 35,000 deaths suggest that VA can ascertain the leading causes of death, reduce the misclassification of causes, and derive the probable underlying cause of death when it has not been reported. VA yields broad classification of the underlying causes in about 90% of deaths before age 70. In old age, however, the proportion of classifiable deaths is lower. By tracking underlying demographic denominators, the study permits quantification of absolute mortality rates. Household case-control, proportional mortality, and nested case-control methods permit quantification of risk factors.
Conclusions
This study will reliably document not only the underlying cause of child and adult deaths but also key risk factors (behavioral, physical, environmental, and eventually, genetic). It offers a globally replicable model for reliably estimating cause-specific mortality using VA and strengthens India's flagship mortality monitoring system. Despite the misclassification that is still expected, the new cause-of-death data will be substantially better than that available previously.
This paper reports the rationale, design, and implementation of the world's largest prospective study of the causes and correlates of mortality.
==== Body
Introduction
About 46 million of the estimated 60 million deaths per year worldwide occur in developing countries [1]. However, there is a dearth of reliable and accurate information on the causes and distribution of mortality in these countries. India has about 9.5 million deaths a year, or about one in six of all deaths worldwide. Over three-quarters of deaths in India occur in the home; more than half of these do not have a certified cause.
To meet these modern challenges of mortality measurement, the world's largest prospective study of the causes and correlates of mortality in India is being undertaken by the Registrar General of India (RGI)'s Sample Registration System (SRS). The study, called the Prospective Study of 1 Million Deaths in India, is implemented in close collaboration with the Centre for Global Health Research at the University of Toronto, leading Indian and overseas academic institutions, and the Indian Council of Medical Research. The study has several innovations that are relevant to other developing countries considering the measurement of mortality and to recent calls for improved health statistics [2–5]. It uses a well-validated household instrument to ascertain causes of death and dual recording methods to improve reliability and consistency. It is national in scale, representative of the population, and–by recording underlying demographics–able to quantify absolute mortality rates. Prospective and retrospective study designs, such as household and proportional mortality and nested case-control methods, permit quantification of correlates of mortality.
In this report, we present the rationale, design, results of validation studies to date, key statistical issues, and expansion to biologic measurement for the Million Death Study in India. We discuss the challenges in implementing modern mortality measurement and the implications for global health.
Rationale: Why Measure Mortality?
Historically reliable, representative, routine, low-cost, and long-term mortality measurements are the key to monitoring trends in health conditions of the population, detecting new epidemics (such as HIV/AIDS), spurring research into avoidable causes of death, evaluating the success of control programs, and improving accountability for expenditures on disease control [6,7]. Routinely collected data have helped to spur further research and public health action and contributed to the enormous increases in life expectancy in the 20th century [8].
Public health in industrialized countries was transformed when vital statistics on age, sex, and socioeconomic distribution of births and deaths became available in the late 19th and 20th centuries. Vital statistics have demonstrated major trends in fertility, child survival, and mortality. They have shown good news, such as the large declines in under-5 mortality and tuberculosis mortality during the 20th century. They have also raised alarm; in the mid-1940s, a dramatic increase in lung cancer deaths in British and American men after World War II led to much research on smoking [9]. In the early 1980s, routine mortality data from San Francisco revealed an exceptional increase in immune-related deaths among young men and signaled the start of the American HIV-1 epidemic [10].
Vital statistics need to keep up with modern patterns of disease. India and other countries have seen consistent decreases in child mortality (under-5 mortality has fallen by about 2% per year since 1971 in India [11]). Adult deaths in middle age (35–69 y) attributable partially to the effects of smoking, sedentary lifestyles, and higher saturated fat intake have been on the rise. More recently, deaths among young adults (15–34 y) have risen from HIV/AIDS. Reliable information on the diseases of adults and their causes are a large gap in global knowledge.
Recorded deaths from most communicable diseases or injuries generally correspond to their causes (for example, malaria deaths are caused by Plasmodium parasites), but the more “chronic” communicable diseases (such as tuberculosis) and most noncommunicable diseases can have multiple causes. For example, a myocardial infarction could be caused by smoking, elevated blood pressure, high lipids, or other factors. The age- and sex-specific importance of established risk factors, or combinations of risk factors, has only recently been reliably documented through appropriately large studies in Western populations, and with surprising results. For example, the association of blood pressure and vascular disease is twice as steep as previously believed [12] if blood pressure is measured reliably and the effects of “regression-dilution” bias [13] are properly considered. There are few epidemiological studies that document the age-, sex-, and region-specific hazards of blood pressure, blood lipids, and smoking in developing populations [14], where most of the world's vascular deaths occur [1].
Anecdotal evidence suggests that in India each disease that is common in one part of the country is relatively uncommon elsewhere, for reasons that are not understood. This means that there are important avoidable causes that still await discovery. Much more remains to be discovered about the novel genomic, proteomic, and other biochemical correlates of respiratory, intestinal, or other infections in general, and of the avoidable causes of chronic diseases such as cancer, heart attack, stroke, and lung disease[15,16] that currently account for most of the adult mortality in India. Even for infections such as HIV-1 and tuberculosis, there may well be genetic causes (such as polymorphisms in genes involved in innate [17,18] or adaptive [19] immune recognition) or environmental causes (such as other infections [20,21] other than the relevant pathogens) that make particular infections, or progression from infection to disease, more probable.
Alternative Designs for Measuring Mortality
The ideal mortality measurement system has several characteristics [7]. It is routine, reproducible, long-term, low-cost, and sustainable. It is reliable and representative of the population (implying that it avoids major selection biases in enrollment). It captures not only the death, but also reliably the cause based on the International Statistical Classification of Diseases and Related Health Problems (ICD-10 [22]). The ideal system includes not only events or “numerators,” but also underlying population demographics (“denominators”).
Three systems measure mortality in India (http://www.censusindia.net/). The first, the Civil Registration System, is currently unreliable due to gross underregistration. While some areas have very good vital registration (Mumbai provides death registration as far back as 1848 [23]), overall, only 3.5 million of the estimated 9.5 million annual deaths were registered in India in 1999. Among registered deaths, cause-of-death data are available for about one in three deaths, but this often merely subdivides deaths as due to accident, violence, or disease, without further details. Civil registration is the accepted mortality measurement system in Western countries where coverage nears 100%. However, access to medical care is far less common in India, and most deaths occur at home rather than in hospitals. Eventually, civil registration will increase, as has happened recently in China. But this may take decades; the United States took the better part of a century to increase death certification, and some states did not have complete coverage until the 1970s [24].
The second system is the Medically Certified Causes of Death. However, this covers only about 0.4 million deaths and is largely confined to selected urban settings that are not representative of the general population. Problems with inconsistent physician attribution of causes of death, especially for senility and ill-defined causes have been noted [6,7].
The third is the SRS (described in detail below). Only the SRS is representative of both urban and rural settings of India, covering some 6,700 to 7,600 units randomly selected from the preceding census. The SRS is much smaller, though, covering only 0.05 million deaths. Thus, its chief drawback is that it cannot yet provide district-level data for local decision making, and it lacks sufficient power to generate yearly rates for less common causes, such as maternal deaths. Until recently, the SRS did not adequately capture information on causes of death. However, we have addressed this gap by developing and implementing verbal autopsy (VA)–an innovative method to estimate cause-specific mortality.
Finally, it is worth noting that econometric models of cause-specific mortality [1] are no replacement for direct measurement. Indirect estimates are only as good as their underlying data. The econometric models have not been well tested in the presence of HIV/AIDS growth. In India the global burden of disease varies considerably depending on the assumptions used. For example, the 1994 global burden of disease estimated 0.78 million cancer deaths in 1990, but cancer registry data suggested a much lower figure of 0.43 million deaths [25]. The 1996 version of the global burden of disease projected 0.95 million deaths from tuberculosis in 2000; the 1999 version estimated 0.42 million deaths in 1998 [26].
Study Objectives
In light of the above, we have the following objectives: (a) to reliably document cause-specific mortality from 2001 to 2003 (3 y; approximately 150,000 deaths) within the SRS to establish regional-, gender-, and age-specific variation and patterns of mortality; (b) to document causes of death with routine use of VA in the new SRS sampling frame from 2004 to 2014 (10 y; approximately 700,000 deaths expected); (c) to improve our understanding of selected risk factors, most notably tobacco use, alcohol use, indoor air pollution, fertility preferences for male children and its effect on female survival, immunization, and migration through linking of mortality outcomes with exposure status, using retrospective and prospective methods; and (d) to expand the new SRS to obtain reliable epidemiological evidence about the relevance of physical (such as blood pressure and peak flow), behavioral (such as migration and HIV-1 risk), and biological (such as blood lipids and gene polymorphisms) measurements to the development of disease in individuals or disease rates in populations.
Methods
Study Setting
This study is conducted within the SRS, a large, routine demographic survey and the primary system for the collection of Indian fertility and mortality data since 1971 [27]. There are two SRS sample frames. The first SRS sample frame covers 6.3 million people (including 2.9 million adults aged 25 y or older) in all 28 states and seven union territories of India. An average of 150 households are drawn from each of 6,671 sample units (4,436 rural and 2,235 urban), which in turn are selected using 1991 census data. The new SRS sample frame covers about 7.6 million people (including 3.5 million adults aged 25 y or older) in all 28 states and seven union territories of India. Households are drawn from 7,597 sample units (4,433 rural and 3,164 urban) selected from the 2001 census.
SRS sample units are randomly selected to be representative of the population at the state level. The sample design is a unistage stratified simple random sample without replacement. The sample size for the first and new SRS sample frames are based on total fertility rates and infant mortality rates, respectively. Within the SRS, selected households are continuously monitored for vital events by two independent surveyors. The first is a part-time enumerator (commonly a local school teacher familiar with the area/village) who visits the home every month. The second is a full-time (nonmedical) RGI surveyor who visits the home every 6 mo. Each independently records the births and deaths in the household for a 6-mo period. A third staff member does a reconciliation of the two reports, arriving at a final list of births and deaths for each household, which completes each half-yearly survey. The RGI surveyors each cover about 150 households with a total average population of 900 (ranging from 700 to 1,500), and report about 50 deaths (and about 150 births) every 6 mo.
Plan of Investigation
Box 1 and Figure 1 provide an overview of the study methods. In brief, the method involves 800 trained (nonmedical) RGI surveyors implementing VA reports among enrolled populations every 6 mo. A random 10% of the VA fieldwork is repeated by an independent audit team. After data entry, field reports are sent electronically to two independent and trained physicians who assign cause of death, based on the ICD-10 [22], using a Web-based system. The two physicians have to agree on the underlying cause of death, and if they do not, such records undergo reconciliation and third-physician adjudication.
Box 1. SRS VA Methods Overview
Design of verbal autopsy questionnaire.Combined open/closed format. Structured questions accompanied by an open-ended narrative. Symptom list to assist attribution of deaths.
Questionnaire layout.One-page, double-sided, scannable forms. Four age-specific forms (neonatal, child, adult, and maternal). Forms available in either English or Hindi.
Interviewers.Nonmedical RGI surveyors (mostly male) with knowledge of local language(s) and trained in VA instrument.
Interview technique.One-on-one interviews during home visits. Duration of 30–45 min.
Respondents.Family members or other informants (usually neighbors or close associates of the deceased).
Recall period.Usually less than 6 mo, but still valid up to 3 y.
Data quality.Random resample of 10% of all deaths to ensure completeness of fieldwork.
Derivation of diagnosis.Central medical review of cause by two independent physicians using modified VA reports (Physician Reports) and an Internet-based Web application. Adjudication of disagreements by an expert physician.
Mortality classification.
The International Classification System of Diseases and Related Health Problems, Tenth Revision (ICD-10).
Sample size.One million deaths (about 0.3 million from 1998–2003, 0.7 million from 2004–2014).
Figure 1 SRS VA Activities
Text S1–S3 provide the details of the field collection methods, resampling, physician coding, data management, and research ethics. The full protocol, field instruments, training manuals, slide presentation of validation results, and procedures are available at http://www.cghr.org/project.htm.
Estimated Sample Size and Distribution
The primary outcomes of interest are all-cause mortality and cause-specific mortality. We expect about 150,000 cause-specific deaths determined through VA and about twice as many deaths (300,000) to study all-cause mortality in the 1998–2003 sample frame. The expected age distribution of cause-specific deaths captured with VA, based on the age-specific SRS death rates in 2001, is shown in Figure 2. Using indirect World Health Organization estimates on causes of death in India [1], we also show the approximate numbers of major categories of deaths expected (in thousands) between ages 25 to 69 y (Table 1).
Figure 2 Expected Number of Deaths by Age Group Using VA (Thousands), 2001 to 2003
Table 1 Expected Number of Deaths between Ages 25 and 69 by Cause (Thousands), 2001 to 2003
As of November 2005, 140,000 VA reports have been collected from all SRS units, and about 35,000 records have undergone double physician coding and reconciliation.
Overall we expect several thousand tuberculosis, vascular, and cancer deaths among adults. Such numbers are not excessively large, particularly if the age- and sex-specific relevance of several risk factors is to be assessed simultaneously. For example, assuming a power of 90% and two-sided α of 0.001, and assuming a 40% smoking rate among male controls, the study would have sufficient power to detect relative risks among men as low as 1.4 for lung cancer, 1.1 for all cancers or for tuberculosis, and 1.1 for cardiovascular disease [28]. Thus, the study has robust statistical power to detect small but significant increases in risk for most key variables of interest.
The main planned analyses involve simple tabulations and standard Cox proportional hazards analyses, calculating relative risks that are standardized for age, educational level, and selected covariates as relevant. Analyses of adult deaths will focus on deaths at ages 25–69 y, as these are much less likely to be misclassified than deaths occurring after age 70.
Sample-size estimates for the 2004–2014 sample frame are less certain, as the cause-specific child and adult mortality will depend on the rapidity of declines in childhood mortality and increases in HIV-1-related mortality [11]. However it is reasonable to assume that with expanding sample size (reflecting the growth of populations in the SRS households), some 700,000 deaths will occur over the 10-y period.
Reliable Measurement of Absolute Rates and Relative Risks
The ability to generate absolute rates depends on completeness of enumeration and being able to calculate underlying demographic denominators, including migration. Evaluations suggest that the SRS has a high ascertainment rate of expected events, although adjustments may be needed for certain age groups. A 1984 study [29] concluded that the system captured 90% of deaths between 1971 and 1980. The SRS, which employs continuous enumeration, is more sensitive at detecting child deaths than are single surveys, and it has recorded more child deaths than those estimated by the National Family Health Survey 2 (NFHS-2), a nationally representative demographic and health survey that interviewed nearly 0.1 million households (http://nfhsindia.org/index.html). Bhat [30] found that adult deaths were underreported by about 13% to 14% (slightly higher in females) and that there is evidence to suggest that the undercount has increased slightly in particular states recently. It is expected that the new SRS sample frame should have corrected the underreporting of adult deaths noted in the first SRS sample frame. Formal demographic evaluation will be done once results from the new SRS are collected. Because household composition is updated every 6 mo, and in-migration and out-migration are traced, it should be possible to calculate absolute mortality rates in the population, including person years at risk. Deaths among in-migrating groups can be excluded, as the death report lists if the person was a usual resident in the SRS unit or not. Some periodic correction for undercount will be needed.
Loss to Follow-Up
The baseline survey for the study was called the Special Fertility and Mortality Survey (SFMS) and was confined to usual residents of households only [31]. Those living in these households in 1998 were a subselection of the SRS baseline sample frame originally done in 1993. Experience from a prospective study in Chennai (V. Gajalakshmi, personal communication) suggests that if people are resident for a few years, they are not likely to move again. Thus, the enrolled group in the current SRS sample frame is less likely to migrate than would be the general population. This is supported by data from our own small pilot examination of 389 randomly chosen SRS records in two northern states, which showed that only 6% (25/389) of the households had moved from 1998 to 2002.
Nonetheless, we do expect loss to follow-up caused by out-migration from the SRS unit. The risk ratios would be biased downward if such loss to follow-up is nondifferential between exposed and unexposed groups. Similarly, risk ratios would be biased downward if impoverished people (who are most likely exposed to smoking and other risk factors) migrate.
Results
Validation Studies of Mortality Outcomes
VA relies on the assumption that most causes of death have distinct symptoms and signs that can be recognized, recalled, and reported by household members or associates of the deceased to a trained, usually nonmedical field-worker. Further, it is assumed that deaths characterized through VA possess a distinct set of features that can be distinguished from other underlying causes of death [32]. Thus, diseases with very distinct symptoms and signs, such as tetanus, that are recognized by the local population may be more suitable for VA than systemic diseases, such as malaria, which is associated with many common symptoms and signs. Factors that influence the validity and reliability of VA include the VA instrument (mortality classification, diagnostic procedures), the data collection procedures (recall period, interviewer's characteristics, respondent's characteristics), and the underlying distribution of cause-specific mortality in a given population [33–35]. Although there is variation in the sensitivity and specificity for specific conditions, VAs are now of established value in helping to classify the broad patterns of mortality for childhood deaths in populations that are not covered by adequate medical services. VAs have also been used to assess the causes of maternal deaths [36].
Background work for this study included two validation studies of adult deaths in India [37–39]. The first study [37,38] developed and tested a VA instrument among 48,000 adult deaths in urban Chennai and 32,000 adult deaths in rural Tamil Nadu, including a 5% random resample. VA conducted by trained nonmedical field-workers resulted in 90% successful reporting on cause of deaths for middle-age adults (25–69 y). The VA instrument reduced the proportion of adult (age 25 or older) deaths attributed to unspecified or unknown causes from 54% to 23% in urban areas and from 41% to 26% in rural areas. VA yielded fewer unspecified causes (only 10%) than the death certificate (37%)–particularly for the deaths that did not occur in hospital–and often yielded somewhat more specific information, for example, about the approximate site of origin of a cancer, or about evidence of tuberculosis, stroke, myocardial infarction, or diabetes (Table 2). The urban Tamil Nadu study also compared VA results to those from a Chennai population-based cancer registry. The VA sensitivity to identify cancer was 95% in the age group 25–69 y, and VA identified 288 deaths that were not registered in the Chennai Cancer Registry.
Table 2 Cause of Death Based on Vital Statistics and VA of 48,000 Adult (>25 y of Age) Deaths in Chennai, India: 1995–1997a
The second validation study compared all-cause mortality determined by VA against hospital-based records for 262 adult deaths in northern India [39]. Deaths characterized with VA were compared to medically certified cause-of-death certificates for patients who had died in a hospital. Cause-specific mortality fractions assigned by the VA method were statistically similar to the causes arrived at by review of hospital records (p > 0.05). Specificity was high (>95%) for all broad cause groups except cardiovascular (79%) diseases. Sensitivity was highest for injuries (85%), and it was in the range of 60% to 65% for circulatory diseases, neoplasms, and infectious diseases. Sensitivity was low (20% to 40%) for respiratory, digestive, and endocrine diseases. These figures are broadly in agreement with the results from a multicenter validation study of VA for adult deaths conducted in Africa, which found a sensitivity and specificity of 82% and 78%, respectively, for all communicable diseases, and a sensitivity and specificity of 71% and 87%, respectively, for all noncommunicable diseases [34].
Preliminary results from two states indicate that the distribution of underlying causes of death based on the random reinterview does not differ substantially from the cause of death derived from the VA reports of the original RGI surveyors (Table 3).
Table 3 Cause of Death Results for RGI Surveyors and Resample Teams in Tamil Nadu, Maharashtra, and Goa Using VA (Deaths >28 d) in 2003
Validation Studies of Exposures Measured at Baseline
Baseline exposures were captured in a one-time SFMS conducted within the SRS in February 1998. Baseline exposures captured in the SFMS included socioeconomic information (education, occupation, household income, household composition), water and sanitation facilities and other living conditions, smoking and its type (bidi, cigarette, hukka, or other), age at onset, alcohol use and frequency per week, past history of various medical conditions, and the type of cooking fuel used (major and secondary, and use of a separate kitchen). The survey also recorded deaths, although not their causes, in 1997 [31].
The 2004 baseline survey in the new SRS sample frame (2004–2014) recorded similar exposures and added history of disability from various medical conditions, recent short-term illnesses, tobacco smoking and chewing, alcohol use, vegetarianism, and maternal history, including contraceptive use and pregnancies.
The reliability and representativeness of the SFMS baseline survey can be assessed, in part, by comparing the age-specific prevalence of smoking and alcohol consumption among males found in the SFMS and other standard surveys, such as the Indian Census or NFHS-2. Text S1 provides additional validation studies, such as birth order of children and measures of indoor air pollution.
Age-Specific Prevalence of Smoking in Adult Males
Smoking is currently common only among Indian males. The SFMS and NFHS-2 report very similar trends in age-specific prevalence of current male smokers. The steepest increase in prevalence of smoking is between 20–30 y of age (Figure 3). Similarly, the prevalence of smoking among males across 25 states shows a strong correlation between the two studies (Pearson correlation coefficient or r
2 of 0.92; data not shown). We plan to remeasure the extent of smoking by each of several thousand living adults so as to correct for regression-dilution bias [13]. Additionally, to verify and validate smoking status, we will use a simple hand-held carbon monoxide breathalyzer on about 1,000 smokers and nonsmokers in select states. These simple breathalyzers appear to be effective at detecting smoking status [40].
Figure 3 Prevalence of Current Male Smokers: SFMS 1998 versus NFHS-2 1998–1999
Age-Specific Alcohol Consumption in Adult Males
As with smoking, a similar pattern in age-specific alcohol consumption is captured by the SFMS and NFHS-2 (Figure 4). The discrepancy in absolute prevalence for each age group may be due to the difference in the sex of the respondents, where females are either overreporting male alcohol consumption in the household in the NFHS-2, or males are underreporting their own consumption in the SFMS. It should also be noted that, unlike the NFHS-2, usually the male head of the household is the respondent of the SFMS.
Figure 4 Prevalence of Current Male Alcohol Drinkers: SFMS 1998 versus NFHS-2 1998–1999
Retrospective Household Case-Control and Proportional Mortality Methods
The determinants of death can be identified by comparing risk factors between the dead and living. Such household case-control studies use the dead as cases and their surviving spouses or close relatives as controls. A retrospective study in Chennai of 43,000 male deaths and 35,000 living controls [41], using these methods, has documented that throughout middle age, the death rates from medical causes of smokers were double those of nonsmokers (standardized risk ratio at ages 25–69 of 2.1, with 95% confidence interval 2.0–2.2, smoking-attributable fraction 31%). A large part of this excess risk was from tuberculosis and vascular deaths. If these hazards are similar across India, then about half of all tuberculosis deaths in India could be accounted for by smoking. We will apply similar retrospective methods for smoking, tobacco chewing, and alcohol, as these exposures are gathered for dead adults and from a living household respondent.
Simply asking about the dead person's risk factors could be useful. A recent retrospective study of 1 million deaths in China compared the proportions of smokers and nonsmokers who have died of tobacco-attributable diseases versus nontobacco-related diseases (chiefly injuries), to calculate the excess in smokers [42]. Our preliminary pilot studies among childhood deaths in Northern India found that 61% of children (16/26) who died of vaccine-preventable diseases were not immunized in comparison to 40% of control children (18/45) who died from injuries. The crude odds ratio of 2.4 suggests that half of the vaccine-preventable child deaths need not have occurred.
Using these two retrospective methods, we are conducting validation studies of selected risk factors such as childhood immunization, childhood malnutrition, alcohol, male time away from home (as a proxy of HIV-1-related sexual risk taking), and other variables.
Discussion
Key Design Challenges
Several challenges stand out in the design and implementation of this massive study. First is its sheer size. A total of 14 million people is a large sample frame, but commensurate with the needs of monitoring health status among the one billion people in India. The study builds upon the routine infrastructure that has been in place for over 30 y within the SRS. It works with the leading demographic research organization in India (for example, the decennial census involves the RGI hiring two million enumerators to survey about 150 million households within 25 d). Second is the need for simplification of methods to make such a scale of study practicable. The study is not the only epidemiological study needed in India, but its unique focus and scale raise important challenges in the development of new methods. The Million Death Study uses a “large simple” design, which places specific demands on ensuring that new procedures (such as the new VA instrument and physical and biological measurements) are rigorously tested, piloted, and simplified. The third challenge is sustainable funding. The current SRS sample frame is being followed on a total incremental budget of US$2 million (or less than 33¢ per person). This does not include the core funding of the RGI surveyors and related infrastructure. With inclusion of those, the overall study can still be done in a highly cost-effective manner at well under US$1 per person. Collection, processing, and long-term storage of biological samples is expected to cost US$2–5 per person for dried blood spots and US$10–20 per person for a 10-ml tube of blood. These costs compare extremely favorably with those existing biobanks in the United Kingdom and elsewhere [43,44]. Most of the infrastructure costs will be sustained by the Government of India. However, we believe that redirecting some of the considerable, and often inefficient, spending on monitoring individual disease projects is required to enhance the SRS.
Key Design Issues for Blood-Based Genetic Epidemiology
Discovery of “new” risk factors should benefit from the recent and extraordinarily rapid progress by many different research groups and biotechnology companies in developing low-cost, miniaturized methods for the simultaneous assay in small volumes of blood (or, perhaps, dried blood spots) of vast numbers of nucleic acid fragments, host genetic factors, proteins, small molecules, and pathogens [45]. This rapid biotechnological progress has been backed by increasingly sophisticated computer software to help interpret the mass of numerical information that can be generated from each person's blood, yielding within the next few years many, as yet unforeseen, qualitatively different analytic capacities. Appropriately large-scale epidemiological studies that acquire blood (or other) samples from individuals and systematically link them to relevant measures of disability and future mortality are required to make such technological progress relevant to human populations. It is particularly important for such studies to address the shortcomings of the existing biological sample collections that are now being undertaken [43,44], including the particular circumstances of India and the specific infectious diseases common to developing countries.
Key issues which arise in moving to blood-based epidemiology include the choice of blood sample, long-term storage and retrieval, and statistical design issues.
Choice of Blood Sample
Biological specimen collection procedures will build on experience in China, India, and elsewhere. We will undertake pilot studies to test the feasibility and acceptability of methods, to ensure that biological samples collected can serve as a long-term source of DNA for genotyping and that material collected is suitable for biochemical, hematological, proteomic, and other assays [46]. Major options include a nonfasting blood sample collected into one 10-ml ethylenediaminetetraacetic acid Vacutainer, which has been shown previously by the University of Oxford laboratory to allow a wide range of assays [47]. This system is used by the ongoing Chinese Kadoorie Study of 500,000 adults and by the UK Biobank project of the same size [43]. Alternatively, dried blood samples on filter paper have the advantages of easy storage and transport, as well as being less intrusive (i.e., by finger prick rather than by venipuncture). Dried blood spots have been recommended by the World Health Organization for use in field HIV-1 investigations and have been used in various studies within India [48,49].
We currently plan pilot studies of dried blood spots or tubes of blood among 4,000–9,000 adults in several SRS units in four to five states. These pilots will focus on standardizing methods to obtain anthropometric, behavioral, and physiologic measurements and to evaluate the alternative methods for collection of biological specimens for their utility, cost, and practicability (dried blood spot or tube of blood). The pilots will focus on feasibility of field methods, simplification of approaches, and quality control. The results of the pilot studies, which are expected by June 2006, will help inform the design of the larger survey. A special survey will help define practicable questions on HIV-1 risk behaviors.
Long-Term Storage, Retrieval, and Analyses
We are undertaking systematic reviews of the literature and reviewing available assays in India to examine which current bioanalytic methods can be used to test current hypotheses (either for correlates of infection or chronic disease). Using the above pilot samples, we will develop testing and analytic methods for biological samples that will permit high-throughput, low-cost, and high-quality assays to be run and also permit the long-term reliable storage and retrieval of such samples. Dried blood spots may well be stable in a basic 4 °C refrigerator, with minimal storage requirements [46]. With a population of one billion, India certainly requires at least two major biorepositories (including splitting samples, which safeguards against loss at one facility). We are developing plans for long-term biorepositories with the ability to store samples for decades. Much will depend on the choice of the final assay.
Nested Case-Control Methods and Genetic Association Studies
There is a long list of biological factors in blood that might be correlates of disease-specific mortality. To be efficient, we will use a nested case-control approach in our prospective study. Biological samples are taken from all adults in a baseline survey and stored long-term; then, when sufficient numbers of cases with the disease of interest have died (based on the 6-mo follow-up of the causes of death), aliquots from those cases are retrieved from storage, plus aliquots from a few matched controls per case; and, finally, the factors of interest are assayed in these cases and these controls. This design is similar to that in the Chinese Kadoorie Study and the UK Biobank project [43]. As noted above, the number of deaths is likely to be substantial for most of the common diseases of interest to make this an efficient strategy.
Under some circumstances, studies that used population-based controls to study gene-disease associations have been biased due to underlying variation in gene frequencies between populations (“population stratification” [50]). Genetic epidemiology can provide reliable population-based estimates of disease–allele frequency, penetrance, and attributable risk, particularly if designs that account for this bias are employed, and accordingly this has led to greater emphasis recently on family-based association designs. However, there is also evidence that well-designed population-based studies are sometimes superior [51]. The SRS offers a unique opportunity to explore statistical design issues, as both general population and family-based sampling is possible. Family-based studies allow for population-specific estimates of clustering of disease and correlation (heritability). Further, families with a large number of siblings could be oversampled to increase power for genetic linkage studies. The use of “genomic controls” where anonymous markers are genotyped to test for population stratification can be employed [52].
Limitations
The study has several limitations. First, despite a very large sample size, the statistical power remains modest for less common causes such as those leading to maternal deaths. However, for most of the major public health conditions of importance, sufficient events should occur to generate plausible absolute rates and relative risks for risk factors. Second, careful attention in design means that most identified biases should be minimized and should ensure high internal consistency of fieldwork and coding. However, periodic revalidation of mortality outcomes against external standards will be needed. Similarly, continuous improvement of exposure measurements, partially through careful pilots, will be needed. Third, VA yields broad classification of the underlying causes in about 90% of deaths before age 70. In old age, however, the proportion of classifiable deaths is lower. For some specific conditions, such as childhood pneumonia, the cause-specific fractions may be difficult to estimate below certain levels [53].
Significance to Global Health
The Million Death Study will reliably document not only the underlying cause of child and adult deaths, but also key risk factors (behavioral, physical, environmental, and eventually, genetic). It offers a globally replicable model for reliably estimating cause-specific mortality, using VA, and strengthens India's flagship mortality monitoring system. Despite the misclassification that is still expected, the new cause-of-death data will be substantially better than that available previously in India. The study builds India's capacity for research and for public health action. It provides a large, representative, low-cost and long-term system to reliably track the health status of one billion people for the next decade or longer.
Supporting Information
Text S1 Supplemental Text October 14
(80 KB DOC).
Click here for additional data file.
Text S2 Ethics Approval ICMR
(132 KB JPG).
Click here for additional data file.
Text S3 Ethics Approval PGI
(117 KB JPG).
Click here for additional data file.
Patient Summary
Background
A few pieces of information are crucial to measure the health status of a population, which permits a sound basis for health policy. One basic question is which diseases people die from. To answer it, one needs to know the numbers and causes of death. The age at which people die is critical to estimate the burden for society and to develop strategies to prevent some of the deaths. In developed countries, doctors determine and certify the causes of most deaths that occur in hospitals. However, the majority of deaths worldwide in developing countries occur at home and without reliable recording of their causes and distribution. The present study takes place in India. About one in six deaths worldwide occurs in India, or about 9.5 million deaths per year. Only about a third of these deaths are registered.
Why Is This Study Being Done?
The Indian government recognized the need for better data on the distribution and causes of deaths across the country and, together with an international group of researchers, is undertaking a large study over 16 years (from 1998 to 2014) to collect representative data for the country.
What Are the Researchers Doing?
In this article, the researchers who came up with the plan for the survey and will oversee its implementation describe and discuss the details of the project. They will monitor, through regular visits and interviews by surveyors, the health status of nearly 14 million people in 2.4 million representative households from all over India. The surveyors, who will be trained for the job but not have a medical background, will collect information about the health status of all household members and about key risk factors for disease such as smoking, alcohol use, childhood immunization, and indoor air pollution. About one million deaths are expected to occur among these people in the study period. As part of their interviews, the surveyors will conduct “verbal autopsies,” that is, ask specific questions about how someone died, to determine the causes of death. Each verbal autopsy will be checked by two independent physicians (and a third if there is disagreement) before the cause of death is recorded. The researchers also describe their plans to extend the survey–they plan to collect simple physical measurements such as blood pressure, height and weight, and blood samples (most likely through a simple spot of blood taken from the finger) from the participants. These biological samples contain information about the genetic makeup of the individuals and, potentially, on other markers that might be connected to the cause of death, such as exposure to environmental toxins, the presence of viral or bacterial infections, and others. This information will help them to understand what caused the diseases that the people died from.
What Does This Mean?
This study will provide the most accurate picture of causes of childhood and adult death in India, as well as document key risk factors. By studying diseases that are common in one part of India but not in another, new risk factors will be identified and can be used to reduce deaths overall. It will also serve as an example for other countries.
Where Can I Find More Information Online?
The following Web sites provide relevant information on this and similar large studies.
Full protocol, field instruments, training manuals, and more details for this study:
http://www.cghr.org/project.htm
Home page of the UK Biobank:
http://www.ukbiobank.ac.uk/
Kadoorie study of chronic disease in China:
http://www.ctsu.ox.ac.uk/~kadoorie/public/
The World Health Organization's Health Metrics Network:
http://www.who.int/healthmetrics/en/
Indepth Network for Continuous Demographic Evaluation of Populations and Their Health in Developing Countries:
http://www.indepth-network.net
The largest proportion of the study costs are met by the Government of India as part of the routine costs of running the SRS. External funding from the study comes from the National Institute of Health Tobacco Research Grant (R01 TW05991–01; Aron Primack), the Canadian Immunization Initiative of the International Developmental and Research Centre (Grant number 102172; Sharmila Mhatre), and the Canadian Institute of Health Research (Establishment Grant number IEG-53506; Mark Bisby). Funding also comes from unrestricted grants from the McLaughlin Centre for Molecular Medicine, University of Toronto (No. 1901643995; Duncan Stewart), St. Michael's Hospital (Arthur Slutsky), and University of Toronto (Harvey Skinner). PJ is supported by a Canada Research Chair of the Government of Canada. We thank Dr. Paul Doherty for editorial assistance. External funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
RGI-CGHR Prospective Study Collaborators
RGI-CGHR National Centre, Office of the Registrar General, RK Puram, New Delhi, India: DK Sikria, RC Sethia,
b, N Dhingraa,
b, DK Dey, M Jain, S Jain, K Lal, and L Sushant.
Indian Academic Partners
Clinical Epidemiology Resource and Training Centre, Trivandarum, India: KB Leena and KT Shenoya.
Department of Community Medicine, Gujarat Medical College, Ahmedabad, India: DV Bala, P Seth, and KN Trivedia.
Department of Community Medicine, Kolkatta Medical College, Kolkatta, India: SK Roya.
Department of Community Medicine, Osmania Medical College, Hyderabad, India: P Bhatiaa.
Department of Community Medicine, Regional Institute of Medical Sciences, Imphal, India: L Usharania.
Department of Community Medicine, SMS Medical College, Jaipur, India: AK Bharadwaja.
Epidemiological Research Centre, Chennai, India: V Gajalakshmia,
b.
Gandhi Medical College, Bhopal, India: R Dikshita and S Sorangi.
Healis-Seskarhia Institute of Public Health, Navi Mumbai, India: PC Guptaa,
b, MS Pednekar, and S Sreevidya.
Institute of Population Health and Clinical Research, St. John's Medical College, Bangalore, India: A Kurpad, P Monya,
b, and M Vaz.
King George Medical College, Lucknow, India: S Awasthia.
North Eastern Indira Gandhi Institute of Regional Medical Sciences, Shillong, Meghalaya, India: FU Ahmeda.
Regional Medical Research Center, ICMR Institute, Bhubaneshwar, India: AS Karkettaa and K Dar.
School of Preventive Oncology, Patna, India: DN Sinhaa.
School of Public Health, Post Graduate Institute of Medical Education and Research, Chandigarh, India: N Kaur, R Kumara,
b, and JS Thakur.
Other Partners
Clinical Trial and Epidemiological Studies Unit, University of Oxford, Oxford, England: Z Chen, R Collins, and Sir R Petoa,
b.
Hospital for Sick Children, University of Toronto, Toronto, Canada: A Patterson and S Schrier.
Indian Council of Medical Research, New Delhi, India: NK Gangulya.
Mt. Sinai Hospital, University of Toronto, Toronto, Canada: J McLaughlin.
McLaughlin Centre for Molecular Medicine, University of Toronto, Toronto, Canada: P Jhaa,
b, K Kain and R Kaul.
United Arab Emirates University, Al Ain, UAE: N Naglekerke.
World Health Organization, Geneva, Switzerland: T Boermaa, T Evansa, and K Shibuyi.
World Health Organization, South East Asia Regional Office, New Delhi, India: N Singh and T Sein.
Global Coordinating Centre, Centre for Global Health Research, St. Michael's Hospital, University of Toronto, Canada: B Jacob, P Jha (Principal Investigator)a,b, R Kadmood, C Major, J Mooreb, P Parra, S Sgaier, H Shadmand, P Vasab, and F Zhang.
aMember Advisory Committee.
bWriting committee for this report.
Citation: Jha P, Gajalakshmi V, Gupta PC, Kumar R, Mony P, et al. (2006) Prospective study of 1 million deaths in India: Rationale, design, and validation results. PLoS Med 3(2): e18
Abbreviations
ICD-10
International Statistical Classification of Diseases and Related Health Problems
NFHS-2National Family Health Survey 2
RGIRegistrar General of India
SFMSSpecial Fertility and Mortality Survey
SRSSample Registration System
VAverbal autopsy
==== Refs
References
World Health Organization Reducing risks: Promoting healthy life: World health report 2002 Geneva (Switzerland) World Health Organization 230
Stumbling around in the dark [editorial] Lancet 2005 365 1983
Horton R The Ellison Institute: Monitoring health, challenging WHO Lancet 2005 366 179 181 16023490
Stansfield S Structuring information and incentives to improve health Bull World Health Organ 2005 83 562 563 16184268
Murray CJ Lopez AD Wibulpolprasert S Monitoring global health: Time for new solutions BMJ 2004 329 1096 1100 15528624
Mitra B Asma S Jha A Gupta PC India's mortality measurement systems Counting the dead in India in the 21st century 1999 Atlanta (Georgia) US Centers for Disease Control B2 B8
Jha P Reliable mortality data: A powerful tool for public health Natl Med J India 2001 14 129 131 11467137
Jha P Slutsky AS Brown D Nagelkerke N Brunham BG Health and economic benefits of an accelerated program of research to combat global infectious diseases CMAJ 2004 171 1203 1208 15534314
Doll R Peto R Boreham J Sutherland I Mortality in relation to smoking: 50 years' observations on male British doctors BMJ 2004 328 1519 15213107
Gottlieb MS Schroff R Schanker HM Weisman JD Fan PT Pneumocystis carinii pneumonia and mucosal candidiasis in previously healthy homosexual men: Evidence of a new acquired cellular immunodeficiency N Eng J Med 1981 305 1425 143l
Jha P Avoidable mortality in India: Past progress and future prospects Natl Med J India 2002 15 Suppl 1 32 36 12047131
Lewington S Clarke R Qizilbash N Peto R Collins R Age-specific relevance of usual blood pressure to vascular mortality: A meta-analysis of individual data for one million adults in 61 prospective studies Lancet 2002 360 1903 1913 12493255
Clarke R Shippley M Lewington S Youngman L Collins R Underestimation of risk associations due to regression dilution in long-term follow-up of prospective studies Am J Epidemiol 1999 150 341 353 10453810
Yusuf S Hawken S Ounpuu S Dans T Avezum A Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): Case-control study Lancet 2004 364 937 952 15364185
Danesh J Collins R Peto R Lipoprotein (a) and coronary heart disease: Meta-analyses of prospective studies Circulation 2000 102 1082 1085 10973834
Knoblauch H Bauerfeind A Toliat M Becker C Luganskaja C Haplotypes and SNPs in 13 lipid-relevant genes explain most of the genetic variance in high-density lipoprotein and low-density lipoprotein cholesterol Hum Mol Genet 2004 13 993 1004 15044381
Gonzalez E Bamshad M Sato N Mummidi S Dhanda R Race-specific HIV-1 disease-modifying effects associated with CCR5 haplotypes Proc Natl Acad Sci U S A 1999 96 1204 1209
Shanmugalakshmi S Pitchappan R Genetic basis of tuberculosis susceptibility in India Indian J Pediatr 2002 69 Suppl 1 S25 S28 12501922
MacDonald K Fowke K Kimani J Dunand VA Nagelkerke NJ Influence of HLA supertypes on susceptibility and resistance to human immunodeficiency virus type 1 infection J Infect Dis 2000 181 1581 1589 10823757
Kaul R Kimani J Nagelkerke NJ Fonck K Ngugi EN Monthly antibiotic chemoprophylaxis and incidence of sexually transmitted infections and HIV-1 infection in Kenyan sex workers: A randomized controlled trial JAMA 2004 291 2555 2562 15173146
Nagelkerke NJ de Vlas SJ MacDonald KS Rieder HL Tuberculosis and sexually transmitted infections Emerg Infect Dis 2004 10 2055 2056 16010749
World Health Organization [WHO] International statistical classification of diseases and related health problems 10th rev. Volume 1 2003 Geneva (Switzerland) World Health Organization Available: http://www3.who.int/icd/vol1htm2003/fr-icd.htm . Accessed 18 October 2005
Banthia J Dyson T Smallpox in 19th century India Popul Dev Rev 1999 24 649 680
Caselli G Schofield R Reher D Bideau A Health transition and cause-specific mortality The decline of mortality in Europe 1991 Oxford Clarendon Press 42 57
Gupta PC Sankaranarayanan R Ferlay J Cancer deaths in India: Is the model-based approach valid? Bull World Health Organ 1994 72 943 944 7867141
Dye C Scheele S Dolin P Pathania V Raviglione MC Consensus statement. Global burden of tuberculosis: Estimated incidence, prevalence, and mortality by country. WHO Global Surveillance and Monitoring Project JAMA 1999 282 677 686 10517722
Office of the Registrar General Compendium of India's fertility and mortality indicators 1971–1999 2001 New Delhi Office of the Registrar General 172
Schlesselman S Case-control studies: Design, conduct, analysis 1982 New York Oxford University Press 354
Preston S Bhat P New evidence on fertility and mortality trends in India Popul Dev Rev 1984 10 481 503
Bhat PN Completeness of India's sample registration system: An assessment using the general growth balance method Popul Stud (Camb) 2003 56 119 134
Registrar General of India Special fertility and mortality survey, 1998: A report of 1.1 million households 2005 New Delhi Registrar General 302
Anker M Black R Coldham C Kalter H Quigley M A standard verbal autopsy for investigating causes of death in infants and children. Geneva (Switzerland): World Health Organization. Report Number WHO/CDS/CRS/ISR/99.4 1999 83
Quigley MA Chandramohan D Rodrigues LC Diagnostic accuracy of physician review, expert algorithms and data-derived algorithms in adult verbal autopsies Int J Epidemiol 1999 28 1081 1087 10661651
Chandramohan D Maude GH Rodrigues LC Hayes RJ Verbal autopsies for adult deaths: Their development and validation in a multicentre study Trop Med Int Health 1998 3 436 446 9657505
Kalter HD Gray RH Black RE Gultiano SA Validation of the diagnosis of childhood morbidity using maternal health interviews Int J Epidemiol 1991 20 193 198 2066219
Kumar R Sharma AK Barik S Kumar V Maternal mortality inquiry in a rural community of north India Int J Gynaecol Obstet 1989 29 313 319 2571532
Gajalakshmi V Richard P Santhanakrishnan K Sivagurunathan B Verbal autopsy of 48,000 adult deaths attributed to medical causes in Chennai (formerly Madras), India BMC Public Health 2002 2 7 12014994
Gajalakshmi V Peto R Verbal autopsy of 80,000 adult deaths in Tamil Nadu, South India BMC Public Health 2004 4 47 15488138
Kumar R Thakur J Rao M Singh M Bhatia P Validity of verbal autopsy in determining causes of adult deaths Indian J Public Health 2005 In press
Cunnington AJ Hormbrey P Breath analysis to detect recent exposure to carbon monoxide Postgrad Med J 2002 78 233 237 11930027
Gajalakshmi V Peto R Kanaka TS Jha P Smoking and mortality from tuberculosis and other diseases in India: Retrospective study of 43000 adult male deaths and 35000 controls Lancet 2003 362 507 515 12932381
Liu BQ Peto R Chen ZM Boreham J Wu YP Emerging tobacco hazards in China: 1. Retrospective proportional mortality study of one million deaths BMJ 1998 317 1411 1422 9822393
Biobank Project Biobank Project (2004) 2004 Available: http://www.biobank.ac.uk . Accessed 8 May 2005
Kaiser J Biobanks. Population databases boom, from Iceland to the U.S Science 2002 298 1158 1161 12424349
Varmus H Genomic empowerment: The importance of public databases Nat Genet 2003 35 Suppl 1 3 14578889
Steinberg K Beck J Nickerson D Garcia-Closas M Gallagher M DNA banking for epidemiologic studies: A review of current practices Epidemiology 2002 13 246 254 11964924
Clark S Youngman LD Palmer A Parish S Peto R Stability of plasma analytes after delayed separation of whole blood: Implications for epidemiological studies Int J Epidemiol 2003 32 125 130 12690023
Solomon SS Solomon S Rodriguez II McGarvey ST Ganesh AK Dried blood spots (DBS): A valuable tool for HIV surveillance in developing/tropical countries Int J STD AIDS 2002 13 25 28 11802926
Ramakrishnan L Reddy KS Jailkhani BL Measurement of cholesterol and triglycerides in dried serum and the effect of storage Clin Chem 2001 47 1113 1115 11375303
Stephens JC Schneider JA Tanguay DA Choi J Acharya T Haplotype variation and linkage disequilibrium in 313 human genes Science 2001 293 489 493 11452081
Cardon LR Palmer LJ Population stratification and spurious allelic association Lancet 2003 361 598 604 12598158
Devlin B Roeder K Genomic control for association studies Biometrics 2000 55 997 1004
Williams BG Gouws E Boschi-Pinto C Bryce J Dye C Estimates of world-wide distribution of child deaths from acute respiratory infections Lancet Infect Dis 2002 2 25 32 11892493
|
16354108
|
PMC1316066
|
CC BY
|
2021-01-05 11:14:06
|
no
|
PLoS Med. 2006 Feb 20; 3(2):e18
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0030018
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1635410610.1371/journal.pmed.0030014Research ArticleInfectious DiseasesNeuroscienceGeriatricsPathologyAnalysis of Prion Strains by PrPSc Profiling in Sporadic Creutzfeldt–Jakob Disease PrPSc ProfilingSchoch Gaby
1
Seeger Harald
1
Bogousslavsky Julien
2
Tolnay Markus
3
Janzer Robert Charles
2
Aguzzi Adriano
1
Glatzel Markus
1
*¤1Institute of Neuropathology, University Hospital of Zürich, Zürich, Switzerland2Department of Neurology and Division of Neuropathology, Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland3Institute of Pathology, Department of Neuropathology, University Hospital Basel, Basel, SwitzerlandCaughey Byron Academic EditorNational Institutes of HealthUnited States of America* To whom correspondence should be addressed. E-mail: [email protected]¤ Current address: Institute of Neuropathology, University Medical Center Hamburg-Eppendorf, Hamburg, Germany
Competing Interests: The authors have declared that no competing interests exist.
Author Contributions: GS, HS, AA, and MG analyzed the data. HS, JB, MT, and RCJ enrolled patients. AA and MG designed the study. GS, HS, MT, RCJ, AA, and MG contributed to writing the paper.
2 2006 20 12 2005 3 2 e142 5 2005 14 10 2005 Copyright: © 2006 Schoch et al.2006This 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.
Toward a Better Understanding of Human Prion Diseases
Background
Prion diseases are a group of invariably fatal neurodegenerative disorders affecting humans and a wide range of mammals. An essential part of the infectious agent, termed the prion, is composed of an abnormal isoform (PrPSc) of a host-encoded normal cellular protein (PrPC). The conversion of PrPC to PrPSc is thought to play a crucial role in the development of prion diseases and leads to PrPSc deposition, mainly in the central nervous system. Sporadic Creutzfeldt–Jakob disease (sCJD), the most common form of human prion disease, presents with a marked clinical heterogeneity. This diversity is accompanied by a molecular signature which can be defined by histological, biochemical, and genetic means. The molecular classification of sCJD is an important tool to aid in the understanding of underlying disease mechanisms and the development of therapy protocols. Comparability of classifications is hampered by disparity of applied methods and inter-observer variability.
Methods and Findings
To overcome these difficulties, we developed a new quantification protocol for PrPSc by using internal standards on each Western blot, which allows for generation and direct comparison of individual PrPSc profiles. By studying PrPSc profiles and PrPSc type expression within nine defined central nervous system areas of 50 patients with sCJD, we were able to show distinct PrPSc distribution patterns in diverse subtypes of sCJD. Furthermore, we were able to demonstrate the co-existence of more than one PrPSc type in individuals with sCJD in about 20% of all patients and in more than 50% of patients heterozygous for a polymorphism on codon 129 of the gene encoding the prion protein (PRNP).
Conclusion
PrPSc profiling represents a valuable tool for the molecular classification of human prion diseases and has important implications for their diagnosis by brain biopsy. Our results show that the co-existence of more than one PrPSc type might be influenced by genetic and brain region–specific determinants. These findings provide valuable insights into the generation of distinct PrPSc types.
Internal standards allow meaningful comparison of PrPSc profiles from different origins. Applying the new method to 50 postmortem samples from patients with sporadic CJD reveals substantial differences in molecular pathology.
==== Body
Introduction
Transmissible spongiform encephalopathies or prion diseases are neurodegenerative disorders characterized by posttranslational conversion and cerebral accumulation of a pathological isoform (PrPSc) of a host-encoded membrane-associated glycoprotein (cellular prion protein, PrPC) [1]. These diseases include scrapie in sheep, bovine spongiform encephalopathy in cattle, chronic wasting disease in cervids, and sporadic, genetic, or acquired human prion diseases [2]. The most common human prion disease, the sporadic form of Creutzfeldt–Jakob disease (sCJD), is clinically characterized by rapidly progressive dementia, neurologic dysfunction including myoclonic involuntary movements, and, finally, a terminal state of severe cognitive impairment leading to death within months from the onset of clinical symptoms [3]. Various clinical phenotypes can be observed, e.g., patients with ataxia opposed to dementia as the initial symptom or patients with anopsia as a prominent clinical feature. This marked clinical heterogeneity observed in sCJD is reflected by the presence of diverse sCJD types [4] and is not yet fully understood.
It is known that various isolates, or strains, of prions may be propagated within genetically identical hosts, leading to distinct clinical and pathological features [5,6]. The apparent diversity of human prion diseases may, at least in part, be attributed to the presence of distinct prion strains in affected individuals. In agreement with this, distinct clinical phenotypes of human prion diseases present with diverse deposition patterns of PrPSc in addition to disparate biochemical properties of PrPSc [7,8]. Since nucleic acids do not copurify with prion infectivity and prion strains do not seem to be encoded by differences in PrP primary structure, the hypothesis that PrPSc itself may encode strain-specific phenotypic properties within its tertiary or even quaternary structure has to be considered [9,10]. This assumption is supported by the presence of strain-specific N-terminal cleavage of PrPSc upon Western blot analysis following limited digestion with proteinase K [9].
Another important factor in phenotype variability is the host genotype of the gene encoding the prion protein (PRNP). It is well established that the polymorphism at codon 129, encoding either methionine or valine in PRNP, influences the disease phenotype [8].
Thus, the classification of human prion diseases is based on the clinical presentation of the affected individual, PRNP status in concert with neuropathological findings, and the biochemical analysis of PrPSc [11–13]. Previous studies have focused on the assessment of the fragment sizes of unglycosylated PrPSc and the ratios of the three glycoforms of PrPSc as seen on Western blot following protease treatment [12,13]. Although there is some disparity in the exact number of mobilities of the unglycosylated PrPSc fragment, researchers agree that this fragment principally migrates between 19 and 21 kDa [12,13]. Since the resolution of one-dimensional gel electrophoresis is limited, it is likely that additional species of N-terminal fragments will be identified if more accurate techniques, such as two-dimensional immunoblot or mass spectrometry, are utilized. In fact, recent studies using two-dimensional gel electrophoresis of human sCJD brain homogenates strongly suggest that this is the case [14,15]. In order to simplify comparability of our study to published work, we decided to distinguish two principal mobilities of unglycosylated PrPSc i.e., 21 kDa (high) and 19 kDa (low), thereby accepting that both the high and the low PrPSc fragments show a certain degree of variation.
In agreement with the fact that the predominant PrPSc type may be indicative of the sCJD type, previous studies have demonstrated that the co-existence of more than one PrPSc type is a rare event [8]. In the light of these data, the finding that the co-existence of more than one PrPSc type within one individual occurs in 30% of all patients if multiple regions of the brain are investigated [16] came as a surprise. Because of the limited sample size in the latter study, the question of the true incidence of co-existence of more that one PrPSc type within one individual has remained unanswered.
It has been shown that the spatial analysis of histopathological changes often referred to as the lesion profile may provide a tool to discriminate various clinically diverse types of prion diseases [8,17]. A major drawback of this method, however, is the considerable inter-observer variability. Thus it has been virtually impossible to compare published lesion profiles [8,13].
Here, we assess the distribution of PrPSc types in nine central nervous system regions in a cohort of 50 patients with sCJD, using PrPSc profiling. Furthermore, we address the question of co-existence of more than one PrPSc type in sCJD-affected individuals.
Methods
Selection of Patients
Cases in this study were derived from an unselected series of patients with clinically, genetically, and neuropathologically proven sCJD (Tables 1 and 2). All tissue specimens originated from patients referred to the Swiss National Reference Centre for Prion Diseases, and the specimens were processed according to established guidelines regarding safety and ethics. Frozen brain tissue was stored at –80 °C and samples were taken from the following brain regions: frontal, parietal, occipital, and temporal cortex, and the putamen, thalamus, midbrain, medulla oblongata, and cerebellum.
Table 1 Demographic Characteristics and Classification of Patients with sCJD Included in the Study
Table 2 PrPSc Quantities in Defined Brain Areas
For positive control, neuropathologically proven sCJD cases (Codon 129 VV, PrPSc type 2 and Codon 129 MM, PrPSc type 1) were chosen. For negative control, a non-demented, age-matched autopsy case was utilized.
Protein Analysis
Brain-tissue homogenates (10% w/v) were prepared in homogenization buffer (100 mM NaCl, 10 mM EDTA, 100 mM Tris-HCl, 0.5% NP40, 0.5% NaDOC [pH 6.9]) using a RiboLyser (Hybaid, Ashford, United Kingdom) and stored at –80 °C until use. Samples, containing 25 μg of protein, were digested with Proteinase K (PK-recombinant PCR-grade solution with a specific activity of 41.3 U/mg, measured by a hemoglobin-activity test [Roche, Basel, Switzerland]) for 30 min at 37 °C with a concentration of 0.03 U (40 μg/ml) per sample. Proteinase K was stored at –80 °C in storage buffer (50% glycerol, 10 mM Tris-HCl [pH 7.5], 2.9 mg/ml CaCl2). Proteins were separated on a 12% SDS-PAGE (10 × 10.5 cm, Hoefer, San Francisco, California, United States) and then transferred (1.5 h at 250 mA) to a nitrocellulose membrane (Protran, Schleicher & Schuell, Dassel, Germany) using a wet-blotting system (Bio-Rad, Hercules, California, United States). Membranes were incubated overnight with monoclonal antibody 3F4 [18] (Signet, Dedham, Massachusetts, United States) at a dilution of 1:2,000. After washing, HRP-conjugated rabbit-anti-mouse-IgG-γ (Zymed, San Francisco, California, United States) served as the secondary antibody at a dilution of 1:20,000. The signal was visualized by enhanced chemiluminescence using a VersaDoc 5000 imaging station (Bio-Rad). Cases were typed according to the size of the protease-resistant unglycosylated PrP fragment [12,13], and relative quantification of the signal was performed employing Quantity One software (Bio-Rad).
To quantify PrPSc, serial dilutions of a standard sCJD brain homogenate (33, 16.5, and 8.25 μg of protein) were used on each Western blot to generate a standard curve. Quantification was performed only if the square of the correlation coefficient of the standard curve was above 0.93 (r2 > 0.93).
Selected cases were separated by SDS-urea-PAGE. For this, we adapted a published protocol [19]. Briefly, the stacking gel was composed of 11.1% of a 60% acrylamid/0.8% bis-acrylamid mixture, 27.7% Tris-HCl 0.6M (pH 6.8), 1.3% SDS, 1.6% APS, 0.2% Temed (N,N,N,N,-tetra-methyl-ethylenediamine [Bio-Rad]), and 8 M urea, and the resolving gel was comprised of 44% of a 60% acrylamid/0.8% bis-acrylamid mixture, 53.9% Tris-HCl 1.875 M (pH 8.8), 1.4% SDS, 0.6% APS, 0.1% Temed (N,N,N,N,-tetra-methyl-ethylenediamine [Bio-Rad]), and 8 M urea.
PRNP Analysis
Genomic DNA was extracted from the buffy-coat fraction of peripheral blood or from brain tissue. The complete PRNP open reading frame was analyzed using established methods [20].
Neuropathology
Tissue from selected patients (n = 32) was fixed in 10% formalin. Tissue blocks from frontal, parietal, occipital, and temporal cortex, and the putamen, thalamus, midbrain, medulla oblongata, and cerebellum, localized adjacent to the specimens used for biochemical analysis, were selected, decontaminated for 1 h with 98% formic acid and embedded in paraffin. Sections (3 μm) were subjected to conventional staining and to immunostaining for glial fibrillary acidic protein (Dako), and for PrP (3F4), upon hydrolytic autoclaving according to published protocols [20].
Spongiosis was evaluated on a 0–4 scale (not detectable, mild, moderate, severe, and status spongiosus). Gliosis and PrPSc content were scored on a 0–3 scale (not detectable, mild, moderate, severe). PrPSc accumulation patterns were described as synaptic, perivacuolar, patchy, or plaque-like. Averaging of the three scores resulted in the value that was employed in order to obtain the lesion profile for individual patients [8]. Histological analysis was performed by two independent investigators blinded to clinical data, codon-129 status, and PrP biochemistry.
Results
Reliable Quantification of PrPSc Using Internal Standards
Routinely, quantification of PrPSc is performed by comparing the signal intensities on Western blot analysis following limited proteinase K digestion. While this method is highly specific and objective, it does not allow for the comparison of samples run on separate gels owing to variations in the blotting process, antibody binding efficiencies, or differences in visualization reaction. In order to circumvent these problems, we developed a system whereby these variables are compensated for by quantifying the signal intensity of a serially diluted PrPSc standard on every Western blot. The resulting standard curve is then used to quantify PrPSc content in defined brain regions. The reliability and feasibility of this method was assessed by analyzing nine different brain areas on a cohort of 50 patients with sCJD (see Table 1; Figure 1). Standard curves were accepted only if the square of the correlation coefficient was above 0.93. Our PrPSc standard is available to the scientific community and should facilitate comparison of PrPSc quantities present in defined samples.
Figure 1 Comparison of PrPSc Profiles to Lesion Profiles
(A) Schematic drawing indicating sampled areas which are used for the generation of PrPSc profiles (red boxes).
(B) Biochemical PrPSc profiles are indicated in the upper panel, whereas histological lesion profiles are shown in the lower panel. Patients were grouped according to [12] in MM1, MV1, MV2, and VV2. Brain regions are shown on the x-axis. Values for PrPSc amounts are given in arbitrary units measured in relation to the PrPSc standard. Values for lesion profiles were obtained by averaging the scores for spongiosis (scored on a scale from 0–4), astrogliosis, and PrP immunoreactivity (both scored on a scale from 0–3). Black dots represent individual patients and black lines within boxes represent medians; boxes encompass 25th and 75th percentiles of distribution. One outlier (denoted by asterisk: MV2, cerebellum, PrPSc content = 84.1) has been omitted in the graphical representation.
Molecular Analysis of sCJD Using PrPSc Profiling
Nine defined brain areas of 50 subjects with sCJD were analyzed. They showed a mean age of 69.1 y (± 8.5) and a gender ratio of 28 males to 22 females. Patients were grouped according to both their genotype at codon 129 of the PRNP gene and to their predominant PrPSc type (Table 1), in line with the classification introduced by Parchi [8], and included MM1 (n = 30), MV1 (n = 5), MV2 (n = 6), VV2 (n = 6), MM2 (n = 2), and VV1 (n = 1).
Analysis of the relative amounts of PrPSc in various brain regions revealed distinct distribution patterns, which seem to be dependent on the type of PrPSc and on the host's codon-129 polymorphism. Cases with a slower-migrating unglycosylated band of PrPSc (∼21 kDa) and homozygosity for methionine on codon 129 (MM type 1) harbor significant amounts of PrPSc in cortical regions (frontal, parietal, temporal, and occipital) and the thalamus, smaller and variable amounts in the putamen and cerebellum, and very low or undetectable amounts in the midbrain and medulla oblongata (Figure 1). PrPSc quantities for this cohort are listed in Table 2.
In contrast, cases with a faster-migrating unglycosylated band of PrPSc (∼19 kDa) and homozygosity for valine on codon 129 (VV type 2) had only scant amounts of PrPSc in the cerebral cortex, but showed marked PrPSc deposition within the putamen, thalamus, the midbrain, the cerebellum and, to a lesser extent, in the medulla oblongata, although this was clearly detectable (Figure1). PrPSc quantities for this cohort are listed in Table 2.
Whereas the deposition patterns described above were relatively uniform in the respective groups, the picture is not so obvious for patients heterozygous for methionine or valine at codon 129. In this group, patients with a slower-migrating unglycosylated band of PrPSc (∼21 kDa, MV1) exhibited deposition patterns reminiscent of MM1 individuals. However, PrPSc deposition was more abundant in the temporal than in the occipital lobe, and deposits within the putamen showed greater variability (Figure 1).
Patients with a faster-migrating unglycosylated band of PrPSc (∼19 kDa, MV2) exhibited abundant PrPSc deposition in the putamen, thalamus, midbrain, and cerebellum, clearly detectable deposits in the medulla, and, in addition, variable deposits within the cerebral hemispheres (Figure 1). PrPSc quantities for these two cohorts and for patients either homozygous for methionine (MM2, n = 2) or valine (VV1, n = 1) are listed in Table 2.
Because this method possesses some similarities with lesion profiling, usually used to discriminate between sCJD types or between prion strains in rodents, we have termed this method “PrPSc profiling” [8,17].
High Proportion of PrPSc Type Co-Existence in Patients Expressing MV at Codon 129
Studies on relatively small cohorts of patients with sCJD evidenced the co-occurrence of different PrPSc types within one individual [16]. We analyzed nine distinct brain regions of 50 patients with sCJD, resulting in a total of 450 analyzed specimens. In patients where routine SDS-PAGE analysis suggested possible co-occurrence of distinct PrPSc types, high-resolution SDS-urea-PAGE was performed (n = 13).
A total of nine sCJD individuals showed non-ambiguous co-occurrence of distinct PrPSc types as evidenced by the presence of a faster- and a slower-migrating unglycosylated band of PrPSc, either in different brain areas or within one brain region (Figure 2). In the remainder of the cases (n = 41), only a single PrPSc type could be detected in all nine different brain areas. In cases where more than one PrPSc type was observed, the topological distribution of distinct strain types revealed no obvious pattern (Table 3). However, the cerebral hemispheres appeared to be less affected than the thalamus, putamen, cerebellum, midbrain, and medulla oblongata. Whereas co-occurrence of both PrPSc types was found in only two cases within the frontal lobe, the midbrain was affected in three cases, the medulla in four cases, and the putamen, thalamus, and cerebellum each in five cases.
Figure 2 Western Blots Illustrating the Principle of PrPSc Profiling and the Co-Occurrence of Multiple PrPSc Types within One Patient with sCJD
(Upper panel) The PrPSc content in various central nervous system regions (indicated below respective lanes) was quantified by plotting the signal intensity on a standard curve created by a serially diluted PrPSc standard (standard 1 to 3). Proteinase K digestion is indicated above lanes.
(Lower panel) Presence of a faster-migrating unglycosylated band of PrPSc (∼19 kDa, PrPSc type 2, frontal and temporal cortex) and a slower-migrating unglycosylated band of PrPSc (∼21 kDa, PrPSc type 1, occipital cortex) within one patient with sCJD (codon 129 MM). Controls include PrPSc type 1 and 2 and a sample from a non prion-diseased individual. Proteinase K digestion is indicated above lanes.
Table 3 Different PrPSc Types in Defined Central Nervous System Regions of sCJD Patients
The most striking feature of patients harboring two PrPSc types was their association with the methionine/valine polymorphism at codon 129. Six out of 11 methionine/valine heterozygotes displayed both PrPSc types, whereas only three of 32 individuals homozygous for methionine produced more than one PrPSc type. In the groups of patients expressing valine/valine on codon 129, we were not able to detect more than one PrPSc type.
Neuropathological Analysis
The analysis of the neuropathological lesion pattern was performed according to published protocols [8,13] with the purpose to: (i) evaluate whether immunohistochemical PrP staining and histological lesions (gliosis, spongiosis) correlate with the absolute amount of biochemically assessed PrPSc, and (ii) compare whether lesion profiles generated by histological assessment parallel biochemically defined PrPSc profiles. In order to facilitate these analyses, patients with sCJD were subdivided into following groups: MM1, MV1, MV2, VV2 [12]. Patients presenting with multiple PrPSc types were allocated according to the dominating PrPSc type.
The MM1 group (n = 20) showed marked involvement of the frontal and temporal cerebral cortex and slightly fewer lesions in the cerebellum, whereas the putamen, thalamus, midbrain, and medulla oblongata showed relatively modest pathological alterations (see Figure 1).
In the MV1 cohort (n = 4), a similar lesion profile could be observed (Figure 1). The observation that this group showed a greater variability might in part be attributable to the smaller sample size.
MV2-predominant patients (n = 4) exhibited a consistently high degree of neuropathological lesions with relative sparing of the occipital cortex and the cerebellum.
In the VV2 group (n = 4), pathological alterations were most pronounced in the cerebellum and in the cerebral cortex, as well as in the thalamus and midbrain. The medulla oblongata did not show a prominent involvement (Table S1).
The correlation between histological lesions as assessed by lesion profiling and the presence of PrPSc evaluated by PrPSc profiling was not striking. Although brain regions with abundant PrPSc tend to show severe histopathological lesions, the reverse does not always seem to be the case. Brain regions with massive histopathological lesions (i.e., cortical regions in VV2 patients) show only scarce depositions of PrPSc.
The comparison of lesion profiles to PrPSc profiles demonstrates that the patterns obtained within the above-mentioned groups are only roughly concurrent. Whereas PrPSc profiles show marked differences between investigated regions, lesion profiles tend to show a more homogeneous picture with fewer marked regional differences.
In cases in which multiple PrPSc types could be observed, we investigated whether the predominant deposition pattern of PrPSc was associated with a particular biochemical type of PrPSc. Areas with a predominance of PrPSc type 1 (∼21-kDa unglycosylated band of PrPSc) displayed diffuse, synaptic PrP immunoreactivity, whereas areas with a PrPSc type 2 (∼19-kDa unglycosylated band of PrPSc), showed mainly perivacuolar or plaque-like deposits of PrP (Figure 3).
Figure 3 Histological Analysis of Central-Nervous-System Sections in One Patient with sCJD with Co-Occurrence of Multiple PrPSc Types
H&E-stained (A and B) and PrP-immunostained (C and D) cortical areas showing pronounced spongiosis and deposition of PrPSc in a plaque-like (C; PrPSc type 2, see Figure 2, frontal area on the lower panel) and synaptic (D; PrPSc type 1, see Figure 2, occipital area on the lower panel) pattern.
Discussion
The precise classification of sCJD by molecular, histopathological, and clinical parameters is important for understanding the underlying pathogenesis, unraveling possible etiological causes, and devising efficient therapeutic protocols. Several classification systems for human prion diseases have been proposed in the past [8,13]. Although the exact number of molecular subtypes is still a matter of debate, there is a consensus that these may be differentiated on the basis of clinical data, histopathological analysis, codon-129 status of PRNP, and biochemical analysis of PrPSc.
The approach described here represents a new method for the discrimination of molecular subtypes of human prion diseases. By quantifying PrPSc in nine brain regions of patients with sCJD, we were able to determine the PrPSc profile of 50 patients with sCJD. Because individual measurements are calibrated to internal standards, this method allows, for the first time, for absolute quantification of PrPSc in prion-diseased individuals, and therefore enables the direct comparison of individual PrPSc profiles, excluding inter-observer and methodological variations. Detailed analysis of different molecular subtypes of sCJD confirmed the feasibility of this method and showed subtype-specific differences in the regional distribution of PrPSc. Besides advancing the molecular analysis of sCJD, these data demonstrate that if confirmation of sCJD by brain biopsy is indispensable, it should only be carried out taking into account the sCJD subtype. In patients belonging to the MM1 groups [8], high amounts of PrPSc are present in all cortical areas and would favor a cortical biopsy site. In patients belonging to the VV2 group [8], minimal amounts of PrPSc in the cerebral cortex would designate the cerebellum or the thalamus as sites for successful confirmation of the disease by biopsy.
An earlier study that attempted to assess the distribution of PrPSc within the central nervous system was performed only on a small patient cohort and did not employ internal PrPSc standards–thus impeding the direct comparison of analyses [16]. In other studies, lesion profiles were based on histological scores [8,13,17]. Although this histological, score-based method has certain advantages and can be carried out on fixed tissue, we believe that PrPSc profiling will be a valuable tool for prion research. Since both our work and other studies show that there is no strict correlation between histopathological alterations and the presence of PrPSc, one could assume that PrPSc profiling is superior since it measures PrPSc, which is believed to constitute an essential component of prion infectivity [4,16].
Even though experiments in rodents performed some years ago demonstrated the phenomenon of multiple PrPSc types within one animal [21], incongruity regarding the frequency of co-occurrence of different PrPSc types in sCJD patients still endures. Initial studies suggested that this is a relatively rare event, occurring in less than 5% of patients [8]. This low incidence may, at least partially, be attributed to the fact that these analyses are routinely performed on a limited range of distinct brain regions per patient [8,13]. The fact that region-specific presence of distinct PrPSc types may occur in sCJD was highlighted by a publication investigating ten defined regions within the central nervous system in 14 patients with sCJD [16]. The authors found more than one PrPSc type in five individuals and hypothesized that the co-occurrence of more than one PrPSc type could be the rule, rather than the exception, if the entire central nervous system was investigated. In our analysis of nine distinct central-nervous-system regions in 50 patients with sCJD using a standardized protocol [22], we detected more than one PrPSc type in nine individuals. The observation that most patients harboring two PrPSc types are codon-129 methionine/valine heterozygotes stresses the significance of this polymorphism in the replicative cycle of PrPSc.
Analysis of the spatial distribution of PrPSc types revealed that the putamen, thalamus, medulla oblongata, and cerebellum are prone to accommodate multiple PrPSc types simultaneously, whereas this was rarely observed in the cerebral cortex. The fact that central nervous system regions which simultaneously harbor two PrPSc types are complex structures containing diverse subsets of neuronal cells indicates that susceptibility of target cells may, at least partially, determine the type of PrPSc that is produced. This finding is compatible with the target-cell hypothesis of prion-strain diversity [23,24].
In conclusion, the present study has introduced a new method for the characterization of human prion diseases. The approach described here will aid in understanding the molecular diversity of this disease entity. Furthermore, these data imply that the subsets of cells present in specific brain areas in concert with the polymorphism on codon 129 of PRNP may have a fundamental role in determining which PrPSc type is preferentially replicated.
Supporting Information
Table S1 Histological Assessment of the VV2 Subgroup of Patients
Cases 1 to 4: spongiosis is scored on a scale from 0–4; astrogliosis and PrP immunoreactivity are scored on a scale from 0–3.
(41 KB DOC).
Click here for additional data file.
Patient Summary
Background
Prions cause neurodegenerative diseases in humans and in animals like cows, sheep, or deer. In most cases, humans get sick sporadically, when, for unknown reasons, normal proteins in the brain (called PrPc) change to harmful prions (called PrPSc). The PrPSc proteins then cause severe degeneration of the brain, which causes death within a matter of months. The most common human prion disease is sporadic Creutzfeld-Jacob Disease, or sCJD. The disease is rare (it affects roughly one in a million people worldwide), and there is no cure or therapy. Scientists still understand little about why and how, in some cases, normal proteins change to the harmful prions that destroy the brain. Although all patients with prion disease die, there are differences between individual cases. Neurological symptoms, particularly early in the disease, are variable. Scientists have also found that there are different versions of the gene that codes for the normal PrPc protein. Moreover, in some patients all of the prions found in the brain after autopsy are alike, whereas in other patients there seem to be several types of prions. In addition, brain autopsies have shown differences in the amount of prions and in the extent of damage present in different parts of the brain.
Why Was This Study Done?
The hope is that understanding the differences between individual cases might eventually provide some insight into the causes of the disease and suggest ways to treat it. To catalogue and examine those differences, however, it is important to come up with standard assays that allow meaningful comparisons. The authors of this study wanted to develop such assays.
What Did the Researchers Do and Find?
They developed an assay, which they called PrPSc profiling. The new assay measures the amount of prions in different parts of the brain in a way that makes it easy to compare autopsy samples from different patients analyzed at different times and by different people. They then used this assay to study autopsy samples from 50 patients for whom they knew the disease symptoms and the genetic make-up. For each patient sample, they examined nine different brain regions for the amount of prions present. They also checked whether all of the prions were alike or whether there were different types of prions. They found that there were distinct prion-distribution patterns in different sCJD subtypes. In about 20% of the patients, they found more than one prion type. Many of these patients had a specific genetic make-up, and the mixture of prion types was mostly seen in a few specific brain regions.
What Do These Findings Mean?
PrPSc profiling should prove to be useful for the classification of human prion diseases and should allow scientists worldwide to compare their samples. The specific results in the 50 patient samples will encourage other researchers to look for correlations between disease subtypes and prion-distribution patterns, and to further explore the link between genetic make-up, specific brain region, and the existence of more than one prion type. The hope is that the combined data will help scientists to understand the disease and to come up with ways to prevent or treat it.
Where Can I Get More Information Online?
The following websites provide information on prion diseases.
The National Creutzfeldt–Jakob Disease Surveillance Unit in Edinburgh, United Kingdom:
http://www.cjd.ed.ac.uk
The European and Allied Countries Collaborative Study Group of CJD:
http://www.eurocjd.ed.ac.uk
US National Institute of Mental Health pages on prion diseases:
http://www.ninds.nih.gov/disorders/tse/tse.htm
US Centers of Disease Control and Prevention pages on prion diseases:
http://www.cdc.gov/ncidod/dvrd/prions
Helpguide pages on Creutzfeld-Jacob disease:
http://www.helpguide.org/elder/creutzfeldt_jakob.htm
Wikipedia pages on prions:
http://en.wikipedia.org/wiki/prions
We thank the University Hospital of Zürich for a generous infrastructural grant enabling the Swiss CJD tissue collection. We thank the EU (TSELAB) for support. We also are grateful to Mauri Peltola and Marianne Koenig for excellent technical help and to Peter Roth for artwork. The National Reference Center for Prion Diseases (Zürich, Switzerland) is funded by the Swiss Federal Office of Public Health (Bern, Switzerland). MG is supported by a career-development award of the University of Zürich and by a grant from the Stammbach foundation (Basle). The funding agencies had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Citation: Schoch G, Seeger H, Bogousslavsky J, Tolnay M, Janzer RC, et al. (2006) Analysis of prion strains by PrPSc profiling in sporadic Creutzfeldt–Jakob disease. PLoS Med 3(2): e14.
Abbreviations
sCJDsporadic Creutzfeldt–Jakob disease
==== Refs
References
Prusiner SB Prions Proc Natl Acad Sci U S A 1998 95 13363 13383 9811807
Glatzel M Ott PM Lindner T Gebbers JO Gmur A Human prion diseases: Epidemiology and integrated risk assessment Lancet Neurol 2003 2 757 763 14636781
Pocchiari M Puopolo M Croes EA Budka H Gelpi E Predictors of survival in sporadic Creutzfeldt-Jakob disease and other human transmissible spongiform encephalopathies Brain 2004 127 2348 2359 15361416
Glatzel M Stoeck K Seeger H Luhrs T Aguzzi A Human prion diseases: Molecular and clinical aspects Arch Neurol 2005 62 545 552 15824251
Dickinson AG Meikle VM Host-genotype and agent effects in scrapie incubation: Change in allelic interaction with different strains of agent Mol Gen Genet 1971 112 73 79 5165642
DeArmond SJ Yang SL Lee A Bowler R Taraboulos A Three scrapie prion isolates exhibit different accumulation patterns of the prion protein scrapie isoform Proc Natl Acad Sci U S A 1993 90 6449 6453 8101989
Telling GC Parchi P DeArmond SJ Cortelli P Montagna P Evidence for the conformation of the pathologic isoform of the prion protein enciphering and propagating prion diversity Science 1996 274 2079 2082 8953038
Parchi P Castellani R Capellari S Ghetti B Young K Molecular basis of phenotypic variability in sporadic Creutzfeldt-Jakob disease Annal Neurol 1996 39 767 778 8651649
Bessen RA Marsh RF Distinct PrP properties suggest the molecular basis of strain variation in transmissible mink encephalopathy J Virol 1994 68 7859 7868 7966576
Tanaka M Chien P Naber N Cooke R Weissman JS Conformational variations in an infectious protein determine prion strain differences Nature 2004 428 323 328 15029196
Hill AF Desbruslais M Joiner S Sidle KC Gowland I The same prion strain causes vCJD and BSE Nature 1997 389 448 450 9333232
Parchi P Giese A Capellari S Brown P Schulz-Schaeffer W Classification of sporadic Creutzfeldt-Jakob disease based on molecular and phenotypic analysis of 300 subjects Ann Neurol 1999 46 224 233 10443888
Hill AF Joiner S Wadsworth JD Sidle KC Bell JE Molecular classification of sporadic Creutzfeldt-Jakob disease Brain 2003 126 1333 1346 12764055
Pedrioli PG Eng JK Hubley R Vogelzang M Deutsch EW A common open representation of mass spectrometry data and its application to proteomics research Nat Biotechnol 2004 22 1459 1466 15529173
Zanusso G Farinazzo A Prelli F Fiorini M Gelati M Identification of distinct N-terminal truncated forms of prion protein in different Creutzfeldt-Jakob disease subtypes J Biol Chem 2004 279 38936 38942 15247220
Puoti G Giaccone G Rossi G Canciani B Bugiani O Sporadic Creutzfeldt-Jakob disease: Co-occurrence of different types of PrP(Sc) in the same brain Neurology 1999 53 2173 2176 10599800
Bruce ME Will RG Ironside JW McConnell I Drummond D Transmissions to mice indicate that “new variant” CJD is caused by the BSE agent Nature 1997 389 498 501 9333239
Kascsak RJ Rubenstein R Merz PA Tonna DeMasi M Fersko R Mouse polyclonal and monoclonal antibody to scrapie-associated fibril proteins J Virol 1987 61 3688 3693 2446004
Wiltfang J Smirnov A Schnierstein B Kelemen G Matthies U Improved electrophoretic separation and immunoblotting of beta-amyloid (A beta) peptides 1–40, 1–42, and 1–43 Electrophoresis 1997 18 527 532 9150936
Glatzel M Rogivue C Ghani A Streffer JR Amsler L Incidence of Creutzfeldt-Jakob disease in Switzerland Lancet 2002 360 139 141 12126826
Safar J Wille H Itri V Groth D Serban H Eight prion strains have PrP(Sc) molecules with different conformations Nat Med 1998 4 1157 1165 9771749
Notari S Capellari S Giese A Westner I Baruzzi A Effects of different experimental conditions on the PrPSc core generated by protease digestion: Implications for strain typing and molecular classification of CJD J Biol Chem 2004 279 16797 16804 14754888
Hecker R Taraboulos A Scott M Pan KM Yang SL Replication of distinct scrapie prion isolates is region specific in brains of transgenic mice and hamsters Genes Dev 1992 6 1213 1228 1628828
Caughey B Prion protein conversions: Insight into mechanisms, TSE transmission barriers and strains Br Med Bull 2003 66 109 120 14522853
|
16354106
|
PMC1316067
|
CC BY
|
2021-01-05 11:14:05
|
no
|
PLoS Med. 2006 Feb 20; 3(2):e14
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0030014
|
oa_comm
|
==== Front
PLoS MedPLoS MedpbioplosbiolPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 10.1371/journal.pmed.0030051SynopsisCell BiologyImmunologyMolecular Biology/Structural BiologyAllergy/ImmunologyDiabetes/Endocrinology/MetabolismPathologyBiochemistryEndocrinologyDiabetesAutoimmune DiseasesImmunology and AllergyDeath by Necrosis: The Early Stages of Type 1 Diabetes Synopsis2 2006 20 12 2005 3 2 e51Copyright: © 2006 PLoS Medicine.2006This 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.
Interleukin 1 Stimulates β-Cell Necrosis and Release of the Immunological Adjuvant HMGB1
==== Body
Diabetes mellitus is a major global health problem. The most common form of diabetes—90%–95% of cases—is type 2 or non-insulin-dependent diabetes, which affects mainly adults. The remaining cases are type 1 or insulin-dependent diabetes. Also known as juvenile diabetes, type 1 diabetes usually develops before the age of 30. It is an autoimmune disease in which the insulin-producing or β-cells of the pancreatic islets of Langerhans (groups of specialized cells that regulate blood sugar levels) are destroyed by lymphocytes. Type 1 diabetes affects about one in 500 American children and adolescents; the only treatment is daily insulin injections.
β-cell death is the hallmark of type 1 diabetes. An early phase of β-cell death, probably triggered by environmental factors (for example, viral infection) in genetically susceptible individuals, releases β-cell-specific antigens; subsequently, T lymphocytes that specifically recognize these antigens mediate widespread β-cell killing. John Corbett and colleagues believe that by studying the early phase of β-cell death, it may be possible to find ways to prevent this destructive autoimmunity from developing in individuals with a family history of type 1 diabetes.
Cytokines, chemical messengers produced by lymphocytes and macrophages, are thought to contribute to the loss of β-cell function and viability early in autoimmune diabetes. The effect of cytokines on β-cell function is mediated by nitric oxide (NO), but it is not clear if the same is true for β-cell death. In their study, Corbett and colleagues asked whether NO mediates the death of rat β-cells induced in vitro by the macrophage-derived cytokine interleukin-1 (IL-1), and whether the cells are killed by apoptosis or necrosis, two different mechanisms of cell death. Apoptosis, or programmed cell death, is a highly organized process that minimizes the leakage of cell contents and the development of inflammation. Necrosis is much less tidy: the dying cells swell and burst, releasing their contents into the extracellular space where they cause inflammation.
The researchers report that 24–48 hours treatment with IL-1 reduced the viability of rat β-cells from two sources—an insulinoma cell line and islets. Then, by inhibiting NO synthesis or by adding an NO donor, they provide evidence that IL-1-induced death of β-cells is mediated in part by NO production. Turning to the mechanism of β-cell death, the researchers show that IL-1 treatment failed to activate caspase 3—an enzyme required for apoptosis—in β-cells, and that a caspase-3 inhibitor did not attenuate IL-1 induced β-cell death. Another marker of apoptotic cell death—lipid accumulation on the cell surface—was also missing in β-cells treated with IL-1.
IL-1 causes nuclear membrane breakdown in rat insulinoma cells
Having discounted death by apoptosis, the researchers then show that IL-1 stimulated the release of HMGB1 (a chromatin-binding protein that is released by cells undergoing necrosis but not apoptosis) by rat β-cells. Finally, because human β-cells behave somewhat differently from rat β-cells, the researchers demonstrate that a combination of cytokines (including IL-1) stimulated HMGB1 release from some preparations of human islets in an NO-dependent manner.
Overall, the authors conclude that macrophage-derived cytokines may participate in the early stages of type 1 diabetes by inducing necrotic death in β-cells. Other researchers believe that apoptotic cell death is more important in these early stages, particularly in human cells. But, based on their results, Corbett and colleagues speculate that cytokine induction of necrosis could kick start type 1 diabetes by causing the release of both β-cell antigens and HMGB1, which stimulates inflammatory responses. Although these findings need to be confirmed within the context of the human pancreas—what cells do in vitro may not reflect what happens in the body—they provide new insights into the early stages of type 1 diabetes that could suggest ways to prevent or stop its development.
|
0
|
PMC1316068
|
CC BY
|
2021-01-05 11:14:05
|
no
|
PLoS Med. 2006 Feb 20; 3(2):e51
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0030051
|
oa_comm
|
==== Front
PLoS MedPLoS MedpbioplosbiolPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 10.1371/journal.pmed.0030054SynopsisGenetics/Genomics/Gene TherapyPhysiologyCardiology/Cardiac SurgeryDiabetes/Endocrinology/MetabolismStatisticsBiochemistryCardiovascular MedicineEndocrinologyDiabetesGeneticsNutrition and MetabolismEndothelial Lipase: A New Risk Factor for Atherosclerosis? Synopsis2 2006 20 12 2005 3 2 e54Copyright: © 2006 PLoS Medicine.2006This 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.
Endothelial Lipase Concentrations Are Increased in Metabolic Syndrome and Associated with Coronary Atherosclerosis
==== Body
Atherosclerosis, or hardening of the arteries, is the most common cause of death in industrialized countries. The formation in the arteries of deposits or plaques containing cholesterol, calcium, and other materials directly causes heart attacks, strokes, and peripheral circulatory problems by blocking the blood flow to the heart, brain, or limbs, respectively. In addition, plaque rupture causes the formation of blood clots, which can block arterial blood flow anywhere in the body.
Major risk factors for atherosclerosis include high blood pressure, obesity, smoking, and levels of certain cholesterol-containing particles in the blood stream. Cholesterol is carried around the body by particles called lipoproteins, molecules that include both lipids (fats) and proteins. Low-density lipoprotein (LDL) particles transport cholesterol to the arteries and other tissues that take up lipids. High blood concentrations of LDL particles are associated with high risk of atherosclerosis; the cholesterol carried by LDL (LDL-C) is “bad” cholesterol. By contrast, high concentrations of high-density lipoprotein (HDL) particles protect against cardiovascular disease by removing cholesterol from the arteries and taking it to the liver for excretion. Consequently, cholesterol carried by HDL (HDL-C) is “good” cholesterol, and individuals with high plasma concentrations of HDL-C and low concentrations of LDL-C are at lowest risk for cardiovascular disease.
Enzymes known as lipases modulate lipoprotein metabolism, and the activity of two of them—lipoprotein lipase and hepatic lipase—is known to affect the risk of atherosclerosis. Karen Badellino and colleagues are investigating whether a third lipase—endothelial lipase (EL)—also affects atherosclerosis risk in people, since this enzyme digests HDL and affects the development of atherosclerosis in mouse models.
Endothelial lipase and atherosclerosis
Badellino and colleagues have used a new immunoassay to measure plasma EL concentrations in 858 unrelated people enrolled in the Study of Inherited Risk of Atherosclerosis, a cross-sectional study designed to investigate biomarkers and genetic factors associated with coronary atherosclerosis. Participants, who were just examined once rather than followed over time, as happens in a prospective, longitudinal study, had a family history of premature coronary artery disease (CAD) but were asymptomatic and had no other major risk factors for CAD. Because EL is normally bound tightly to heparin and is largely unavailable for immunoassay, the authors also measured EL concentrations in 510 individuals after heparin treatment to get an accurate measure of total EL mass. They then checked whether pre and postheparin EL levels were linked in any way to the concentrations of specific lipoproteins, components of the metabolic syndrome (a syndrome associated with cardiovascular disease that includes high blood pressure and obesity), or coronary arterial calcification, a noninvasive measure of early coronary atherosclerosis.
The researchers found that although the average postheparin concentration of EL was about three times the average preheparin concentration, pre and postheparin levels in individuals were strongly correlated. This result suggests that measurements of EL levels in untreated individuals will provide a good indication of the total vascular expression of EL for future studies. Then, they discovered that EL plasma levels both before and after heparin treatment correlated positively with components of the metabolic syndrome. Furthermore, as in mice, high levels of EL correlated with low levels of HDL-C, or “good” cholesterol. This last inverse association was small but statistically significant, as was the positive association between EL concentrations and signs of early atherosclerosis, even after taking into account other risk factors for atherosclerosis.
Overall, the authors suggest that EL is a proatherogenic factor in people, particularly those with metabolic syndrome. EL concentrations, they suggest, may modulate the composition of lipoproteins in the blood and thus atherosclerosis risk. But, they note, only prospective, longitudinal studies in which EL concentrations are correlated with subsequent heart attacks, strokes, and other cardiovascular problems can show whether EL concentration really is a risk factor for atherosclerosis, and thus whether interventions designed to reduce EL activity can prevent atherosclerosis.
|
0
|
PMC1316069
|
CC BY
|
2021-01-05 11:22:50
|
no
|
PLoS Med. 2006 Feb 20; 3(2):e54
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0030054
|
oa_comm
|
==== Front
PLoS MedPLoS MedpbioplosbiolPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 10.1371/journal.pmed.0030055SynopsisInfectious DiseasesNeuroscienceGeriatricsPathologyToward a Better Understanding of Human Prion Diseases Synopsis2 2006 20 12 2005 3 2 e55Copyright: © 2006 PLoS Medicine.2006This 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.
Analysis of Prion Strains by PrP
Sc Profiling in Sporadic Creutzfeldt-Jakob Disease
==== Body
Misfolding of a single protein, the cellular prion protein (PrP
c) into the disease-associated form PrP
Sc is believed to cause fatal prion diseases in humans and other mammals, including sheep, cattle, and deer. The misfolding can occur sporadically or after contact—through inoculation or ingestion—with PrP
Sc from an external source. There are also familial forms of prion diseases that are associated with certain mutations in
PNRP, the gene encoding PrP
c. These abnormal PrP
c proteins are thought to have a higher probability to misfold than normal PrP
c. The idea is that a few misfolded molecules can initiate a chain reaction and cause transformation of many of the other PrP
c molecules into harmful PrP
Sc versions. The presence of PrP
Sc proteins causes widespread cell death, leading to the characteristic spongiform degeneration of the brain that kills patients, most of them within a matter of months.
The most common human prion disease is sporadic Creutzfeld –Jacob Disease (sCJD). The disease is rare (affecting roughly one to two individuals per a million people, worldwide), and its etiology is unclear; neither exogenous nor endogenous causes have been identified. sCJD is inevitably fatal, but the disease is clinically, pathologically, and genetically heterogeneous. Most patients have rapidly progressing dementia, often accompanied by involuntary muscle spasms, and death occurs within months of the first clinical symptoms. However, for some patients ataxia is the first clinical sign, while others develop sight problems, and for some the disease duration can be more than two years. In the hope that understanding the heterogeneity will help them to understand what causes sCJD, researchers are trying to systematically collect and catalog data from patients. To do this in a meaningful way, standardized assays that allow results from different patients and different laboratories to be compared in a meaningful way are necessary.
PrP
Sc profiling was done in nine different central nervous system areas
Markus Glatzel and colleagues have developed such an assay, and applied it as part of the detailed molecular characterization to autopsy samples from 50 patients with sCJD. The new assay, which the researchers call PrP
Sc profiling, measures the amount of PrP
Sc in defined brain regions. In the past, PrP
Sc amounts were routinely only measured in one or two regions by a variety of assays. PrP
Sc profiling quantifies the amount of PrP
Sc in nine defined brain regions relative to internal standards, and thereby allows for direct comparison of individual profiles.
The researchers determined PrP
Sc profiles of 50 patients, and tried to correlate the profiles with information on disease types of the patients and prion types present in the different brain areas. sCJD types are determined by a patients'
PRNP genotype at the polymorphic position 129 of
PRNP and by the relative resistance of PrP
Sc to proteolytic degradation. It is thought that most patients only have one prion type, but previous reports have described coexistence of two different types in some samples.
Analysis of this wealth of data revealed correlations between distinct PrP
Sc distribution patterns and sCJD subtypes. These results have implications for confirmation of sCJD by brain biopsy. Before doing such biopsies, Glatzel and colleagues suggest, the sCJD subtype should be determined so that the correct brain area is examined. The researchers also found coexistence of two different prion types in 20% of their overall samples, and in more than 50% of the samples from patients who were heterozygous for the 129 polymorphism in the
PRNP gene. These data lend further support to a link between molecular signature and clinical heterogeneity of the disease.
While many questions remain, this study underlines that the systematic analysis of prion cases can reveal links between molecular pathology, genetic makeup of patients, and disease symptoms. Glatzel and colleagues believe that “PrP
Sc profiling will be a valuable tool for prion research.” In the hope that it will “facilitate comparisons of PrP
Sc quantities present in defined samples,” the researchers will make their PrP
Sc standard available to the scientific community.
|
0
|
PMC1316070
|
CC BY
|
2021-01-05 11:22:50
|
no
|
PLoS Med. 2006 Feb 20; 3(2):e55
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0030055
|
oa_comm
|
==== Front
PLoS MedPLoS MedpbioplosbiolPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 10.1371/journal.pmed.0030056SynopsisInfectious DiseasesEpidemiology/Public HealthStatisticsCohort studiesEpidemiologyPublic HealthMeasuring Mortality in Developing Countries Synopsis2 2006 20 12 2005 3 2 e56Copyright: © 2006 PLoS Medicine.2006This 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.
Prospective Study of One Million Deaths in India: Rationale, Design, and Validation Results
==== Body
Some of the biggest public health successes in the Western world have come about because of simple records of people's deaths—their age at death, where they lived, and what they died of. Soaring lung cancer rates in the United Kingdom and United States around World War II, for example, led to life-saving research into the effects of smoking.
More than two-thirds of deaths worldwide are in developing countries, yet little is known about the causes of death in these nations. In India, for example, just one-third of deaths are registered, and of these, only one-third provide data on the cause of death. India's HIV/AIDS epidemic is rising—it may already have surpassed South Africa for the highest number of people infected. And like many other developing nations, the numbers of people dying from noncommunicable diseases such as heart disease and cancer are growing. Unlike most infectious diseases, the causes of noncommunicable ones can be the result of several risk factors, such as smoking, elevated blood pressure, or inherited genetic mutations.
With a population of 1 billion and growing, India urgently needs better data on the causes of death in its people if it is to take further steps to improve public health. To address this need, Prabhat Jha and colleagues designed a prospective study of 1 million deaths in India to run until 2014. They will monitor an expected 1 million deaths in nearly 14 million people across 2.4 million households to find patterns of disease according to gender, age, and region, and to better understand how risk factors such as tobacco and alcohol use and indoor air pollution are related to disease.
The study uses one of India's existing frameworks for measuring mortality, called the Sample Registration System (SRS). At present, the system covers 50,000 deaths every year. Two independent workers visit the households; one visiting every month and the other every six months. Their reports are collated and any discrepancies reconciled by a third person. To improve the system, Jha and colleagues are using an innovative method called a “verbal autopsy” to record details of death as reported by family or friends to a trained but nonmedical fieldworker. To ensure the robustness of the method, a random 10% of the fieldwork will be repeated by an independent audit team.
After validating the verbal autopsy method, the researchers began the first phase of the study, which ran from 1998 to 2003, and recorded deaths in 6.3 million people across 1.1 million urban and rural households nationwide. As of November 2005, the researchers have collected 140,000 verbal autopsy reports, and 35,000 have been coded and reconciled by two independent and trained physicians. They expect to record a total of about 300,000 deaths in the first phase and 700,000 in the second phase in 2004–2014, which will look at 7.6 million people in 1.3 million households.
Better knowledge of genetic risk factors—about which little is known in developing or developed countries—requires collection of biological samples. Jha and colleagues are also planning to test the feasibility of this by collecting dried blood spots or tubes of blood in SRS units in four to five Indian states.
Studying mortality in 14 million people is a huge challenge, but one that is necessary in view of India's vast population. As the researchers point out, direct measurement of the causes of death is a great deal more reliable than indirect estimation. By studying diseases that are common in one part of India but not in another, new risk factors should be discovered, and these are likely relevant worldwide. Mortality measurements will be key to the success of one of the world's largest public health initiatives—the Millennium Development Goals, which were set in 2000 when countries worldwide pledged to reduce by half or more the incidence of many diseases in poor countries. We will only know whether these goals have been met if we have reliable mortality statistics.
|
0
|
PMC1316071
|
CC BY
|
2021-01-05 11:14:06
|
no
|
PLoS Med. 2006 Feb 20; 3(2):e56
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0030056
|
oa_comm
|
==== Front
BMC BiochemBMC Biochemistry1471-2091BioMed Central London 1471-2091-6-251630574710.1186/1471-2091-6-25Research ArticleCharacterization of the interactions between the active site of a protein tyrosine kinase and a divalent metal activator Lin Xiaofeng [email protected] Marina K [email protected] Gongqin [email protected] Department of Cell and Molecular Biology, University of Rhode Island, Kingston, RI 02881, USA2005 23 11 2005 6 25 25 31 8 2005 23 11 2005 Copyright © 2005 Lin and Sun; licensee BioMed Central Ltd.2005Lin and Sun; 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
Protein tyrosine kinases are important enzymes for cell signalling and key targets for anticancer drug discovery. The catalytic mechanisms of protein tyrosine kinase-catalysed phosphorylation are not fully understood. Protein tyrosine kinase Csk requires two Mg2+ cations for activity: one (M1) binds to ATP, and the other (M2) acts as an essential activator.
Results
Experiments in this communication characterize the interaction between M2 and Csk. Csk activity is sensitive to pH in the range of 6 to 7. Kinetic characterization indicates that the sensitivity is not due to altered substrate binding, but caused by the sensitivity of M2 binding to pH. Several residues in the active site with potential of binding M2 are mutated and the effect on metal activation studied. An active mutant of Asn319 is generated, and this mutation does not alter the metal binding characteristics. Mutations of Glu236 or Asp332 abolish the kinase activity, precluding a positive or negative conclusion on their role in M2 coordination. Finally, the ability of divalent metal cations to activate Csk correlates to a combination of ionic radius and the coordination number.
Conclusion
These studies demonstrate that M2 binding to Csk is sensitive to pH, which is mainly responsible for Csk activity change in the acidic arm of the pH response curve. They also demonstrate critical differences in the metal activator coordination sphere in protein tyrosine kinase Csk and a protein Ser/Thr kinase, the cAMP-dependent protein kinase. They shed light on the physical interactions between a protein tyrosine kinase and a divalent metal activator.
==== Body
Background
Protein tyrosine kinases (PTK)1 are a large family of enzymes that transfer the γ-phosphate of ATP to tyrosine hydroxyl groups in proteins. By phosphorylation, PTKs regulate the conformation and function of their protein substrates [1]. This covalent modification is a fundamental mechanism of signal transduction in mammalian cells. Aberrant activation of many specific protein tyrosine kinases causes mishaps in cell signalling, and results in proliferative diseases, such as cancer [2]. Many protein tyrosine kinases are considered as important targets for drug development against such diseases [3]. For full understanding of phosphorylation-mediated signalling and to provide a knowledge base for anti-PTK drug discovery, it is important to understand the catalytic mechanisms of protein tyrosine kinases.
C-terminal Src kinase (Csk) is a cytoplasmic PTK that phosphorylates Src family kinases (SFKs) and down-regulates their kinase activities [4,5]. The mechanistic basis of catalysis by Csk and PTKs in general is still poorly understood. Csk-catalyzed phosphorylation reaction obeys a ternary complex mechanism, likely with rapid and random binding of ATP-Mg and the phosphate-accepting substrate [6]. In addition to a Mg2+ cation (M1) as part of the ATP-Mg complex, Csk requires another Mg2+ ion (M2) for optimal kinase activity [7,8]. Kinetic studies demonstrate that M2 is an essential activator [7]. Because the affinity of Csk for the metal activator at 2.3 mM falls within the range of the cellular Mg2+ concentration, this activation may play a regulatory role in the kinase function [7,9].
Even though Mg2+ is likely the physiological activator, several other divalent metal cations can substitute for Mg2+ and activate Csk to various levels [8,10]. For example, Mn2+ can replace Mg2+ and results in higher activity of Csk, while Co2+, Ni2+ are not as effective as Mg2+ as an activator. Zn2+ can also substitute for Mg2+ in binding to the M2 binding site, but it cannot serve as an activator. Thus, Zn2+ acts as an inhibitor of Csk activity competitive against M2 [8]. Another intriguing property of the Csk-metal interaction is that these substitution metals all bind to Csk considerably stronger than the physiological activator, Mg2+. While Csk binds to Mg2+ with an AC50 of 2.3 mM, the other metal cations all bind to Csk with AC50 or IC50 in the low μM range. Among all divalent metal cations tested, Zn2+ has the highest affinity for Csk, with an IC50 of 0.5 μM [8].
The requirement of two divalent metal cations for full activity by Csk appears to represent a general catalytic requirement by all PTKs. Several PTKs from different families, such as v-Fps [11], Yes [12], Src [13], Lck [14], insulin receptor kinase [15] and epidermal growth factor receptor [16], all require two Mg2+ cations for full activity. The insulin receptor kinase has been co-crystallized with both a peptide substrate and an ATP analog [17]. In the active site, two Mg2+ are observed, providing direct structural evidence for the presence of two Mg2+ ions in PTK catalysis. Kinetic analysis reveals that the metal cation activator might participate in catalysis by different mechanisms for different PTKs. For example, M2 activates Csk and Src by increasing the kcat without affecting the Km for ATP [7]. However, M2 activates IRK [18] and v-Fps [11] by decreasing the Km for ATP without affecting the kcat. The mechanistic basis for such kinetic differences has not been determined.
Interestingly, a protein Ser/Thr kinase, the cAMP-dependent protein kinase (PKA), also binds to two divalent metal cations in the active site during catalysis [19]. However, the second Mg2+ inhibits the kinase activity [20]. Crystallization of PKA complexed with catalytic ligands reveals that two Mg2+ cations are present in the active site [21].
In the current study, we characterized the parameters for the interactions between Csk and M2, such as activity sensitivity to pH, required physical parameters of the divalent metal cation activators, and potential M2 coordinating residues. Mutagenic studies eliminated a residue as a potential ligand for M2, but could not determine if two other residues are involved due to lack of activity in all mutants varying these residues.
Results
Csk activity is sensitive to pH in the range of 6 to 7
Because many PTKs are molecular targets for drug discovery, it is of high interest to understand the mechanisms of PTK catalysis. Like most protein tyrosine kinases, the catalytic mechanism of Csk is not fully understood. Csk has a bell-shaped pH response curve, with a pH optimum of around 8 (Figure 1). In the acidic arm of the pH curve, the activity is highly sensitive to pH. At pH 6, the enzyme shows very little activity, but at pH 7, the enzyme is nearly fully active. We hypothesize that this sensitivity likely reflects certain catalytically essential step(s) that is carried out by a functional group with an apparent pKa in this range. Elucidation of such catalytically essential step(s) may shed light on the catalytic mechanism.
Figure 1 pH optimum of Csk. The Csk kinase activity is assayed as described in the Methods section. The activity at different pH is normalized to that at pH 8.
To test the above hypothesis, we characterized how steady state catalytic parameters responded to pH changes in this range (Figure 2). We first determined the catalytic parameters using ATP as the variable substrate. For this purpose, polyE4Y, a random polymer of Glu and Tyr (4:1), was used as the phosphate-accepting substrate at a fixed concentration of 1 mg ml-1. The apparent Km for ATP decreased from 200 μM at pH 6 to 91 μM at pH 7.2 (relative Km from 1 to 0.45 in Figure 2), but the kcat increased approximately 16 fold. The pKa for this kcat change is estimated to be 6.2. This indicates that phosphoryl transfer but not ATP binding to Csk is sensitive to pH. This pattern is different from those for PKA, whose apparent Km for ATP gradually decreases by a factor of 7 when pH increases from 5.5 to 7 [22]. This contrast in kinetic patterns likely reflects differences in participation by pH-sensitive functional groups in catalysis by these two enzymes. In PKA, the sensitivity to pH is due to increased binding of ATP-M1 at a higher pH. Asp184 is a coordination ligand for M1 [21]. For Csk, the apparent Km for ATP is also dependent on the presence of M1, with M1 resulting in lower Km for ATP [7]. However, because the apparent Km of Csk for ATP was largely independent of pH, the function of M1 is likely not affected by pH in this range. The contrast between Csk and PKA suggests that M1 coordination in Csk and PKA is different.
Figure 2 Effect of pH on Csk kinetics with ATP as the variable substrate. (A) Double reciprocal plot of Csk activity using ATP as the variable substrate at pH 6, 6.1 and 6.2. (B) Plots of relative kcat and Km for ATP as a function of pH. The phosphate-accepting substrate was polyE4Y at 1 mg ml-1. The kcat (9.1 min-1) and Km (200 μM) at pH 6 are taken as 1 and all other values are relative to those.
Steady state kinetics was then performed using the phosphate-accepting substrate as the variable substrate. The physiological substrates for Csk are the Src family kinases. Because SFKs are themselves PTKs that autophosphorylate, we used a kinase-defective mutant of Src (kdSrc) as the substrate. KdSrc contains a point mutation in the active site (Lys295Met), which abolishes Src kinase activity but does not affect its ability to serve as a Csk substrate [23,24]. The Km of Csk for kdSrc did not change significantly in response to pH in the range of 6 to 7. However, the kcat increased over 100-fold (Figure 3A). A similar kinetic pattern was observed for polyE4Y, a commonly used artificial substrate for PTK activity assays (Figure 3B) [7,8]. This result indicates that the recognition of the phosphate-accepting substrate is not affected by pH changes in this range. Because the kcat may be a function of both the phosphoryl transfer step and the ADP release [25], one of these steps is likely affected by pH in this range.
Figure 3 Effect of pH on Csk catalytic parameters using phosphate-accepting substrate as the variable substrate. (A) KdSrc is used as the variable substrate. (B) PolyE4Y is used as the variable substrate.
M2 binding to Csk is sensitive to pH
Because phosphoryl transfer, not substrate binding is sensitive to pH, it is likely that certain functional groups that catalyze phosphoryl transfer are subject to ionization in this pH range. Two types of functional groups could fall within this category. A functional group may directly participate in catalysis by interacting with the transition state, or it may interact with an essential activator. If only one ionization state of such a group is functional, it would render Csk phosphoryl transfer sensitive to pH. Because M2 acts as an essential activator, we determined if M2 binding to Csk was sensitive to pH in this range. The affinity of Csk for M2 can be roughly measured by the AC50, the concentration of Mg2+ that activates Csk to 50% of its full activity [7]. For example, at the optimal pH (8.0), Mg2+ activates Csk with an AC50 of 2.3 mM. We determined if the AC50 of Csk for Mg2+ activation was sensitive to pH. As shown in Figure 4A, the ability of Mg2+ to activate Csk was indeed sensitive to pH. At pH 6.9, the AC50 was 3 mM, close to the optimal AC50. As pH decreased, progressively higher concentration of Mg2+ was required for Csk activation. At pH 6.4, 64 mM MgCl2 nearly saturated Csk, and the AC50 was estimated to be 20 mM. At pH 6.3 or below, highest activity was detected at 64 mM, making it difficult to estimate an AC50, but it is clear AC50 continued to increase as pH decreased. It is known that Csk activity is highly sensitive to ionic inhibition [26], making it difficult to separate the effect of Mg2+ as an activator from ionic inhibition at high MgCl2 concentrations. Despite the lack of an accurate determination of the relationship between AC50 and pH, it is clear that the AC50 of Csk for Mg2+ is dependent on pH. The pH dependence of metal binding is further illustrated by the pH dependence profile in Figure 4B. The pKa of this function was estimated in the range of 6.2 to 6.5, which correlated to the pH range where Csk activity was most sensitive to pH. This result strongly suggests that the sensitivity of phosphoryl transfer to pH is at least partly due to sensitivity of M2 binding to pH. Because the activity is gained with increased pH, the deprotonated form of the functional group is responsible for M2 binding.
Figure 4 Mg2+ activation of Csk as a function of pH. (A) Mg2+ activation of Csk is determined at different pH. The pH values are labeled in the graph. (B) Plot of AC50 as a function of pH.
Characterization of potential metal-coordinating residues in the active site of Csk
Only one protein tyrosine kinase, the insulin receptor kinase, has been co-crystallized with substrate analogs and divalent metal activators [17]. We compared the structures of Csk and IRK to identify Csk residues potentially involved in M2 coordination. In IRK, three residues are involved in metal cation coordination, Asp1150, Glu1047, and Asn1137. All three residues are conserved among PTKs, corresponding to Asp332, Glu236 and Asn319 in Csk [27]. Even though Csk and IRK displayed some differences in the kinetic patterns of Mg2+ activation, it is likely that the conserved residues are playing similar roles in Mg2+ coordination. We performed site-specific mutagenesis on these residues to determine if they are involved in metal activator coordination in Csk.
Asn319 is located in the catalytic loop and a universally conserved residue in all protein kinases, including Ser/Thr kinases. In IRK, the equivalent residue is involved in coordinating M1. Asn319 was mutated to Asp, His, Gln and Ser. Only one mutant, Asn319Ser, could be expressed as an active enzyme. Representative double reciprocal plots determining catalytic parameters of Asn319Ser mutant are presented in Figure 5 and summarized in Table 1. Overall, the catalytic efficiency measured by kcat decreased by a factor of approximately 10,000, while the apparent Km values for ATP, polyE4Y and kdSrc did not change significantly (within a factor of 2). The Mg2+ activation profile of this mutant was nearly identical to that of wt Csk (Figure 6) indicating that even though this residue is crucial for Csk catalysis, it is unlikely to be responsible for the M2 coordination. Because of the large decrease in kcat due to the Asn319 mutation, there is a possibility that the mutation may have changed the rate-limiting step in Csk catalysis, making a direct conclusion about the role of Asn319 in M2 binding more complicated. The identical Mg2+ responses by Asn319Ser and wild type Csk argue against this possibility.
Figure 5 Kinetic analysis of Asn319Ser. Double reciprocal plots of Asn319Ser, using ATP (A), kdSrc (B) or polyE4Y (C) as the variable substrate. For the determination in (A), polyE4Y is used as the phosphate-accepting substrate.
Table 1 Catalytic parameters of Csk and Asn319Ser mutant
Parametera Wt Csk Asn319Ser
kcat-ATP (min-1) 160 ± 10 0.01 ± 0.007
Km-ATP (μM) 140 ± 12 150 ± 25
kcat-polyE4Y (min-1) 82 ± 12 0.01 ± 0.001
Km-polyE4Y (μg ml-1) 156 ± 30 220 ± 48
kcat-kdSrc (min-1) 109 ± 3 0.01 ± 0.001
Km-kdSrc (μM) 6.4 ± 0.1 3.4 ± 0.1
aAll assays were performed at least three times. The standard errors were calculated from three most consistent assays.
Figure 6 Activation of wt and Asn319Ser mutant of Csk by Mg2+. The maximum activities for both wt and Asn319Ser are set as 100%.
Mutation of Glu236 and Asp332 to a number of residues, Ala, Asp, Gln for Glu236, Ala, Asn, Glu for Asp332, produced inactive mutants, thus kinetic analysis of their role in M2 binding is precluded. These two residues remain likely candidates for coordinating M2, but confirmation awaits further study by other methods.
Divalent Metal cations of certain size bind to and activate Csk
To characterize the physical properties required of the metal activator, it has been previously determined that Mn2+, Co2+ and Ni2+ could also serve as activators, while 14 other commercially available divalent metal cations could not [8,10]. To determine if the ability to activate Csk correlates to any specific physical attributes, we plotted the coordination number versus ionic radius [28] and determined where the activating metal cations were located on this map (Figure 7). The four divalent metal cations, Mg2+, Mn2+, Co2+ and Ni2+, that support Csk activity clustered together with a coordination number of 6 and an ionic radius of 0.65 to 0.8 Angstroms. Two other metal cations also met these criteria but apparently did not support the kinase activity: Cr2+ and Fe2+. Both of these metal ions have multiple valencies, and interfere with the kinase assay. Fe2+ forms brown precipitates likely with the enzyme and protein substrate, while Cr2+ forms precipitates with ATP in the kinase assay. Such interferences prevent a definitive analysis if they could support Csk activity. Zn2+ binds to Csk tightly but does not support Csk activity, thus it inhibits Csk activity as an inhibitor competitive against M2. Zn2+ also falls within the range of the ionic radius, but has a coordination number of 4 or 5. This suggests that a coordination number of 6 may be required of the metal cation activator. Although this analysis suggests that the size and coordination number are likely important factors in determining if a metal cation could activate Csk catalysis, other factors may also be important.
Figure 7 Plot of coordination number versus cation radius for commercially available divalent metal cations. Those cations that support Csk activity are shown in blue. Zn2+, which also binds to Csk but does not support Csk activity, is shown in red. Fe2+ and Cr2+ are also labelled. Their ability to support Csk activity is discussed in the text. The coordination numbers and the cation radii are taken from reference [28] and the inhibition data is taken from reference [8].
Discussion
In this communication, we investigated the molecular basis of a commonly observed catalytic property of Csk. First, Csk activity is sensitive to pH change in the range of 6 to 7. Steady state kinetics demonstrates that the sensitivity is not due to the binding of Csk to either ATP-Mg or the protein substrate. The sensitivity is due to the sensitivity of M2 binding to pH in this range. Second, several residues that have the potential for M2 binding were studied by mutagenesis. These studies eliminated Asn319 in the active site as a potential ligand for M2 binding, but were inconclusive about the role of Asp332 and Glu236, because mutants at these two positions were inactive. Third, commercially available divalent metal cations were surveyed for their ability to support Csk activity. A strong correlation between the ability of divalent metal cation to support Csk activity and its physical parameters (ionic radius and the coordination number) was identified. Divalent metal cations with a coordination number of 6 and an ionic radius of 0.65–0.8 Å were able to support the activity while ions outside of this range were not. Overall this investigation provided insights into the kinase-divalent metal interaction in the active site.
The sensitivity of kinase activity to pH has been previously investigated in the cAMP-dependent protein kinase [22]. Interestingly, the binding of ATP was sensitive to pH for PKA while the binding of the second Mg2+ to PKA is not sensitive to pH. This pattern is the opposite of that observed for Csk, likely reflecting different coordination patterns for M1 and M2 in Csk and PKA. This is also consistent with the structural information on IRK and PKA. Although in both PKA and IRK, three conserved residues (Glu1047, Asn1137, and Asp1150 in IRK, Glu91, Asn171 and Asp184 in PKA) are involved in coordinating M1 and M2, the positions of M1 and M2 are switched in the two kinases. In IRK, M1 is coordinated with Asn1127, while M2 is coordinated with Asp1150 directly and Glu1047 through two water molecules. In PKA, M2 is coordinated with Asn171, while M1 is coordinated with Asp184. Because M1 binding to PKA is sensitive to pH, it is likely due to deprotonation of Asp184. In this case, M2 binding to IRK would likely be sensitive to pH. This pattern is observed in Csk. This suggests that Csk and IRK likely uses a similar M2 binding site. In this case, Glu236 and Asp332 would be expected to be key ligands for M2 coordination.
Our effort to pinpoint the residues for coordinating M2 in Csk by mutagenic and kinetic studies is not fully successful. We were able to eliminate Asn319 as responsible for binding to M2, but our results are inconclusive regarding Asp332 and Glu236 due to the inability to generate active mutants at these two positions. This highlights the limitation of mutagenic approach to study catalytically essential residues. Further studies by other tools are required to solve these issues.
Methods
Generation of Csk mutants
Glutathione S transferase (GST)-Csk fusion proteins were generated and purified as previously described [29]. Csk point mutants were generated using QuikChange (Stratagene) in the parental plasmid and were confirmed by DNA sequencing. Kinase-defective Src (kdSrc) was produced as described previously [23,24].
Enzyme purification
Bacteria harbouring appropriate plasmids were cultured in LB medium at 37°C with shaking at 250 rpm overnight. The overnight culture was then mixed with an equal volume of fresh LB medium, cooled down to about 20°C. IPTG (0.2 mM) was added to the culture to induce recombinant protein expression at 20°C for 12 hours. The GST fusion proteins were purified by glutathione affinity chromatography as previously described [29]. The purified enzymes were desalted on a Sephadex G25 column equilibrated with the storage buffer (100 mM Tris-Cl, pH 8.0, and 0.1% β-mercaptoethanol). Glycerol was added to the purified fractions to 30% and the enzymes were stored at -20°C. Protein concentration was determined by the Bradford assay and the purity of purified proteins was assessed by SDS-PAGE with coomassie blue staining.
Kinase activity assay
For assaying PTK activity, phosphorylation of polyE4Y and kdSrc was measured using the acid precipitation assay as previously described [7]. Standard kinase assay buffer contains 100 mM EPPS, pH 8, 10% glycerol, 0.1% triton X-100 and 0.1% β-mercaptoethanol. Reaction time for the assays was 10 min. Standard assays used polyE4Y at 1 mg ml-1, or kdSrc at 10 μM as the phosphate-accepting substrate and ATP at 0.2 mM as the phosphate-donating substrate. To determine the kinase activity at different pH, the kinase buffer contained all the standard buffer components except EPPS was replaced by 100 mM MES or Tris at designated pH. When Km and kcat were determined with regard to one substrate, the kinase activity was determined at various concentrations of that substrate in the range of 20 to 200 μg ml-1 for polyE4Y, 1 to 10 μM for kdSrc or 20 to 200 μM for ATP. When the phosphate-accepting substrate (either polyE4Y or kdSrc) was the variable substrate, ATP concentration was 0.2 mM. PolyE4Y at 1 mg ml-1 was used when ATP was the variable substrate. The kcat and Km values were determined by Lineweaver-Burk plots with linear regression using Microsoft Excel. All steady state kinetic assays were performed in duplicate, and repeated at least once. Standard errors were calculated if an assay was performed at least three times.
Abbreviation used
AC50, the concentration of a divalent metal cation that activates Csk to 50% of its full activity; Csk, C-terminal Src kinase; IRK, insulin receptor kinase; PKA, the camp-dependent protein kinase; PTK, protein tyrosine kinase(s); SFK, Src family kinase(s).
Authors' contributions
XL designed and performed the experiments and analyzed the results. MKA designed and performed experiments. GS designed the experiments, analyzed the results and wrote the paper.
Acknowledgements
This work was supported by grants from NIH (1RO1CA111687 and 1 P20 RR16457), and the American Cancer Society (RSG-04-247-01-CDD). DNA sequencing was performed at URI Genomics and Sequencing Center.
#G. Sun is an American Cancer Society Research Scholar.
==== Refs
Hubbard SR Till JH Protein tyrosine kinase structure and function Annu Rev Biochem 2000 69 373 398 10966463 10.1146/annurev.biochem.69.1.373
Hunter T The Croonian Lecture 1997. The phosphorylation of proteins on tyrosine: its role in cell growth and disease Philos Trans R Soc Lond B Biol Sci 1998 353 583 605 9602534 10.1098/rstb.1998.0228
Krause DS Van Etten RA Tyrosine kinases as targets for cancer therapy N Engl J Med 2005 353 172 187 16014887 10.1056/NEJMra044389
Okada M Nada S Yamanashi Y Yamamoto T Nakagawa H CSK: a protein-tyrosine kinase involved in regulation of src family kinases J Biol Chem 1991 266 24249 24252 1722201
Sun G Sharma AK Budde RJ Autophosphorylation of Src and Yes blocks their inactivation by Csk phosphorylation Oncogene 1998 17 1587 1595 9794236 10.1038/sj.onc.1202076
Cole PA Burn P Takacs B Walsh CT Evaluation of the catalytic mechanism of recombinant human Csk (C-terminal Src kinase) using nucleotide analogs and viscosity effects J Biol Chem 1994 269 30880 30887 7527038
Sun G Budde RJ Requirement for an additional divalent metal cation to activate protein tyrosine kinases Biochemistry 1997 36 2139 2146 9047313 10.1021/bi962291n
Sun G Budde RJ Substitution studies of the second divalent metal cation requirement of protein tyrosine kinase CSK Biochemistry 1999 38 5659 5665 10220355 10.1021/bi982793w
Maguire ME Magnesium: a regulated and regulatory cation Metal Ions Biol Syst 1990 26 135 153
Grace MR Walsh CT Cole PA Divalent ion effects and insights into the catalytic mechanism of protein tyrosine kinase Csk Biochemistry 1997 36 1874 1881 9048573 10.1021/bi962138t
Saylor P Wang C Hirai TJ Adams JA A second magnesium ion is critical for ATP binding in the kinase domain of the oncoprotein v-Fps Biochemistry 1998 37 12624 12630 9730835 10.1021/bi9812672
Sun G Budde JA Expression, purification, and initial characterization of human Yes protein tyrosine kinase from a bacterial expression system Arch Biochem Biophys 1997 345 135 142 9281320 10.1006/abbi.1997.0236
Budde JA Ramdas L Ke S Recombinant pp60c-src from baculovirus-infected insect cells: purification and characterization Prep Biochem 1993 23 493 515 7504254
Wang QM Srinivas PR Harrison ML Geahlen RL Partial purification and characterization of the lck protein-tyrosine kinase from bovine thymus Biochem J 1991 279 567 574 1953650
White MF Haring HU Kasuga M Kahn CR Kinetic properties and sites of autophosphorylation of the partially purified insulin receptor from hepatoma cells J Biol Chem 1984 259 255 264 6368536
Koland JG Cerione RA Activation of the EGF receptor tyrosine kinase by divalent metal ions: comparison of holoreceptor and isolated kinase domain properties Biochim Biophys Acta 1990 1052 489 498 2354210 10.1016/0167-4889(90)90160-F
Hubbard SR Crystal structure of the activated insulin receptor tyrosine kinase in complex with peptide substrate and ATP analog EMBO J 1997 16 5572 5581 9312016 10.1093/emboj/16.18.5572
Vicario PP Saperstein R Bennun A Role of divalent metals in the kinetic mechanism of insulin receptor tyrosine kinase Arch Biochem Biophys 1988 261 336 345 2833165 10.1016/0003-9861(88)90349-9
Mildvan AS Rosevear PR Fry DC Bramson HN Kaiser ET NMR studies of the mechanism of action and regulation of protein kinase Curr Top Cell Regul 1985 27 133 144 3004816
Armstrong RN Kondo H Granot J Kaiser ET Mildvan AS Magnetic resonance and kinetic studies of the manganese(II) ion and substrate complexes of the catalytic subunit of adenosine 3',5'-monophosphate dependent protein kinase from bovine heart Biochemistry 1979 18 1230 1238 218617 10.1021/bi00574a018
Zheng J Knighton DR ten Eyck LF Karlsson R Xuong N Taylor SS Sowadski JM Crystal structure of the catalytic subunit of cAMP-dependent protein kinase complexed with MgATP and peptide inhibitor Biochemistry 1993 32 2154 2161 8443157 10.1021/bi00060a005
Yoon MY Cook PF Chemical mechanism of the adenosine cyclic 3',5'-monophosphate dependent protein kinase from pH studies Biochemistry 1987 26 4118 4125 2820483 10.1021/bi00387a056
Wang D Huang XY Cole PA Molecular determinants for Csk-catalyzed tyrosine phosphorylation of the Src tail Biochemistry 2001 40 2004 2010 11329267 10.1021/bi002342n
Lee S Lin X Nam NH Parang K Sun G Determination of the substrate-docking site of protein tyrosine kinase C-terminal Src kinase Proc Natl Acad Sci USA 2003 100 14707 14712 14657361 10.1073/pnas.2534493100
Shaffer J Sun G Adams JA Nucleotide release and associated conformational changes regulate function in the COOH-terminal Src kinase, Csk Biochemistry 2001 40 11149 11155 11551213 10.1021/bi011029y
Sondhi D Xu W Songyang Z Eck MJ Cole PA Peptide and protein phosphorylation by protein tyrosine kinase Csk: insights into specificity and mechanism Biochemistry 1998 37 165 172 9425036 10.1021/bi9722960
Ogawa A Takayama Y Sakai H Chong KT Takeuchi S Nakagawa A Nada S Okada M Tsukihara T Structure of the carboxyl-terminal Src kinase, Csk J Biol Chem 2002 277 14351 14354 11884384 10.1074/jbc.C200086200
Glusker JP Structural aspects of metal liganding to functional groups in proteins Adv Protein Chem 1991 42 1 75 1793004
Sun G Budde RJ A modified pGEX expression system that eliminates degradation products and thrombin from the recombinant protein Anal Biochem 1995 231 458 460 8595003 10.1006/abio.1995.0081
|
16305747
|
PMC1316873
|
CC BY
|
2021-01-04 16:26:24
|
no
|
BMC Biochem. 2005 Nov 23; 6:25
|
utf-8
|
BMC Biochem
| 2,005 |
10.1186/1471-2091-6-25
|
oa_comm
|
==== Front
BMC GastroenterolBMC Gastroenterology1471-230XBioMed Central London 1471-230X-5-381630955110.1186/1471-230X-5-38Research ArticleRole of rapid urease test and histopathology in the diagnosis of Helicobacter pylori infection in a developing country Yakoob Javed [email protected] Wasim [email protected] Shahab [email protected] Nadim [email protected] Zaigham [email protected] Saeed [email protected] Muhammad [email protected] Kashif [email protected] Hasnain Ali [email protected] Hizbullah [email protected] Department of Gastroenterology and Pathology, Aga Khan University Hospital, Karachi, Pakistan2005 25 11 2005 5 38 38 20 5 2005 25 11 2005 Copyright © 2005 Yakoob 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 aim of this study was to determine the effect of commonly self-prescribed proton pump inhibitors (PPI) on the results of rapid urease test and histology for the diagnosis of H. pylori infection.
Methods
One hundred-nine consecutive patients with dyspeptic symptoms attending the endoscopy suite were enrolled in this study. Antrum biopsy specimens were collected at endoscopy for the rapid urease test (Pronto Dry, Medical Instrument Corp, France) and histopathology. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and like-hood ratio of a positive and negative of Pronto Dry test were compared against histology. The gold standard test for the diagnosis of H. pylori infection was histopathology.
Results
Sixty-one percent (66/109) patients were males with mean age of 43 ± 14.1 years and age range 17–80 years. Fifty-two percent (57/109) were not on any medications while 48% (52/109) used PPI before presentation to the outpatients. Pronto Dry was positive in 40% (44/109) and negative in 60% (65/109). Histopathology was positive for H. pylori in 57% (62/109) and negative in 43% (47/109). The sensitivity, specificity, PPV, NPV and like-hood ratio of a positive and negative Pronto Dry test with and without PPI were 43.3%, 86.4%, 81.3%, 3.18, 0.656 and 52.8% vs 71.9%, 80%, 82.1%, 69%, 3.59 and 0.35.
Conclusion
This study shows that the sensitivity, specificity, NPV and PPV of rapid urease test was reduced in patients who are on PPI. The exclusive use of the rapid urease test for the diagnosis of Helicobacter pylori cannot be recommended in patients with prior PPI use.
==== Body
Background
Helicobacter pylori (H. pylori) infection occurs worldwide. It results in chronic gastritis, ulcer, mucosal associated lymphomas and gastric carcinomas [1,2]. The diagnostic methods available for detecting H. pylori infection include serology (IgG ELISA), rapid urease test, histopathology, 13 C-urea breath test (UBT) and polymerase chain reaction (PCR) [3-5]. Rapid urease test is highly specific for H. pylori infection and is commonly used for the detection of H. pylori infection at endoscopy. It requires a high density of bacteria, and anything that reduces the bacterial load may produce false-negative tests. The diagnostic yield of rapid urease test is enhanced by increasing the number of biopsies taken and the number of sites in the stomach that are biopsied [6]. The sensitivity of urease test is reduced in patients who are taking proton pump inhibitors (PPI), antibiotics or bismuth compounds [7,8]. Any antibiotic active against H. pylori will cause a reduction in the numbers of bacteria in the stomach [9]. Also, if the patient has received a drug that reduces the acid in the stomach and raises the pH, this will affect the area of the stomach to be biopsied [10]. H2-receptor antagonists (ranitidine and cimetidine) raise the gastric pH, but PPI such as omeprazole and lansoprazole, raise the gastric pH to a higher level. Proton pump inhibitors are known to decrease the activity of H. pylori within the stomach and to shift their distribution proximally [8]. H2-receptor antagonists differ from proton pump inhibitors as high intragastric pH may cause a reduction in urease activity, unrelated to a reduced bacterial load [11]. This effect may reduce the sensitivity of histological examination and rapid urease testing for H. pylori on biopsies taken from recommended sites [8]. In Pakistan a third world country self-prescription is common and medications are available on counter of pharmacies for sale without prescriptions [12,13]. Data from 66 pharmacies evaluated 1231 over-the-counter (OTC) encounters, of which 43% were instances of self-medication, while the rest were given on the advice of pharmacy staff [12]. Self-medication increased with the level of socioeconomic status [13]. Proton pump inhibitors are much cheaper than anti-H2 receptor blockers (H2-RB) costing as much as 10 cents/pill. In a local tertiary care hospital the prescriptions for PPI in 2003 alone numbered 31086 and 399189 tablets/injections were dispensed on prescriptions. The aim of this study was to determine the effect of commonly self-prescribed proton pump inhibitors (PPI) on the results of the rapid urease test (Pronto Dry) and histology.
Methods
One hundred-nine consecutive patients with dyspeptic symptoms attending the endoscopy suite of gastroenterology section of Aga Khan University Hospital, Karachi, Pakistan from April 2004–January 2005 were enrolled. There were sixty-six males and forty-three females (age range 17–80 years, mean age 40.89 ± 12 years; Table 1). Clinical symptoms at the time of presentation, diagnosis, drug treatment dosage and duration were noted with endoscopic findings. An informed consent was taken from all patients and study was approved by the ethics review committee. Four antral biopsy specimens were collected at endoscopy from each patient two each for the Pronto Dry (a commercially manufactured rapid urease test by Medical Instrument Corp, France) and histopathology. Pronto Dry consists of a dry filter paper containing urea, phenol red (a pH indicator), buffers and a bacteriostatic agent in a sealed plastic slide. If the urease enzyme of H. pylori is present in an inserted tissue sample, the resulting decomposition of urea causes the pH to rise and the color of the dot turns from yellow to a bright magenta. Pronto Dry results were read in 30 minutes and one hour after sampling as directed by the manufacturer. The color change from yellow to pink was considered positive result and no color change as negative for Pronto Dry. In this study sensitivity, specificity, positive predictive value, negative predictive value and accuracy of Pronto Dry were compared against histology in the presence and absence of PPI. Histopathologist was kept blinded about the results of Pronto Dry. Two gastric antral biopsy specimens for histopathology were stained with Hematoxylin and eosin stain for the detection of H. pylori and degree of gastritis. In doubtful cases Giemsa staining was carried out, to ascertain presence of H. pylori. The degree of gastritis as determined on Hematoxylin and eosin (H & E) stain was scored in accordance with the Sydney system, representing absence of gastritis and minimal, mild, moderate chronic active and severe chronic active gastritis, respectively [14].
Statistical Analysis
Data was entered and analyzed in Statistical Package for Social Sciences (SPSS) ver13.0. Results are presented as mean ± standard deviation for quantitative variables and number (percentages) for qualitative variables. Sensitivity, specificity, positive and negative predictive values of urease test was calculated by two by two standard method. Likelihood ratio of a positive test was equal to sensitivity/(1-specificity) and of a negative test was (1 - sensitivity)/specificity.
Results
There were 57% (66/109) males and 43% (43/109) females with age range 17–80 years and mean age of 43 ± 14.1 years (Table 1). 52% (57/109) were not on any medications while 48% (52/109) used PPI before presentation to the outpatients.
Clinical Features
Abdominal pain was present in 75% (82/109), dyspepsia 8% (9/109), vomiting 6% (7/109), heartburn in 6% (7/109), and weakness 5% (4/109) (Table 1). The main endoscopic findings were gastritis 65% (71/109), gastroesophageal reflux disease 16% (17/109) and duodenitis in 10% (11/109) (Table 1).
Urease test
Pronto Dry was positive in 40% (44/109) and negative in 60% (65/109) (Table 1). The sensitivity, specificity, NPV and PPV of Pronto Dry with and without PPI was 43.3%, 86.4%, 81.3% and 52.8% vs. 71.9%, 80%, 82.1% and 69% (Table 2).
Histopathology
Histopathology was positive for H. pylori in 57% (62/109) and negative in 43% (47/109) (Table 1).
Comparison of Urease test and Histopathology
The sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of urease test was reduced on PPI (Table 2). The likelihood of a positive urease test with and without PPI was 3.18 and 3.59 and negative urease test with and without PPI was 0.65 and 0.35.
Discussion
The rapid urease test is the most frequently used test for the diagnosis of H. pylori infection in routine gastrointestinal endoscopy practice. It is extremely valuable because it gives a positive result for H. pylori infection before the patient leaves the endoscopic suite. Histological diagnosis of H. pylori infection is usually reserved for patients with a negative biopsy urease test or when histology was required for another reason such as exclusion of malignancy. In an earlier study rapid urease test (Pronto Dry) had the sensitivity, specificity, positive predictive value, negative predictive value and diagnostic accuracy were 98%, 100%, 100%, 98% and 99%, respectively [15]. The sensitivity and specificity of Pronto Dry against culture were 98% and 97% [16].
In this study treatment with a PPI before endoscopy reduced the sensitivity of urease test from antral biopsies for H. pylori detection. Ideally PPI should be discontinued before the endoscopy [8]. However, in our practice patients quite frequently self medicate. Even referrals from primary care service cannot be discontinuing PPIs for an adequate period before endoscopy. In patients on PPI the biopsy specimen may contain low bacterial density of viable cells giving a negative urease test. This also leads to lack of H. pylori identification on histology. Of the various tests that are available for H. pylori detection, histological examination of gastric biopsy is considered the most accurate method of diagnosis [6]. In a previous study even histological examination sensitivity, specificity, positive predictive value, negative predictive value and diagnostic accuracy were demonstrated to be reduced on acid reducing drugs [8]. It was demonstrated that after 4 weeks of omeprazole treatment, the histological density of H. pylori in the antrum and corpus was reduced, while that in the fundus was increased [17]. The migration of H. pylori from the antrum to the fundus was also associated with a corresponding decrease in the activity of antral gastritis and matched by a progressive fall in the excretion of 13C urea breath test [17]. In an animal model of H. pylori infection antrum-body transitional zone was identified as a sanctuary site in treatment failure [18]. If more than one gastric biopsy tissue is used to inoculate the rapid urease test a positive test might appear thus improving the test sensitivity without compromising its specificity. The diagnostic yield is said to be increased by over 5% by taking more than a single biopsy [6]. However, this also prolongs the endoscopy time. Our study showed the likelihood of a negative urease test with and without PPI was 0.65 and 0.35, respectively. Hence, in patients on PPI additional biopsies should be taken from the body of the stomach beside the antrum for the detection of H. pylori. This will be consistent with previous studies which recommend obtaining biopsies both from antrum and body of the stomach in patients on PPIs for the diagnosis of H. pylori infection [6,8]. This is also important in view of the high prevalence of H. pylori in the region. As rapid urease test can miss a low-level infection with H. pylori, a negative test should not be the sole criterion for either absence or cure of H. pylori infection. A negative diagnosis on PPI might be backed up with a serological test which should not be affected by PPI. Also, in view of the small sample size of our study, the result needs to be confirmed in a larger population of patients. In conclusion it is of particular relevance to know if a PPI has been used before the patient undergoes diagnostic endoscopy. If PPI can not be discontinued for an adequate period before the endoscopy multiple biopsies should be taken from both the antrum and the body of the stomach for H. pylori detection.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
JY conceived, designed and coordinated the study, WJ, SA, NJ, ZA, SH, HAS, KA, HS participated in its design and coordination MI participated in the design of the study and performed the statistical analysis. 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
Table 1 Clinical, endoscopic and histopathology findings of the patients (n = 109)
Factors With PPI n (%) Without PPI n (%)
Gender
Male 35 (53) 31 (47)
Female 17 (40) 26 (60)
Age (in years) mean ± SD 42 ± 12.2 44 ± 15.6
Clinical feature
Abdominal pain 44 (54) 38 (46)
Heart burn 4 (57) 3 (43)
Vomiting 2 (29) 5 (71)
Dyspepsia 1 (11) 8 (89)
Weakness 1 (25) 3 (75)
Endoscopic findings
Gastritis 36(51) 35(49)
Gastritis and GERD 2(100) -
GERD 6(35) 11(65)
Gastric Ulcer - 3(100)
Duodenitis 7(64) 4(36)
Duodenal Ulcer 1(20) 4(80)
Urease Test
Positive 16 (36) 28 (64)
Negative 36 (55) 29 (45)
Histopathology
Positive 30 (48) 32 (52)
Negative 22 (47) 25 (53)
Results are presented as mean ± standard deviation and number (percentages).
SD = Standard Deviation
Table 2 The comparison of Urease test and histopathology with the presence and absence of PPI.
Test Sensitivity (95% C.I.) Specificity (95% C.I.) PPV (95% C.I.) NPV (95% C.I.)
Urease Test with PPI 43.3% (30.9–50.6%) 86.4% (69.5–96.2%) 81.3% (58–94.8%) 52.8% (42.4–58.3%)
Urease Test without PPI 71.9% (59.2–80.7%) 80% (63.7–91.3%) 82.1% (67.6–92.3%) 69% (55–78.7%)
*95% confidence interval (CI) are given in brackets
==== Refs
Blaser MJ Ecology of Helicobacter pylori in the human stomach J Clin Invest 1997 100 759 762 9259572
Blaser MJ Perez-Perez GI Kleanthous H Cover TL Peek RM Chyou PH Stemmermann GN Nomura A Infection with Helicobacter pylori strains possessing cagA is associated with an increased risk of developing adenocarcinoma of the stomach Cancer Res 1995 55 2111 2115 7743510
Goossens H Glupczynski Y Burette A Van den Borre C Butzler JP Evaluation of commercially available second generation immunoglobulin G enzyme immunoassay for detection of Helicobacter pylori infection J Clin Microbiol 1992 30 176 180 1734050
Klein PD Malaty HM Martin RF Graham KS Genta RM Graham DY Noninvasive detection of Helicobacter infection in clinical practice: The 13C urea breath test Am J Gastroenterol 1996 91 690 696 8677930
Brooks HJ Ahmed D McConnell MA Barbezat GO Diagnosis of Helicobacter pylori infection by polymerase chain reaction: is it worth it? Diagn Microbiol Infect Dis 2004 50 1 5 15380272 10.1016/j.diagmicrobio.2003.11.010
Lam SK Talley NJ Consensus conference on the management of Helicobacter pylori infection J Gastroenterol Hepatol 1998 13 1 12 9737564
Laine L Estrada R Trujillo M Knigge K Fennerty MB Effect of proton pump inhibitor therapy on diagnostic testing for Helicobacter pylori Ann Intern Med 1998 129 547 550 9758575
Dickey W Kenny BD McConnell JB Effect of proton pump inhibitors on the detection of Helicobacter pylori in gastric biopsies Aliment Pharmacol Ther 1996 10 289 293 8791953
Marshall BJ Treatment strategies for Helicobacter pylori infection Gastrenterol Clin North Am 1993 22 183 198
Stolte M Bethke B Elimination of Helicobacter pylori under treatment with omeprazole Zeitschrift Fur Gastroenterologie 1990 28 271 274 2238754
Graham DY Opekun AR Jogi M Yamaoka Y Lu H Reddy R El-Zimaity HM False negative urea breath tests with H2-receptor antagonists: interactions between Helicobacter pylori density and pH Helicobacter 2004 9 17 27 15156900 10.1111/j.1083-4389.2004.00191.x
Siddiqi S Hamid S Rafique G Chaudhry SA Ali N Shahab S Sauerborn R Prescription practices of public and private health care providers in Attock District of Pakistan Int J Health Plann Manage 2002 17 23 40 11963441 10.1002/hpm.650
Sturm AW van der Pol R Smits AJ van Hellemondt FM Mouton SW Jamil B Minai AM Sampers GH Over-the-counter availability of antimicrobial agents, self-medication and patterns of resistance in Karachi, Pakistan J Antimicrob Chemother 1997 39 543 547 9145830 10.1093/jac/39.4.543
Price AB The Sydney System: histological division J Gastroenterol Hepatol 1991 6 209 222 1912431
Said RM Cheah PL Chin SC Goh KL Evaluation of a new biopsy urease test: Pronto Dry for the diagnosis of Helicobacter pylori infection Eur J Gastroenterol Hepatol 2004 16 195 199 15075994 10.1097/00042737-200402000-00012
Schnell GA Schubert TT Barnes WG Rupani MK Comparison of urease, H & E and culture tests for Helicobacter pylori Gastroenterology 1998 94 A 410
Logan RP Walker MM Misiewicz JJ Gummett PA Karim QN Baron JH Changes in the intragastric distribution of Helicobacter pylori during treatment with omeprazole Gut 1995 36 12 16 7890214
Veldhuyzen van Zanten SJO Leung V O'Rourke JL Lee A Gastric Transitional Zones, Areas where Helicobacter Treatment Fails: Results of a Treatment Trial Using the Sydney Strain Mouse Model Antimicrob Agents Chemother 2003 47 2249 2255 12821476 10.1128/AAC.47.7.2249-2255.2003
|
16309551
|
PMC1316874
|
CC BY
|
2021-01-04 16:03:27
|
no
|
BMC Gastroenterol. 2005 Nov 25; 5:38
|
utf-8
|
BMC Gastroenterol
| 2,005 |
10.1186/1471-230X-5-38
|
oa_comm
|
==== Front
BMC Med GenetBMC Medical Genetics1471-2350BioMed Central London 1471-2350-6-411632421510.1186/1471-2350-6-41Research ArticlePolymorphisms of the Flavin containing monooxygenase 3 (FMO3) gene do not predispose to essential hypertension in Caucasians Dolan Ciara [email protected] Denis C [email protected] Alice [email protected]'Brien Eoin [email protected] Deborah M [email protected]'Brien John K [email protected] Eileen P [email protected] Department of Clinical Pharmacology, Royal College of Surgeons in Ireland, 123 St Stephens Green, Dublin 2, Ireland2 Conway Institute, University College Dublin, Dublin 4, Ireland3 ADAPT Centre, Beaumont Hospital and Department of Clinical Pharmacology, Royal College of Surgeons in Ireland, Dublin 2, Ireland4 Department of Genetics, Children's University Hospital, Temple St, Dublin 1, Ireland5 National Centre for Inherited Metabolic Disorders, Children's University Hospital, Temple St., Dublin 1, Ireland2005 2 12 2005 6 41 41 27 4 2005 2 12 2005 Copyright © 2005 Dolan 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 recessive disorder trimethylaminuria is caused by defects in the FMO3 gene, and may be associated with hypertension. We investigated whether common polymorphisms of the FMO3 gene confer an increased risk for elevated blood pressure and/or essential hypertension.
Methods
FMO3 genotypes (E158K, V257M, E308G) were determined in 387 healthy subjects with ambulatory systolic and diastolic blood pressure measurements, and in a cardiovascular disease population of 1649 individuals, 691(41.9%) of whom had a history of hypertension requiring drug treatment. Haplotypes were determined and their distribution noted.
Results
There was no statistically significant association found between any of the 4 common haplotypes and daytime systolic blood pressure in the healthy population (p = 0.65). Neither was a statistically significant association found between the 4 common haplotypes and hypertension status among the cardiovascular disease patients (p = 0.80).
Conclusion
These results suggest that the variants in the FMO3 gene do not predispose to essential hypertension in this population.
==== Body
Background
Essential hypertension is the most common reason for adult visits to office-based physicians and for the use of medication [1-3]. It is estimated that 24% of the US adult population (70% in persons over 70) and at least 50% of the Irish adult population aged 50 or over are hypertensive[4,5]. Ethnic differences exist in the prevalence of hypertension between populations, for example between Africans and Northern Europeans [6-8]. Mendelian forms of hypertension resulting from single gene defects are rare [9-11]. The major contribution to the etiology of this disorder is proposed to result from the combined effects of genes that modify the response of blood pressure to environmental stresses such as diet and 'environmental susceptibility genes' [12-14]. Essential hypertension does not follow a clear pattern of inheritance but exhibits familial aggregation of cases. This multifactorial trait increases the affected individuals' risk of myocardial infarction, stroke and end-stage renal disease, and is one of the leading causes of morbidity and mortality in adults[8,15]. Population-wide application of preventative measures and candidate gene analysis to predict modifiable risks in addition to treatment are thus very worthwhile [16-19].
Trimethylaminuria (TMAuria) is an inborn error of metabolism (MIM # 275700) resulting from diminished oxidation of the tertiary amine trimethylamine to trimethylamine N-oxide resulting in a severe body odour in affected individuals. The microsomal NADPH-dependent flavin-containing monooxygenases (FMOs) (E.C. 1.14.12.8) are a family of chemical and drug metabolising enzymes that catalyze the phase 1 oxygenation of a wide variety of nucleophilic heteroatom containing compounds including catecholamines[20,21]. FMO3, a phase 1 drug metabolising gene, is the main liver dependent human isoform[22]. We and others have shown that recessivity for mutations of FMO3 cause TMAuria [20-24].
We have previously noted that a number of patients with TMAuria have idiopathic hypertension [21,22]. In addition, Cashman et al [25] have reported a proposed link in African-American males between hypertension and increased excretion of trimethylamine. We have identified increased excretion of catecholamines in a proband with TMAuria who is homozygous for a deletion of the FMO3 gene [20]suggesting a possible association between abnormal catecholamine metabolism and variants of the FMO3 gene. Endogenous substrates for FMO3 include tyramine and phenylethylamine [26,27] and tyramine has a known pressor effect. As high levels of circulating catecholamines contribute to hypertension, it is proposed that polymorphisms of FMO3 gene could contribute to impaired catecholamine metabolism and hypertension.
We have previously described the population frequencies of a number of common polymorphisms of the FMO3 gene (E158K, V257M, E308G) in a North American population[25,28]. Expression studies indicate that the polymorphisms E158K and V257M exhibit decreased tyramine oxidation in vitro[25]. We and others have noted that the E308G polymorphism shows a substantial decrease in FMO3 activity in the presence of the E158K polymorphism causing mild TMAuria[22,29]. While these common polymorphisms exhibit some difference in expression of the functional FMO3 enzyme, this effect is minimal in comparison to that of the null mutations.
We hypothesised that common FMO3 polymorphisms might predispose to essential hypertension. While null alleles in the FMO3 gene may be too rare to cause any appreciable effect on the population burden of hypertension, prevalent polymorphisms with documented in vitro evidence of variation in catecholamine metabolism may however potentially associate with and predispose to this condition.
Herein we report our investigations into a possible association of haplotypes of three polymorphic variants of the FMO3 gene with the phenotype of blood pressure in an occupational Irish adult population and with presence of essential hypertension in an Irish cardiovascular disease (CVD) population.
Methods
Study populations
There were 2 population groups in this study. The first population was recruited from the Allied Irish Bank Phase II (AIB Phase II) study that commenced in 1989 and consists of 387 bank employees (Table 1). Blood pressure was measured every 30 minutes for 24 hours using a validated monitor – the SpaceLabs 90207. The mean daytime and night time systolic (SBP) and diastolic (DBP) values were used for analysis in each case. Hypertension was defined as having daytime blood pressure of 135/85 mm Hg or greater. This group was collected between 1998 and 2002 and was between the ages of 30–70 years. It was noted if individuals consumed more than 2 alcohol-containing drinks per day.
The second were a group of 1649 patients with coronary disease (CHD), 1313 of whom were ascertained on the basis of having acute coronary syndromes (ACS) (myocardial infarction (MI) or unstable angina) and 336 as having stable angina (Table 2). MI was defined as chest pain of at least 20 minutes duration, along with previous or current electrocardiogram or serum enzyme changes diagnostic of MI. Unstable angina was defined as chest pain typical of angina occurring at rest or lasting at least 20 minutes and requiring hospitalization in a patient with known coronary artery disease based on a positive stress test or a coronary angiogram. Inclusion criteria for stable angina were chest pain occurring with exercise typical of angina in a patient with known coronary artery disease based on a coronary angiogram or a positive treadmill test. These patients were collected between the years 1999–2002 and were between the ages of 32 and 85 years. They were classified as having a history of hypertension requiring drug treatment or having no history of hypertension.
Informed consent and ethics approval from the Beaumont Hospital Ethics (Medical Research) Committee was obtained for all samples collected.
FMO3 polymorphism genotyping
Three single nucleotide polymorphisms in the FMO3 gene were genotyped in the above populations. These were E158K (G472A), V257M (G769A) and E308G (A923G). The E158K polymorphism was genotyped using the primer GAAGGTGACCAAGTTCATGCTTGGCCTTACCTGGAAAGGACTT for the G allele and GAAGGTCGGAGTCAACGGATTTTTGGCCTTACCTGGAAAGGACT for the A allele. The V257M polymorphism was genotyped using the primer GAAGGTGACCAAGTTCATGCTCAGCCATCTCTGACTGGTTGTACA for the G allele and GAAGGTCGGAGTCAACGGATTAGCCATCTCTGACTGGTTGTACG for the A allele. The E308G polymorphism was genotyped using the primer GAAGGTGACCAAGTTCATGCTGCCTAACGTGAAGGAATTCACAGA for the A allele and GAAGGTCGGAGTCAACGGATTGCCTAACGTGAAGGAATTCACA for the G allele.
Genotyping was carried out using the Amplifluor™ method by K Biosciences . Genomic DNA was isolated from blood. Genotyping was performed in 384-well microplates using a fluorescence resonance energy transfer (FRET)-based genotyping method. Amplification was initiated using allele-specific primers and a common downstream primer. The allele-specific primers were tailed with unique sequences that create corresponding complementary sequences in the two amplicons. In the second round of amplification, quenched Universal Amplifluor™ primers (in a hairpin formation) were used. These primers contain 3' tails that specifically bind to the unique tailed sequences in the amplicons and continue amplification. In the final round of amplification, the action of the DNA polymerase opened up the hairpin structure and the quencher and reporter moieties are spatially separated. The excited reporter moiety emitted either red or green fluorescence, the colour of which depends on which nucleotide is at the polymorphism site. The fluorescence was quantified by a microplate reader and then analysed via an Excel macro to provide genotypes for each SNP. Confirmation of the validation of the mutation detection method was based on the use of known positive DNA controls (which had been sequenced) for the three polymorphisms provided by McGill University (Montreal).
Haplotype analysis
Haplotypes for the 3 SNPs were inferred using a maximum likelihood approach and their association with the rank of the various mean blood pressure measurements in the AIB Phase II study group and their association with hypertension in the CVD group was determined using "haplo.score", a function written for the statistical package SPLUS 6.0[30]. This method used score tests for association between a quantitative trait and the haplotypes. The result of the association was given as a haplotype score. The haplotypes score was calculated using a general linear model, which permitted adjustment for age, sex and alcohol consumption.
Statistical analysis
The 2 primary analyses were (1) the association between haplotypes of the 3 SNPs and mean daytime systolic blood pressure in the AIB Phase II study population and (2) the association between haplotypes and the hypertension status in the CVD patients.
The secondary analyses included (1) the association between genotypes and daytime systolic blood pressure in the AIB Phase II study, (2) the association between genotypes and hypertensive status in the CVD group, (3) the comparison of genotypic frequencies between the AIB Phase II study group, the hypertensives of the CVD group and the non-hypertensives of the CVD group, (4) the effect of haplotypes on mean daytime diastolic blood pressure, on mean night time systolic blood pressure and on mean night time diastolic blood pressure in the AIB Phase II study group, (5) the association between polymorphisms in the FMO3 gene and cardiovascular disease.
Haplo.score was used to look for any associations between haplotypes and daytime systolic blood pressure/hypertensive status. We only tested for relatively common haplotypes, and defined as those at an inferred frequency of greater than 5% in the populations. For all of the above analyses, statistical significance was determined when p < 0.05.
ANOVA was used to look for an association between the genotypes and mean daytime systolic blood pressure in the AIB Phase II study group. Both haplo.score and ANOVA were adjusted for age, sex and the number of alcohol units consumed per week.
Chi-squared analysis was used to look for the difference in genotype frequencies between the hypertensive and non-hypertensive patients in the CVD group. Linear regression was then used to adjust for age and sex. Chi-squared analysis was also used to compare genotypic frequencies between the AIB Phase II study group, the hypertensives from the CVD group and the non-hypertensives of the CVD group. Logistic regression analysis was used to predict if any of the SNPs influence the risk of cardiovascular disease, including diabetes, BMI, hypercholesterolemia, smoking status, age and gender in the model. All of the above analyses were repeated stratifying according to gender.
Results
The AIB Phase II population consisted of 387 subjects, 224 (57.8%) of which were male. In comparison to the females, the males had on average higher blood pressure measurements, triglycerides and total cholesterol. Slightly more males in this group were smokers (Table 1).
The CVD group consisted of 1649 subjects, 691 (41.9%) of which were classified as having a history of hypertension requiring drug treatment. A greater percentage of those classified as hypertensive also had hypercholesterolemia and diabetes mellitus in comparison to those who were classified as non-hypertensive. As might be expected from a CVD group, most of the subjects were taking medication, with almost the same frequency of hypertensives and non-hypertensives taking each group of medications (Table 2). Blood pressure measurements are only presented to characterise, rather than compare, the groups, since medication is lowering blood pressure in most of the CHD group, but not in most of the AIB Phase II study group.
The most notable difference between the AIB Phase II study group and the CVD group is that there were a much greater proportion of current or ex smokers in the CVD group than in the AIB Phase II study group (73% vs. 30.1%).
The 3 SNPs were found to be in Hardy-Weinberg Equilibrium. 4 common haplotypes were inferred in the two populations (Table 3), with similar frequencies in the AIB Phase II study group, the hypertensives and in the non-hypertensives of the CVD group. However the haplotype KVG, which is the 2nd most frequent haplotype in the AIB Phase II study group, was the 3rd most frequent haplotype in the CVD group. However the difference in frequency was only 4%, which is most likely due to chance.
No statistically significant associations were found between any of the four common haplotypes and mean daytime systolic blood pressure in the AIB Phase II study (overall p = 0.65, 3 d.f) (See table 3). Nor was this statistically significant when adjusted for age, sex and daily alcohol intake (overall p = 0.729, 3 d.f) (See table 3). Neither did we find any statistically significant associations between any of the haplotypes and hypertension status in all subjects with and without adjustment of sex and age (see table 3). When stratified by gender, there were still no statistically significant associations between any of the specific haplotypes and daytime SBP in the AIB Phase II study (males overall p = 0.91, 3 df and females overall p = 0.61, 3 df) or with hypertension status in the CVD patients (males overall p = 0.59, 3 df and females overall p = 0.82, 3 df). Table 4 displays the genotypic frequencies of the 3 SNPs in the AIB Phase II population, in the hypertensives of the CVD group and in the non-hypertensives of the CVD group. When genotypic frequencies were compared among 3 groups, there were no statistically significant differences (E158K p = 0.30, V257M p = 0.47, E308G p = 0.15). There were no statistically significant associations between any of the individual genotypes with daytime SBP in the AIB Phase II study group, even when adjusted for age, sex and daily alcohol consumption (E158K p = 0.91, V257M p = 0.85, E308G p = 0.68).
The mean daytime SBP and mean daily alcohol consumption are broken down by genotype in table 5 and it is clear that there is not much difference between the genotype groups. Stratifying the analysis by gender, there was still no association between genotypic frequencies in the AIB Phase II study group (females E158K p = 0.92, V257M p = 0.86, E308G p = 0.13, males E158K p = 0.37, V257M p = 0.87, E308G p = 0.38).
There was no genotypic difference between hypertensive and non-hypertensive CVD subjects, even after adjustment for age and sex (E158K p = 0.84, V257M p = 0.44, E308G p = 0.91). When this analysis was stratified by gender, still no statistically significant differences are noted (females E158K p = 0.79, V257M p = 0.54, E308G p = 0.29, males E158K p = 0.71, V257M p = 0.22, E308G p = 0.66).
Testing for any associations between haplotype and daytime diastolic blood pressure, night time systolic blood pressure and night time diastolic blood pressure in the AIB Phase II study yielded no statistically significant associations (Table 6). Likewise it was shown by logistic regression that none of the SNPs were significant predictors for cardiovascular disease when the CVD and the AIB Phase II study groups were compared (E158K z = 0.60 p = 0.55, V257M z = -0.31 p = 0.76, E308G z = 0.49 p = 0.63), though diabetes was found to be a significant predictor.
Discussion
The observation that a number of patients with TMAuria are hypertensive was the basis to the hypothesis that there may be an association between variants in the FMO3 gene and hypertension. The FMO3 enzyme has a broad substrate specificity, which includes catecholamines as a minor pathway[21,26]. Catecholamines are produced in response to stress and modulate heart rate and blood pressure. Variants of the FMO3 gene resulting in decreased enzymatic activity could result in decreased catabolism of catecholamines, which impact on blood pressure homeostasis. As catecholamines are at peak levels during the day, we elected to study the association between daytime SBP and variants in the FMO3 gene as a primary endpoint. In our analysis, we adjusted for alcohol use as tyramine is a substrate for FMO3 oxidation[26]. The FMO3 enzyme is also modulated by the sex hormones[28]. Females with moderate to severe TMAuria note exacerbations perimenstrually and it is proposed that decreases in FMO3 enzymatic activity resulting from polymorphisms could also be influenced by hormonal events[31]. For this reason the analysis was stratified by treating males and females as separate groups.
In this study, by constructing haplotypes based on the 3 SNPs, we discovered that none of the haplotypes were statistically significantly associated with either daytime systolic blood pressure in the AIB Phase II study group or with hypertensive status in the CVD group. There was also a lack of association between any of the genotypes of the SNPs with daytime SBP or with hypertension status. Neither did the stratified analyses show any statitically significant associations.
While the definition of hypertension in the CVD population provides a rather arbitrary estimation of hypertension, the 24-hour blood pressure measurements in the AIB Phase II study were carefully controlled. Even so, the results from both groups showed that the variants had no association with blood pressure or hypertension status.
Why are FMO3 variants maintained in population, and in particular the E158K variant? It is proposed that genes such as FMO3, a drug and toxin detoxicating gene, may have survived as balanced polymorphisms in an ancient human or primate population to neutralise the effect of harmful exposures[32,33]. Evolutionary changes in FMO3 may have evolved in different geographic locations to buffer changes in the highly polymorphic drug metabolising genes, which show substantial geographic and ethnic variation. It is possible that 'low penetrance' polymorphisms of FMO3 alone may not independently affect blood pressure homeostasis but severe loss of function causing mutations may unmask pressor effects of variation in other drug metabolising enzymes previously buffered by FMO3[34]. This hypothesis is supported by the observation that only some, but not all, TMAuria patients exhibit hypertensive symptoms. Therefore other common FMO genes may be potential hypertension candidates. There could be identified by typing other FMO genes in TMAuria patients with hypertensive symptoms or by directly assessing the candidate genes in a large study of hypertension.
Conclusion
In conclusion, these data indicate that these three polymorphisms of the FMO3 gene inherited singly or in specific haplotype combinations do not represent an independent susceptibility risk for hypertension as an absolute trait or continuous variable in a representative Irish population.
Authors' contributions
CD carried out the statistical analysis and drafted the manuscript. DCS assisted in the statistical analysis and participated in the study design. AS contributed to the interpretation of the hypertension data and participated in the study design. EOB designed the data collection for the AIB Phase II study and participated in the study design. DML helped draft the manuscript. JKOB led the genotypic analysis component. EPT conceived the study and helped to draft the manuscript. All authors read and approved the final manuscript.
Pre-publication history
The pre-publication history for this paper can be accessed here:
Acknowledgements
This work was supported by the Programme for Research in Third Level Institutions administered by the Irish Higher Education Authority. With special thanks to Allied Irish Banks and their workers who participated in the study.
Figures and Tables
Table 1 Characteristics of AIB Phase II study group
Male Mean/SD* (n = 224) Female Mean/SD* (n = 163)
Age (years) 52.5 +- 9.2 47.4 +- 10.0
Weight (KG) 85.7 +- 16.1 66.3 +- 13.7
Height (cm) 177.5 +- 12.6 163.1 +- 11.3
Hypertensive 3.1% 1.2%
Daytime systolic BP (mmHg) 128.2 +- 10.3 118.3 +- 9.9
Daytime diastolic BP (mmHg) 80.8 +- 7.7 75.6 +- 7.3
Nighttime systolic BP (mmHg) 108.7 +- 9.8 101.5 +- 8.6
Nighttime diastolic BP (mmHg) 62.5 +- 7.5 58.1 +- 6.3
Total cholesterol (mmol/L) 5.7 +- 1.2 5.3 +- 1.2
Triglycerides (mmol/L) 1.5 +- 0.9 1.0 +- 0.6
Diabetes mellitus (%) 0.8% 0.6%
Smokers (Current/Ex) (%) 32.4% 27.9%
Medication:
Antihypertensive (%) 3.6% 3.7%
Hormone replacement therapy 0% 21%
Anti-arrhythmic 0.5% 0%
*Standard deviation
Table 2 Characteristics of Cardiovascular disease group
Hypertensive Mean/SD* (n = 691) Non-hypertensive Mean/SD* (n = 958)
Age (years) 62.5 +/- 7.9 60.4 +/- 8.7
Weight (KG) 79.1 +/- 19.7 77.5 +/- 18.4
Height (cm) 160.3 +/- 34.9 163.1 +/- 34.3
Gender (males) 66.8% 83.6%
Hypercholesterolemia (%) 68.3% 51.7%
Diabetes mellitus (%) 12.9% 6.8%
Smokers (Current/Ex) (%) 69.9% 76.9%
Aspirin (%) 99.0% 99.2%
Statins (%) 96.3% 98.5%
B-Blockers (%) 95.9% 97.9%
Nitrates (%) 91.7% 95.4%
Calcium channel blockers (%) 92.2% 92.7%
ACE inhibitors (%) 92.2% 91.9%
*Standard deviation
Table 3 Associations between haplotypes in the AIB Phase II group with daytime systolic blood pressure and in the CVD group with hypertension status.
AIB Phase II group CVD group
Haplotypes Freq Haplotype scoreb p-value Haplotype scoreb* p-value* Hypertensive freq Non-hypertensive freq Haplotype scoreb p-value Haplotype scoreb† p-value†
EVEa 50.5% 0.497 0.62 0.392 0.70 53.5% 54.4% -0.578 0.56 -0.715 0.47
KVGa 22.0% 0.638 0.52 0.629 0.51 18.0% 17.8% 0.206 0.84 0.433 0.66
KVEa 21.2% -0.669 0.50 -0.687 0.49 21.5% 21.7% -0.004 0.99 -0.073 0.94
EMEa 6.3% -0.954 0.34 -0.725 0.45 7.0% 6.1% 0.921 0.36 0.948 0.34
Overall p-value 0.65 0.73 0.80 0.79
a haplotypes defined for positions 158, 257 and 308 respectively
b from the haplo.score method[30]
* Adjusted for age, sex and alcohol intake
† Adjusted for age and sex
Table 4 Tabulation of genotype percentages and frequencies in groups
Genotype (amino acid) AIB Phase II study (387) Hypertensive (775) Non-hypertensive (984)
E158K KK 19.4% (75) 14.6% (101) 16.0% (154)
EK 45.7% (177) 47.7% (330) 45.6% (439)
EE 33.8% (131) 35.1% (243) 37.1% (357)
Missing 1.1% (4) 2.6% (18) 1.3% (12)
V257M MM 1.0% (4) 0.4% (3) 0.3% (3)
VM 11.1% (43) 12.3% (85) 11.3% (109)
VV 86.3% (334) 86.4% (598) 87.6% (843)
Missing 1.5% (6) 0.9% (6) 0.7% (7)
E308G GG 5.7% (22) 3.3% (23) 3.3% (32)
GE 31.0% (120) 28.2% (195) 28.5% (274)
EE 61.2% (237) 66.9% (463) 66.8% (643)
Missing 2.1% (8) 1.6% (11) 1.4% (13)
Table 5 Mean daytime systolic blood pressure and mean alcohol consumption broken down by genotype in the AIB Phase II study group.
Genotype (amino acid) Number subjects Mean daytime sbp/SD* Alcohol consumption +/- SD*
E158K KK 75 126.71 +- 13.1 12.87 +- 13.8
EK 177 125.87 +- 12.6 12.83 +- 12.4
EE 131 126.61 +- 13.2 14.81 +- 14.1
V257M MM 4 126.67 +- 9.3 14.00 +- 12.3
VM 43 121.25 +- 14.7 15.50 +- 16.7
VV 334 126.02 +- 12.7 13.24 +- 12.9
E308G GG 22 129.32 15.65
GE 120 125.82 11.87
EE 237 126.18 14.26
* Standard deviation
Table 6 Summary of haplotype scores representing associations between blood pressure and haplotypes in the AIB phase II population showing overall statistical significance values
Haplotypes (amino acid) EVEa KVGa KVEa EMEa Overall p-val
Daytime DBP Haplotype scoreb -0.136 -0.266 0.890 -0.726 0.76
Haplotype scoreb (adjusted*) -0.361 -0.192 0.985 -0.565 0.77
Nighttime SBP Haplotype scoreb -0.657 1.676 -0.576 -0.586 0.40
Haplotype scoreb (adjusted*) -0.803 1.676 -0.573 -0.294 0.42
Nighttime DBP Haplotype scoreb -1.313 1.313 0.665 -0.699 0.41
Haplotype scoreb (adjusted*) -1.453 1.177 0.706 -0.260 0.47
* Adjusted for age, sex and alcohol intake
a haplotypes defined from positions 128, 257 and 308 respectively
b from the haplo.score method
==== Refs
Rudnick KV Sackett DL Hirst S Holmes C Hypertension in a family practice Can Med Assoc J 1977 117 492 497 902191
Rose VL National Heart, Lung, and Blood Institute releases new guidelines for the treatment of hypertension Am Fam Physician 1998 57 362 4, 366 9456995
Kaplan NM A cardioprotective approach to the management of hypertension J Hypertens Suppl 2000 18 S29 31 10952085 10.1038/sj.jhh.1001798
Burt VL Whelton P Roccella EJ Brown C Cutler JA Higgins M Horan MJ Labarthe D Prevalence of hypertension in the US adult population. Results from the Third National Health and Nutrition Examination Survey, 1988-1991 Hypertension 1995 25 305 313 7875754
Creagh D Neilson S Collins A Colwell N Hinchion R Drew C O'Halloran D Perry IJ Established cardiovascular disease and CVD risk factors in a primary care population of middle-aged Irish men and women Ir Med J 2002 95 298 301 12542011
Kurtz TW Spence MA Genetics of essential hypertension Am J Med 1993 94 77 84 8420305 10.1016/0002-9343(93)90124-8
Kaplan NM Ethnic aspects of hypertension Lancet 1994 344 450 452 7914569 10.1016/S0140-6736(94)91774-4
Whelton PK Klag MJ JDS Hypertension in westernised populations Textbook of Hypertension 1994 Oxford, Blackwell Scientific Publications 11 19
Jeunemaitre X Soubrier F Kotelevtsev YV Lifton RP Williams CS Charru A Hunt SC Hopkins PN Williams RR Lalouel JM Molecular basis of human hypertension: role of angiotensinogen Cell 1992 71 169 180 1394429 10.1016/0092-8674(92)90275-H
Lifton RP Gharavi AG Geller DS Molecular mechanisms of human hypertension Cell 2001 104 545 556 11239411 10.1016/S0092-8674(01)00241-0
Wilson FH Disse-Nicodeme S Choate KA Ishikawa K Nelson-Williams C Desitter I Gunel M Milford DV Lipkin GW Achard JM Feely MP Dussol B Berland Y Unwin RJ Mayan H Simon DB Farfel Z Jeunemaitre X Lifton RP Human hypertension caused by mutations in WNK kinases Science 2001 293 1107 1112 11498583 10.1126/science.1062844
Hamet P Environmentally-regulated genes of hypertension Clin Exp Hypertens 1996 18 267 278 8743020
Hamet P Pausova Z Adarichev V Adaricheva K Tremblay J Hypertension: genes and environment J Hypertens 1998 16 397 418 9797185 10.1097/00004872-199816040-00001
Perry IJ Whincup PH Shaper AG Environmental factors in the development of essential hypertension Br Med Bull 1994 50 246 259 8205457
Andersson OK Almgren T Persson B Samuelsson O Hedner T Wilhelmsen L Survival in treated hypertension: follow up study after two decades Bmj 1998 317 167 171 9665894
Halushka MK Fan JB Bentley K Hsie L Shen N Weder A Cooper R Lipshutz R Chakravarti A Patterns of single-nucleotide polymorphisms in candidate genes for blood-pressure homeostasis Nat Genet 1999 22 239 247 10391210 10.1038/10297
Smithies O Maeda N Gene targeting approaches to complex genetic diseases: atherosclerosis and essential hypertension Proc Natl Acad Sci U S A 1995 92 5266 5272 7777495
Holtzman NA Marteau TM Will genetics revolutionize medicine? N Engl J Med 2000 343 141 144 10891526 10.1056/NEJM200007133430213
Senior V Marteau TM Weinman J Impact of genetic testing on causal models of heart disease and arthritis: analogue studies Phsychol Health 2000 14 1077 1088
Treacy EP Akerman BR Chow LM Youil R Bibeau C Lin J Bruce AG Knight M Danks DM Cashman JR Forrest SM Mutations of the flavin-containing monooxygenase gene (FMO3) cause trimethylaminuria, a defect in detoxication Hum Mol Genet 1998 7 839 845 9536088 10.1093/hmg/7.5.839
Cashman JR Bi YA Lin J Youil R Knight M Forrest S Treacy E Human flavin-containing monooxygenase form 3: cDNA expression of the enzymes containing amino acid substitutions observed in individuals with trimethylaminuria Chem Res Toxicol 1997 10 837 841 9282831 10.1021/tx9700533
Akerman BR Lemass H Chow LM Lambert DM Greenberg C Bibeau C Mamer OA Treacy EP Trimethylaminuria is caused by mutations of the FMO3 gene in a North American cohort Mol Genet Metab 1999 68 24 31 10479479 10.1006/mgme.1999.2885
Akerman BR Forrest S Chow L Youil R Knight M Treacy EP Two novel mutations of the FMO3 gene in a proband with trimethylaminuria Hum Mutat 1999 13 376 379 10338091 10.1002/(SICI)1098-1004(1999)13:5<376::AID-HUMU5>3.0.CO;2-A
Dolphin CT Janmohamed A Smith RL Shephard EA Phillips IR Missense mutation in flavin-containing mono-oxygenase 3 gene, FMO3, underlies fish-odour syndrome Nat Genet 1997 17 491 494 9398858 10.1038/ng1297-491
Cashman JR Akerman BR Forrest SM Treacy EP Population-specific polymorphisms of the human FMO3 gene: significance for detoxication Drug Metab Dispos 2000 28 169 173 10640514
Lin J Cashman JR Detoxication of tyramine by the flavin-containing monooxygenase: stereoselective formation of the trans oxime Chem Res Toxicol 1997 10 842 852 9282832 10.1021/tx970030o
Elfarra AA Potential role of the flavin-containing monooxygenases in the metabolism of endogenous compounds Chem Biol Interact 1995 96 47 55 7720104 10.1016/0009-2797(94)03582-S
Lambert DM Mamer OA Akerman BR Choiniere L Gaudet D Hamet P Treacy EP In vivo variability of TMA oxidation is partially mediated by polymorphisms of the FMO3 gene Mol Genet Metab 2001 73 224 229 11461189 10.1006/mgme.2001.3189
Zschocke J Kohlmueller D Quak E Meissner T Hoffmann GF Mayatepek E Mild trimethylaminuria caused by common variants in FMO3 gene Lancet 1999 354 834 835 10485731
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
Zhang AQ Mitchell SC Smith RL Exacerbation of symptoms of fish-odour syndrome during menstruation Lancet 1996 348 1740 1741 8973460 10.1016/S0140-6736(05)65872-2
Julius S The defense reaction: a common denominator of coronary risk and blood pressure in neurogenic hypertension? Clin Exp Hypertens 1995 17 375 386 7735282
Nebert DW Polymorphisms in drug-metabolizing enzymes: what is their clinical relevance and why do they exist? Am J Hum Genet 1997 60 265 271 9012398
Hartman JLT Garvik B Hartwell L Principles for the buffering of genetic variation Science 2001 291 1001 1004 11232561 10.1126/science.291.5506.1001
|
16324215
|
PMC1316875
|
CC BY
|
2021-01-04 16:03:32
|
no
|
BMC Med Genet. 2005 Dec 2; 6:41
|
utf-8
|
BMC Med Genet
| 2,005 |
10.1186/1471-2350-6-41
|
oa_comm
|
==== Front
BMC Public HealthBMC Public Health1471-2458BioMed Central London 1471-2458-5-1251631367110.1186/1471-2458-5-125Research ArticleComparative levels and time trends in blood pressure, total cholesterol, Body Mass Index and smoking among Caucasian and South-Asian participants of a UK primary-care based cardiovascular risk factor screening programme Lyratzopoulos Georgios [email protected] Patrick [email protected] Richard F [email protected] Margaret [email protected] Philip S [email protected] Directorate of Public Health and Clinical Services, Norfolk, Suffolk and Cambridgeshire Strategic Health Authority, Cambridge, UK2 Hunter New England Population Health, Hunter New England Area Health Service, Wallsend, Australia3 Evidence for Population Health Unit, University of Manchester, Manchester, UK4 Former Stockport Community Health Trust, Stockport, UK5 Department of Cardiology, Stockport NHS Trust, Stockport, UK2005 28 11 2005 5 125 125 2 8 2005 28 11 2005 Copyright © 2005 Lyratzopoulos et al; licensee BioMed Central Ltd.2005Lyratzopoulos 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
Individuals of South-Asian origin have a comparatively higher cardiovascular disease burden, but there is uncertainty about whether this is due to differences in risk factor levels and trends. We therefore studied comparative levels and time trends in blood pressure (BP), total cholesterol, body mass index (BMI) and current smoking among UK Caucasian and South-Asian individuals.
Methods
Repeatable cross-sectional survey of men and women aged 35–60 attending for first screening as part of a primary-care based cardiovascular risk factor screening programme 1989 and 1999.
Results
Of 34,122 men and 37,294 women participants, 499 men (1.5%)and 381 women (1%) were of South-Asian origin. South-Asian men had lower systolic [(-4.91 mmHg (95% Confidence Iterval (CI): -3.58 to -6.23)] and diastolic BP [-2.87 mmHg (-2.02 to -3.72)], with no significant differences in cholesterol and BMI. South-Asian women had lower systolic BP [-1.77 mmHg, 95% (-0.21 to -3.33)], diastolic BP [-1.87 mmHg (-0.92 to -2.82)], cholesterol [-0.24 mmol/l (-0.08 to -0.39)]; and higher BMI [+0.78 kg/m2 (0.25 to 1.3)]. South-Asian men and women had significantly lower prevalence of self-reported current smoking (29.0% and 1.8% respectively). With the exception of self-reported current smoking, between ethnic group risk factor trends were not converging.
Conclusion
With the exception of women's BMI, South-Asian individuals had either lower or similar levels of the examined cardiovascular risk factors, compared with Caucasian individuals. Although time trends in smoking were converging, other risk factors trends were similar between the two ethnic groups. Overall the findings do not support the hypothesis that the relatively high cardiovascular disease burden in UK South-Asians is due to higher levels exposure to the examined risk factors. Other hypotheses, such as higher frequency of diabetes and increased genetic predisposition, require further exploration.
==== Body
Background
Cardiovascular disease is the leading cause of death in the UK and responsible for approximately 238,000 deaths annually, or 39% of total mortality [1]. Some UK ethnic groups have a higher cardiovascular disease burden, and this is particularly true for individuals of South-Asian origin [2]. The causes of higher disease burden among individuals of South-Asians origin are not fully understood, but may include excess exposure to known risk factors, excess exposure to as yet unknown risk factors, greater biological susceptibility to cardiovascular disease, or a lower risk of competing mortality and morbidity from other diseases (such as cancer) [2].
In the UK and over the past two decades, the mean population values of total cholesterol and blood pressure have been decreasing, and the same has been true for the prevalence of current smoking; however, mean values of body mass index (BMI) manifest an increasing trend [3]. Whilst cardiovascular risk factor levels for the UK population, including for ethnic groups [4], are relatively well described, there is a relative paucity of information about cardiovascular risk factor time trends in UK ethnic groups, particularly for the period before 1999. Evidence about recent trends in cardiovascular risk factors in UK individuals of South-Asian origin is therefore required [5].
Although a previous meta-analysis of primary studies reported between 1977 and 1996 has questioned the overall efficacy of multiple cardiovascular risk factor screening as a means of preventing the development of cardiovascular disease[6], one potentially important secondary use of population-based screening programmes is as public health surveillance tools, to monitor population risk factor trends. The importance of best using information from routine sources to help give a comprehensive picture of population health and its determinants has been highlighted recently by both the most recent "Wanless Report" [7], and the subsequent UK Department of Health White Paper on Public Health [8]. The Stockport Cardiovascular Disease Risk Factor Screening Programme, originally introduced in 1989, provides an example for the potential secondary use of routine data sources for public health surveillance purposes [9]. Further details about operating protocols and population coverage have been previously described [10,11]. Briefly the Programme was introduced in 1989, and used a call-recall system operated by the Stockport Health Authority, where all Stockport residents aged between 35 and 60 were invited every five years to book a screening appointment at their GP surgery. Cardiovascular risk factors including blood pressure, total cholesterol, BMI and smoking status were assessed, usually by a practice nurse. Between 1989 and 1993, about 10.8% of all patients registered with a GP were excluded from the screening invitation, as they were already known to suffer from hypertension (3.9%), diabetes (1.2%) and conditions including history of any cardiovascular disease, and terminal illness (6.6%).10 Over the whole 11-year period 1989–1999, the population coverage for one (first) screening was about 72.2% for Stockport men aged 35–60 and about 78.4% for Stockport women (see Additional File 1). Individual data on risk factor levels were collated by the Health Authority and anonymised into a usable electronic dataset, which was used in the present study [10,11]. We therefore conducted a study to examine recent risk factor levels and risk factor trends in cardiovascular risk factors among individuals of Caucasian and South Asian origin in a UK population using the Stockport Cardiovascular Risk Factor Screening Programme dataset 1989–1999.
Stockport is a borough of Northwest England. In recent years, general population health has been slightly better than the general UK population, with a Standardised Mortality Ratio from all causes (all ages) of 96 (95% CI 94–98)[12]. In the 1991 Census 1.12% of all Stockport residents were of South Asian origin, and this proportion rose to 2.1% in the 2001 Census [13].
Methods
Information on 34,122 men and 37,294 women who had a first screening episode during the 11-year period 1989–1999 was analysed.
Attribution of ethnicity
Information on ethnicity was imputed by the use of the Nam Pahcham software programme version 1.1 [14]. For data protection purposes, the process of assigning ethnic group was completely separated by the data held on each individual, although anonymous linkage occurred at a later stage, by the use of a key identifier, by Stockport Primary Care Trust, who were the legal custodians of the data. The programme was run against the full list of surnames of people included in the database, in order to identify individuals belonging to non-Caucasian ethnic groups. Adopting a method originally described by Cummins et al. [15] the initial yield of names identified as belonging to ethnic group individuals was subsequently visually inspected for inaccuracies by two investigators independently. One of the investigators was of Caucasian and one of South-Asian descent, and results were compared before a final decision was made as to whether names identified as belonging to individuals of ethnic background corresponded to South-Asian individuals. Subsequently the same pair of investigators independently visually inspected all names not initially identified as of probably ethnic origin, to identify any "false negatives". A person was assigned to the South-Asian ethnic group if their name matched the typology of known names of Pakistani, Bangladeshi or Indian origin. Therefore, in this paper South-Asian ethnicity includes a heterogeneous group of persons born either in or outside the UK, and belonging to one of the many ethnic and religious groups originating from Bangladesh, India, Pakistan and Sri Lanka.
Risk factor measurements
Details are included in Additional File 2, as described previously.10 Briefly, total serum cholesterol was measured at the biochemistry laboratory of the Stepping Hill Hospital (a typical UK District General Hospital) by an enzymatic colorimetric assay using cholesterol esterase and cholesterol oxidase [16]. The Biochemistry laboratory participated in external quality assurance schemes. Fresh blood samples were sent to the laboratory and processed within 24 hours of sampling. Systolic and diastolic BP were measured according to a standardised protocol conforming to the 1987 "Recommendations on Blood Pressure Measurement" of the British Hypertension Society [17]. Training for all physical measurements was provided by a visiting nurse facilitator, employed by the Stockport Health Authority, whose role was to quality assure and co-ordinate the implementation of the screening activity. Smoking status was ascertained by direct questioning (self-reported).
Statistical analysis
For each ethnic group, simple counts and proportions were calculated for ascertainment completeness of each risk factor.
Comparison of risk factor levels
To assess whether there were differences in the average level of a risk factor between ethnic groups, by sex, firstly age-standardised risk factor levels were calculated by the direct method [18-20]. Due to the relatively small number of individuals in the South-Asian group, means for four two-year sub-periods (1989–90, 1991–2, 1993–4, 1995–6) and one three-year sub-period (1997–9) were calculated. Associated 95% confidence intervals were calculated using the normal approximation to the binomial distribution for each age stratum and combining all the strata using weighted averages.
To inform the interpretation of standardised rates, and to examine the significance of differences in risk factor levels between ethnic groups, regression models were constructed. For each continuous risk factor (as the dependent variable), a linear regression model was fitted, with age, test year and ethnic group as the independent variables (Model 1a). In these models, the co-efficient for ethnic group denotes the (age- and test year – adjusted) difference in risk factor levels between the Caucasian and South-Asian groups. For smoking, similarly, a logistic regression model was constructed, with current smoking status (current smoker vs. non-smoker) as the dependent binary variable, and age, test year and ethnic group as independent variables (Model 1b). In this model, the exponential of the co-efficient for ethnic group denotes the (age- and test year – adjusted) odds ratio of current smoking status in the South-Asian group compared to the Caucasian.
Comparison of risk factor trends
To examine changes in risk factor levels over time by ethnic group, for each continuous risk factor (as the dependent variable), a linear regression model was fitted, with age and test year as the independent variables, and the model was fitted for the two ethnic groups separately (Model 2a). Similarly, for smoking, a logistic regression model was constructed, with the binary outcome of current smoking as the dependent variable, and adjusted for age and test year, and applied separately for each ethnic group (Model 2b). The coefficients for test year from these models denote the mean annual age-adjusted change in the risk factor level (continuous variables), or mean annual change in the logit of the probability of smoking status, specific to each ethnic group.
Whether trends in risk factors for the two ethnic groups studied were different (e.g. converging) was first examined empirically, by observing whether there was an overlap in the 95% confidence intervals of the test year co-efficients for each ethnic group. Secondly, the significance of any differentials in slope over time was assessed with the significance level of the co-efficient for an interaction term ethnic group X test year which was included in the models already containing the age, test year, and ethnic group variables both for continuous (Model 3a) and the smoking variable (Model 3b). Interaction variables were centred to avoid possible co-linearity.
Results
There were 34,122 men and 37,294 women aged 35–60 with a first screening episode during 1989 and 1999. Following the steps described below 33,597 men and 36,890 women were assigned Caucasian and 499 men and 381 women were assigned South-Asian origin.
Attribution of ethnicity
Of all participants, 1.8% were initially identified as of probable South-Asian descent by the Nam Pahcham software. Of all persons initially identified as of probable ethnic origin, 69.2% (or 1.23% of all cases) were judged to truly correspond to South-Asian names. Furthermore 27.3% of individuals initially identified as of ethnic origin (or 0.48% of all cases) were judged as definitely non-South-Asian (Caucasian), 2.1% (or 0.03% of all cases) were judged as Chinese, and 1.4% (or 0.02% of all cases) were judged as of ethnic origin but not certain if South-Asian or other. Subsequent inspection of all surnames not initially identified as of probable ethnic origin by the Nam Pahcham software, identified fewer than 20 (or <0.02% of all cases) additional individuals with a South-Asian (Hindu) surname/origin that were missed out by the software; their ethnicity was re-assigned. Further analysis was restricted to individuals in the Caucasian and South-Asian groups.
Descriptive comparisons
The mean age of Caucasian individuals (both men and women respectively) was greater than of South-Asian individuals (Table 1). Diastolic blood pressure ascertainment was complete (100%), and systolic blood pressure nearly complete for all groups and both sexes. In women, 68.9% of all first screening episodes occurred during the "prevalence round" of the programme (1989–1993), whilst in men the same proportion was 65.1%. The relative percentages were lower in South-Asian individuals. Ascertainment completeness for BMI, smoking status and cholesterol was systematically lower in South-Asian individuals, as also previously reported [11], although the both the absolute and relative magnitude of this difference was small. There were only small differences over time between ethnic groups in ascertainment completeness for any of the above three variables (Additional File 3, Figures 1-2).
Table 1 Main characteristics of participants, by ethnic group
Men Women
Baseline characteristic Caucasian (n = 36,890) South-Asian (n = 381) Caucasian (n = 33,597) South-Asian (n = 499)
Mean Age (years) 45.6 (IQR: 39.0–52.0) 44.2 (IQR: 37.0–50.0) 45.7 (IQR: 39.0–53.0) 43.7 (IQR: 38.0–50.0)
% of first screening episodes curing "prevalence round" 1989–1993 65.3% 53.3% 69.0% 54.6%
Ascertainment completeness (%)
BMI 83.2% 79.4% 80.5% 77.7%
Total Cholesterol 58.8% 52.3% 53.5% 47.2%
Smoking status 83.7% 79.2% 81.0% 76.1%
IQR: Inter-quartile range, BMI: Body Mass Index.
Comparisons of risk factor levels
The 95% confidence intervals of values relating to South-Asian individuals however, due to small numbers, often overlap with the confidence intervals of values for Caucasian individuals relating to the same time period. Table 2 below presents summary information showing the difference in risk factor levels between the two ethnic groups, for the two most distant time periods in the study (1989–1990 and 1997–9), along with the proportion ascertained and the number of individuals contributing data to each time period and for each individual risk factor. The most substantial changes in both absolute and relative terms relate to a decline in total cholesterol (for both sexes and ethnic groups) and current smoking status, the latter only for the Caucasian group.
Table 2 Summary information about risk factor levels between study baseline (1989–90) and end period (1997–9), by ethnic group. Percentage ascertainment and sample size presented, along with absolute and % difference.
Men Women
1989–90 1997–9 Absolute difference % difference 1989–90 1997–9 Absolute difference % difference
Cholesterol Caucasian % ascertained 51.7% 50.5% 47.6% 43.8%
n 4,096 2,591 4,454 2,130
Mean (mmol/l) (95% CI) 6.168 (6.133 – 6.203) 5.628 (5.578 – 5.678) -0.54 -8.8% 6.034 (6.002–6.066) 5.409 (5.358 – 5.460) -0.625 -10.4%
South-Asian % ascertained 40.9% 50.0% 41.8% 45.5%
n 38 55 28 35
Mean (mmol/l) (95% CI) 6.539 (6.209 – 6.869) 5.614 (5.293 – 5.935) -0.925 -14.1% 5.629 (5.181–6.077) 5.217 (4.935 – 5.499) -0.412 -7.3%
SBP Caucasian % ascertained 100% 100% 99.9% 99.9%
n 7,923 5,128 9,352 4,862
Mean (mmHg) (95% CI) 129.9 (129.6–130.2) 131.5 (131.0 – 132.0) 1.6 1.2% 126.3 (126.0 – 126.6) 125.6 (125.1 – 126.1) -0.7 -0.6%
South-Asian % ascertained 100% 100% 100% 100%
n 93 110 67 77
Mean (mmHg) (95% CI) 125.7 (121.8–129.6) 124.8 (121.7–127.9) -0.9 -0.7% 128.9 (123.7 – 134.1) 125.1 (121.4 – 128.8) -3.8 -2.9%
DBP Caucasian % ascertained 100% 100% 100% 100%
n 7,924 5,130 9,357 4,865
Mean (mmHg) (95% CI) 80.4 (80.2 – 80.6) 81.8 (81.5 – 82.1) 1.4 1.7% 77.4 (77.2 – 77.6) 77.9 (77.6 – 78.2) 0.5 0.6%
South-Asian % ascertained 100% 100% 100% 100%
n 93 110 67 77
Mean (mmHg) (95% CI) 78.1 (75.8 – 80.4) 79.3 (77.4 – 81.2) 1.2 1.5% 78.3 (76.0 – 80.6) 77.5 (75.4 – 79.6) -0.8 -1.0%
BMI Caucasian % ascertained 73.9% 88.4% 71.8% 86.9%
n 5,856 4,533 6,722 4,228
Mean (kg/m2) (95% CI) 25.84 (25.75 – 25.93) 26.65 (26.52 – 26.78) 0.81 3.1% 25.06 (24.95 – 25.17) 25.96 (25.77 – 26.15) 0.9 3.6%
South-Asian % ascertained 63.4% 87.3% 65.7% 84.4%
n 59 96 44 84.4
Mean (kg/m2) (95% CI) 24.83 (24.1 – 25.6) 25.94 (25.1 – 26.8) 1.11 4.5% 26.16 (24.9 – 27.40) 26.42 (25.23 – 27.61) 0.26 1.0%
Smoking Caucasian % ascertained 74.6% 88.5% 72.6% 87.0%
n 5,911 4,542 6,793 4,231
Mean (% current smokers) (95% CI) 51.9% (50.5 – 53.2) 43.6% (41.9 – 45.3) -8.3 -16.0% 38.7% (37.5 – 39.9) 34.4% (32.7 – 36.1) -4.3 -11.1%
South-Asian % ascertained 62.4% 86.4% 59.7% 80.5%
n 58 95 40 62
Mean (% current smokers) (95% CI) 25.9% (14.3 – 37.5) 27.4% (16.5 – 38.3) 1.5 5.8% 0% (0.0 – 0.0) 7% (0.0 – 14.0) 7 ...
SBP: Systolic Blood Pressure, DBP: Diastolic Blood Pressure, BMI: Body Mass Index.
In regression analysis, adjusting for age and test year, South Asian men and women throughout the study period appear to have comparatively either lower or similar risk factor levels, with the exception of BMI for women (Additional File 4, Figures 1-10). In particular, South Asian men had significantly lower values of systolic BP [-4.91 mmHg (95% CI: -3.58 to -6.23)] and diastolic BP [-2.87 mmHg (-2.02 to -3.72)], with no significant difference in total cholesterol and BMI (Models 1a-b, Tables 3, 4). South Asian women had significantly lower systolic BP [-1.77 mmHg (-0.21 to -3.33)], diastolic BP [-1.87 mmHg (-0.92 to -2.82)], cholesterol [-0.24 mmol/l (-0.08 to -0.39 mmol/l)]; and higher BMI [+0.78 kg/m2 (+0.25 to +1.30)]. South Asian men and women had significantly age-standardised lower prevalence of current smoking [29.0% (24.1%–33.9%)] and 1.8% (0.1%–3.5% respectively], compared to men and women of Caucasian origin [49.3% (48.7%–49.9%) and 36.6% (36.1%–37.2%) respectively].
Table 3 Mean annual change and overall difference in continuous risk factor levels by ethnic group stratum, adjusted for age and test year.
Men Women
Mean UCI LCI p Mean UCI LCI p
Systolic BP
Model 1a (overall difference between South-Asian and Caucasian group)
Ethnic group difference^ -4.91 -6.23 -3.58 <0.001 -1.77 -3.33 -0.21 0.026
Model 2a (trends by ethnic group)
Caucasian* 0.25 0.19 0.30 <0.001 0.02 -0.04 0.07 0.614
South-Asian* -0.03 -0.46 0.39 0.874 -0.28 -0.79 0.23 0.277
Model 3a (Time trend difference between South-Asian and Caucasian group)
Interaction ethnic group × time** -0.27 -0.71 0.17 0.234 -0.29 -0.82 0.23 0.276
Diastolic BP
Model 1a (overall difference between South-Asian and Caucasian group)
Ethnic group difference^ -2.87 -3.72 -2.02 <0.001 -1.87 -2.82 -0.92 <0.001
Model 2a (trends by ethnic group)
Caucasian* 0.19 0.15 0.22 <0.001 0.08 0.04 0.12 <0.001
South-Asian* 0.18 -0.11 0.46 0.228 -0.13 -0.44 0.17 0.387
Model 3a (Time trend difference between South-Asian and Caucasian group)
Interaction ethnic group × time** -0.22 -0.54 0.11 0.187 -0.22 -0.54 0.11 0.187
Cholesterol
Model 1a (overall difference between South-Asian and Caucasian group)
Ethnic group difference^ -0.04 -0.18 0.10 0.555 -0.24 -0.39 -0.08 0.002
Model 2a (trends by ethnic group)
Caucasian* -0.07 -0.08 -0.06 <0.001 -0.06 -0.07 -0.05 <0.000
South-Asian* -0.08 -0.12 -0.03 0.001 -0.06 -0.10 -0.004 0.033
Model 3a (Time trend difference between South-Asian and Caucasian group)
Interaction ethnic group × time** -0.01 -0.06 0.04 0.763 0.02 -0.04 0.07 0.581
BMI
Model 1a (overall difference between South-Asian and Caucasian group)
Ethnic group difference^ -0.23 -0.59 0.14 0.223 0.78 0.25 1.30 0.004
Model 2a (trends by ethnic group)
Caucasian* 0.10 0.08 0.11 <0.001 0.11 0.09 0.13 <0.001
South-Asian* 0.12 -0.02 0.25 0.083 0.00 -0.16 0.17 0.968
Model 3a (Time trend difference between South-Asian and Caucasian group)
Interaction ethnic group × time** 0.05 -0.07 0.18 0.401 -0.12 -0.29 0.06 0.190
BP: Blood pressure, BMI: Body Mass Index, S-A: South-Asian
^adjusted for age, year, stratified by ethnic group
*adjusted for age, year and ethnic group
**adjusted for age, test year, ethnic group and interaction term "ethnic group × test year"
Table 4 Mean overall difference in probability of current smoking between ethnic groups.
Men Women
OR LCI UCI p OR LCI UCI p
Model 1b (overall difference between South-Asian and Caucasian group)
Ethnic group difference^ 0.48 0.38 0.60 <0.001 0.03 0.01 0.08 <0.001
Model 2b (trends by ethnic group)
Caucasian* 0.96 0.95 0.96 <0.001 0.97 0.96 0.98 <0.001
South-Asian* 1.01 0.93 1.09 0.858 1.42 0.99 2.04 0.059
Model 3b (Time trend difference between South-Asian and Caucasian group)
Interaction ethnic group × time** 1.07 0.99 1.15 0.093 1.49 1.04 2.14 0.032
OR: Odds Ratio, UCI: Upper Confidence Interval, LCI: Lower Confidence Interval
^ adjusted for age and test year
* adjusted for age, stratified by ethnic group
**adjusted for age, test year, ethnic group and interaction term "ethnic group × test year"
Comparisons of risk factor trends
Men of all ethnic groups manifested significant increasing trends in sysstolic BP [+0.23 mmHg/year (95% CI: +0.18 to +0.29)], diastolic BP [+0.18 mmHg/year (+0.15 to +0.22)], and BMI [+0.10 kg/m2/year (+0.08 to +0.11)]; and decreasing trends in total cholesterol [-0.07 mmol/l/year (-0.06 to -0.07)] (Models 2a-b, Tables 3, 4). Women of all groups manifested significant increasing time trends in diastolic BP [+0.08 mmHg/year (+0.04 to +0.11)] and BMI [+0.11 kg/m2/year (+0.09 to +0.13)], and decreasing time trends in total cholesterol (-0.06 mmol/l/year (-0.06 to -0.07)]. For smoking, overall there was a strong negative time trend in the probability of prevalence of current smoking for men [OR: 0.96 (0.95 to 0.96)]; and women [OR: 0.97 (0.96 to 0.97)]. Although current smoking prevalence was increasing in South-Asian individuals of both sexes, these trends were non-significant.
Empirically, compared with Caucasians, men and women of South-Asian origin did not have statistically significant differences in risk factor trends for systolic and diastolic BP, BMI and total cholesterol, as evidenced by overlapping confidence intervals for the test year co-efficient. Furthermore, for all continuous risk factors, the interaction term ethnic group X test year was non-significant, indicating that there is no statistically significant difference in time trend slopes between the groups (i.e. where significant differences between groups existed, they remain fixed during the study period) (Table 3, Model 3a). For smoking, there was a positive interaction between test year and ethnicity in women (p = 0.032), indicating a convergence in smoking prevalence rates between the two groups (Table 4, Model 3b), whilst for men, the p value was (0.093), which may be suggestive of an overall trend.
Discussion
The study included UK individuals of middle age and with no previous diagnosis of hypertension, diabetes and cardiovascular disease. The findings indicate that during the 1990s and in the study setting, individuals of South-Asian origin did not have an adverse risk factor profile compared with the Caucasian ethnic group. In fact for some risk factors the inverse is true (i.e. more favourable risk factor levels among South-Asians). The findings also appear to show a similar pattern of time trends in risk factor levels for the examined five risk factors between Caucasian and South Asian groups. Individuals of South Asian origin are at higher risk of impaired glycose tolerance and diabetes, in itself a strong independent risk factor of cardiovascular disease, but no data on diabetic status were available for analysis. Nevertheless, the findings would contrast with the hypothesis that increased cardiovascular disease burden in UK South-Asians is due to higher level of exposure to conventional cardiovascular risk factors (other than impaired glucose tolerance and diabetes). Other hypotheses, such as greater genetic susceptibility to cardiovascular disease, may be relevant. Further research should also continue to explore the likely importance of novel and as yet unknown risk factors, and the likely relevance of lower risk of competing morbidity.
The lower levels of total cholesterol and smoking prevalence levels observed in the South-Asian group are in general agreement with previous findings from UK cross-sectional surveys [20,21], including the 1999 Health Survey for England (HSE) which focused on ethnic minority groups [4]. In particular, both the 1999 HSE and the present study indicate significantly lower prevalence of current smoking among South-Asian individuals particularly for women (although it should be noted that the 1999 HSE found higher smoking prevalence in one of the South-Asian ethnic subgroups – Bangladeshi men); similar or lower levels of cholesterol (particularly in relation to lower levels in South-Asian women); and higher levels of mean BMI in women. Conflictingly, a number of UK studies, previously expertly reviewed [22], have shown both higher (particularly in London) and lower levels of hypertension in South Asian compared to Caucasian populations -the present study being in broad agreement with the latter. In relation to BP, comparisons of the findings of the present study with the 1999 Health Survey for England area difficult, due to the variable findings in the latter survey, depending on South-Asian particular ethnic subgroup and sex.
The risk factor trends for total cholesterol, BMI and current smoking prevalence observed in this study are in agreement with those observed in UK population-based studies of the similar period (e.g. Health Survey for England (HSE) [3], Health Survey for Scotland [23]) i.e. reflecting downward trends for cholesterol and smoking status, and increasing trends for BMI. However, the lack of any considerable decrease in systolic and diastolic blood pressure for both men and women observed in the present study is in contrast to information from population-based epidemiological surveys [3,23]. This difference raises the question whether the difference can be due to bias, a "real" effect or chance. Selection bias is theoretically possible, as patients with known diagnosis of hypertension and other cardiovascular conditions were excluded from the study, unlike population-based surveys such as the HSE that randomly include both individuals with and without known diagnosis of hypertension (or cardiovascular disease). This hypothesis would have meant that the downward population trends in mean BP observed in HSE surveys of the same period are largely due to strong treatment effects and secular improvement in BP control among known hypertensive participants, and this is judged unlikely. Artefactual explanations are also likely, for example systematic differences in measurement progressively taking place during the study period (e.g. timing allowed to achieve a resting state, body posture, observer technique and training, number of BP readings, calculation of mean values from more than one readings).
A crucial question in relation to the generalisability of the study findings is whether screening participants were representative of the Stockport population of 35–60 year olds who were free of hypertension, diabetes and other cardiovascular disorders. However, if the rigour and accuracy with which individuals were identified and excluded because of known hypertension, diabetes and other cardiovascular conditions was differential between the two ethnic groups, then this potentially could have biased the findings, and this possibility cannot be dismissed with the data that are available.
The study used a computerised package to ascribe ethnicity. Reliance on the same software for identification of ethnic names has been proven not fully reliable before, particularly for South-Asian populations residing outside Northern England [15]. Nevertheless the computerised process of ethnic group assignment was supplemented by visual inspection of both the positive and the negative cases by two independent researchers of diverse ethnic background, which has been shown to improve specificity and specificity [15]. Overall, 1.23% of all cases were assigned South-Asian origin, which corresponds with estimates of the Stockport population of South Asian origin (all ages) of 1.12% in the 1991 and 2.1% in the 2001 Census [13]. In any case, any misclassification errors in the ascertainment of ethnicity would tend to make the two ethnic groups similar, and hence would have made true differences between ethnic groups more difficult to detect is such differences truly existed. Therefore although there is a possibility of undetected differences, those ethnic differences that have been identified in this study are unlikely to be biased by the method used to assign ethnicity.
The fact that Stockport has a relatively small South-Asian community, in combination with the fact that a relatively greater proportion of South-Asian population are younger than 30, is reflected in the small proportion of South Asians included in the study. It is theoretically possible that because Stockport is neither affluent nor deprived, Stockport South Asians are not representative of the wider South Asian communities in the UK.
An important limitation of the study is that the South-Asian ethnic group category contains an important degree of heterogeneity in terms of ethnic group, religion, culture and country of origin. Previous research has showed considerable heterogeneity in relation to exposure to cardiovascular risk factors between UK South-Asian ethnic groups [24]. Nevertheless, aggregation of all South-Asian groups has helped increase the statistical precision of estimates. In the 1991 census, within Stockport South-Asian individuals aged over 30 years of age, the relative proportion of those categorised as of Indian, Pakistani and Bangladeshi origin was 43.4 %, 49.5% and 7.1% respectively (data available from MIMAS, ). In the 2001 census a similar overall breakdown of South-Asian groups was observed (31.3%, 49.3% and 5.8%). It is therefore likely that the relative proportion of South-Asian subgroups in the dataset is a reflection of the proportions of subgroups of South-Asian origin in the community, but no further detailed study of such subgroups was possible. More importantly, it is unlikely that the observed time trends in risk reflect changes in the ethnic sub-group composition of the South-Asian group. Individuals belonging to ethnic subgroups are known to have an overall lower access rate to preventive interventions. We therefore acknowledge that it is possible that screening participants of South-Asian origin (and of some South-Asian subgroups in particular) may not be representative of the overall community. Unfortunately quantifying this potential problem is not possible with the available data.
Conclusion
In a UK district the exposure of individuals of South-Asian origin to conventional cardiovascular risk factors excluding diabetes status was found to be overall either favourable or similar to individuals of Caucasian origin. Recent time trends in risk factor levels in South-Asians appear to mirror those observed in the Caucasian population, with the exception of smoking.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
GL had identified the research subject and defined the research question with support from and RFH. GL has developed the statistical methodology, with supervisory support by PMcE. PL and MH have over a number of years helped run the Stockport Cardiovascular Risk Factor Screening Programme and collect and collate data that enabled the analysis to be carried out. GL wrote the first draft of the manuscript. The work leading to this report has been carried out as part fulfilment for the study for the degree of MD, University of Manchester, for GL. RFH is his supervisor for the named degree and P McElduff his advisor.
Acknowledgements
We would wish to acknowledge the help, advice and practical support of the following individuals: Professor Deborah Baker, Dr Stephen Watkins, Ms Jane Jefferson, Ms Jane Pilkington, Dr Gill Greenhough, Ms Jane Bowdenleigh, Ms Bernadette Ryan-Wooley, Ms Barbara Shallaker, Mr Dan Byrne. Dr Umesh Chauhan, Dr Robert (Mark) Hann UC and MH were responsible for the attribution of ethnic group.
Pre-publication history
The pre-publication history for this paper can be accessed here:
Supplementary Material
Additional File 1
'Supplementary information about the population coverage of the Stockport Cardiovascular Risk Factor Screening Programme 1989–1999, and its representativeness. This file includes information about population coverage (screening Programme uptake) among Stockport residents.
Click here for file
Additional File 2
'Details about risk factor measurements'. Describes details about methods used in risk factor measurement.
Click here for file
Additional File 3
'Risk factor ascertainment completeness (%) by ethnic group and sex, by test year period.' Describes the degree of ascertainment completeness by ethnic group.
Click here for file
Additional File 4
'Age-standardised mean risk factor levels by ethnic group and sex, by test year period.' Self-explanatory.
Click here for file
==== Refs
Deaths by cause, 2002, United Kingdom British Heart Foundation Statistics Website "Heartstats" (Accessed 14.04.2005)
Bhopal R Epidemic of cardiovascular disease in South Asians BMJ 2002 324 625 6 11895809 10.1136/bmj.324.7338.625
Health Survey for England 2003 Risk factors for cardiovascular disease 2 (Accessed 08.10.2005)
Erens B Primatesta P Prior G Eds Health Survey for England 1999 The Health of Ethnic Minority Groups (Accessed 08.10.2005)
Bhopal R Hayes L White M Unwin N Harland J Ayis S Alberti G Ethnic and socio-economic inequalities in coronary heart disease, diabetes and risk factors in Europeans and South Asians J Public Health Med 2002 24 95 105 12141592 10.1093/pubmed/24.2.95
Ebrahim S Smith GD Systematic review of randomised controlled trials of multiple risk factor interventions for preventing coronary heart disease BMJ 1997 314 1666 74 9193292
Wanless D Treasury HM Securing good health for the whole population: Final report – February 2004 (Accessed 14.04.2005)
Department of Health Choosing Health Making healthy choices easier November HMSO, London 2004 (Accessed 14.04.2005)
Edgar MA Schneiden H The economics of mild hypertension programmes Soc Sci Med 1989 28 211 22 2493162 10.1016/0277-9536(89)90264-5
Lyratzopoulos G McElduff P Heller RF Hanily M Lewis PS Mid-term Body Mass Index increase among obese and non-obese individuals in middle life and deprivation status: A cohort Study BMC Public Health 2005 5 32 15811178 10.1186/1471-2458-5-32
Bartys S Baker D Lewis P Middleton E Inequity in recording of risk in a local population-based screening programme for cardiovascular disease Eur J Cardiovasc Prev Rehab 2005 12 63 67 10.1097/00149831-200502000-00010
Office for National Statistics Compendium of clinical indicators, 1998–2000 2001 HMSO, London
National Statistics. Census 2001. Stockport (accessed 14.04.2005)
Nam Pehcham computer software (Version 1.1). Bradford Health Authority, New Mill, Victoria Road, Saltaire, Shipley BD18 3LD. Computer Services, City of Bradford Metropolitan Council (Dept. 13), Britannia House, Bradford D1 1HX
Cummins C Winter H Cheng KK Maric R Silcocks P Varghese C An assessment of the Nam Pehchan computer program for the identification of names of south Asian ethnic origin J Public Health Med 1999 21 401 6 11469361 10.1093/pubmed/21.4.401
Allain CC Poon LS Chan CS Richmond W Fu PC Enzymatic determination of total serum cholesterol Clin Chem 1974 20 470 5 4818200
British Hypertension Society Recommendations on Blood Pressure Measurements BMJ 1987
Armitage P Berry G Statistical methods in medical research 1994 3 Oxford: Blackwell Scientific publications
Anand SS Yusuf S Vuksan V Devanesen S Teo KK Montague PA Kelemen L Yi C Lonn E Gerstein H Hegele RA McQueen M Differences in risk factors, atherosclerosis, and cardiovascular disease between ethnic groups in Canada: the Study of Health Assessment and Risk in Ethnic groups (SHARE) Lancet 2000 356 279 284 11071182 10.1016/S0140-6736(00)02502-2
Cappuccio FP Cook DG Atkinson RW Strazzullo P Prevalence, detection, and management of cardiovascular risk factors in different ethnic groups in south London Heart 1997 78 555 563 9470870
France MW Kwok S McElduff P Seneviratne CJ Ethnic trends in lipid tests in general practice Q J Med 2003 96 929 923
Bhopal RS Agyemang C Is the blood pressure of South Asian adults in the UK higher or lower than that in European white adults? A review of cross-sectional data J Hum Hypertension 2002 16 739 751 10.1038/sj.jhh.1001488
The Scottish Health Survey 1998 The Scottish Executive Department of Health 1 (accessed 14.04.2005)
Bhopal R Unwin N White M Yallop J Walker L Alberti KGM Harland J Patel S Ahmad N Turner C Watson B Kaur D Kulkarni A Laker M Tavridou A Heterogeneity of coronary heart disease risk factors in Indian, Pakistani, Bangladeshi, and European origin populations: cross sectional study BMJ 1999 319 215 20 10417082
|
16313671
|
PMC1316876
|
CC BY
|
2021-01-04 16:28:59
|
no
|
BMC Public Health. 2005 Nov 28; 5:125
|
utf-8
|
BMC Public Health
| 2,005 |
10.1186/1471-2458-5-125
|
oa_comm
|
==== Front
Microb Cell FactMicrobial Cell Factories1475-2859BioMed Central London 1475-2859-4-311628392710.1186/1475-2859-4-31ReviewRole of cultivation media in the development of yeast strains for large scale industrial use Hahn-Hägerdal Bärbel [email protected] Kaisa [email protected] Christer U [email protected] Marie [email protected]örgens Johann [email protected] Zyl Willem H [email protected] Applied Microbiology, LTH/Lund University, P O Box 124, SE-221 00 Lund, Sweden2 Department of Process Engineering, Stellenbosch University, Private Bag X1, Stellenbosch, 7602, South Africa3 Department of Microbiology, Stellenbosch University, Private Bag X1, Stellenbosch2005 10 11 2005 4 31 31 30 9 2005 10 11 2005 Copyright © 2005 Hahn-Hägerdal 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.
The composition of cultivation media in relation to strain development for industrial application is reviewed. Heterologous protein production and pentose utilization by Saccharomyces cerevisiae are used to illustrate the influence of media composition at different stages of strain construction and strain development. The effects of complex, defined and industrial media are compared. Auxotrophic strains and strain stability are discussed. Media for heterologous protein production and for bulk bio-commodity production are summarized.
==== Body
Introduction
The composition of the medium used for cultivation of micro-organisms is directly reflected in their physiological phenotype and their fermentation performance, which in turn affects the results of strain analyses and strain performance in industrial applications. For this reason, the successful development of strains for large scale industrial production of heterologous proteins [1,2] and low-value fuels, chemicals and materials [3,4] merits the composition of cultivation media in various steps of strain development to be reconsidered.
Introducing novel recombinant strains into industrially relevant cultivation media may reveal that the strain has not been properly designed for this environment. For example, it was found that a strain of the lactic acid bacterium Streptococcus thermophilus engineered for enhanced exopolysaccharide production – a trait highly desirable in yoghurt production – failed to express the phenotype in milk without the addition of an extra nitrogen source [5]. Similarly, a genetically modified strain of the yeast Saccharomyces cerevisiae, which had been communicated as the ultimate solution to the fermentation of lignocellulose derived xylose [6,7], was found to require yeast extract, additional hexose sugar and oxygenation to efficiently ferment the xylose fraction in spent sulphite liquor [8]. Furthermore, heterologous protein production in yeast is strongly influenced by the nitrogen-composition of the production medium [9,10]. Thus the final industrial environment must be considered throughout the strain development process to avoid unfounded expectation and – more importantly – to prevent costly investment into premature production facilities.
A cultivation medium is designed to reflect the elemental composition and the biosynthetic capacity of a given microbial cell (see e.g. [11]). While the elemental composition of microbial cells is relatively similar, their biosynthetic capacity varies widely. The yeast S. cerevisiae and the bacterium Escherichia coli have extensive biosynthetic capacity and grow in defined mineral media [12]. In contrast, the biosynthetic capacity of many lactic acid bacteria is limited and they require rich or extensively supplemented medium for efficient growth [13]. Furthermore, economic constraints make the large-scale production of low-cost products reliant on cheap sources of carbon and nitrogen, such as molasses from the sugar industry, corn steep liquor from the starch industry, spent sulphite liquor from the forest products industry and cheese whey from the dairy industry [14,15]. In addition to providing carbon, nitrogen, vitamins and trace elements necessary for cell growth and metabolite production, such industrial media may also contain substances which inhibit growth and metabolite production.
This paper reviews the influence of the composition of cultivation media on the development of novel industrial production strains with the view that it is necessary to consider the final cultivation conditions in every stage of strain development. Primarily two types of recombinant strains of S. cerevisiae are used as examples: strains which produce heterologous proteins and strains with an expanded substrate range to include pentose sugars. Experience from other organisms is included to complement the discussion. Finally, genetic engineering approaches to overcome industrial media constraints are also exemplified.
Metabolic engineering, evolutionary engineering and systems biology in strain development
Traditionally, novel production strains have been developed by mutagenesis [16], breeding [17], and the lately revived concept of evolutionary engineering (Figure 1; [18-20]). Strains with novel traits are now also developed by life science technologies including genetic and metabolic engineering (Figure 2; [21-23]). In recent years these engineering concepts have been expanded in the context of systems biology to also include information and system science technologies [24-26]. In metabolic engineering, cells are iteratively analyzed, designed and synthesized (Figure 2) using molecular tools such as recombinant DNA technology and genomic information, [27,28]. Evolutionary engineering (Figure 1) relies on carefully designed selection protocols, i. e. media and cultivation conditions [18-20] for the development of strains with industrially interesting characteristics. Metabolic and evolutionary engineering technologies may also be combined to generate novel traits [29-33]. The multitude of data generated in the analysis of the genome, transcriptome, proteome and metabolome [25] requires the use of information and system science technologies to translate these data into design strategies for next rounds of metabolic and evolutionary engineering [24]. Several studies have pointed out that the cultivation conditions and media composition used for the analysis of novel engineered strains strongly influence the data generated [34-39]. Since such data form the basis for the design strategy for the following rounds of strain development, it is evident that choice of cultivation media is a fundamental and integral part of strain development.
Media and strain stability
Whereas strain development by recombinant techniques is usually performed in genetically defined laboratory strains harboring markers suitable for selection of transformed cells in chemically defined cultivation media, the typical industrial production strain is genetically undefined and adapted to perform in rather poor, toxic, viscous, and nutrient-limited media. Once desired novel traits have been established in recombinant laboratory strains, the novel strains are either directly transferred to the industrial production environment or – as occurs much more frequently – a potential production strain has to undergo a new round of metabolic engineering procedures. In both cases, the medium in which the novel pathways are developed differs substantially from the medium in which the final production strain is expected to perform.
The genetic stability of strains is an absolute requirement for utilization in industrial processes. Due to the adaptive nature of microorganisms, attention should also be directed towards the stability of any novel traits in recombinant or mutant strains. In industry, rich or undefined media are often used, which may result in unexpected loss of plasmids and even chromosomal modifications. Prolonged cultivation, for example in continuous fermentation set-ups, increases the probability of detrimental genetic instability. Even in mineral medium, loss of plasmids with auxotrophic marker has been reported in prolonged continuous cultures [40]. This was caused by released amino acids from the dying cells, and probably also by spontaneous chromosomal insertion of the marker gene [40].
In S. cerevisiae, both episomal plasmids (YEp; [41]) and integrative plasmids (YIp; [42]) are used as expression vectors for heterologous protein expression and metabolic engineering. The advantage of YIp vectors, despite their low copy numbers, is their robust genetic stability even in unselective medium due to the integration of the vector into the yeast genome [43-51]. The benefit of using YEp plasmids is the high gene copy number of up to 70 copies per cell [52] resulting in high expression levels of the desired proteins, although their high segregational instability often results in plasmid loss especially in rich medium [53,54]. However, the stability of YEp-type vectors can be improved by autoselection systems, such as the fur1 ura3 system [55], where the deletion of FUR1 together with the use of a plasmid containing the URA3 marker results in stable plasmid expression even in continuous culture [56]. Without such autoselection systems it is necessary to use a selective medium to overcome the instability of YEp plasmids, which may pose a limitation to the industrial use of such strains especially with low-cost products.
Rich complex media versus defined media
Generally, microorganisms grow more vigorously in rich media than in mineral media, because rich media contain biosynthetic precursors that can be channeled directly into anabolic pathways, reducing the need to produce biosynthetic precursors and saving metabolic energy. This has a significant effect on growth and production characteristics.
For example, a three-fold increase in production levels of heterologous laccase by recombinant Yarrowia lipolytica was reported when switching from yeast nitrogen base (YNB) to complex medium [10]. When autoselective strains of S. cerevisiae expressing heterologous xylanase or α-L-arabinofuranosidase genes were cultivated in complex YPD medium, 24-fold higher xylanase and up to 70-fold higher levels of α-L-arabinofuranosidase were produced [57,58]. Similarly, production levels of the potent thrombin-specific inhibitor, hirudin, by recombinant S. cerevisiae was improved 20 fold in complex medium [59], demonstrating the substantial impact of medium composition on heterologous protein production.
Aoki et al. [60] elegantly demonstrated single-step purification of recombinant cysteine proteinase (NsCys) from Pichia pastoris by switching medium composition during cultivation. The recombinant P. pastoris was first cultivated in glycerol complex medium to generate biomass in a short time. The cells were harvested and resuspended in minimal medium for induction of NsCys production. The minimal medium faciliated protein secretion and subsequent purification.
Strains of E. coli with altered levels of pyruvate decarboxylase and alcohol dehydrogenase displayed a reduced flux of pyruvate into the native fermentation pathways when cultivated in defined medium [61]. In addition, the flow of carbon skeletons into the 2-ketoglutarate arm of the tricarboxylic acid pathway and biosynthesis was restricted, which dramatically reduced growth yields in defined medium compared with complex medium. The observations demonstrated that inherent limitations in the metabolism of engineered strains can be masked by the presence of complex nutrients in the medium and are often not observed without cultivation in defined medium.
To illustrate the influence of media composition on strain performance, we compared the growth of baker's yeast and two recombinant strains: a laboratory strain of S. cerevisiae, TMB3001 [62] and an industrial strain of S. cerevisiae, TMB3400 [29]. Both recombinant strains have been engineered for xylose utilization with the introduction of the XYL1 [63] and XYL2 [64] genes encoding xylose reductase and xylitol dehydrogenase, respectively, from the yeast Pichia stipitis. In addition the endogenous gene XKS1 [65] encoding xylulokinase has been overexpressed.
We evaluated the influence of four commonly used media on growth and product formation under aerobic and oxygen limited conditions: yeast extract-peptone (YP [66,67]); a defined mineral medium (DM; [68]); yeast nitrogen base (YNB; [12]); and synthetic complete (SC) medium equivalent to supplemented YNB [12] (Table 1). YP is an undefined rich complex medium composed of yeast extract (YE) and peptone. YE is prepared by autolysis of whole yeast cells at around 50°C [66,69-71] and peptone is an acid- or enzymatic hydrolysate of a protein-rich by-product from the food and feed industry [67]. YP contains all components necessary for propagation of yeast cells, including biosynthetic building blocks, and it is frequently used in the initial stages of fermentation when a large inoculum is required. YNB is a chemically defined medium that can be supplemented to satisfy auxotrophic requirements of yeast mutants used in metabolic engineering, then referred to as SC medium. DM medium contains almost all components of YNB medium (Table 1), however, some components are present in higher and even an order of magnitude higher concentration than in YNB medium. The DM medium and variants thereof are commonly used to obtain quantitative physiological data for yeast strains. It has been designed to assure that concentrations of vitamins and trace elements do not exercise growth limitation [68]. Sodium chloride, riboflavin and folic acid were not found to be necessary for growth of S. cerevisiae, whereas cobalt apparently supported growth (Table 1). EDTA seems to be required to dissolve elevated concentrations of trace elements.
YP supported growth of a commercial baker's yeast strain even in the absence of additional carbon source (Figure 3). The maximum specific growth rate under these conditions was 0.29 h-1 with a final OD620 of 3–4 after 24 hours. With additional 20 g/l glucose an OD620 of 22 was reached at a maximum specific growth rate of 0.45 h-1. Sugars present in YE may explain this phenomenon. Yeast accumulates storage carbohydrates such as glycogen and trehalose, the amount of which is strongly dependent on cultivation conditions [72]. During the preparation of YE these compounds are fully or partially hydrolyzed to monomer glucose. YE also contains lactate, which can serve as carbon source in yeast cultivation [9]. Lactate is a consequence of non-sterile cultivation conditions in baker's and brewer's yeast production [73]. In addition to these auxiliary carbon sources, YE also contains a number of other compounds, which strongly influence fermentation performance [9].
The maximum specific growth rate of TMB3001 and TMB3400 in the four different media (Table 2) under aerobic and oxygen limited conditions varied considerably. Highest growth rates were obtained in aerobic YP medium and significantly lower growth rates were observed in the three defined media both for glucose and xylose as carbon source (Table 2). The results support previous observations that complex media components can mask inherent limitations in the metabolism of recombinant strains as demonstrated for E. coli [61] and S. cerevisiae [8].
In all media TMB3400 displayed significantly higher specific growth rate on xylose than TMB3001 confirming previous results [29,32]. Reducing the oxygen supply emphasized this difference. Both strains displayed significantly lower specific growth rate on xylose than on glucose (Table 2) confirming previous observations [29]. Also in this comparison, oxygen limitation increased the difference. DM and YNB medium resulted in almost identical specific growth rates indicating that the level of trace elements and vitamins in YNB were not limiting growth under the presently chosen cultivation conditions. Amino acid supplementation in YNB slightly increased the specific growth rates during xylose utilization but not under glucose utilization (Table 2).
The volumetric xylose uptake mirrored the growth rate in all four media with both strains and under the two levels of oxygenation (Table 2). Figure 4 shows the time course for xylose consumption and product formation for TMB3400 under oxygen limited conditions using YP and YNB medium. Results from only one of the mineral media are displayed since growth and product formation was identical in all three mineral media. Whereas the results emphasize the strong growth promoting influence of the YP medium they also show that the media composition did not influence the distribution of products under the chosen conditions (Figure 4).
Buffers
Strain development may require large numbers of strains to be evaluated for their performance in simple screening set-ups [74,75], where neither oxygen availability nor pH is controlled. When microorganisms grow in defined mineral medium with ammonium as the sole nitrogen source, the medium quickly acidifies due to proton excretion during active transport of nutrients into the cell [76]. Acidification quickly inhibits cell growth and metabolism [77]. Therefore, the media must be buffered around the optimal pH for the microorganism to be investigated. For example in industrial yeast fermentations, it is relevant to maintain pH around 5.5. We compared the influence of 50 mM buffering salts on growth of TMB3001 in YP and in DM media (Figure 5). The suitability of citrate, citrate/phosphate, phosphate and phthalate to buffer the growth medium at pH 5.5 were compared.
YP had an inherent buffering capacity, while pH in non-buffered DM decreased to 2.5 when maximum OD620 was reached. The presence of citrate and citrate/phosphate severely inhibited growth in YP, whereas the inhibition was somewhat less severe in DM. With three carboxyl groups, citrate is a chelating compound and complexes with trace elements in YP. In DM, where the concentration of trace elements has been enhanced (Table 1), the inhibition of citrate was less severe. Phthalate showed the best buffering capacity, however, the price of this buffering compound may limit its use in large amounts. With phosphate buffer, pH of DM medium dropped to around 3 in the late stationary phase, but no growth inhibition was observed. Thus depending on the scale of strain screening either phthalate or phosphate buffer should be used for yeast development work.
Auxotrophic markers: pros and cons
The construction of recombinant strains requires selectable marker genes for efficient detection and selection of transformed cells. For S. cereviciae, mutant and deletion strains having one or several auxotrophic requirements are the most commonly used tools in the development of recombinant strains [78]. The use of auxotrophic mutants relies on the assumption that complementing auxotrophy by plasmid expression makes the strain equivalent to its prototrophic counterpart. However, this is not always the case, as was shown for strains carrying the LEU2 gene on a multicopy plasmid [34].
In addition to the auxotrophic markers used for plasmid retention, uncomplemented auxotrophic mutations often remain present in the transformed yeast strains, requiring the addition of the necessary amino or nucleic acids to the cultivation medium. The use of such auxotrophic strains has recently been critically reviewed [36]. Based on the complications involved in translating experimental data obtained with auxotrophic strains into quantitative physiological data, the author concluded that auxotrophic strains should be avoided unless auxotrophy itself was under investigation. A solution to this problem is genetic complementation of the remaining auxotrophic markers, which is quite simple (see e.g. [32,37]) and recovers the prototrophic genotype.
Uncomplemented auxotrophic mutations can also affect production levels of recombinant proteins [36]. This was recently confirmed when growth and extracellular protein production were compared for an auxotrophic and a prototrophic S. cerevisiae strain expressing the Trichoderma reesei β-1, 4-xylanase XYN2 gene [37]. Only excessive amino acid supplementation allowed the auxotrophic strain to produce the heterologous protein at levels comparable to the prototrophic strain. Other studies have confirmed that excessive auxotrophic markers in transformed S. cerevisiae strains often result in overconsumption of the required metabolite and decreased growth, protein production and genetic stability [9,34,45,79-84]. These studies clearly demonstrated that physiological data obtained with auxotrophic strains have to be evaluated with great caution and should not form the basis for future strain design strategies.
Media requirements/supplements for heterologous protein production
Expression of proteins is an inherent strategy of metabolic engineering, whether it is performed for the production of the protein itself or for redirecting a metabolic pathway. It was early observed that high-level protein expression influenced cell physiology of both prokaryotic and eukaryotic microorganisms (lit. reviewed in [37,38,85]). The most prominent effect was reduced cell growth. The phenomenon was named "metabolic" [53,86] and later "protein" [40,87] burden since it was discriminated from the influence of the catalytic activity of the protein [87]. The magnitude and the cause of the protein burden were estimated by comparing recombinant strains of S. cerevisiae expressing T. reesei β-1, 4-xylanase encoded by the XYN2 gene [85]. Both the introduction of the glycolytic promoter without the structural gene, and the structural gene itself, exercised a metabolic burden on the host. The reduction of the maximum specific growth rates, the biomass yields and the specific glucose consumption rates were much larger than expected from the amount of heterologous protein produced [85]. When the cultivation medium was supplemented with a balanced mixture of preferred amino acids (Ala, Arg, Asn, Glu, Gln and Gly) or succinate, the detrimental metabolic effect could partially be relieved [38]. Amino acids enhanced cell growth and heterologous protein production, which supported the observation that recombinant yeast expressing heterologous proteins experience depletion of amino acids and biosynthetic precursors [88].
The latter observation was the basis for genome-wide transcription analysis of two isogenic strains of S. cerevisiae harboring either a multicopy plasmid with the T. reesei XYN2 gene under control of the S. cerevisiae PGK1 promoter [58] or the plasmid with neither the structural gene nor the promoter. Transcription data (available at [89,90]) are summarized in Table 3. Transcriptional profiles during the expression of the heterologous xylanase strongly resembled severe amino acid limitation resulting in up-regulation of amino acid transport and synthesis, complemented by the induction of the general stress response and respiration, and the repression of ribosomal and glycolytic gene expression. The transcriptional response to heterologous xylanase expression thus closely resembled the stringent stress response, apparently due to amino acid limitation [38,91,92]. Similar stringent stress response has been reported for strains of the bacterium E. coli overproducing heterologous proteins. The E. coli stringent stress response normally involves the repression of ribosome synthesis and the derepression of respiration, amino acid uptake and amino acid biosynthesis due to nutrient limitation [93]. The stringent stress response in S. cerevisiae has previously been associated with nitrogen limitation and a nutritional downshift [94-97]. The fact that amino acid supplementation of the cultivation medium also improved heterologous protein production under control of an oxygen regulated promoter in the yeast Pichia stipitis [39] seems to further support that the induction of a nitrogen starvation response due to heterologous protein expression is general.
The choice of nitrogen source in cultivation media for the production of heterologous proteins is crucial as has been amply illustrated by observations with various complex nitrogen sources for industrial protein production. Inconsistency in complex components such as yeast extract can limit the reproducibility of industrial fermentation performance, resulting in 2–3 fold differences in heterologous protein production levels [9]. For industrial production, the proteins of complex cheese whey can be hydrolysed by proteases to allow for utilization by micro-organisms [98], which has been shown to improve the heterologous protein production compared with mineral medium [99,100]. However, other reports have indicated slower growth and lower production of heterologous proteins in cheese whey compared to mineral medium containing lactose [101].
The presence of nitrogen components in cultivation media may also be important to protect heterologous proteins in the extracellular medium from proteolysis. Extracellular proteolysis of heterologous proteins is affected by nutritional conditions, and may increase due to glucose exhaustion or carbon starvation [45,102,103]. The addition of complex nitrogen sources, such as casamino acids, peptides, amino acids, skim milk or bovine serum albumin has been shown to decrease the degradation of the heterologous proteins by S. cerevisiae and P. pastoris, probably by providing large amounts of protein substrate or reducing the production of extracellular proteases [81,104-110]. Addition of the amino acids arginine and lysine to cultures of S. cerevisiae in defined medium has decreased proteolysis of extracellular recombinant proteins, most likely due to the inhibition of proteolytic enzymes specific for peptide bonds including basic amino acid [111,112]. Buffering the cultivation medium to a pH where protein degradation is minimized can also reduce the breakdown of heterologous proteins [104,107,113-116].
Particular carbon sources may also be required to support heterologous protein production during specific growth phases. The production of recombinant antigens during gluconeogenesis in S. cerevisiae required additional medium components such as lactate and trehalose to ensure sufficient availability of metabolic energy [9]. In addition, too high concentrations of salts may reduce heterologous protein production [117].
For industrial production of high-value heterologous proteins such as bio-pharmaceuticals, the higher costs associated with the use of a defined mineral medium may be justified on the basis of increased reproducibility, productivity, and requirements for regulatory approval [45]. For both defined and complex media, the negative effect of nutrient limitations can be minimized by optimizing the concentrations of the medium components. This should preferably be done by the response surface methodology (see e.g. [118,119]). Such an empirical procedure is required separately for each heterologous protein, for each of which the optimal medium compositions may differ substantially.
Industrial media for bulk bio-commodity production
Media components have a very strong impact on economics of industrial fermentation processes and can account for up to 30% of the total production cost [120,121]. Large scale production of cheap commodities such as fuels, chemicals and materials requires very cheap raw material [14,15]. Such processes use by-products from the agricultural, forestry and chemical industry as carbon and nitrogen sources. Carbon sources include sugar beet and sugar cane molasses, residues from sugar production, spent sulphite liquor (SSL) from the paper pulping industry, and cheese whey from the dairy industry. Spent yeast biomass can be processed to obtain valuable medium supplements (see e.g. [120,121]) and may serve as replacement for the more expensive yeast extract. A frequently used nitrogen source in industrial fermentation processes is corn steep liquor formed during starch production from corn [14,15].
Economic constraints in large-scale industrial processes rigorously limit the utilization of additives and pretreatments prior to cultivation. For the production of cell mass and ethanol using molasses and SSL, salts of ammonia are often the only additive providing the desired pH, a nitrogen source and possibly phosphate. Production of cell mass and ethanol are usually performed under non-sterile conditions [73] at a low pH, which allows yeast to grow while the growth of contaminating microorganisms is inhibited. It is obvious that production strains working efficiently in such media are widely different from laboratory-strains initially used to develop novel metabolic traits. The environmental constraints of industrial fermentation media will be summarized under the following headings: (i) multiple sugar substrates to be converted into the final product; (ii) by-product formation; (iii) nutrient limitation; and (iv) inhibitors.
(i) Multiple substrates
In addition to easily metabolized sugars, industrial substrates may also contain a mixture of more unusual sugars. For example, beet and cane molasses contain galactose, raffinose, and melibiose; starch derived substrates contain maltose; and hemicellulose-derived substrates contain galactose, mannose, xylose and arabinose. For maximum process economy all sugars should be converted to the desired product. The simultaneous presence of multiple sugars in the industrial media may pose limitations such as incomplete substrate utilization and inhibition of sugar utilization pathways. Some sugars such as galactose [122] and mannose are metabolized by S. cerevisiae, whereas the utilization of other sugars such as raffinose [123-125], melibiose [126], xylose [127] and arabinose [30,32] requires that a new metabolic pathway is genetically introduced. In addition, "natural" sugar utilization by S. cerevisiae is governed by carbon catabolite repression [128] and pathway induction [129], such that glucose and mannose are utilized first and other sugars are consumed only when these carbon sources are depleted. To circumvent this phenomenon, recombinant yeasts engineered in key signaling elements of the carbon catabolite repression cascade have been developed [130,131], which resulted in enhanced total sugar consumption rate. However, the feasibility of such engineered strains in industrial environments remains to be demonstrated. Carbon catabolite repression can also be overcome by using fed-batch fermentation regimes [132], which are easily applicable in industrial processes.
(ii) By-products
Glycerol is formed in relatively small amounts during anaerobic ethanolic fermentation [133]. However, considering the scale of ethanol production this unwanted by-product represents product losses in the million € range. Glycerol production during ethanolic fermentation is a consequence of surplus NADH formation in biosynthetic reactions [134-136]. During anaerobic growth of S. cerevisiae in the absence of an active respiratory pathway, biosynthetic NADH can only be oxidized through the reduction of dihydroxyacetone-phosphate to glycerol 3-phosphate, which ultimately leads to glycerol secretion. It was experimentally demonstrated that glycerol secretion is directly linked to amino acid synthesis in S. cerevisiae [135], Glycerol production was reduced when ammonium in the cultivation medium was substituted with amino acids. However, amino acid supplementation of industrial substrates for large-scale ethanol production is presently not considered economically viable even with relatively cheap protein hydrolyzates such as yeast extract and peptone.
Reducing glycerol formation during ethanolic fermentation has also been approached with metabolic engineering strategies. Bacteria harbor transhydrogenase enzymes, which convert NADH into NADPH in response to cellular requirements. Attempts to express these enzymes in S. cerevisiae have met with limited success [137,138]. Instead, endogenous redox reactions of the ammonia and amino acid metabolism in S. cerevisiae have been engineered to create artificial transhydrogenase functions [139,140]. In anaerobic cultivation, ethanol formation increased at the expense of glycerol formation in the engineered strains [138]. The use of such engineered strains in industrial applications remains to be demonstrated.
Anaerobic fermentation of xylose results in xylitol formation as a consequence of the difference in co-factor preference in the xylose reductase and xylitol dehydrogenase reactions, respectively (reviewed in [127]). Xylose reductase can use both NADPH and NADH as cofactor, whereas xylitol dehydrogenase exclusively uses NAD+. Xylitol is secreted and lost for ethanol production as a consequence of intracellular NAD+ depletion. Several strain design strategies have been explored to increase ethanol formation during xylose fermentation including modulations of intracellular co-factor availability [141-143] and expression of a mutated xylose reductase with reduced affinity for NADPH [144]. None of the engineered strains have so far been reported to be exposed to an industrial substrate.
An unexpected, yet fully explainable observation is that industrial cultivation media sometimes decrease unwanted by-product formation. For natural xylose fermenting yeast it was recognized that the reduction of an external electron acceptor such as acetoin provided NAD+ for the xylitol dehydrogenase reaction, which prevented xylitol formation [145-147]. For recombinant S. cerevisiae the same phenomenon was quantified with metabolic flux analysis [148]. The fact that recombinant laboratory strains of S. cerevisiae produced more ethanol in a lignocellose hydrolysate was interpreted in terms of lignocellulose derived components acting as external electron acceptors ([149]; see further discussion below).
Ethanol is an unwanted by-product in baker's yeast production [150,151]. Baker's yeast is industrially produced using a fed-batch regime, where the carbon substrate is fed into the production vessel at a rate which prevents "overflow" metabolism at the level of pyruvate and thus limits ethanol formation [152]. Since S. cerevisiae is also used for large-scale heterologous protein production [153], the unwanted ethanol formation during cell mass production has been approached by genetic engineering. The affinity of a S. cerevisiae hexose transporter has been reduced by gene shuffling [154] as well as by chemostat selection [155]. For both yeast strains, reduced ethanol formation during batch growth at high glucose concentration could be demonstrated as a consequence of the reduced glucose uptake rate. Such engineered strains are advantageous in the field of heterologous protein production, but it is more doubtful whether such strains ever can replace the simple fed-batch fermentation regime in baker's yeast production. For baker's yeast the characteristics of the product inherently include efficient carbon dioxide formation under non-growing oxygen limited conditions and it has not yet been demonstrated that this feature remains un-impaired in strains with reduced glycolytic rate.
(iii) Nutrient limitation
Nutrient limitation and starvation with respect to industrial yeast fermentation has mainly been discussed in relation to the classical processes of beer, wine and baker's yeast production. It may lead to "stuck" fermentation, which translates into large economic losses to the industry. New insight into the molecular mechanisms of nutrient limitation and starvation [156] makes this field of research develop rapidly [156-158]. Whereas media and strain modification in the production of beer, wine and baker's yeast may be limited by legislation and the final organoleptic quality of the product, the large scale fuel, materials and chemical industry is limited by economic constraints. Therefore, it remains to be demonstrated that recent research on nutrient starvation in yeast can be translated into novel fermentation strategies and novel industrial fermentation substrates. In ethanol production for the fuel and chemical markets, one rather relies on natural strain isolates, which have fully adapted to nutritional variation.
(iv) Inhibitors
Fermentation substrates for the production of fuels, materials and chemicals will be produced from lignocellulosic raw materials rather than from starch and sugar. The liquefaction of lignocellulose inherently leads to the formation of weak acids, furan derivatives and phenolic derivatives [74,159]. It is well known that weak acids can act as uncouplers and stimulate ethanol production [68,160]. Similarly, furan and phenolic compounds often appear carbonylated and as such function as external electron acceptors, which in the case of xylose fermentation is beneficial for ethanol formation (see (ii) By-products; [148,149]). However, the beneficial effect of these compounds is strongly concentration dependent and they more often act synergistically to inhibit yeast fermentation [159]. Therefore, the majority of reports on the fermentation of hydrolysates derived from lignocellulosics deal with the inhibitory characteristics of such fermentation substrates.
Lignocellulose hydrolysates have to be detoxified prior to fermentation [161,162]. However, the detoxification adds cost to the process [163] and should therefore be avoided. An elegant solution was demonstrated by applying a fed-batch regime to the fermentation of lignocellulosic hydrolysate [164]. Numerous yeast strains have been evaluated for their ability to ferment non-detoxified lignocellulose hydrolysate [165-169] The results of these studies are not always coherent, which reflects the profound influence of fermentation conditions such as media composition, oxygenation and fermentor set-up. However, it emerges that most laboratory strains used in the early stages of strain development cannot be used for an industrial raw material such as hydrolyzed lignocellulosics, whereas strains isolated from industrial environments generally perform much better. This was illustrated by comparing three different recombinant Saccharomyces strains expressing the XYL1, XYL2 and XKS1 genes for their ability to grow and ferment sugars in non-detoxified northern spruce hydrolysates (NSH). The laboratory strain TMB3001 [62] could not even tolerate 20% NSH, whereas the industrial strain TMB3400 [29] could grow in 33% NSH after supplementation with yeast extract. TMB3006 [170] derived from an acetic acid-tolerant Saccharomyces industrial strain isolated from a continuous spent sulfite liquor fermentation plant [171], could sustain growth in 40% NSH.
Strains tolerant to industrial media can be further improved by evolutionary engineering [18-20]. After exposing TMB3006 to continual selection to a NSH gradient of 40 – 70%, a NSH-adapted strain was obtained that could sustain growth in 70% NSH. This strain could be maintained in steady state at a dilution rate of D = 0.1 h-1 with an ethanol yield of 0.41 g/g on consumed glucose, which illustrated the importance of strain background to achieve the necessary robustness to ferment harsh sugar syrups, such as NSH.
Concluding remarks
The current literature on media composition in different stages of strain development for large scale industrial yeast fermentation has been summarized with a view that media composition is an integral part of strain development. In particular the final industrial environment must be carefully considered throughout the strain development process in order to assure the successful introduction of novel engineered strains into large-scale industrial processes.
Acknowledgements
The Swedish foundation for International Cooperation in Research and Higher Education (STINT), the Swedish Energy Agency (STEM), the Swedish Institute (SI), the International Foundation for Science (IFS) and the National Research Foundation (NRF; South Africa) are gratefully acknowledged for financial support.
Figures and Tables
Figure 1 Principle of Evolutionary Engineering [18-20].
Figure 2 Principe of Metabolic Engineering (adapted from [23]).
Figure 3 Aerobic growth of baker's yeast in YP medium buffered with 50 mM Phthalate. With glucose -◊- (μmax = 0.45), Without glucose -×- (μmax = 0.29)
Figure 4 Xylose utilisation and product formation during oxygen-limited cultivation of TMB 3400 [29] in YP and YNB media. Xylose YP-●-, Xylose YNB -○-, Ethanol YP -▴-, Ethanol YNB -△-, OD620 YP -■-, OD620 YNB -□- Sugars and products were analyzed with HPLC (Bio-Rad, Aminex 87-H column).
Figure 5 a-d. Influence of 50 mM buffer on aerobic growth of TMB 3001 [62] in YP and DM media. a, c: glucose 20 g/l in YP, b, d: glucose 20 g/l in DM. Non buffered -○-, Phthalate -□-, Phosphate -△-, Citrate-Phosphate -×-, Citrate -◊-
Table 1 Media composition
Components (g l-1) YP DM YNB SC (YNB+Suppl)
Yeast extract 10 - - -
Peptone 20 - - -
(NH4)2SO4 - 5 5 5
KH2PO4 - 3 1 1
MgSO4·7H2O - 0.5 0.5 0.5
NaCl - - 0.1 0.1
Vitamins (mg l-1)
Biotin - 0.05 0.002 0.002
D-Pantothenic Acid - 1 0.4 0.4
Nicotinic Acid - 1 0.4 0.4
myo-Inositol - 25 2 2
Thiamine - 1 0.4 0.4
Pyridoxine - 1 0.4 0.4
p-Aminobenzoic Acid - 0.2 0.2 0.2
Riboflavin - - 0.2 0.2
Folic Acid - - 0.002 0.002
Trace Elements (mg l-1)
H3BO3 - 1 0.5 0.5
CuSO4·5H2O - 0.3 0.04 0.04
KI - 0.1 0.1 0.1
Na2MoO4·2H2O - 0.4 0.2 0.2
ZnSO4·7H2O - 4.5 0.4 0.4
FeSO4·7H2O - 3 - -
FeCl3·6H2O - - 0.2 0.2
MnCl2·2H2O - 1 - -
MnSO4·4H2O - - 0.4 0.4
EDTA - 15 - -
CoCl2·6H2O - 0.3 - -
CaCl2·2H2O - 4.5 100 100
Supplements (mg l-1)
Adenine (hemisulfate salt) - - - 40
L-arginine (HCl) - - - 20
L-aspartic acid - - - 100
L- glutamic acid (hemisulfate salt) - - - 100
L-histidine - - - 20
L-leucine - - - 60
L-lysine (mono-HCl) - - - 30
L-methionine - - - 20
L-phenylalanine - - - 50
L-serine - - - 375
L-threonine - - - 200
L-tryptophan - - - 40
L-tyrosine - - - 30
L-valine - - - 150
Uracil - - - 20
Table 2 Growth on glucose and xylose, respectively, and xylose uptake rate for recombinant xylose-utilizing S. cerevisiae strains, TMB 3001 [62] and TMB 3400 [29]. Cells were pre-grown (16 h) in DM medium with 20 g l-1 glucose, harvested by centrifugation and washed three times before inoculation at 0.01 g l-1. Aerobic growth: 100 ml medium in 1000 ml baffled E-flask at 180 rpm. Oxygen-limited growth: 78 ml medium including 1.25 ml l-1 Ergosterol/Tween® 80 in 80 ml flask with rubber septum and cotton-filled needle for CO2 outlet, stirring speed 150 rpm by a 2 cm magnetic stirrer bar. Media containing 20 g/l glucose and 20 g/l xylose, respectively, were buffered to pH 5.5 with 50 mM phthalate and the temperature was set at 30°C.
Aerobic Ox-lim
YP DM YNB SC YP DM YNB SCP
Glucose μmax (h-1) TMB 3001 0.56 0.42 0.42 0.45 0.40 0.32 0.32 0.32
TMB 3400 0.59 0.43 0.45 0.49 0.47 0.36 0.35 0.38
Xylose μmax (h-1) TMB 3001 0.32 0.14 0.14 0.14 0.009 0.006 0.006 0.006
TMB 3400 0.40 0.26 0.27 0.30 0.021 0.012 0.013 0.015
Vol. Xylose rate (g l-1 h-1) TMB 3001 0.10 0.013 0.016 0.018 0.025 0.013 0.013 0.013
TMB 3400 1.13 0.56 0.31 0.39 0.12 0.09 0.09 0.09
Table 3 Summary of overall transcriptional changes in cellular processes of xylanase-producing S. cerevisiae compared to reference strain
Cellular Process Order of magnitude change
μmax (measured rate) ↓↓ [85]
Glucose uptake rate (measured rate) ↓↓ [85]
Glycolysis ↓
Amino acid requirement ↑↑
NH4+ utilisation ↓
Respiration and TCA ↑
Iron uptake ↑
Transcription machinery ↑
Protein synthesis ↓
General stress response ↑↑
==== Refs
Porro D Mattanovich D Recombinant protein production in yeasts Methods Mol Biol 2004 267 241 258 15269428
Macauley-Patrick S Fazenda ML McNeil B Harvey LM Heterologous protein production using the Pichia pastoris expression system Yeast 2005 22 249 270 15704221
Chotani G Dodge T Hsu A Kumar M LaDuca R Trimbur D Weyler W Samford K The commercial production of chemicals using pathway engineering Biochem Biophys Acta 2000 1543 434 455 11150618
Ostergaard S Olsson L Nielsen J Metabolic engineering of Saccharomyces cerevisiae Microbiol Mol Biol Rev 2000 64 34 50 10704473
Svensson M Waak E Svensson U Rådström P Metabolically improved exopolysaccharide production by Streptococcus thermophilus and its influence on the reological properties of fermented milk Appl Environ Microbiol 2005 71 6398 6400 16204566
Ho NW Chen Z Brainard AP Sedlak M Successful design and development of genetically engineered Saccharomyces yeasts for effective cofermentation of glucose and xylose from cellulosic biomass to fuel ethanol Adv Biochem Eng Biotechnol 1999 65 163 92 10533435
Sedlak M Ho NW Characterization of the effectiveness of hexose transporters for transporting xylose during glucose and xylose co-fermentation by a recombinant Saccharomyces yeast Yeast 2004 21 671 84 15197732
Helle SS Murray A Lam J Cameron DR Duff SJ Xylose fermentation by genetically modified Saccharomyces cerevisiae 259ST in spent sulfite liquor Bioresour Technol 2004 92 163 71 14693449
Zhang J Reddy J Buckland B Greasham R Toward consistent and productive complex media for industrial fermentations: Studies on yeast extract for a recombinant yeast fermentation process Biotechnol Bioeng 2003 82 640 652 12673763
Madzak C Otterbein L Chamkha M Moukha S Asther M Gaillardin C Beckerich J-M Heterologous production of a laccase from the basidomycete Pycnoporus cinnabarinus in the dimorphic yeast Yarrowia lipolytica FEMS Yeast Res 2005 5 635 646 15780663
Prescott LM Harley JP Klein DA Microbiology 2002 5 New York: McGraw-Hill Higher Education
BD Diagnostics
van Niel EWJ Hahn-Hägerdal B Nutrient requirements of Lactococci in defined growth media Appl Microbiol Biotechnol 1999 52 617 627
Miller T Churchill BW Demian AL, Solomon NA Substrates for large-scale fermentations Manual of industrial microbiology and biotechnology 1986 Washington DC: ASM Press 122 136
Dahod SK Demain AL, Davies JE Raw material selection and medium development for industrial fermentation processes Manual of industrial microbiology and biotechnology 1999 2 Washington DC: ASM Press 213 220
Lawrence CW Classical Mutagenesis Techniques, Methods in Enzymology Guide to Yeast Genetics and Molecular and Cell biology 2002 350 189 199
Walker GM Yeast physiology and biotechnology 1998 John Wiley and Sons
Francis JC Hansche PE Directed evolution of metabolic pathways in microbial populations. I. Modification of the acid phosphatase pH optimum in S. cerevisiae Genetics 1972 70 59 73 4552227
Dykhuizen DE Hartl DL Selection in chemostats Microbiol Rev 1983 47 150 168 6308409
Sauer U Evolutionary engineering of industrially important microbial phenotypes Adv Biochem Eng Biotechnol 2001 73 129 170 11816810
Bailey JE Towards a science of metabolic engineering Science 1991 252 1668 91 2047876
Stephanopoulos G Aristidou A Nielsen J Metabolic engineering principles and methodologies 1998 Academic Press
Nielsen J Metabolic Engineering Appl Microbiol Biotechnol 2001 55 263 283 11341306
Kitano H Systems biology: a brief overview Science 2002 295 1662 1664 11872829
Lee SY Lee DY Kim TY Systems biotechnology for strain improvement Trends Biotechnol 2005 23 349 35 15923052
The yeast systems biology network
Arber W Linn S DNA modification and restriction Annu Rev Biochem 1969 38 467 500 4897066
Goffeau A The yeast genome directory Nature 1997 387 5 6
Wahlbom CF van Zyl WH Jönsson LJ Hahn-Hägerdal B Cordero Otero RR Generation of the improved recombinant xylose-utilizing Saccharomyces cerevisiae TMB 3400 by random mutagenesis and physiological comparison with Pichia stipitis CBS 6054 FEMS Yeast Res 2003 3 319 326 12689639
Becker J Boles E A modified Saccaharomyces cerevisiae strain that consumes L-arabinose and produces ethanol Appl Envirom Microbiol 2003 69 4144 4150
Sonderegger M Sauer U Evolutionary engineering of Saccharomyces cerevisiae for anaerobic growth on xylose Appl Environ Microbiol 2003 69 1990 1998 12676674
Karhumaa K Hahn-Hägerdal B Gorwa-Grauslund MF Investigation of limiting metabolic steps in the utilization of xylose by recombinant Saccharomyces cerevisiae using metabolic engineering Yeast 2005 22 359 368 15806613
Çakar ZP Seker UO Tamerler C Sonderegger M Sauer U Evolutionary engineering of multiple-stress resistant Saccharomyces cerevisiae FEMS Yeast Res 2005 5 569 578 15780656
Çakar ZP Sauer U Bailey J Metabolic engineering of yeast: the perils of auxotrophic hosts Biotechnol Lett 1999 21 611 616
Piper MD Daran-Lapujade P Bro C Regenberg B Knudsen S Nielsen J Pronk JT Reproducibility of oligonucleotide transcriptome analyses. An interlaboratory comparison of using chemostat cultures of Saccharomyces cerevisiae J Biol Chem 2002 277 37001 37008 12121991
Pronk J Auxotrophic yeast strains in fundamental and applied research Appl Environ Microbiol 2002 68 2095 2100 11976076
Görgens JF Planas J van Zyl WH Knoetze JH Hahn-Hägerdal B Comparison of three expression systems for heterologous xylanase production by Saccharomyces cerevisiae in defined medium Yeast 2004 21 1205 1217 15515128
Görgens JF van Zyl WH Knoetze JH Hahn-Hägerdal B Amino acid supplementation improves heterologous protein production by Saccharomyces cerevisiae in defined medium Appl Microbiol Biotechnol 2005 67 684 691 15630584
Görgens JF Passoth V van Zyl WH Knoetze JH Hahn-Hägerdal B Amino acid supplementation, controlled oxygen limitation and sequential double induction improves heterologous xylanase production by Pichia stipitis FEMS Yeast Res 2005 5 677 683 15780668
Meinander B Hahn-Hägerdal B Fed-batch xylitol production with two recombinant Saccharomyces cerevisiae strains expressing XYL 1 at different levels, using glucose as co-substrate: a comparison of production parameters and strain stability Biotechnol Bioeng 1997 54 391 399
Rose AB Broach JR Propagation and expression of cloned genes in yeast: 2-microns circle-based vectors Methods Enzymol 1990 185 234 79 2199781
Myers AM Tzagoloff A Kinney DM Lusty CJ Yeast shuttle and integrative vectors with multiple cloning sites suitable for construction of lacZ fusions Gene 1986 45 299 310 3026915
Shuster JR Moyer DL Lee H Dennis A Smith B Merryweather JP Yeast mutants conferring resistance to toxic effects of cloned human insulin-like growth factor I Gene 1989 83 47 55 2687116
Weber JM Ponti CG Kaeppeli O Reiser J Factors affecting homologous overexpression of the Saccharomyces cerevisiae lanosterol 14 α-demethylase gene Yeast 1992 8 519 533 1523885
Mendoza-Vega O Sabatie J Brown SW Industrial production of heterologous proteins by fed-batch cultures of the yeast Saccharomyces cerevisiae FEMS Microbiol Rev 1994 15 369 410 7848660
Lopes TS Wijs IJ Steenhauer SI Verbakel J Planta RJ Factors affecting the mitotic stability of high-copy-number integration into ribosomal DNA of S. cerevisiae Yeast 1996 12 467 477 8740420
Gellissen G Hollenberg CP Application of yeasts in gene expression studies: a comparison of Saccharomyces cerevisiae, Hansenula polymorpha and Kluyveromyces lactis – a review Gene 1997 190 87 97 9185853
Bae CS Yang DS Chang KR Seong BL Lee J Enhanced secretion of human granulocyte colony stimulating factor directed by a novel hybrid fusion peptide from recombinant Saccharomyces cerevisiae at high cell concentration Biotechnol Bioeng 1998 57 600 609 10099239
Harashima S Murooka Y, Imanaka T Heterologous protein production by yeast host-vector systems Recombinant microbes for industrial and agricultural applications 1998 New York: Marcel Dekker 137 158
Ljubijankic G Storici F Glisin V Bruschi CV Synthesis and secretion of Providencia rettgeri and Escherichia coli heterodimeric penicillin amidases in Saccharomyces cerevisiae Gene 1999 228 225 232 10072775
Park EH Shin YM Lim YY Kwon TH Kim DH Yang MS Expression of glucose oxidase by using recombinant yeast J Biotechnol 2000 81 35 44 10936658
Ugolini S Tosato V Bruschi CV Selective fitness of four episomal shuttle-vectors carrying HIS3, LEU2, TRP1, and URA3 selectable markers in Saccharomyces cerevisiae Plasmid 2002 47 94 107 11982331
Janes M Meyhack B Zimmermann W Hinnen A The influence of GAP promoter variants on hirudin production, average plasmid copy number and cell growth in S. cerevisiae Curr Genet 1990 18 97 103 2225146
Nacken V Achstetter T Degryse E Probing the limits of expression levels by varying promoter strength and plasmid copy number in Saccharomyces cerevisiae Gene 1996 175 253 260 8917107
Loison G Nguyen-Juilleret M Alouani S Marquet M Plasmid-transformed URA3 FUR1 double-mutants of S. cerevisiae: An autoselection system applicable to the production of foreign proteins Bio/Technol 1986 4 433 437
Marquet M Alouani S Haas ML Loison G Brown SW Double mutants of Saccharomyces cerevisiae harbour stable plasmids: stable expression of a eukaryotic gene and the influence of host physiology during continuous culture J Biotechnol 1987 6 135 145
La Grange DC Pretorius IS van Zyl WH Expression of a Trichoderma reesei α-xylanase gene (XYN2) in Saccharomyces cerevisiae Appl Environ Microbiol 1996 62 1036 1044 8975597
Choi ES Sohn JH Rhee SK Optimization of the expression system using galactose-inducible promoter for the production of anticoagulant hirudin in Saccharomyces cerevisiae Appl Microbiol Biotechnol 1994 42 587 594 7765734
Crous JM Pretorius IS Van Zyl WH Cloning and expression of the α-L-arabinofuranosidase gene (ABF2) of Aspergillus niger in Saccharomyces cerevisiae Appl Microbiol Biotechnol 1996 46 256 260 8933843
Aoki H Nazmul Ahsan M Watabe S Heterologous expression in Pichia pastoris and single-step purification of a cysteine proteinase from northern shrimp Protein Expr Purif 2003 31 213 221 14550639
Underwood SA Buszko ML Shanmugam KT Ingram LO Flux through citrate synthase limits the growth of ethanologenic Escherichia coli KO11 during xylose fermentation Appl Environ Microbiol 2002 68 1071 1081 11872452
Eliasson A Wahlbom CF Christensson C Hahn-Hägerdal B Anaerobic xylose fermentation by recombinant Saccharomyces cerevisiae harbouring XYL1, XYL2 and XKS1 in mineral media chemostat cultivations Appl Environ Microbiol 2000 66 3381 3386 10919795
Hallborn J Walfridsson M Airaksinen U Ojamo II Hahn-Hägerdal B Penttilä M Keränen S Xylitol production by recombinant Saccharomyces cerevisiae Bio/technology 1991 9 1090 1099 1367625
Kötter P Ciriacy M Xylose fermentation by Saccharomyces cerevisiae Appl Microbiol Biotechnol 1993 38 776 783
Deng XX Ho NW Xylulokinase activity in various yeasts including Saccharomyces cerevisiae containing the cloned xylulokinase gene Appl Biochem Biotechnol 1990 24–25 193 199
Merck
Oxoid
Verduyn C Postma E Scheffers WA van Dijken JP Effect of benzoic acid on metabolic fluxes in yeast: a continuous-culture study on the regulation of respiration and alcoholic fermentation Yeast 1992 8 501 517 1523884
Chao KC McCarthy EF McConaghey GA Yeast autolysis process US Patent #4 1980 218 481
Akin C Murphy RM Methods for accelerating autolysis of yeast US Patent #4 1981 285 976
Kelly M Godfrey T, Reichelt M Yeast extract Industrial enzymology, the application of of enzymes in industry 1983 New York: Nature Press 457 464
Panek AD Rose AH, Harrison JS Storage carbohydrates The yeasts 1991 4 2 San Diego: Academic Press 655 678
Rose AH Vijayalakshmi G Rose AH, Harrison JS Baker's yeast Yeast technology 1993 5 San Diego: Academic Press 357 397
Larsson S Palmqvist E Hahn-Hägerdal B Tengborg C Stenberg K Zacchi G Nilvebrant N-O The generation of fermentation inhibitors during dilute acid hydrolysis of softwood Enzyme Microb Technol 1999 24 151 159
Warringer J Blomberg A Automated screening in environmental arrays allows analysis of quantitative phenotypic profiles in Saccharomyces cerevisiae Yeast 2003 20 53 67 12489126
Hensing MCM Bangma KA Raamsdonk LM De Hulster E Van Dijken JP Pronk JT Effects of cultivation conditions on the production of heterologous α-galactosidase by Kluyveromyces lactis Appl Microbiol Biotechnol 1995 43 58 64
Yu S Jeppsson H Hahn-Hägerdal B Xylulose fermentation by Saccharomyces cerevisiae and xylose-fermenting yeast strains Appl Microbiol Biotechnol 1995 44 314 320 8597536
Entian KD Kötter P Brown JPA, Tuite MF Yeast mutant and plasmid collections Yeast gene analysis 1998 Academic Press 431 449
Beretta I Sanglard D Käppeli O Fiechter A Optimisation of Candida tropicalis cytochrome P450alk gene expression in Saccharomyces cerevisiae with continuous cultures Appl Microbiol Biotechnol 1991 36 48 60 1367775
Korogodin VI Korogodina VL Fajszi C Chepurnoy AI Mikhova-Tsenova N Simonyan NV On the dependence of spontaneous mutation rates on the functional state of genes Yeast 1991 7 105 117 2063623
Kozlov DG Prahl N Efremov BD Peters L Wambut R Karpychev IV Eldarov MA Benevolensky SV Host cell properties and external pH affect proinsulin production by Saccharomyces yeast Yeast 1995 11 713 724 7668041
VanDusen WJ Fu J Bailey J Burke CJ Herber WK George HA Adenine quantitation in yeast extracts and fermentation media and its relationship to protein expression and cell growth in adenine auxotrophs of Saccharomyces cerevisiae Biotechnol Prog 1997 13 1 7 9041705
Chopra R Sharma VM Ganesan K Elevated growth of Saccharomyces cerevisiae ATH1 null mutants on glucose is an artifact of nonmatching auxotrophies of mutant and reference strains Appl Environ Microbiol 1999 65 2267 2268 10224035
Shiba Y Ono C Fukui F Watanabe I Serizawa N Gomi K Yoshikawa H High-level secretory production of phospholipase A1 by Saccharomyces cerevisiae and Aspergillus oryzae Biosci Biotechnol Biochem 2001 65 94 101 11272851
Görgens JF van Zyl WH Knoetze JH Hahn-Hägerdal B The metabolic burden of the PGK1 and ADH2 promoter systems for heterologous xylanase production by Saccharomyces cerevisiae in defined medium Biotechnol Bioeng 2001 73 238 245 11257606
Bentley WE Mirjalili N Andersen DC Davis RH Kompala DS Plasmid-encoded protein: The principle factor in the "metabolic burden" associated with recombinant bacteria Biotechnol Bioeng 1990 35 668 681
Snoep JL Yomano LP Westerhoff HV Ingram LO Protein burden in Zymomonas mobilis : negative flux and growth control due to overproduction of glycolytic enzymes Microbiology 1995 141 2329 2337
Jin S Ye K Shimizu K Metabolic flux distributions in recombinant Saccharomyces cerevisiae during foreign protein production J Biotechnol 1997 54 161 174 9208486
University of Stellenbosch – Chemical Engineering
Lund University – Applied Microbiology
Moehle CM Hinnebusch AG Association of RAP1 binding sites with the stringent control of ribosomal protein gene transcription in Saccharomyces cerevisiae Mol Cell Biol 1991 11 2723 2735 2017175
Mattanovich D Gasser B Hohenblum H Sauer M Stress in recombinant protein producing yeasts J Biotechnol 2004 113 121 135 15380652
Chatterji D Kumar Ojha A Revisiting the stringent response, ppGpp and starvation signaling Curr Opin Microbiol 2001 4 160 165 11282471
Backhus LE DeRisi J Brown PO Bisson LF Functional genomic analysis of a commercial wine strain of Saccharomyces cerevisiae under differing nitrogen conditions FEMS Yeast Res 2001 1 111 125 12702356
Yang R Wek SA Wek RC Glucose limitation induces GCN4 translation by activation of Gcn2 protein kinase Mol Cell Biol 2000 20 2706 2717 10733573
Natarajan K Meyer MR Jackson BM Slade D Roberts C Hinnebusch AG Marton MJ Transcriptional profiling shows that Gcn4p is a master regulator of gene expression during amino acid starvation in yeast Mol Cell Biol 2001 21 4347 4368 11390663
Perez-Ortin JE Garcia-Martinez J Alberola TM DNA chips for yeast biotechnology. The case of wine yeasts J Biotechnol 2002 98 227 241 12141989
Rubio-Texeira M Arevalo-Rodriguez M Lequerica JL Polaina J Lactose utilization by Saccharomyces cerevisiae strains expressing Kluyveromyces lactis LAC genes J Biotechnol 2000 84 97 106
Maullu C Lampis G Basile T Ingianni A Rossolini GM Pompei R Production of lysozyme-enriched biomass from cheese industry by-products J Appl Microbiol 1999 86 182 186 10063615
Maullu C Lampis G Desogus A Ingianni A Rossolini GM Pompei R High-level production of heterologous protein engineered yeasts grown in cottage cheese whey Appl Environ Microbiol 1999 65 2745 2747 10347071
Domingues L Teixera JA Penttilä M Lima N Construction of a flocculant Saccharomyces cerevisiae strain secreting high levels of Aspergillus niger β-galactosidase Appl Microbiol Biotechnol 2002 58 645 650 11956748
Vad R Moe E Saga K Kvinnsland AM Oyen TB High-level production of human parathyroid hormone (hPTH) by induced expression in Saccharomyces cerevisiae Protein Expr Purif 1998 13 396 402 9693065
Gimenez JA Monkovic DD Dekleva ML Identification and monitoring of protease activity in recombinant Saccharomyces cerevisiae Biotechnol Bioeng 2000 67 245 251 10592523
Nomura N Yamada H Matsubara N Horinouchi S Beppu T High level secretion by Saccharomyces cerevisiae of human apolipoprotein E as a fusion to Rhizomucor rennin Biosci Biotechnol Biochem 1995 59 382 387 7766173
Aho S Arffman A Korhola M Saccharomyces cerevisiae mutants selected for increased production of Trichoderma reesei cellulases Appl Microbiol Biotechnol 1996 46 36 45 8987533
Sreekrishna K Brankamp RG Kropp KE Blankenship DT Tsay J-T Smith PL Wierschke JD Subramaniam A Birkenberger LA Strategies for optimal synthesis and secretion of heterologous proteins in methylotropic yeast Pichia pastoris Gene 1997 190 55 62 9185849
Shiba Y Fukui F Ichikawa K Serizawa N Yoshikawa H Process development for high-level secretory production of carboxypeptidase Y by Saccharomyces cerevisiae Appl Microbiol Biotechnol 1998 50 34 41 9720198
Boze H Celine L Patrick C Fabien R Christine V Yves C Guy M High-level secretory production of recombinant porcine follicle-stimulating hormone by Pichia pastoris Process Biochemistry 2001 36 907 913
Goodrick JC Xu M Finnegan R Schilling BM Schiavi S Hoppe H Wan NC High-level expression and stabilization of recombinant human chitinase produced in a continuous constitutive Pichia pastoris expression system Biotechnol Bioeng 2001 74 492 497 11494216
Werten MWT Van den Bosch TJ Wind RD Mooibroek H De Wolf FA High-yield secretion of recombinant gelatins by Pichia pastoris Yeast 1999 15 1087 1096 10455232
Chung BH Park KS Simple approach to reducing proteolysis during secretory production of human parathyroid hormone in Saccharomyces cerevisiae Biotechnol Bioeng 1998 57 245 249 10099200
Kang HA Choi ES Hong WK Kim JY Ko SM Sohn JH Rhee SK Proteolytic stability of recombinant human serum albumin secreted in the yeast Saccharomyces cerevisiae Appl Microbiol Biotechnol 2000 53 575 582 10855718
Juge N Williamson G Puigserver A Cummings NJ Connerton IF Faulds CB High-level production of recombinant Aspergillus niger cinnamoyl esterase (FAEA) in the methylotrophic yeast Pichia pastoris FEMS Yeast Res 2001 1 127 132 12702357
Cox H Mead D Sudbery P Eland RM Mannazzu Evans L Constitutive expression of recombinant proteins in the methylotrophic yeast Hansenula polymorpha using the PMA1 promoter Yeast 2000 16 1191 1203 10992283
Kobayashi K Kuwae S Ohya T Ohda T Ohyama M Ohi H Tomomitsu K Ohmura T High-level expression of recombinant human serum albumin from the methylotrophic yeast Pichia pastoris with minimal protease production and activation Journal of Bioscience and Bioengineering 2000 89 55 61 16232698
Rosenfeld SA Use of Pichia pastoris for expression of recombinant protein Methods in Enzymology 1999 306 154 169 10432453
Becerra M Prado SD Siso MIG Cerdan ME New secretory strategies for Kluyveromyces lactis β-galactosidase Protein Engineering 2001 14 379 386 11438761
Ha MY Kim SW Lee YW Kim MJ Kim SJ Kinetic analysis of growth and lactic acid production in pH-controlled batch cultures of Lactobacillus casei KH-1 using yeast extract/corn steep liquor/glucose medium J Biosci Bioeng 2003 96 134 140 16233498
Bustos G Moldes AB Alonso JL Vazquez M Optimization of D-lactic acid production by Lactobacillus coryniformis using response surface methodology Food Microbiology 2004 21 143 148
Rivas B Moldes AB Dominguez JM Parajo JC Development of culture medium containing spent yeast cells of Debaromyces hansenii and corn steep liquor for lactic acid production with Lactobacillus rhamnosus Int J Food Microbiol 2004 97 93 98 15527923
Lee K A media design program for lactic acid production coupled with extraction by electrodialysis Bioresource Technology 2005 96 1505 1510 15939279
Ostergaard S Olsson L Nielsen J In vivo dynamics of galactose metabolism in Saccharomyces cerevisiae: metabolic fluxes and metabolite levels Biotechnol Bioeng 2001 73 412 25 11320512
Vincent SF Bell PJ Bissinger P Nevalainen KM Comparison of melibiose utilizing baker's yeast strains produced by genetic engineering and classical breeding Lett Appl Microbiol 1999 28 148 152 10063644
Rönnow B Olsson L Nielsen J Mikkelsen JD Derepression of galactose metabolism in melibiase producing bakers' and distillers' yeast J Biotechnol 1999 72 213 228 12680392
Yoon SH Mukerjea R Robyt JF Specificity of yeast (Saccharomyces cerevisiae) in removing carbohydrates by fermentation Carbohydr Res 2003 338 1127 1132 12706980
Ostergaard S Roca C Ronnow B Nielsen J Olsson L Physiological studies in aerobic batch cultivations of Saccharomyces cerevisiae strains harboring the MEL1 gene Biotechnol Bioeng 2000 68 252 259 10745193
Hahn-Hägerdal B Wahlbom CF Gardonyi M van Zyl WH Cordero Otero RR Jönsson L Metabolic engineering of Saccharomyces cerevisiae for xylose fermentation Adv Biochem Eng/Biotechnol 2001 73 53 84
Gancedo JM Yeast carbon catabolite repression Microbiol Mol Biol Rev 1998 62 334 361 9618445
Ostergaard S Olsson L Johnston M Nielsen J Increasing galactose consumption by Saccharomyces cerevisiae through metabolic engineering of the GAL gene regulatory network Nat Biotechnol 2000 18 1283 1286 11101808
Klein CJ Olsson L Ronnow B Mikkelsen JD Nielsen J Alleviation of glucose repression of maltose metabolism by MIG1 disruption in Saccharomyces cerevisiae Appl Environ Microbiol 1996 62 4441 4449 8953715
Roca C Haack MB Olsson L Engineering of carbon catabolite repression in recombinant xylose fermenting Saccharomyces cerevisiae Appl Microbiol Biotechnol 2004 63 578 583 12925863
Belem MA Lee BH Production of bioingredients from Kluyveromyces marxianus grown on whey: an alternative Crit Rev Food Sci Nutr 1998 38 565 598 9813735
Oura E Reaction products of yeast fermentation Process Biochem 1977 12 19 21
van Dijken JP Scheffers WA Redox balances in the metabolism of sugars by yeast FEMS Microbiol Rev 1986 32 199 224
Albers E Larsson C Lidén G Niklasson C Gustafsson L Influence of the nitrogen source on Saccharomyces cerevisiae anaerobic growth and product formation Appl Environ Microbiol 1996 62 3187 3195 8795209
Wang ZX Zhuge J Fang H Prior BA Glycerol production by microbial fermentation: a review Biotechnol Adv 2001 19 201 223 14538083
Anderlund M Nissen TL Nielsen J Villadsen J Rydström J Hahn-Hägerdal B Kielland-Brandt MC Expression of the E. coli pntA and pntB genes encoding nicotinamide nucleotide transhydrogenase in Saccharomyces cerevisiae and its effect on product formation during anaerobic glucose fermentation Appl Environ Microbiol 1999 65 2333 2340 10347010
Nissen TL Anderlund M Nielsen J Villadsen J Kielland-Brandt MC Expression of a cytoplasmic transhydrogenase in Saccharomyces cerevisiae results in formation of 2-oxoglutarate due to depletion of the NADPH pool Yeast 2001 18 19 32 11124698
Nissen TL Kielland-Brandt MC Nielsen J Villadsen J Optimization of ethanol production in Saccharomyces cerevisiae by metabolic engineering of the ammonium assimilation Metab Eng 2000 2 69 77 10935936
Moreira dos Santos M Thygesen G Kotter P Olsson L Nielsen J Aerobic physiology of redox-engineered Saccharomyces cerevisiae strains modified in the ammonium assimilation for increased NADPH availability FEMS Yeast Res 2003 4 59 68 14554197
Roca C Nielsen J Olsson L Metabolic engineering of ammonium assimilation in xylose-fermenting Saccharomyces cerevisiae improves ethanol production Appl Environ Microbiol 2003 69 4732 4736 12902265
Jeppsson M Johansson B Ruhdal-Jensen P Hahn-Hägerdal B Gorwa-Grauslund MF The level of glucose 6-phosphate dehydrogenase activity strongly influences xylose fermentation and inhibitor sensitivity in recombinant Saccharomyces cerevisiae strains Yeast 2003 20 1263 1272 14618564
Verho R Londesborough J Penttila M Richard P Engineering redox cofactor regeneration for improved pentose fermentation in Saccharomyces cerevisiae Appl Environ Microbiol 2003 69 5892 5897 14532041
Jeppsson M Bengtsson O Franke K Lee H Hahn-Hägerdal B Gorwa-Grauslund MF The expression of a Pichia stipitis xylose reductase mutant with higher KM for NADPH increases ethanol production from xylose in recombinant Saccharomyces cerevisiae Biotechnol Bioeng
Alexander NJ Acetone stimulation of ethanol production from D-xylose by Pachysolen tannophilus Appl Microbiol Biotechnol 1986 25 203 207
Bruinenberg P de Bot P van Dijken J Scheffers A The role of redox balances in the anaerobic fermentation of xylose by yeasts Eur J Appl Microbiol Biotechnol 1983 18 287 292
Ligthelm ME Prior BA du Preez JC Effect of hydrogen acceptors on D-xylose fermentation by anaerobic culture of immobilized Pachysolen tannophilus cells Biotechnol Bioeng 1989 32 839 844
Wahlbom CF Hahn-Hagerdal B Furfural, 5-hydroxymethyl furfural, and acetoin act as external electron acceptors during anaerobic fermentation of xylose in recombinant Saccharomyces cerevisiae Biotechnol Bioeng 2002 78 172 178 11870608
Johansson B Christensson C Hobley T Hahn-Hägerdal B Xylulokinase overexpression in two strains of Saccharomyces cerevisiae also expressing xylose reductase and xylitol dehydrogenase and its effect on fermentation of xylose and lignocellulose hydrolysate Appl Environ Microbiol 2001 67 4249 4255 11526030
von Meyenburg HK Energetics of the budding cycle of Saccharomyces cerevisiae during glucose-limted aerobic growth Arch Microbiol 1969 66 289 303
Petrick M Käppeli O Fiechter A An expanded concept for glucose effect in the yeast Saccharomyces uvarum: Involvement of short- and long-term regulation J Gen Microbiol 1983 129 43 49
Beudecher RF van Dam HW van der Plaat JB Vellega K Verachtert H, de Mot R Developments in baker's yeast production Yeast Biotechnology and Biocatalysis 1990 New York: Marcel Dekker 103 146
Byrne LJ O'Callaghan KJ Tuite MF Heterologous gene expression in yeast Methods Mol Biol 2005 308 51 64 16082025
Otterstedt K Larsson C Bill RM Stahlberg A Boles E Hohmann S Gustafsson L Switching the mode of metabolism in the yeast Saccharomyces cerevisiae EMBO Rep 2004 5 532 537 15071495
Jansen ML Diderich JA Mashego M Hassane A de Winde JH Daran-Lapujade P Pronk JT Prolonged selection in aerobic, glucose-limited chemostat cultures of Saccharomyces cerevisiae causes a partial loss of glycolytic capacity Microbiology 2005 151 1657 1669 15870473
Cooper TG Transmitting the signal of excess nitrogen in Saccharomyces cerevisiae from the Tor proteins to the GATA factors: connecting the dots FEMS Microbiol Rev 2002 26 223 238 12165425
Tate JJ Rai R Cooper TG Methionine sulfoximine treatment and carbon starvation elicit Snf1-independent phosphorylation of the transcription activator Gln3 in Saccharomyces cerevisiae J Biol Chem 2005 280 27195 27204 15911613
Thomsson E Svensson M Larsson C Rapamycin pre-treatment preserves viability, ATP level and catabolic capacity during carbon starvation of Saccharomyces cerevisiae Yeast 2005 22 615 623 16034823
Palmqvist E Hahn-Hägerdal B Fermentation of lignocellulosic hydrolysates. II:Inhibitors and mechanisms of inhibition Bioresource Technol 2000 74 25 33
Pampulha ME Loureiro-Dias MC Combined effect of acetic acid, pH and ethanol on intracellular pH of fermenting yeast Appl Microbiol Biotechnol 1989 31 547 550
Larsson S Reimann A Nilvebrant N-O Jönsson LJ Comparison of different methods for the detoxification of lignocellulose hydrolysates of spruce Appl Biochem Biotechnol 1999 77–79 91 103
Palmqvist E Hahn-Hägerdal B Fermentation of lignocellulosic hydrolysates. I:Inhibition and detoxification Bioresource Technol 2000 74 17 24
Von Sivers M Zacchi G Olsson L Hahn-Hägerdal B Cost analysis of ethanol production from willow using recombinant Escherichia coli Biotechnol Progr 1994 10 555 560
Taherzadeh M Niklasson C Lidén G Conversion of dilute-acid hydrolysates of spruce and birch to ethanol by fed-batch fermentation Biores Technol 1999 69 59 66
Martin C Jönsson LJ Comparison of the resistance of industrial and laboratory strains of Saccharomyces and Zygosaccharomyces to lignocellulose-derived fermentation inhibitors Enzyme Microbial Technol 2003 32 386 395
Brandberg T Franzen CJ Gustafsson L The fermentation performance of nine strains of Saccharomyces cerevisiae in batch and fed-batch cultures in dilute-acid wood hydrolysate J Biosci Bioeng 2004 98 122 125 16233676
Garay-Arroyo A Covarrubias AA Clark I Nino I Gosset G Martinez A Response to different environmental stress conditions of industrial and laboratory Saccharomyces cerevisiae strains Appl Microbiol Biotechnol 2004 63 734 741 12910327
Hahn-Hägerdal B Pamment N Microbial pentose metabolism Appl Biochem Biotechnol 2004 113–116 1207 1209
Sonderegger M Jeppsson M Larsson C Gorwa-Grauslund MF Boles E Olsson L Spencer-Martins I Hahn-Hägerdal B Sauer U Fermentation performance of engineered and evolved xylose-fermenting Saccharomyces cerevisiae strains Biotechnol Bioeng 2004 87 90 98 15211492
Johansson B Metabolic engineering of the pentose phosphate pathway of xylose fermenting Saccharomyces cerevisiae PhD thesis 2001 Lund University, Applied Microbiology
Lindén T Peetre J Hahn-Hägerdal B Isolation and characterization of acetic acid-tolerant galactose-fermenting strains of Saccharomyces cerevisiae from a spent sulfite liquor fermentation plant Appl Environ Microbiol 1992 58 1661 1669 1622236
|
16283927
|
PMC1316877
|
CC BY
|
2021-01-04 16:05:47
|
no
|
Microb Cell Fact. 2005 Nov 10; 4:31
|
utf-8
|
Microb Cell Fact
| 2,005 |
10.1186/1475-2859-4-31
|
oa_comm
|
==== Front
BMC GenomicsBMC Genomics1471-2164BioMed Central London 1471-2164-6-1591628198310.1186/1471-2164-6-159Research ArticlePreferential attachment in the evolution of metabolic networks Light Sara [email protected] Per [email protected] Arne [email protected] Stockholm Bioinformatics Center, Department of Biochemistry and Biophyhsics, Albanova University Center, Stockholm University, Stockholm SE-10691, Sweden2005 10 11 2005 6 159 159 7 7 2005 10 11 2005 Copyright © 2005 Light et al; licensee BioMed Central Ltd.2005Light 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
Many biological networks show some characteristics of scale-free networks. Scale-free networks can evolve through preferential attachment where new nodes are preferentially attached to well connected nodes. In networks which have evolved through preferential attachment older nodes should have a higher average connectivity than younger nodes. Here we have investigated preferential attachment in the context of metabolic networks.
Results
The connectivities of the enzymes in the metabolic network of Escherichia coli were determined and representatives for these enzymes were located in 11 eukaryotes, 17 archaea and 46 bacteria. E. coli enzymes which have representatives in eukaryotes have a higher average connectivity while enzymes which are represented only in the prokaryotes, and especially the enzymes only present in βγ-proteobacteria, have lower connectivities than expected by chance. Interestingly, the enzymes which have been proposed as candidates for horizontal gene transfer have a higher average connectivity than the other enzymes. Furthermore, It was found that new edges are added to the highly connected enzymes at a faster rate than to enzymes with low connectivities which is consistent with preferential attachment.
Conclusion
Here, we have found indications of preferential attachment in the metabolic network of E. coli. A possible biological explanation for preferential attachment growth of metabolic networks is that novel enzymes created through gene duplication maintain some of the compounds involved in the original reaction, throughout its future evolution. In addition, we found that enzymes which are candidates for horizontal gene transfer have a higher average connectivity than other enzymes. This indicates that while new enzymes are attached preferentially to highly connected enzymes, these highly connected enzymes have sometimes been introduced into the E. coli genome by horizontal gene transfer. We speculate that E. coli has adjusted its metabolic network to a changing environment by replacing the relatively central enzymes for better adapted orthologs from other prokaryotic species.
==== Body
Background
Recent studies indicate that metabolic networks evolve at the local level through patchwork evolution and retrograde evolution [1-3]. Patchwork evolution, which is likely to be more important, occurs when an enzyme evolves from a broad spectrum enzyme to an enzyme with a highly specialized activity [4]. Retrograde evolution is a process where the depletion of a substrate from the environment leads to the evolution of an enzyme which can accept a new substrate and catalyze the production of the depleted substance [5].
Metabolic networks and other complex networks such as the film actor collaboration network, the world wide web, protein domain networks and protein-protein interaction networks are small-world networks with some properties which are consistent with scale-free networks [6-8]. The small-worldness of the metabolic network of E. coli has recently been contested for an alternative network representation where carbon atomic traces in metabolic reactions were used [9]. A small-world network is characterized by 1) short path lengths between any two nodes in the network and 2) a high clustering coefficient, which means that the neighbors of a certain node of the network are often connected to each other thereby forming clusters. A scale-free network, in this context, has a power-law connectivity (degree) distribution, i.e. there are many nodes which have very low connectivities and a handful of nodes with much higher connectivities (hubs), see Figure 1a. Scale-free networks are robust networks in the sense that they often remain intact when a large fraction of randomly chosen nodes is eliminated from the network [10]. However, if a small fraction of the hubs of the network is eliminated the network is likely to become fragmented into several components. It has been suggested that the scale-free character of biological networks has evolved through natural selection for the advantage of robustness and error-tolerance that the scale-free network topology confers to the organism [6]. A study by Gleiss et al showed that chemical reaction networks, which clearly have not been subjected to natural selection, also show scale-free characteristics thereby showing that scale-free networks can arise without natural selection and may be a general feature of chemical reaction networks [11].
Figure 1 Properties of scale-free networks and network representation. a) The figure shows a network where most nodes have very low connectivities (k = 1) but two nodes have connectivities which are far higher than the connectivity of most nodes in the network (k = 12). Scale-free networks, among other networks, have this general property. b) The upper part of the figure shows a common network representation where the substrates and products of the reactions represent the nodes in the network and the enzymes represent the edges. Our network representation is shown in the lower part of the figure where the enzymes represent the nodes in the network and the substrates and products represent the edges (reaction graph).
Networks with scale-free properties have been shown to evolve when two simple rules are applied: 1) The network grows by the addition of new nodes. 2) Preferential attachment: New nodes are more likely to become connected to well connected nodes in the network [12]. While preferential attachment is often at the root of scale-freeness, a network with an power-law degree distribution might be produced through other mechanisms. Preferential attachment in the context of genetic networks may take place partly through gene duplication [13,14]. In agreement with preferential attachment Eisenberg and Levanon [15] showed that the proteins which have homologs in all 3 domains of life, which are likely to be of ancient origin, have higher connectivities in the protein-protein interaction network of S. cerevisiae. In contrast, Kunin et al [16] recently showed that the most highly connected proteins date to after the evolution of primordial eukaryotes but before the radiation of eukaryotes to Plants, Metazoa and Protista. Here, we investigate the evidence for preferential attachment and the role of horizontal gene transfer in the metabolic network evolution of E. coli.
Results
Connectivity and phylogenetic group
If preferential attachment is an important mechanism in the evolution of metabolic networks older enzymes should have a higher average connectivity (k) than younger enzymes. In order to investigate this prediction we extracted the enzymes and the reactions in E. coli from the EcoCyc [17] and KEGG [18] databases. The network representation of the metabolic network of E. coli was constructed using EcoCyc, see methods. The nodes in our graph represent the enzymes (complete EC numbers) catalyzing the reactions and the edges represent one or more compounds involved in the reactions. There is an edge from enzyme El to enzyme E2 if El catalyzes a reaction where compound A is produced and then E2 uses A as substrate. There can be at most one edge in each direction between the nodes in the graph. The connectivity of a node is defined as the number of edges connecting the node to other nodes in the network.
Enzymes (complete EC numbers) were collected from KEGG orthology [18] and were found in 163 different organisms (11 eukaryotes, 17 archaea and 135 bacteria). Among these the E. coli enzymes with representatives in 74 organisms (11 eukaryotes, 17 archaea and 46 bacteria) of reasonably well understood phylogenies were extracted for further studies, see Figure 2. The enzymes were divided into five age groups, see Table 1. The enzymes in group 1 are likely to be among the oldest since they have representatives in eukaryotes, archaea and bacteria. Group 2 contains the enzymes with representatives in eukaryotes and bacteria but not in archaea while group 3 contains the enzymes with representatives in archaea and bacteria but not in eukaryotes. Group 4 contains E. coli enzymes without representatives in eukaryotes or archaea but with representatives in bacteria other than γβ-proteobacteria and group 5 contains enzymes with representatives in γβ-proteobacteria only. Enzymes belonging to group 1–3 are probably ancient enzymes since they exist in at least two domains of life. Enzymes which are found in group 5 are only found in bacteria which are comparatively close relatives of E. coli, in the γβ-proteobacteria group, see Figure 2, which indicates that they are relatively recent additions to the metabolic repertory of E. coli. It is possible that some of the enzymes in groups 1–3 could have evolved relatively recently and subsequently been horizontally transferred to the other domains of life but the evidence of horizontal gene transfer between organisms belonging to different domains of life is not abundant. Horizontal gene transfer between bacterial species is believed to be more common [19] and therefore there may be enzymes in group 4 which have been transferred from γβ-proteobacteria. Gene loss, which is estimated to be three times more common than horizontal gene transfer in prokaryotes [20], is probably a more important source of error in this study since genes which serve an important function in bacteria but not in archaea or eukaryotes may have been lost in the archaeal and eukaryotic lineages and as a consequence groups 4 and 5 may not exclusively contain relatively recently evolved enzymes.
Figure 2 Schematic representation of the phylogenetic tree for most of the organisms used in this analysis. The tree was redrawn from Gough et al [35] using Drawgram from the Phylip package . Eukaryotes: Homo sapiens, Caenorhabditis elegans, Drosophila melanogaster, Saccharomyces cerevisiae, Arabidopsis thaliana, archaea: Aeropyrum pernix, Sulfolobus solfataricus, Thermoplasma volcanium, Thermoplasma acidophilum, Methanopyrus kandleri, Methanococcus jannaschii, Methanobacterium thermoautotrophicum, Archaeoglobus fulgidus, Pyrococcus abyssi, Pyrococcus horikoshii, Halobacterium sp. and bacteria: Campylobacter jejuni; Thermatogales: Thermotoga maritima; Parasitic proteobacteria: Rickettsia conorii, Rickettsia prowazekii, Buchnera aphidicola; Chlamydiae: Chlamydophila pneumoniae, Chlamydia trachomatis; Spirochetes: Borrelia burgdorferi, Treponema pallidum; Mycoplasmas: Mycoplasma genitalium, Mycoplasma pneumoniae, Ureaplasma urealyticum, Mycoplasma pulmonis; Bacillus/Clostridium-group: Caulobacter crescentus, Staphylococcus aureus, Bacillus halodurans, Bacillus subtilis, Lactococcus lactis, Streptococcus pneumoniae, Streptococcus pyogenes, Clostridium acetobutylicum; Cyanobacteria: Nostoc sp., Synechocystis sp.; Thermus/Deinococcus-group: Deinococcus radiodurans; Actinobacteria: Mycobacterium tuberculosis CDC1551, Mycobacterium tuberculosis H37Rv, Mycobacterium leprae; free-living α-proteobacteria: Agrobacterium tumefaciens, Mesorhizobium loti, Sinorhizobium meliloti; ε-proteobacteria: Helicobacter pylori-J99, Helicobacter pylori-26695; Aquificales: Aquifex aeolicus and βγ-proteobacteria: Escherichia coli:0157:H7, Escherichia coli:0157:H7:EDL933, Escherichia coli:k-12, Salmonella typhimurium, Yersinia pestis, Vibrio cholerae, Neisseria meningitidis 72491, Neisseria meningitidis MC58, Xylella fastidiosa, Pseudomonas aeruginosa, Ralstonia solanacearum, Pasteurella multocida, Haemophilus influenzae. Furthermore, 6 additional eukaryotes (Schizosaccharomyces pombe, Plasmodium falciparum, Encephalitozoon cuniculi, Mus musculus, Rattus norvegicus, Danio rerio) and 6 additional archaea (Methanosarcina mazei, Methanosarcina acetivorans, Nanoarchaeum equitans, Pyrobaculum aerophilum, Pyrococcus furiosus, Sulfolobus tokodaii) were used in the analysis.
Table 1 Description of the phylogenetic groups 1–5 and the number of E. coli enzymes in each group. For instance, an E. coli enzyme which has at least one representative in one or more eukaryotes but not in archaea is a group 2 enzyme. The fourth column contains the number of enzymes which are proposed examples of horizontal gene transfer. The phylogenetic classification is based on the phylogenetic tree in Figure 2.
GROUP ORGANISMS NO. ENZYMES NO. HGT ENZYMES
1 E. coli, eukaryotes and archaea 262 45
2 E. coli, eukaryotes but not archaea 71 14
3 E. coli and archaea 50 8
4 E. coli and bacteria other than βγ-proteobacteria 75 14
5 βγ-proteobacteria 28 4
The average connectivities for the E. coli enzymes with representatives in the five groups were calculated, see Figure 3a. The enzymes which are represented in all 3 domains of life (group 1) have the highest average connectivity together with the enzymes which occur in bacteria and eukaryotes (group 2) while the enzymes that occur only in γβ-proteobacteria (group 5) have the lowest connectivities. The average connectivity was 40% higher for group 1 than for group 5 enzymes. In order to estimate the significance of the results 100 000 randomized networks were generated through shuffling the group numbers while preserving the network topology. The Z-score, see methods, of the enzymes that occur in eukaryotes is substantially higher than the Z-score for the enzymes which only occur in archaea and bacteria, see Figure 3b.
Figure 3 The connectivity of enzymes belonging to different phylogenetic groups. a) Average connectivity for enzymes in phylogenetic groups 1–5 in the metabolic network where the 15 most promiscuous compounds have been removed. b) The average Z-score (between 15–20 compounds removed from the network) is plotted for the phylogenetic groups 1–5.
In a similar manner 147 other organisms were investigated. We collected the metabolic reactions for these organisms from KEGG. It should be noted that these metabolic networks have not been investigated to the extent of the E. coli metabolic network. Therefore, the connectivities of the enzymes and the results are probably not of the same reliability. We found that the larger prokaryotic genomes often have a particularly strong correlation between connectivity and domain presence while the correlation in smaller prokaryotes and eukaryotes is weaker, see Figure 4. Many of the smaller prokaryotes are obligate intracellular parasites or symbionts. The genomes of obligate parasites and symbionts have been metabolically reduced and many metabolic functions, such as the amino acid metabolism, are frequently provided by the host. Since enzymes involved in amino acid metabolism are often ancient proteins with high connectivities, see Figure 5a and 5b, the absence of genes coding for these enzymes in the genomes of the obligate symbionts/parasites can account for the lack of correlation between connectivity and domain presence in smaller genomes.
Figure 4 Average Z-score and number of protein coding genes. The average Z-score (between 15–20 compounds removed from the network) for enzymes which occur in 1 (black), 2 (red) and 3 (green) domains of life is plotted against the number of protein coding genes contained in the genome.
Figure 5 The phylogenetic distribution and connectivity of enzymes in different functional classes. a) Function and phylogenetic distribution. The E. coli enzymes were classified into 7 functional classes (amino acid metabolism, energy metabolism, lipid metabolism, nucleotide metabolism, sugar metabolism, other biosynthesis and other degradation) and divided into enzymes which are represented in 1, 2 or 3 domains of life. 100 000 randomized networks were generated for comparison and Z-score calculation. b) Connectivity and function. The enzymes were divided into functional classes and the Z-scores for the connectivities of each functional class were calculated for the network where the 15 most promiscuous compounds have been removed.
In conclusion we found that E. coli enzymes which have representatives in all domains of life, and in eukaryotes but not archaea, have a higher average connectivity in the metabolic network of E. coli than the presumably younger enzymes which only have representatives in γβ-proteobacteria. This finding lends support for one of the predictions of the mechanism of preferential attachment.
Connectivity and horizontal gene transfer
It has been suggested that the scale-free properties of biological networks may arise, at least partially, as a result of preferential attachment of new nodes to highly connected nodes through gene duplication [13,14]. Preferential attachment by gene duplication may take place according to the following scenario; Initially, the duplicated gene has exactly the same function and position in the network as the template gene. Since many genes are connected to the hub of the network, the duplicated gene is by chance likely to be connected to the hub of the network. Subsequently, the duplicate gene may evolve towards another functionality but it could retain some of its original function. For instance, a multi-domain protein could loose one of its domains through deletion but retain the other domains and possibly part of its original functionality. In such a scenario the older proteins are more likely to be highly connected than the younger proteins.
An alternative scenario is preferential attachment by horizontal gene transfer (HGT); A new, or alternative, enzyme is introduced through HGT. The new enzyme is more likely to be retained in the metabolic repertory if it confers a new or improved function at a central, rather than peripheral, position of the metabolism – such as if it is connected to a highly connected enzyme or if it is itself highly connected. This is a consideration which may be particularly important in the metabolism of bacteria since some bacteria are prone to delete dispensable genes from their genomes [21]. Arguably, connectivity is a measure which indicates the centrality and importance of an enzyme in which case horizontally transferred genes should frequently be highly connected or be connected to highly connected enzymes. According to this scenario, horizontally transferred enzymes would be preferentially attached to highly connected enzymes and/or be preferentially replacing highly connected enzymes.
Although the extent of the evolutionary impact of HGT is still under debate [22-24], it is generally accepted as an important evolutionary process in microbial species [19]. Roughly 18% of the protein coding genes in E. coli are likely to have been introduced into the E. coli genome by HGT since the species diverged from the Salmonella lineage according to an analysis by Lawrence and Ochman where base composition and codon usage patterns were used to identify the horizontally transferred sequences [25]. Using this data set we found that 85 of the 486 E. coli enzymes used in this study are likely to be examples of HGT (HGT enzymes). The average connectivity for the HGT enzymes in the metabolic network of E. coli is 42.2 while the other enzymes (non-HGT enzymes) have an average connectivity of 33.1. The most striking difference between the connectivity distributions of the HGT enzymes and the non-HGT enzymes is that only 15% of the HGT enzymes have connectivities between 0–9 and 37% of the HGT enzymes have higher connectivities than 60 while 28% of the non-HGT enzymes have connectivities of 0–9 and 26% have connectivities higher than 60, see Figure 6. These results indicate that horizontally transferred enzymes in the metabolic network of E. coli are either introduced into the repertory of the organism as a comparatively high connectivity enzyme or acquires many connections during the evolution of the network.
Figure 6 Connectivity of HGT and non-HGT enzymes. The proportion of the number of enzymes in each connectivity group is plotted against the binned connectivities for the enzymes which are candidates for horizontal gene transfer (white bars) and enzymes which are not candidates for HGT (black bars). The results are shown for the network where the 15 most promiscuous compounds had been removed.
We classified the HGT enzymes into the five phylogenetic groups and determined the average connectivities for each group, see Figure 7 and Table 1. Only HGT enzymes belonging to group 1 and 2 have higher connectivities than the non-HGT enzymes belonging to the same group. From Figure 7a it is clear that the high average connectivities of group 1 and 2 enzymes, which was seen in Figure 3a, is partly but not solely due to HGT enzymes. It should be noted that the horizontal transfers of enzymes which are specific for βγ-proteobacteria only (group 5) may be underestimated since HGT events between closely related species are hard to detect [24].
Figure 7 Horizontal gene transfer, phylogenetic group and connectivity. a) Average connectivities for enzymes in phylogenetic groups 1–5 in the metabolic network where the 15 most promiscuous compounds have been removed. The white bars represent the enzymes which have probably not been transferred to E. coli through horizontal gene transfer (HGT) and the black bars represent the enzymes which are likely to be examples of HGT. The numbers above the bars signify the number of enzymes in each group. b) The connectivity distribution for enzymes which are the result of horizontal transfer (red dots) and enzymes which are not (black circles).
Most horizontally transferred genes go through the process of amelioration, the adjustment of the transferred sequence to the base composition and codon usage of the resident genome. Therefore, most detectable HGTs have taken place relatively recently in the history of E. coli [25]. Consequently, we can conclude that while it is true that the highly connected enzymes in the metabolic network of E. coli are often old in the sense that they are enzymes with representatives in eukaryotes, and which therefore probably originated in the last common ancestor of eukaryotes and bacteria, they are also overrepresented among the enzymes which have been introduced recently into the E. coli genome through HGT. These findings suggest that horizontally transferred genes are introduced and retained preferentially at central positions of the metabolism of E. coli.
The connectivity of essential enzymes and isozymes
Jeong et al [26] showed that the highly connected proteins in the protein-protein interaction network of S. cerevisiae are more likely to be indispensable to the organism than less well connected proteins. We wished to study if there was a similar correlation in the metabolic network of E. coli. We used the essentiality classification from the study of Gerdes et al [27] of E. coli under aerobic growth in nutrition rich medium. We calculated the mean connectivity for the essential and the dispensable enzymes respectively and found that the essential enzymes do not show a higher connectivity than expected (, for networks where 15 compounds have been removed). It is possible that the relatively small size of the metabolic networks compared to the protein-protein interaction network is the reason a similar correlation could not be found in the metabolic network of E. coli.
The hubs are the most important nodes for the integrity of the network. If a fraction of the hubs are removed the network is likely to become fragmented into smaller components. Since these enzymes are very important for the robustness of the network it might be suspected that the EC numbers with the highest connectivities could have more than one representatives in the genome, i.e. that there are two or more isozymes representing these highly connected nodes. Isozymes in multicellular organisms are often active in different tissues while isozymes in single cellular organisms frequently have different substrate specificities or are activated in different environments (such as aerobic or anaerobic environments). We here designated a pair of enzymes as isozymes if they catalyze the same reaction but are coded for by different genes, which are not part of the same enzyme complex.
We used Expasy [28], SGD [29] and EcoCyc [17] to determine which enzymes in the metabolic networks of E. coli and S. cerevisiae occur as isozymes. We found 77 EC numbers that were associated with isozymes in E. coli and 97 EC numbers that were associated with isozymes in S. cerevisiae, see additional files. The mean connectivities for the isozymes and the non-isozymes were determined and the result was compared to randomized networks. We found that the isozymes do not have a noticeably higher mean connectivity than non-isozymes (, for networks where 15 compounds have been removed). The result may indicate that isozymes are not necessarily crucial for the integrity of the metabolic network. In accordance with our result it has recently been shown that the isozymes of S. cerevisiae are not overrepresented among essential enzymes [30].
Connectivity and function
Kunin et al showed that the functional classes in the protein-protein network of S. cere-visiae display distinctly different connectivity levels [16]. In a similar manner we investigated whether enzymes belonging to different functional groups are characterized by distinct connectivities.
We classified the enzymes into 7 functional classes according to EcoCyc [17]; lipid metabolism, nucleotide metabolism, amino acid metabolism, sugar metabolism, energy metabolism, other biosynthesis and other degradation and calculated the mean connectivities for the different functional classes, see Figure 5b and Table 2. The mean connectivities for the enzymes involved in nucleotide, amino acid, other degradation and energy metabolism are higher than expected. The amino acid metabolism and nucleotide metabolism enzymes are clearly over represented in 3 domains of life while enzymes involved in energy metabolism are slightly more common in 3 domains of life than expected by chance. Many of the pathways involved in energy metabolism, such as the citric acid cycle and glycolysis, are believed to be very old. However, there are substantial variations in the energy metabolism between different species and domains of life [31]. Therefore, the observation that energy metabolism enzymes are not overrepresented in three domains of life is not surprising.
Table 2 The number of E. coli enzymes belonging to 7 functional EcoCyc classes.
FUNCTION NO. ENZYMES
Sugar metabolism 76
Amino acid metabolism 93
Lipid metabolism 17
Nucleotide metabolism 46
Energy metabolism 46
Other biosynthesis 140
Other degradation 53
Contrastingly, enzymes involved in lipid and sugar metabolism are on average half as well connected as the enzymes involved in nucleotide, amino acid and energy metabolism. The group of enzymes involved in lipid metabolism is less than half the size of the second smallest functional group and due to its small size the Z-score for this functional group is less reliable than for the other functional groups. The sugar metabolism enzymes are clearly over represented among the enzymes that occur in bacteria only, see Figure 5a, which was anticipated since there are many bacterial specific enzymes involved in sugar transportation [32].
Network growth through preferential attachment
According to the mechanism of preferential attachment new enzymes in the network should be preferentially attached to already well connected nodes. We do not have access to the last universal common ancestor (LUCA) that existed before the 3 domains of life evolved. However, a rough representation of the metabolic network of that organism was created by extracting the enzymes that occur in all domains of life. The connectivities of the enzymes in that derived network were determined. We then calculated the number of enzyme nodes that have been added to an enzyme by subtracting the connectivity of the enzyme in the current E. coli network by the connectivity of the enzyme in the ancient network. We found that the enzymes which have higher connectivities in the ancient network gain new connections at a higher rate than the enzymes with lower connectivities, see Figure 8. The correlation between connectivity in the ancient network and the connectivity increase appears to be linear (r = 0.87) following the equation f(x) = 2.6 + 0.41x, where x is the connectivity in the ancient network. We can therefore conclude that the addition of new nodes to the metabolic network of E. coli occurs in a manner which is consistent with preferential attachment.
Figure 8 Connectivity increase and connectivity in the ancient network. The number of edges gained from the ancient network to the current E. coli metabolic network is plotted against the connectivity in the ancient network.
Discussion
We have investigated two predictions generated from the mechanism of preferential attachment in the evolution of the metabolic network of E. coli. First, if preferential attachment is of any significance in the evolution of the metabolic network of E. coli, the older enzymes in the network should have a higher average connectivity. We have found that E. coli enzymes which are represented in three domains of life, and in eukaryotes but not archaea, have a higher average connectivity than expected by chance. Second, another prediction generated from the hypothesis of network evolution through preferential attachment is that highly connected nodes should gain new edges at a faster rate than nodes with low connectivities. To investigate this prediction we extracted the enzymes with representatives in 3 domains of life and determined the network representing LUCA's metabolic network. In accordance with the mechanism of preferential attachment we found a positive linear correlation between connectivity in the ancient network and number of connections gained through evolution.
Further, we found that the E. coli enzymes which are believed to have undergone horizontal gene transfer (HGT enzymes) have a higher average connectivity than other enzymes (non-HGT enzymes). This is especially true for the HGT enzymes with representatives in eukaryotes, which is the most highly connected group of E. coli enzymes. This result suggests that the highly connected enzymes are often old in the sense that they are likely to have originated in LUCA and been part of the bacterial metabolic repertory for a long time. However, these ancient enzymes are sometimes relatively recent additions to the metabolic network of E. coli. It is possible that bacteria such as E. coli are adjusting their metabolic networks to a changing environment by replacing the relatively central enzymes, with high connectivities, for better adapted orthologs from other prokaryotic species.
Conclusion
It is well known that many novel functions in organisms are obtained through gene duplication, followed by subfunctionalization and neofunctionalization. Therefore, a possible biological explanation for the preferential attachment growth of metabolic networks, which we have now found some support for, could be that novel enzymes, which are created through gene duplication, maintain some compounds involved in the reaction catalyzed by the original enzyme throughout its future evolution. As a supplementary explanation we propose that horizontally transferred enzymes are introduced preferentially at central positions of the metabolic network of E. coli.
Methods
Databases and representation framework
We built a representation of the metabolic network of E. coli by using EcoCyc [17] (downloaded in March 2004) to gather the EC assigned enzymes and to determine the connectivities of the enzymes. An alternative network based on KEGG was also produced and the study was performed which generated similar results, results not shown. The connectivity of an enzyme is defined as the number of edges connecting the enzyme to other enzymes. Only one edge in each direction between any two enzymes was allowed. Furthermore we used KEGG orthology (KO) assignments [18] (downloaded in May 2004) to determine in which organisms the different EC numbers are represented.
The nodes in our graph represent the enzymes (complete EC numbers) catalyzing the reactions and the edges represent one or more compounds involved in the reactions. There is an edge from enzyme El to enzyme E2 if El catalyzes a reaction where compound A is produced and then E2 uses A as substrate. The network representation used in our study has been used before for metabolic network analysis where it has been referred to as 'protein-centric' graphs [33] or 'reaction graphs' [7], see Figure 1b. Our representation of the full metabolic network of E. coli consists of 486 nodes and 99 917 edges.
One problematic aspect with metabolic network analysis is how promiscuous compounds, such as H2O, should be handled. One may argue that the network would become more biochemically meaningful if these compounds are removed because the promiscuous compounds are usually not limiting factors of reactions [34]. In this study, we have chosen to apply a simple network-based criterion. We count the number of times a compound occurs as part of an edge in the network. The most common compounds were then considered as promiscuous compounds [2,3]. We performed our studies on different networks where up to 40 compounds have been removed.
Statistical analysis
For the statistical analysis 100 000 randomized networks were generated through shuffling the group numbers while preserving the network topology. Subsequently, Z-scores were calculated. The Z-score expresses how far the average connectivity of the enzymes belonging to a certain phylogenetic group differs from the average connectivity of randomly sampled enzymes, measured in units of the random sampling distribution's standard deviation. The larger the Z-score, the less likely that the difference between phylogenetic group's average and the random group's average is by chance.
For the calculations of the Z-score for the average connectivity for phylogenetic groups the Z-score is defined as , where i is the phylogenetic group and is the average connectivity.
For the calculations of the Z-score for the average connectivity for functional groups the Z-score is defined as , where f is the functional group and is the avera ge connectivity.
For the calculations of the Z-score for the number of members belonging to functional groups per domain of life the Z-score is defined as , where d is the number of domains of life and is the number of enzymes belonging to each functional class.
Authors' contributions
SL performed the analysis as a graduate student under the supervision of PK and AE.
Supplementary Material
Additional File 1
E. coli isozymes Flat file containing a simple list of the isozymes in E. coli.
Click here for file
Additional File 2
S. cerevisiae isozymes Flat file containing a simple list of the isozymes in S. cerevisiae.
Click here for file
Acknowledgements
This work was supported by the Foundation for Strategic Research (SSF).
==== Refs
Rison SC Teichmann SA Thornton JM Homology, pathway distance and chromosomal localisation of the small molecule metabolism enzymes in Escherichia coli J Mol Biol 2002 318 911 932 12054833 10.1016/S0022-2836(02)00140-7
Alves R Chaleil RA Sternberg MJ Evolution of enzymes in metabolism: a network perspective J Mol Biol 2002 320 751 770 12095253 10.1016/S0022-2836(02)00546-6
Light S Kraulis P Network analysis of metabolic enzyme evolution in Escherichia coli BMC Bioinformatics 2004 5 15113413
Jensen RA Enzyme recruitment in evolution of new function Annu Rev Microbiol 1976 30 409 425 791073 10.1146/annurev.mi.30.100176.002205
Horowitz NH On the evolution of biochemical syntheses Proc Natl Acad Sci USA 1945 31 153 157 16578152
Jeong H Tombor B Albert R Oltvai ZN Barabasi AL The large-scale organization of metabolic networks Nature 2000 407 651 654 11034217 10.1038/35036627
Wagner A Fell DA The small world inside large metabolic networks Proc R Soc Lond B Biol Sci 2001 268 1803 1810 10.1098/rspb.2001.1711
Wuchty S Scale-free behavior in protein domain networks Mol Biol Evol 2001 18 1694 1702 11504849
Arita M The metabolic world of Escherichia coli is not small Proc Natl Acad Sci U S A 2004 101 1543 1547 14757824 10.1073/pnas.0306458101
Albert R Jeong H Barabasi A-L Error and attack tolerance of complex networks Nature 2000 406 378 382 10935628 10.1038/35019019
Gleiss PM Stadler PF Wagner A Fell DA Relevant cycles in chemical reaction networks Adv Complex Syst 2001 4 207 226 10.1142/S0219525901000140
Barabasi A-L Albert R Emergence of scaling in metabolic networks Science 1999 286 509 512 10521342 10.1126/science.286.5439.509
Qian J Luscombe NM Gerstein M Protein family and fold occurrence in genomes: power-law behaviour and evolutionary model J Mol Biol 2001 313 673 681 11697896 10.1006/jmbi.2001.5079
Barabasi AL Oltvai ZN Network biology: Understanding the cell's functional organization Nat Rev Genet 2004 5 101 113 14735121 10.1038/nrg1272
Eisenberg E Levanon EY Preferential attachment in Protein Network Evolution Phys Rev Lett 2003 91 138701 14525344 10.1103/PhysRevLett.91.138701
Kunin V Pereira-Leal JB Ouzounis CA Functional Evolution of the Yeast Protein Interaction Network Mol Biol Evol 2004 21 1171 1176 15071090 10.1093/molbev/msh085
Karp PD Riley M Saier M Paulsen IT Collado-Vides J Paley SM Pellegrini-Toole A Bonavides C Gama-Castro S The EcoCyc Database Nucleic Acids Res 2002 30 56 58 11752253 10.1093/nar/30.1.56
Kanehisa M Goto S Kawashima S Nakaya A The KEGG databases at GenomeNet Nucleic Acids Res 2002 30 42 46 11752249 10.1093/nar/30.1.42
Boucher Y Douady CJ Papke RT Walsh DA Boudreau ME Nesbo CL Case RJ Doolittle WF Lateral gene transfer and the origins of prokaryotic groups Annu Rev Genet 2003 37 283 328 14616063 10.1146/annurev.genet.37.050503.084247
Kunin V Ouzounis CA The balance of driving forces during genome evolution in prokaryotes Genome Res 2003 13 1589 1594 12840037 10.1101/gr.1092603
Andersson JO Andersson SG Insights into the evolutionary process of genome degradation Curr Opin Genet Dev 1999 9 664 671 10607609 10.1016/S0959-437X(99)00024-6
Kurland CG Canback B Berg OG Horizontal gene transfer: a critical view Proc Natl Acad Sci U S A 2003 100 9658 9662 12902542 10.1073/pnas.1632870100
Snel B Bork P Huynen MA Genomes in flux: the evolution of archaeal and proteobacterial gene content Genome Res 2002 12 17 25 11779827 10.1101/gr.176501
Lawrence JG Ochman H Amelioration of bacterial genomes: rates of change and exchange J Mol Evol 1997 44 383 397 9089078
Lawrence JG Ochman H Molecular archaeology of the Escherichia coli genome Proc Natl Acad Sci U S A 1998 95 9413 9417 9689094 10.1073/pnas.95.16.9413
Jeong H Mason SP Barabasi AL Oltvai ZN Lethality and centrality in protein networks Nature 2001 411 41 42 11333967 10.1038/35075138
Gerdes SY Scholle MD Campbell JW Balazsi G Ravasz E Daugherty MD Somera AL Kyrpides NC Anderson T Gelfand MS Bhattacharya A Kapatral V D'Souza M Baev MV Grechkin Y Mseeh F Fonstein MY Overbeek R Barabasi A-L Oltvai ZN Osterman AL Experimental determination and system level analysis of essential genes in Escherichia coli MG1655 J Bacteriol 2003 185 5673 5684 13129938 10.1128/JB.185.19.5673-5684.2003
Gasteiger E Gattiker A Hoogland C Ivanyi I Appel RD Bairoch A ExPASy: the proteomics server for in-depth protein knowledge and analysis Nucleic Acids Res 2003 31 3784 3788 12824418 10.1093/nar/gkg563
Cherry JM Adler C Ball C Chervitz SA Dwight SS Hester ET Jia Y Juvik G Roe T Schroeder M Weng S Botstein D SGD: Saccharomyces Genome Database Nucleic Acids Res 1998 26 73 79 9399804 10.1093/nar/26.1.73
Papp B Pal C Hurst LD Metabolic network analysis of the causes and evolution of enzyme dispensability in yeast Nature 2004 429 661 664 15190353 10.1038/nature02636
Huynen MA Dandekar T Bork P Variation and evolution of the citric-acid cycle: a genomic perspective Trends Microbiol 1999 7 281 291 10390638 10.1016/S0966-842X(99)01539-5
Paulsen IT Sliwinski MK Saier MHJ Microbial genome analyses: global comparisons of transport capabilities based on phylogenies, bioenergetics and substrate specificities J Mol Biol 1998 277 573 592 9533881 10.1006/jmbi.1998.1609
Gerrard JA Sparrow AD Wells JA Metabolic databases – what next? Trends Biochem Sci 2001 26 137 140 11166573 10.1016/S0968-0004(00)01759-X
Ma H Zeng AP Reconstruction of metabolic networks from genome data and analysis of their global structure for various organisms Bioinformatics 2003 19 270 277 12538249 10.1093/bioinformatics/19.2.270
Gough J Convergent evolution of domain architectures (is rare) Bioinformatics 2005 21 1464 1471 15585523 10.1093/bioinformatics/bti204
|
16281983
|
PMC1316878
|
CC BY
|
2021-01-04 16:32:46
|
no
|
BMC Genomics. 2005 Nov 10; 6:159
|
utf-8
|
BMC Genomics
| 2,005 |
10.1186/1471-2164-6-159
|
oa_comm
|
==== Front
Respir ResRespiratory Research1465-99211465-993XBioMed Central London 1465-9921-6-1391630954710.1186/1465-9921-6-139ResearchInvasive versus noninvasive measurement of allergic and cholinergic airway responsiveness in mice Glaab Thomas [email protected] Michaela [email protected] Ralf [email protected] Regina [email protected] Armin [email protected] Jens M [email protected] Wayne [email protected] Norbert [email protected] Heinz G [email protected] Fraunhofer Institute of Toxicology and Experimental Medicine (ITEM), Nikolai-Fuchs Str.1, 30625 Hannover, Germany2 Hannover Medical School, Department of Respiratory Medicine, Carl-Neuberg Str.1, 30625 Hannover, Germany3 Division of Physiology, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, Maryland 21205, USA2005 25 11 2005 6 1 139 139 18 1 2005 25 11 2005 Copyright © 2005 Glaab et al; licensee BioMed Central Ltd.2005Glaab 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
This study seeks to compare the ability of repeatable invasive and noninvasive lung function methods to assess allergen-specific and cholinergic airway responsiveness (AR) in intact, spontaneously breathing BALB/c mice.
Methods
Using noninvasive head-out body plethysmography and the decrease in tidal midexpiratory flow (EF50), we determined early AR (EAR) to inhaled Aspergillus fumigatus antigens in conscious mice. These measurements were paralleled by invasive determination of pulmonary conductance (GL), dynamic compliance (Cdyn) and EF50 in another group of anesthetized, orotracheally intubated mice.
Results
With both methods, allergic mice, sensitized and boosted with A. fumigatus, elicited allergen-specific EAR to A. fumigatus (p < 0.05 versus controls). Dose-response studies to aerosolized methacholine (MCh) were performed in the same animals 48 h later, showing that allergic mice relative to controls were distinctly more responsive (p < 0.05) and revealed acute airway inflammation as evidenced from increased eosinophils and lymphocytes in bronchoalveolar lavage.
Conclusion
We conclude that invasive and noninvasive pulmonary function tests are capable of detecting both allergen-specific and cholinergic AR in intact, allergic mice. The invasive determination of GL and Cdyn is superior in sensitivity, whereas the noninvasive EF50 method is particularly appropriate for quick and repeatable screening of respiratory function in large numbers of conscious mice.
==== Body
Background
Asthma is a complex disease associated with reversible airway obstruction of variable degree, airway inflammation, airway hyperresponsiveness (AHR) and airway remodeling. These hallmarks of asthma are being examined in murine models, with the goal of understanding the basic cellular and genetic mechanisms of allergic inflammation that underlie the immunologic basis of the disease [1]. To investigate the functional consequences of in vitro findings in the lung in vivo, determination of pulmonary function is an essential tool. Existing methods for measuring respiratory function in mice in vivo include invasive and noninvasive approaches [2,3]. The invasive recording of pulmonary resistance (RL) or pulmonary conductance (1/RL), and dynamic compliance (Cdyn) is the gold standard for precise and specific determinations of pulmonary mechanics [2,3]. Limitations of traditional invasive methodologies commonly involve surgical tracheostomy, anesthesia, and mechanical ventilation, all of which are procedures that may generate significant artifacts [2]. In addition, when tracheostomy is done, this method is limited to single-point measurements only, usually precluding the possibility of performing follow-up studies. A novel modification to this invasive technology has enabled repetitive invasive recordings of pulmonary mechanics in conjunction with local aerosol delivery in anesthetized, orotracheally intubated, spontaneously breathing mice [4].
Noninvasive determination of respiratory parameters in conscious mice is a convenient, repeatable approach for screening respiratory function in large numbers of animals. Here, the application of the empiric variable enhanced pause (Penh) has gained widespread popularity. A recent correspondence written by leading experts [5] has emphasized the danger of the increasing uncritical use of Penh, with potentially misleading assessment of pulmonary function in animal models of lung disease. Although noninvasive measurement of murine respiratory function has virtually become synonymous with the recently questioned Penh method [5-9], a variety of other noninvasive methods have been established [10-12]. We and others have described the utility of midexpiratory flow, as measured by head-out body plethysmography, as a physiologically meaningful, noninvasive parameter of bronchoconstriction for mice and rats [13-17]. No report has as yet directly investigated the ability and utility of repetitive invasive and noninvasive lung function methods to assess allergen-specific EAR and cholinergic airway hyperresponsiveness (AHR) in intact mice. The primary objective of this study in a mouse model of fungal asthma was to compare the capability of noninvasive EF50 measurements to reflect the allergen-specific and cholinergic AR as observed with invasive determination of pulmonary mechanics. Moreover, to support the argument that noninvasive EF50 measurement is more valid than Penh we sought to examine whether EF50, unlike Penh [18], parallels the actual changes in pulmonary mechanics in response to hyperoxia in C57BL/6 mice. Our results showed that, while the noninvasive measurement of EF50 presented greater variability than the classical invasive measurements of RL and Cdyn, the correlation was sufficiently strong to support the use of such noninvasive testing in repetitive measurements in invividual mice.
Methods
Animals and sensitization protocol
Pathogen-free, female BALB/c mice, 12–14 weeks of age, and female C57BL/6 mice (used only for hyperoxia exposures), 7–8 weeks of age (Charles River, Sulzfeld, Germany), were kept in a pathogen-free rodent facility and were provided food and water ad libitum. All animal experiments conformed to NIH guidelines and were approved by the appropriate governmental authority (Bezirksregierung Niedersachsen, Germany). Allergic BALB/C mice (n = 8) received an intraperitoneal and subcutaneous injection of soluble A. fumigatus antigens (5 μg each, Greer Laboratories Inc, Lenoir, NC, USA), dissolved in incomplete Freund's adjuvant in a volume of 0.1 ml given on day 0 and were boosted noninvasively by inhalation over 10 min in a closed chamber with 1 % of A. fumigatus aerosol dissolved in saline on day 14 (jet nebulizer, LC Star, 2.8 μm mass median aerodynamic diameter (MMAD), Pari GmbH, Starnberg, Germany).
On day 21, allergic mice were challenged once with aerosolized A. fumigatus followed by methacholine (MCh, Sigma, Deisenhofen, Germany) dose-response exposure 48 h later (d 23). The control group (n = 8) received the same treatment schedule but was boosted and challenged with saline before MCh exposure. This protocol was chosen to maximize the difference between allergic and control groups. For the noninvasive measurement of pulmonary function separate groups of A. fumigatus-sensitized and control mice were used (n = 8 each group).
Noninvasive measurement of pulmonary function in conscious mice
Noninvasive respiratory function was assessed with a glass-made head-out body plethysmograph system for four mice as previously described [14,17,19]. Briefly, mice were placed in the body plethysmographs while the head of each animal protruded through a neck collar (9 mm ID, dental latex dam, Roeko, Langenau, Germany) into a ventilated head exposure chamber. Monitoring of respiratory function was started when animals and individual measurements settled down to a stable level. For airflow measurement, a calibrated pneumotachograph (capillary tube PTM 378/1.2, HSE-Harvard, March-Hugstetten, Germany) and a differential pressure transducer (Validyne DP 45-14, range ± 2 cm H2O, HSE-Harvard) coupled to an amplifier were attached to the top port of each plethysmograph. For each animal the amplified analog signal from the pressure transducer was digitized via an analog-to-digital converter (DT 302, Data Translation, Marlboro, MA). The pneumotachograph tidal flow signal was integrated with time to obtain tidal volume (VT). From these signals the parameters tidal midexpiratory flow (EF50), time of expiration (TE), tidal volume (VT) and respiratory rate (f) were calculated for each breath and were averaged in 5 s segments with a commercial software (HEM 3.4, Notocord, Paris, France).
During airway constriction the main changes in the tidal flow signal occur during the midexpiratory phase. We defined EF50 (ml/s) as the tidal flow at the midpoint (50 %) of expiratory tidal volume, and we used this as a measure of bronchoconstriction [12,14,17]. A reduction in EF50 of more than 1.5 Standard deviation (SD) of mean baseline value (which translates to a reduction of more than 20% versus baseline) is considered to indicate airway constriction. The degree of bronchoconstriction to inhalation challenge was determined from minimum values of EF50 and was expressed as percent changes from corresponding baseline values.
Invasive measurement of pulmonary function
AR was assessed as an increase in RL or decreases in Cdyn and EF50 in response to aerosolized A. fumigatus or MCh in anesthetized, spontaneously breathing mice as previously described in detail [4]. Briefly, mice were anesthetized with intraperitoneal injections of metomidate (total dose: 38–60 mg/kg) and fentanyl (total dose: 0.02 – 0.06 mg/kg) with minimal supplementations as required. When an appropriate depth of anesthesia was achieved, mice were suspended by their upper incisors from a rubber band on a Plexiglas support. The trachea was transilluminated below the vocal cords by a halogen light source and a standard 20G × 32 mm Abbocath®-T cannula (Abbott, Sligo, Ireland) was gently inserted into the tracheal opening. The intubated, spontaneously breathing animal was then placed in supine position in a thermostat-controlled whole-body plethysmograph (type 871, HSE-Harvard, designed in cooperation with Fraunhofer ITEM). The orotracheal tube was directly attached to a pneumotachograph (capillary tube PTM T16375, HSE-Harvard) installed in the front part of the chamber. Tidal flow was determined by the pneumotachograph connected to a differential pressure transducer (Validyne DP 45-14, HSE-Harvard). To measure transpulmonary pressure (PTP) a water-filled PE-90 tubing was inserted into the esophagus to the level of the midthorax and coupled to a pressure transducer (model P75, HSE-Harvard). The amplified analog signals from the pressure transducers were digitized as described above for noninvasive measurements.
Pulmonary resistance (RL) and dynamic compliance (Cdyn) were calculated over a complete respiratory cycle using an integration method over flows, volumes and pressures as previously described [4,20]. The resistance of the orotracheal tube (0.63 cm H2O·s·ml-1) was subtracted from all RL measurements. RL, Cdyn, EF50 together with other basic respiratory parameters were continuously recorded with a commercial software (HEM 3.4, Notocord). For easier comparison of trends among all variables, RL was expressed as pulmonary conductance GL (GL = 1/RL).
Respiratory parameters were averaged in 5 s segments and minimum GL, Cdyn and EF50 values were taken and expressed as percent changes from corresponding baseline values. After the measurements on day 21, mice were removed from the chamber and extubated as soon as they began recovering from anesthesia.
Administration of aerosols
After recording of baseline values, airway responsiveness (AR) to A. fumigatus 2 % or saline (control group) was determined in separate groups of conscious and intubated mice on day 21. On day 23, dose-response studies to aerosolized MCh were performed in the same mice.
For intubated mice, dried aerosols of A. fumigatus 2 % (inhaled dose: 8 μg) and MCh 5 % (inhaled doses: 0.05–2.5 μg) were generated by a computer-controlled, jet-driven aerosol generator system (Bronchy III, particle size 2.5 μm MMAD, Fraunhofer ITEM, licensed by Buxco, Troy, NY) as previously described (15, 21).
Conscious mice placed in the head-out body plethysmographs were exposed noninvasively to A. fumigatus (2 %, inhaled dose 32 μg) and MCh aerosols (0.5–3 %, cumulative inhaled doses: 3–14 μg) delivered by a Pari jet nebulizer as previously described [13,14,22]. In both systems, aerosol concentrations were determined by a gravimetrically calibrated photometer. The total inhalation doses of A. fumigatus and MCh were calculated based on the continuously measured aerosol concentrations and respiratory volume per min [4,21]. The results of the bronchoconstrictor response to MCh were expressed as PD50 which is the dose of MCh required to reduce either GL, Cdyn or EF50 to 50 % of their respective baseline values and was calculated from the dose-response curves.
Exposure to oxygen
C57BL/6 mice were randomly assigned to two groups: The mice in the control group (n = 8 each) were kept in room air whereas the other group of 8 mice was exposed to 100 % oxygen for 48 h. Exposure to 100 % oxygen was performed in a sealed (25 L) Plexiglas chamber with a flow of 2 L/min as similarly described earlier [18]. The CO2 level in the chamber was maintained at 1 % by using a CO2 absorber (Drägersorb 800 plus, Dräger, Lübeck, Germany). Food and water were provided ad libitum.
Bronchoalveolar lavage (BAL) cell counts
At the end of this protocol, total and differential cell counts from BAL samples using 2 × 0.8 ml aliquots of saline were determined as previously described (14), except that, recovery of BAL fluids was performed from the distal trachea in intubated animals.
Statistics
Comparisons of baseline values between groups and intraindividual comparisons were analyzed by the Student's two-sided t-test, allergic responses of the group of allergic mice versus control mice were analyzed by one-sided t-test. P values < 0.05 were considered significant. Descriptive results were expressed as means ± SE unless indicated otherwise. Comparison of a new measurement technique with an established one is needed to see whether they agree sufficiently. A plot of the difference against the standard measurements will often appear to show a relation between difference and magnitude when there is none. A plot of the difference against the average of the standard and new measurements is unlikely to mislead in this way. Accordingly, the agreement between the invasive and noninvasive lung function methods was analyzed by the method of Bland and Altman [23]. Graphically, the difference of each pair of measurement was plotted against their mean values. Agreement was expressed as the mean differences over all measurements and their corresponding 95% confidence intervals (95% CI). The limits of agreement were expressed as the mean differences ± 2 SD of the differences, together with their 95% confidence intervals (95% CI). Statistics was performed with SPSS 11.5.
Results
Baseline values for respiratory parameters in conscious and anesthetized mice
To illustrate the impact of anesthesia on respiratory function, baseline respiratory parameters were measured in anesthetized and conscious mice. Table 1 presents the baseline values of respiratory parameters obtained from conscious and anesthetized BALB/c mice. There were significant differences in f, TE and EF50 values between anesthetized and conscious animals at baseline. In addition, no differences in respiratory parameters were observed between allergic and control mice at baseline when separated into conscious and anesthetized groups.
Table 1 Baseline values for respiratory parameters from allergic and control BALB/c mice
Respiratory parameters Definition Control mice conscious Allergic mice conscious Control mice anesthetized Allergic mice anesthetized
VT, ml tidal volume 0.21 ± 0.05 0.19 ± 0.04 0.14 ± 0.02 0.13 ± 0.02
f, breaths/min respiratory frequency 198 ± 41 220 ± 23 129 ± 20* 124 ± 29*
TE, s time of expiration 0.17 ± 0.06 0.14 ± 0.02 0.3 ± 0.04* 0.3 ± 0.05*
EF50, ml/s tidal midexpiratory flow 2.05 ± 0.89 2.26 ± 0.46 0.93 ± 0.14* 1.12 ± 0.43*
GL, ml·s-1·cmH2O-1 pulmonary conductance - - 1.05 ± 0.36 1.29 ± 0.69
Cdyn, ml·cmH2O-1 dynamic compliance - - 0.037 ± .007 0.030 ± .008
Baseline values are means ± SD obtained from 8 animals per group during a 5 min control period from conscious and anesthetized, orotracheally intubated BALB/c mice. In comparison with conscious mice, EF50, TE and f values were significantly altered in anesthetized mice. No difference was found between allergic animals and control groups when separated into conscious and anesthetized mice. *P < 0.05 versus conscious mice.
Comparison of invasive and noninvasive lung function measurements of EAR
The allergen-specific early airway response (EAR) to A. fumigatus was investigated in allergic mice on day 21 (Fig. 1 and 2). To avoid unbalanced challenges with allergen or saline, each group was separated into two subgroups for invasive and noninvasive measurement of pulmonary function.
Figure 1 Early airway responsiveness. Invasive vs. noninvasive assessment of early airway responsiveness (EAR) to aerosolized Aspergillus fumigatus 2 %. Allergic (black columns) and control mice (white columns) were separated into groups of invasively and noninvasively monitored animals. The allergic mice showed significant reductions in simultaneously measured GL, Cdyn and EF50an (an: anesthetized), compared with control animals. Noninvasive determination of EF50con (con: conscious) elicited significant decreases in EF50 to inhaled A. fumigatus compared with control animals. EAR was expressed as % change from corresponding baseline values, which were taken as 0 %. Values are means ± SE, n = 8 per group, *p < 0.01 vs. control.
Figure 2 Example of EAR. Example of an early airway response (EAR) to inhaled A. fumigatus 2 % in an orotracheally intubated allergic mouse. Decreases in GL, Cdyn, and EF50 values were associated with small declines in VT, f and TE. The ordinate at the bottom indicates the photometric signal of the allergen aerosol challenge.
Invasive recordings of EAR in allergic mice showed significant decreases in simultaneously measured GL, Cdyn, and EF50 compared with controls thus indicating an allergen-specific EAR to A. fumigatus. As shown in Figure 1, the most prominent alteration was shown for GL with a reduction by -62.1 ± 5.1 % (P < 0.001 vs. control) compared with a reduction by -48.8 ± 8.3 % in Cdyn (P < 0.001 vs. control), and a decrease by -34.5 ± 5.1 % in EF50 (P < 0.001 vs. control). The bronchoconstrictive response started within 7 ± 4 minutes (mean ± SD) after start of exposure and reached its maximum within 14 ± 3 min (mean ± SD). Figure 2 illustrates a characteristic time-response course of the EAR in an anesthetized, orotracheally intubated allergic mouse.
To determine if decreases in invasively monitored EF50, relate to changes in GL and Cdyn, we analyzed the agreement between these measurements by the method of Bland and Altman. Although all three parameters, Cdyn, GL and EF50, adequately reflected the pronounced EAR in allergic mice there was enhanced variation between GL vs. EF50, GL vs. Cdyn and EF50 vs. Cdyn in response to specific allergen challenge. As shown in Table 2, EF50 tended to underestimate the decreases in GL by -27.6 %, and by -14.3 % for Cdyn in allergic animals. In contrast, a very good agreement between EF50, GL and Cdyn values was found for control mice, with mean differences ranging from -2.4 to -6.1 %.
Table 2 Bland-Altman analysis of the differences in GL, EF50 and Cdyn.
Early AR Cholinergic AR
Group Parameters Mean ± SD
(95% CI) Upper limit (95% CI)
Lower limit (95% CI) Mean ± SD
(95% CI) Upper limit (95% CI)
Lower limit (95% CI)
Allergic EF50 vs. GL -27.6 ± 17.8
(-42.6/-12.7) 8.0 (-17.8/33.9)
-63.3 (-89.2/-37.5) -0.7 ± 0.7
(-1.3/0.1) 0.7 (-0.3/1.8)
-2.1 (-3.2/-1.1)
GL vs. Cdyn 13.3 ± 21.9
(-5/31.7) 57.1 (25.4/88.8)
-30.5 (-62.2/1.2) 0 ± 0.2
(-0.2/0.2) 0.4 (0.1/0.7)
-0.4 (-0.7/-0.1)
EF50 vs. Cdyn -14.3 ± 29.5
(-39/10.3) 44.7 (2/87.4)
-73.3 (-116/-30.6) -0.7 ± 0.9
(-1.4/0) 1 (-0.2/2.3)
-2.5 (-3.7/-1.2)
Control EF50 vs. GL -2.4 ± 9.5
(-10.4/5.5) 16.6 (2.8/30.5)
-21.5 (-35.3/-7.7) -2.9 ± 3.3
(-5.7/-0.2) 3.7 (-1.1/8.5)
-9.5 (-14.3/-4.7)
GL vs. Cdyn -3.7 ± 10.4
(-12.2/5.1) 17.2 (2.1 to 32.3)
-24.6 (-39.7/-9.4) 1.4 ± 1.8
(-0.2/2.9) 5 (2.4/7.7)
-2.3 (-4.9/0.4)
EF50 vs. Cdyn -6.1 ± 9.1
(-13.8/-1.5) 12.2 (-1.1/25.4)
-24.4 (-37.6/-11.2) -1.5 ± 3.5
(-4.5/1.4) 5.5 (0.4/10.7)
-8.6 (-13.8/-3.5)
Differences in simultaneous invasive measurements of GL, EF50 and Cdyn for allergic and control mice during EAR and cholinergic AR. Values are means ± SD (95 % confidence intervals (CI) in brackets) for 8 animals per group. The upper and lower limits of agreement (means ± 2 SD) as well as the corresponding 95 % CI intervals (in brackets) are shown. Values for the EAR represent the % change from baseline, whereas the values for cholinergic AR show the absolute PD50 values in μg MCh.
Noninvasive measurements of pulmonary function in allergic mice also demonstrated a marked allergen-specific EAR as manifested by a significant decline by -44.6 ± 6.2 % in EF50 compared with that in control animals (P = 0.002, Fig. 1). The magnitude of the response was similar to the decline observed with invasively recorded EF50. Reduced EF50values were accompanied by decreased VT values and – in contrast to invasive measurements – by decreased f and increased TE values.
Invasive vs. noninvasive determination of cholinergic AHR
To further characterize the utility of noninvasive vs. noninvasive pulmonary function tests, AR to increasing doses of aerosolized MCh, was investigated 48 h after EAR recordings in the same animals. Baseline GL, Cdyn and EF50 values were not significantly different from initial baseline values.
MCh exposure elicited a dose-related reduction in GL, Cdyn, and EF50 values in the intubated animals that was significantly enhanced in allergic mice (p < 0.05 vs. control group). The magnitude of cholinergic AR was significantly higher for GL and Cdyn compared with simultaneously measured EF50 (P = 0.027). Accordingly, the mean PD50 causing a decrease in Cdyn, EF50 and GL to 50 % baseline was 0.4 ± 0.1 for GL, 0.4 ± 0.1 for Cdyn, and 1.2 ± 0.4 μg MCh for EF50 in allergic mice (Fig. 3). The respective mean PD50 values for control animals were significantly higher: 2 ± 0.4 for GL (P = 0.001), 3.4 ± 0.7 for Cdyn (P = 0.002), and 4.9 ± 1.2 μg MCh for EF50 (P = 0.008). The dose-related decreases in EF50 were accompanied by increases in esophageal pressures. At the level of the 50% decline in EF50 (PD50), the peak esophageal pressure increased 121 ± 13 % for the allergic mice and 104 ± 16 % for the control group.
Figure 3 Cholinergic AR. Magnitudes of cholinergic AHR, 48 h after EAR, expressed as PD50 values, which is the dose of MCh required to reduce either GL, Cdyn or EF50 to 50 % of their respective baseline values) of invasively measured GL, Cdyn and EF50 (A) as well as of noninvasively recorded EF50 (B). Allergic mice (black columns) showed significantly lower PD50 values compared with controls (white columns). Baseline values were not significantly different from initial baseline values 48 h before and were within the means ± SD as listed in Table 1. Values are means ± SE, n = 8 per group, *p < 0.05 vs. control.
The peak responses for GL, Cdyn and EF50 occurred within 1 min after challenge and recovered to within 10–20 % of the baseline before MCh exposure during 1–3 min. Agreements between Cdyn, EF50 and GL were excellent, the mean ranging from 0 to -0.71 μg MCh for the allergic group and from -2.9 to 1.38 μg MCh for the control group (Table 2). Figure 4 shows the corresponding Bland-Altman plots of the differences between EF50 vs. GL and between EF50 vs. Cdyn against the mean of both values in allergic animals.
Figure 4 Bland-Altman plots. Individual differences in the degree of MCh-induced bronchoconstriction between invasively measured EF50 and GL and between EF50 and Cdyn, are plotted against the average corresponding values (expressed as PD50, μg MCh). The solid line represents the mean of the differences, the dashed lines show the upper and lower limits of agreement.
Noninvasive determination of EF50 also showed that allergic mice were significantly more responsive to MCh, as indicated by significantly lower PD50 values for EF50 when compared with controls (P = 0.032) (Fig. 3).
Allergic airway inflammation
The A. fumigatus-sensitized and boosted animals showed significant increases in eosinophils and lymphocytes in BAL fluid (Table 3) compared with control mice. This indicates the presence of an inflammatory response in the lungs of allergic mice. The intubated animals receiving aerosols directly via the orotracheal tube had slightly higher numbers of leukocyte populations compared with conscious mice (statistically not significant).
Table 3 Cellular composition of BAL fluid
Control mice conscious Allergic mice conscious Control mice anesthetized Allergic mice anesthetized
Eosinophils, × 104 < 1 7.9 ± 5.6* < 1 13.4 ± 9.3*
Lymphocytes, × 104 < 1 3.2 ± 2.2* 0.5 ± 0.4 1.8 ± 1.6*
Neutrophils, × 104 < 1 1.3 ± 1 1.9 ± 1.3 2.7 ± 4.1
Macrophages, × 104 12.3 ± 3.3 13.7 ± 3.1 22 ± 9.2 16.7 ± 6.1
Values are means ± SD from 8 animals per group. Eosinophils and lymphocytes recovered from bronchoalveolar lavage (BAL) fluid 48 hours after allergen challenge were increased in both conscious and intubated allergic mice. *P < 0.05 vs. control mice.
Impact of hyperoxia on EF50 measurements in C57BL/6 mice
To examine how EF50 correlates with direct lung resistance measurements, C57BL/6 mice were exposed to 100% oxygen for 48 h. Table 4 lists the hyperoxia-induced changes detected by invasive and noninvasive lung function measurements compared with control animals. Noninvasive recordings revealed no significant differences in breathing rate, TE, VT, and EF50 between control and hyperoxia mice after 48 h of hyperoxia. Likewise, direct measurements of pulmonary mechanics in the same animals did not show any differences in EF50, Cdyn and RL values, thus confirming the absence of airway constriction in both groups.
Table 4 Impact of hyperoxia over 48 h on invasively and noninvasively measured respiratory parameters
Noninvasive measurement Invasive measurement
EF50 TE VT f RL Cdyn GL EF50 TE VT* f
Control 2.36 ± 0.12 0.13 ± 0.01 0.20 ± 0.01 251 ± 14 1.44 ± 0.27 0.017 ± 0.004 0.72 ± 0.15 1.01 ± 0.13 0.3 ± 0.03 0.11 ± 0.02 106 ± 9
Hyperoxia 2.30 ± 0.41 0.14 ± 0.02 0.20 ± 0.02 245 ± 41 1.27 ± 0.29 0.018 ± 0.007 0.85 ± 0.18 0.93 ± 0.15 0.32 ± 0.04 0.14 ± 0.02 99 ± 15
Values are means ± SD from 8 C57BL/6 mice per group. *P < 0.05 vs. control mice. VT: tidal volume, EF50: tidal midexpiratory flow, TE: time of expiration, f: respiratory rate, RL: pulmonary resistance, Cdyn: dynamic compliance, GL: pulmonary conductance (GL = 1/RL).
Discussion
In the present study we have evaluated the sensitivity and reliability of repeatable noninvasive versus invasive pulmonary function tests to sequentially measure AR in response to specific allergen and cholinergic challenge in spontaneously breathing mice. Our results demonstrate that both systems reflect the allergen-specific early AR and cholinergic AHR of allergic compared with control mice.
The ability to manipulate the mouse genome has opened up new opportunities to develop mouse models of allergic asthma that demonstrate spontaneous or chronic disease [24]. For a proper phenotyping of AR in experimental models it is crucial to monitor pulmonary function as reliably as possible. One way to achieve this is a novel in-vivo method that combines repetitive recordings of classical pulmonary mechanics with cholinergic aerosol challenges in orotracheally intubated mice [4]. Despite being an accurate measurement of classical pulmonary function on multiple occasions, this invasive method does not readily allow for rapid screening of pulmonary function in large numbers of animals.
In contrast, noninvasive head-out body plethysmography has been shown to yield stable and reliable on-line measurements of AR in several conscious mice at a time and serves as a suitable and valid tool to complement the traditional measures of pulmonary mechanics [13,14,16,22,25]. Limitations of previous EF50 validation studies in mice particularly have included pleural catheterization with the inability to conduct reproducible measurements, the contribution of upper airway resistance and intravenous rather than aerosol challenge [14,17]. These methodological shortcomings introduced variability into the results which made them difficult to compare with other invasive techniques [10].
The current report intended to overcome such problems in that GL, Cdyn and EF50 were measured simultaneously in intact mice including local aerosol challenges via an orotracheal tube. In parallel, noninvasive determinations of EF50 were performed in allergic and control mice. The noninvasive experiments relied on methodologies identical to those used in our previous mice studies to facilitate comparisons [14,17,22].
The values for respiratory parameters measured from both conscious and anesthetized BALB/c mice were reproducible and comparable with those reported previously for this strain (Table 1) [4,14,26]. The changes in respiratory patterns observed in anesthetized mice were associated with increased expiratory time, decreased f, and decreased EF50 values, events likely related to anesthetic effects on neural respiratory control. The independence of EF50 recordings from changes in frequency has been demonstrated in previous investigations [14,15].
To examine the sensitivity of noninvasive and invasive indices of bronchoconstriction, we monitored allergen-specific EAR and, 48 h later, performed MCh dose-response studies in the same allergic animals compared with controls. Challenge with aerosolized A. fumigatus resulted in significant reductions in Cdyn, GL and in EF50 values in allergic mice compared with (sham-exposed) control animals. Demonstration of allergen-specific EAR in allergic mice was followed by cholinergic AHR that was linked with a pronounced influx of neutrophils and eosinophils in BAL fluid. Consistent with previous results, invasively recorded EF50 was slightly less sensitive in detecting the maximum degree of bronchoconstriction to A. fumigatus and MCh compared with GL and Cdyn recordings [15].
Agreement between invasively measured EF50, GL and Cdyn during EAR and cholinergic AHR was good, although there was increased variability at the time of EAR in allergic mice (Table 2). This variability may reflect different sensitivities of GL, EF50 and Cdyn to the airway and tissue components of total pulmonary resistance [3,16]. Related to this issue, is a previous study indicating that mice with airway inflammation experience quite heterogeneous airway narrowing and airway closure during airway smooth muscle contraction [27].
Nevertheless, despite this variability, it is important to emphasize that the noninvasive measurement of EF50 still reflected the enhanced AR to A. fumigatus and MCh in allergic relative to control mice (Figs. 1, 3). Thus, although the calculated inhalation doses for A. fumigatus and MCh in conscious mice may be not as accurate as in intubated mice, the observed EF50 responses still reflect airway constriction. These findings indicate that EF50 can distinguish between different magnitudes of AR and reflects the changes with GL and Cdyn during bronchoconstriction at least under the conditions of this study. Moreover, the relation of the cholinergic EF50 response between allergic and control animals was similar for invasive and noninvasive measurements (Figure 3). The higher PD50 values for EF50 in conscious compared with intubated animals to MCh challenge can be explained by methodological issues. Administration of aerosols directly into the lungs via an orotracheal tube results in aerosol deposition mainly in the parenchyma. In conscious animals there will be substantial deposition in the nasal passages and upper airway, which should lead to the higher PD50 values observed. The AR, as measured noninvasively by EF50, may also be partly affected by altered upper airway resistance. However, because of the rapid onset and resolution of the response, it seems unlikely that edema or mucus hypersecretion in these upper airways was responsible for the increased AR.
In agreement with other investigations, decreases in EF50, as measured by noninvasive head-out body plethysmography, were linked with decreased frequency and VT values and increasing values for TE [12,14,15]. In contrast, no relevant impact on frequency and TE was found in anesthetized, intubated mice during bronchoconstriction.
Concerns with noninvasive EF50 recordings include the uncertainty about the exact degree and localization of bronchoconstriction as well as the potential contribution of upper airway resistance. Due to methodological differences, comparisons between invasive and noninvasive measures are of indirect, qualitative nature. A quantitative comparison, however, is directly available from the intraindividual differences between simultaneously measured EF50 and GL in unconscious mice. Because EF50 tends to underestimate the magnitude of bronchoconstriction (discussed below) it is still unclear whether this limits its use in detecting less marked changes in airway hyperresponsiveness than those induced in high-reponder models. As a result, EF50 measures should be confirmed with direct assessments of pulmonary resistance under these circumstances. Despite these methodological restrictions, the observed EF50 responses still reflected the enhanced AR to ACh and allergen under the conditions of this study.
In comparison with the widely used Penh method, EF50 differs substantially in several important ways: EF50 decreases with bronchoconstriction and in line with invasively measured lung resistance or conductance is linked with a decline in VT during bronchoconstriction [7,28]. Even more importantly, EF50 has a physical meaning (ml/s), allows direct comparison from one animal to another and is closely related to airway resistance. Indeed, if it were possible to know the esophageal pressure in the conscious animals, one could calculate a precise lung resistance. If we assume that esophageal pressure does not change, then changes in the EF50 would be directly proportional to the lung resistance. However, in the anesthetized animals, we found that the esophageal pressure actually increased as the airways constricted, perhaps in response to the increased resistance and lower air flow. This suggests that the EF50 in conscious animals may underestimate the actual changes in lung resistance. Despite this quantitative limitation, the method seems far more representative of changes in resistance than other noninvasive methods, and the approach allows for direct quantitative comparisons from animal to animal. The commonly measured Penh has no theoretical linkage to lung resistance, and its usefulness was further weakened by recent reports, one of which showed that changes in Penh were no better than simply measuring TE to assess AR in common strains of laboratory mice [6]. It is also known that a decline in noninvasively measured EF50 is associated with an increase in TE [12,14]. However, it is important to note that conditions entirely unrelated to bronchoconstriction, such as sensory irritation, will also result in increasing TE values [12,29].
Another report demonstrated that Penh was inadequate for characterization of pulmonary mechanics in the context of hyperoxia-induced changes in C57BL/6 mice [18]. These authors pointed out that Penh may significantly overestimate the actual changes in lung resistance after 24 and 48 h of hyperoxia. Interestingly, increases in Penh were accompanied by decreased TE and rising VT and f. This contrasts with the above-mentioned observation of decreased VT during bronchoconstriction as observed with EF50 and invasive pulmonary function methods [4,28]. Our study in C57BL/6 mice showed a consistent relationship between EF50 and lung resistance measurements in reponse to 48 h hyperoxia, thus indicating non-constricted airways. These data support the concept that EF50 more reliably reflects airway resistance than Penh, which is largely a function of respiratory timing.
Conclusion
In conclusion, this study investigated the utility of repetitive invasive vs. noninvasive techniques to determine AR to allergen and cholinergic challenge in intact, spontaneously breathing mice. We demonstrated allergen-specific EAR to A. fumigatus followed by cholinergic AHR in allergic mice compared with controls. Our results show that the noninvasive EF50 method is directly related to lung resistance, and is thus particularly appropriate for quick and repeatable phenotyping of airway function in large numbers of conscious mice.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
TG participated in the design and coordination of the study and drafted the manuscript. MZ and RB carried out the lung function experiments. RK participated in the data analysis of all experiments, AB carried out the cytological and ELISA tests. WM helped to draft the manuscript. JMH and NK participated in the coordination and analysis of the study. HGH conceived of the study, and participated in its design and analysis. All authors read and approved the final manuscript
Acknowledgements
We greatly thank Prof. H. Hecker, Biometrics of Hannover Medical School, for statistical support and Dr. C. Nassenstein, Fraunhofer ITEM, for excellent technical support.
==== Refs
Epstein MM Do mouse models of allergic asthma mimic clinical disease? Int Arch Allergy Immunol 2004 133 84 100 14726635 10.1159/000076131
Drazen JM Finn PW De Sanctis GT Mouse models of airway responsiveness: physiological basis of observed outcomes and analysis of selected examples using these outcome indicators Annu Rev Physiol 1999 61 593 625 10099703 10.1146/annurev.physiol.61.1.593
Irvin CG Bates JH Measuring the lung function in the mouse: the challenge of size Respir Res 2003 4 4 12783622 10.1186/rr199
Glaab T Mitzner W Braun A Ernst H Korolewitz R Hohlfeld JM Krug N Hoymann HG Repetitive measurements of pulmonary mechanics to inhaled cholinergic challenge in spontaneously breathing mice J Appl Physiol 2004 97 1104 1111 15121749 10.1152/japplphysiol.01182.2003
Bates J Irvin C Brusasco V Drazen J Fredberg J Loring S Eidelman D Ludwig M Macklem P Martin J Milic-Emili J Hantos Z Hyatt R Lai-Fook S Leff A Solway J Lutchen K Suki B Mitzner W Paré P Pride N Sly P The use and misuse of Penh in animal models of lung disease Am J Respir Cell Mol Biol 2004 31 373 374 15317683
Adler A Cieslewicz G Irvin CG Unrestrained plethysmography is an unreliable measure of airway responsiveness in BALB/c and C57BL/6 mice J Appl Physiol 2004 97 286 292 15033965 10.1152/japplphysiol.00821.2003
Hamelmann E Schwarze J Takeda K Oshiba A Larsen GL Irvin CG Gelfand EW Noninvasive measurement of airway responsiveness in allergic mice using barometric plethysmography Am J Respir Crit Care Med 1997 156 766 775 9309991
Lundblad LK Irvin CG Adler A Bates JH A reevaluation of the validity of unrestrained plethysmography in mice J Appl Physiol 2002 93 1198 1207 12235015
Mitzner W Tankersley C Interpreting Penh in mice J Appl Physiol 2003 94 828 831 12531918
Flandre TD Leroy PL Desmecht DJ Effect of somatic growth, strain, and sex on double-chamber plethysmographic respiratory function values in healthy mice J Appl Physiol 2003 94 1129 1136 12571140
Hessel EM Zwart A Oostveen E Van Oosterhout AJ Blyth DI Nijkamp FP Repeated measurement of respiratory function and bronchoconstriction in unanesthetized mice J Appl Physiol 1995 79 1711 1716 8594033
Vijayaraghavan R Schaper M Thompson R Stock MF Alarie Y Characteristic modifications of the breathing pattern of mice to evaluate the effects of airborne chemicals on the respiratory tract Arch Toxicol 1993 68 478 499 8239997 10.1007/s002040050101
Finotto S De Sanctis GT Lehr HA Herz U Buerke M Schipp M Bartsch B Atreya R Schmitt E Galle PR Renz H Neurath MF Treatment of allergic airway inflammation and hyperresponsiveness by antisense-induced local blockade of GATA-3 expression J Exp Med 2001 193 1247 1260 11390432 10.1084/jem.193.11.1247
Glaab T Daser A Braun A Neuhaus-Steinmetz U Fabel H Alarie Y Renz H Tidal midexpiratory flow as a measure of airway hyperresponsiveness in allergic mice Am J Physiol Lung Cell Mol Physiol 2001 280 L565 L573 11159041
Glaab T Hoymann HG Hohlfeld JM Korolewitz R Hecht M Alarie Y Tschernig T Braun A Krug N Fabel H Noninvasive measurement of midexpiratory flow indicates bronchoconstriction in allergic rats J Appl Physiol 2002 93 1208 1214 12235016
Hantos Z Brusasco V Assessment of respiratory mechanics in small animals: the simpler the better? J Appl Physiol 2002 93 1196 1197 12235014
Neuhaus-Steinmetz U Glaab T Daser A Braun A Lommatzsch M Herz U Kips J Alarie Y Renz H Sequential development of airway hyperresponsiveness and acute airway obstruction in a mouse model of allergic inflammation Int Arch Allergy Immunol 2000 121 57 67 10686510 10.1159/000024298
Peták F Habre W Donati YR Hantos Z Barazzone-Argiroffo C Hyperoxia-induced changes in mouse lung mechanics: forced oscillations vs. barometric plethysmography J Appl Physiol 2001 90 2221 2230 11356786 10.1063/1.1389334
Vijayaraghavan R Schaper M Thompson R Stock MF Boylstein LA Luo JE Alarie Y Computer assisted recognition and quantitation of the effects of airborne chemicals acting at different areas of the respiratory tract in mice Arch Toxicol 1994 68 490 499 7802589 10.1007/s002040050101
Roy R Powers SR JrKimball WR Estimation of respiratory parameters by the method of covariance ratios Comput Biomed Res 1974 7 21 39 4817283 10.1016/0010-4809(74)90040-8
Hoymann HG Heinrich U Uhlig S, Taylor AE Measurement of lung function in rodents in vivo Methods in Pulmonary Research 1998 Basel, Birkhäuser Verlag 1 28
Path G Braun A Meents N Kerzel S Quarcoo D Raap U Hoyle GW Nockher WA Renz H Augmentation of allergic early-phase reaction by nerve growth factor Am J Respir Crit Care Med 2002 166 818 826 12231491 10.1164/rccm.200202-134OC
Bland JM Altman DG Statistical methods for assessing agreement between two methods of clinical measurement Lancet 1986 1 307 310 2868172
Kumar RK Foster PS Modeling allergic asthma in mice: pitfalls and opportunities Am J Respir Cell Mol Biol 2002 27 267 272 12204888
Braun A Lommatzsch M Neuhaus-Steinmetz U Quarcoo D Glaab T McGregor GP Fischer A Renz H Brain-derived neurotrophic factor (BDNF) contributes to neuronal dysfunction in a model of allergic airway inflammation Br J Pharmacol 2004 141 431 440 14718253 10.1038/sj.bjp.0705638
Tomioka S Bates JH Irvin CG Airway and tissue mechanics in a murine model of asthma: alveolar capsule vs. forced oscillations J Appl Physiol 2002 93 263 270 12070213
Evans KL Bond RA Corry DB Shardonofsky FR Frequency dependence of respiratory system mechanics during induced constriction in a murine model of asthma J Appl Physiol 2003 94 245 252 12486022 10.1063/1.1579128
Lai YL Chou H Respiratory mechanics and maximal expiratory flow in the anesthetized mouse J Appl Physiol 2000 88 939 943 10710389
Alarie Y Computer-based bioassay for evaluation of sensory irritation of airborne chemicals and its limit of detection Arch Toxicol 1998 72 277 282 9630013 10.1007/s002040050502
|
16309547
|
PMC1316879
|
CC BY
|
2021-01-04 16:36:27
|
no
|
Respir Res. 2005 Nov 25; 6(1):139
|
utf-8
|
Respir Res
| 2,005 |
10.1186/1465-9921-6-139
|
oa_comm
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.